new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
package queue
import (
"bytes"
"context"
"errors"
"fmt"
"io"
"net"
"os"
"strings"
"time"
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promauto"
"github.com/mjl-/adns"
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
"github.com/mjl-/bstore"
"github.com/mjl-/mox/dns"
"github.com/mjl-/mox/dsn"
"github.com/mjl-/mox/mlog"
"github.com/mjl-/mox/mox-"
"github.com/mjl-/mox/mtasts"
"github.com/mjl-/mox/mtastsdb"
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
"github.com/mjl-/mox/smtp"
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
"github.com/mjl-/mox/smtpclient"
"github.com/mjl-/mox/store"
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
"github.com/mjl-/mox/tlsrpt"
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
)
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
var (
metricDestinations = promauto . NewCounter (
prometheus . CounterOpts {
Name : "mox_queue_destinations_total" ,
Help : "Total destination (e.g. MX) lookups for delivery attempts, including those in mox_smtpclient_destinations_authentic_total." ,
} ,
)
metricDestinationsAuthentic = promauto . NewCounter (
prometheus . CounterOpts {
Name : "mox_queue_destinations_authentic_total" ,
Help : "Destination (e.g. MX) lookups for delivery attempts authenticated with DNSSEC so they are candidates for DANE verification." ,
} ,
)
metricDestinationDANERequired = promauto . NewCounter (
prometheus . CounterOpts {
Name : "mox_queue_destination_dane_required_total" ,
Help : "Total number of connections to hosts with valid TLSA records making DANE required." ,
} ,
)
metricDestinationDANESTARTTLSUnverified = promauto . NewCounter (
prometheus . CounterOpts {
Name : "mox_queue_destination_dane_starttlsunverified_total" ,
Help : "Total number of connections with required DANE where all TLSA records were unusable." ,
} ,
)
metricDestinationDANEGatherTLSAErrors = promauto . NewCounter (
prometheus . CounterOpts {
Name : "mox_queue_destination_dane_gathertlsa_errors_total" ,
Help : "Total number of connections where looking up TLSA records resulted in an error." ,
} ,
)
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// todo: recognize when "tls-required-no" message header caused a non-verifying certificate to be overridden. requires doing our own certificate validation after having set tls.Config.InsecureSkipVerify due to tls-required-no.
metricTLSRequiredNoIgnored = promauto . NewCounterVec (
prometheus . CounterOpts {
Name : "mox_queue_tlsrequiredno_ignored_total" ,
Help : "Delivery attempts with TLS policy findings ignored due to message with TLS-Required: No header. Does not cover case where TLS certificate cannot be PKIX-verified." ,
} ,
[ ] string {
"ignored" , // mtastspolicy (error getting policy), mtastsmx (mx host not allowed in policy), badtls (error negotiating tls), badtlsa (error fetching dane tlsa records)
} ,
)
metricRequireTLSUnsupported = promauto . NewCounterVec (
prometheus . CounterOpts {
Name : "mox_queue_requiretls_unsupported_total" ,
Help : "Delivery attempts that failed due to message with REQUIRETLS." ,
} ,
[ ] string {
"reason" , // nopolicy (no mta-sts and no dane), norequiretls (smtp server does not support requiretls)
} ,
)
2023-10-24 14:09:48 +03:00
metricPlaintextFallback = promauto . NewCounter (
prometheus . CounterOpts {
Name : "mox_queue_plaintext_fallback_total" ,
Help : "Delivery attempts with fallback to plain text delivery." ,
} ,
)
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
)
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
// todo: rename function, perhaps put some of the params in a delivery struct so we don't pass all the params all the time?
func fail ( qlog * mlog . Log , m Msg , backoff time . Duration , permanent bool , remoteMTA dsn . NameIP , secodeOpt , errmsg string ) {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// todo future: when we implement relaying, we should be able to send DSNs to non-local users. and possibly specify a null mailfrom. ../rfc/5321:1503
// todo future: when we implement relaying, and a dsn cannot be delivered, and requiretls was active, we cannot drop the message. instead deliver to local postmaster? though ../rfc/8689:383 may intend to say the dsn should be delivered without requiretls?
// todo future: when we implement smtp dsn extension, parameter RET=FULL must be disregarded for messages with REQUIRETLS. ../rfc/8689:379
2023-11-01 19:55:40 +03:00
if permanent || m . MaxAttempts == 0 && m . Attempts >= 8 || m . MaxAttempts > 0 && m . Attempts >= m . MaxAttempts {
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
qlog . Errorx ( "permanent failure delivering from queue" , errors . New ( errmsg ) )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
deliverDSNFailure ( qlog , m , remoteMTA , secodeOpt , errmsg )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
if err := queueDelete ( context . Background ( ) , m . ID ) ; err != nil {
qlog . Errorx ( "deleting message from queue after permanent failure" , err )
}
return
}
qup := bstore . QueryDB [ Msg ] ( context . Background ( ) , DB )
qup . FilterID ( m . ID )
if _ , err := qup . UpdateNonzero ( Msg { LastError : errmsg , DialedIPs : m . DialedIPs } ) ; err != nil {
qlog . Errorx ( "storing delivery error" , err , mlog . Field ( "deliveryerror" , errmsg ) )
}
if m . Attempts == 5 {
// We've attempted deliveries at these intervals: 0, 7.5m, 15m, 30m, 1h, 2u.
// Let sender know delivery is delayed.
qlog . Errorx ( "temporary failure delivering from queue, sending delayed dsn" , errors . New ( errmsg ) , mlog . Field ( "backoff" , backoff ) )
retryUntil := m . LastAttempt . Add ( ( 4 + 8 + 16 ) * time . Hour )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
deliverDSNDelay ( qlog , m , remoteMTA , secodeOpt , errmsg , retryUntil )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
} else {
qlog . Errorx ( "temporary failure delivering from queue" , errors . New ( errmsg ) , mlog . Field ( "backoff" , backoff ) , mlog . Field ( "nextattempt" , m . NextAttempt ) )
}
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Delivery by directly dialing (MX) hosts for destination domain of message.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
//
// The returned results are for use in a TLSRPT report, it holds success/failure
// counts and failure details for delivery/connection attempts. The
// recipientDomainResult is for policies/counts/failures about the whole recipient
// domain (MTA-STS), its policy type can be empty, in which case there is no
// information (e.g. internal failure). hostResults are per-host details (DANE, one
// per MX target).
func deliverDirect ( cid int64 , qlog * mlog . Log , resolver dns . Resolver , dialer smtpclient . Dialer , ourHostname dns . Domain , transportName string , m Msg , backoff time . Duration ) ( recipientDomainResult tlsrpt . Result , hostResults [ ] tlsrpt . Result ) {
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// High-level approach:
// - Resolve domain to deliver to (CNAME), and determine hosts to try to deliver to (MX)
// - Get MTA-STS policy for domain (optional). If present, only deliver to its
// allowlisted hosts and verify TLS against CA pool.
// - For each host, attempt delivery. If the attempt results in a permanent failure
// (as claimed by remote with a 5xx SMTP response, or perhaps decided by us), the
// attempt can be aborted. Other errors are often temporary and may result in later
// successful delivery. But hopefully the delivery just succeeds. For each host:
// - If there is an MTA-STS policy, we only connect to allow-listed hosts.
// - We try to lookup DANE records (optional) and verify them if present.
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// - If RequireTLS is true, we only deliver if the remote SMTP server implements it.
// - If RequireTLS is false, we'll fall back to regular delivery attempts without
// TLS verification and possibly without TLS at all, ignoring recipient domain/host
// MTA-STS and DANE policies.
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Resolve domain and hosts to attempt delivery to.
// These next-hop names are often the name under which we find MX records. The
// expanded name is different from the original if the original was a CNAME,
// possibly a chain. If there are no MX records, it can be an IP or the host
// directly.
origNextHop := m . RecipientDomain . Domain
ctx := context . WithValue ( mox . Context , mlog . CidKey , cid )
haveMX , origNextHopAuthentic , expandedNextHopAuthentic , expandedNextHop , hosts , permanent , err := smtpclient . GatherDestinations ( ctx , qlog , resolver , m . RecipientDomain )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
if err != nil {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// If this is a DNSSEC authentication error, we'll collect it for TLS reporting.
// Hopefully it's a temporary misconfiguration that is solve before we try to send
// our report. We don't report as "dnssec-invalid", because that is defined as
// being for DANE. ../rfc/8460:580
var errCode adns . ErrorCode
if errors . As ( err , & errCode ) && errCode . IsAuthentication ( ) {
// Result: ../rfc/8460:567
reasonCode := fmt . Sprintf ( "dns-extended-error-%d-%s" , errCode , strings . ReplaceAll ( errCode . String ( ) , " " , "-" ) )
fd := tlsrpt . Details ( tlsrpt . ResultValidationFailure , reasonCode )
recipientDomainResult = tlsrpt . MakeResult ( tlsrpt . NoPolicyFound , origNextHop , fd )
recipientDomainResult . Summary . TotalFailureSessionCount ++
}
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
fail ( qlog , m , backoff , permanent , dsn . NameIP { } , "" , err . Error ( ) )
return
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
tlsRequiredNo := m . RequireTLS != nil && ! * m . RequireTLS
2023-10-14 23:30:43 +03:00
// Check for MTA-STS policy and enforce it if needed.
// We must check at the original next-hop, i.e. recipient domain, not following any
// CNAMEs. If we were to follow CNAMEs and ask for MTA-STS at that domain, it
// would only take a single CNAME DNS response to direct us to an unrelated domain.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
var policy * mtasts . Policy // Policy can have mode enforce, testing and none.
2023-10-14 23:30:43 +03:00
if ! origNextHop . IsZero ( ) {
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
cidctx := context . WithValue ( mox . Shutdown , mlog . CidKey , cid )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
policy , recipientDomainResult , _ , err = mtastsdb . Get ( cidctx , resolver , origNextHop )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
if err != nil {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
if tlsRequiredNo {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
qlog . Infox ( "mtasts lookup temporary error, continuing due to tls-required-no message header" , err , mlog . Field ( "domain" , origNextHop ) )
metricTLSRequiredNoIgnored . WithLabelValues ( "mtastspolicy" ) . Inc ( )
} else {
qlog . Infox ( "mtasts lookup temporary error, aborting delivery attempt" , err , mlog . Field ( "domain" , origNextHop ) )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
recipientDomainResult . Summary . TotalFailureSessionCount ++
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
fail ( qlog , m , backoff , false , dsn . NameIP { } , "" , err . Error ( ) )
return
}
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// note: policy can be nil, if a domain does not implement MTA-STS or it's the
// first time we fetch the policy and if we encountered an error.
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// We try delivery to each host until we have success or a permanent failure. So
// for transient errors, we'll try the next host. For MX records pointing to a
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
// dual stack host, we turn a permanent failure due to policy on the first delivery
// attempt into a temporary failure and make sure to try the other address family
// the next attempt. This should reduce issues due to one of our IPs being on a
// block list. We won't try multiple IPs of the same address family. Surprisingly,
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// RFC 5321 does not specify a clear algorithm, but common practice is probably
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
// ../rfc/3974:268.
var remoteMTA dsn . NameIP
var secodeOpt , errmsg string
permanent = false
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
nmissingRequireTLS := 0
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
// todo: should make distinction between host permanently not accepting the message, and the message not being deliverable permanently. e.g. a mx host may have a size limit, or not accept 8bitmime, while another host in the list does accept the message. same for smtputf8, ../rfc/6531:555
for _ , h := range hosts {
// ../rfc/8461:913
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
if policy != nil && policy . Mode != mtasts . ModeNone && ! policy . Matches ( h . Domain ) {
// todo: perhaps only send tlsrpt failure if none of the mx hosts matched? reporting about each mismatch seems useful for domain owners, to discover mtasts policies they didn't update after changing mx. there is a risk a domain owner intentionally didn't put all mx'es in the mtasts policy, but they probably won't mind being reported about that.
// Other error: Surprising that TLSRPT doesn't have an MTA-STS specific error code
// for this case, it's a big part of the reason to have MTA-STS. ../rfc/8460:610
// Result: ../rfc/8460:567 todo spec: propose adding a result for this case?
fd := tlsrpt . Details ( tlsrpt . ResultValidationFailure , "mtasts-policy-mx-mismatch" )
fd . ReceivingMXHostname = h . Domain . ASCII
recipientDomainResult . Add ( 0 , 0 , fd )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
var policyHosts [ ] string
for _ , mx := range policy . MX {
policyHosts = append ( policyHosts , mx . LogString ( ) )
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
if policy . Mode == mtasts . ModeEnforce {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
if tlsRequiredNo {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
qlog . Info ( "mx host does not match mta-sts policy in mode enforce, ignoring due to tls-required-no message header" , mlog . Field ( "host" , h . Domain ) , mlog . Field ( "policyhosts" , policyHosts ) )
metricTLSRequiredNoIgnored . WithLabelValues ( "mtastsmx" ) . Inc ( )
} else {
errmsg = fmt . Sprintf ( "mx host %s does not match enforced mta-sts policy with hosts %s" , h . Domain , strings . Join ( policyHosts , "," ) )
qlog . Error ( "mx host does not match mta-sts policy in mode enforce, skipping" , mlog . Field ( "host" , h . Domain ) , mlog . Field ( "policyhosts" , policyHosts ) )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
recipientDomainResult . Summary . TotalFailureSessionCount ++
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
continue
}
} else {
qlog . Error ( "mx host does not match mta-sts policy, but it is not enforced, continuing" , mlog . Field ( "host" , h . Domain ) , mlog . Field ( "policyhosts" , policyHosts ) )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
qlog . Info ( "delivering to remote" , mlog . Field ( "remote" , h ) , mlog . Field ( "queuecid" , cid ) )
cid := mox . Cid ( )
nqlog := qlog . WithCid ( cid )
var remoteIP net . IP
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
enforceMTASTS := policy != nil && policy . Mode == mtasts . ModeEnforce
2023-10-14 23:30:43 +03:00
tlsMode := smtpclient . TLSOpportunistic
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
tlsPKIX := false
if enforceMTASTS {
tlsMode = smtpclient . TLSRequiredStartTLS
tlsPKIX = true
// note: smtpclient will still go through PKIX verification, and report about it, but not fail the connection if not passing.
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Try to deliver to host. We can get various errors back. Like permanent failure
// response codes, TCP, DNSSEC, TLS (opportunistic, i.e. optional with fallback to
// without), etc. It's a balancing act to handle these situations correctly. We
// don't want to bounce unnecessarily. But also not keep trying if there is no
// chance of success.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
//
// deliverHost will report generic TLS and MTA-STS-specific failures in
// recipientDomainResult. If DANE is encountered, it will add a DANE reporting
// result for generic TLS and DANE-specific errors.
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// Set if TLSA records were found. Means TLS is required for this host, usually
// with verification of the certificate, and that we cannot fall back to
// opportunistic TLS.
var tlsDANE bool
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
2023-10-24 14:09:48 +03:00
var badTLS , ok bool
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
var hostResult tlsrpt . Result
permanent , tlsDANE , badTLS , secodeOpt , remoteIP , errmsg , hostResult , ok = deliverHost ( nqlog , resolver , dialer , cid , ourHostname , transportName , h , enforceMTASTS , haveMX , origNextHopAuthentic , origNextHop , expandedNextHopAuthentic , expandedNextHop , & m , tlsMode , tlsPKIX , & recipientDomainResult )
var zerotype tlsrpt . PolicyType
if hostResult . Policy . Type != zerotype {
hostResults = append ( hostResults , hostResult )
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
2023-11-01 19:55:40 +03:00
// If we had a TLS-related failure when doing TLS, and we don't have a requirement
// for MTA-STS/DANE, we try again without TLS. This could be an old server that
// only does ancient TLS versions, or has a misconfiguration. Note that
2023-10-24 14:09:48 +03:00
// opportunistic TLS does not do regular certificate verification, so that can't be
// the problem.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// ../rfc/7435:459
2023-11-01 19:55:40 +03:00
// We don't fall back to plain text for DMARC reports. ../rfc/7489:1768 ../rfc/7489:2683
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// We queue outgoing TLS reports with tlsRequiredNo, so reports can be delivered in
// case of broken TLS.
if ! ok && badTLS && ( ! enforceMTASTS && tlsMode == smtpclient . TLSOpportunistic && ! tlsDANE && ! m . IsDMARCReport || tlsRequiredNo ) {
2023-10-24 14:09:48 +03:00
metricPlaintextFallback . Inc ( )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
if tlsRequiredNo {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
metricTLSRequiredNoIgnored . WithLabelValues ( "badtls" ) . Inc ( )
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// todo future: add a configuration option to not fall back?
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
nqlog . Info ( "connecting again for delivery attempt without tls" , mlog . Field ( "enforcemtasts" , enforceMTASTS ) , mlog . Field ( "tlsdane" , tlsDANE ) , mlog . Field ( "requiretls" , m . RequireTLS ) )
permanent , _ , _ , secodeOpt , remoteIP , errmsg , _ , ok = deliverHost ( nqlog , resolver , dialer , cid , ourHostname , transportName , h , enforceMTASTS , haveMX , origNextHopAuthentic , origNextHop , expandedNextHopAuthentic , expandedNextHop , & m , smtpclient . TLSSkip , false , & tlsrpt . Result { } )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
if ok {
nqlog . Info ( "delivered from queue" )
if err := queueDelete ( context . Background ( ) , m . ID ) ; err != nil {
nqlog . Errorx ( "deleting message from queue after delivery" , err )
}
return
}
remoteMTA = dsn . NameIP { Name : h . XString ( false ) , IP : remoteIP }
if permanent {
break
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
if secodeOpt == smtp . SePol7MissingReqTLS {
nmissingRequireTLS ++
}
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// In theory, we could make a failure permanent if we didn't find any mx host
// matching the mta-sts policy AND the policy is fresh AND all DNS records leading
// to the MX targets (including CNAME) have a TTL that is beyond the latest
// possible delivery attempt. Until that time, configuration problems can be
// corrected through DNS or policy update. Not sure if worth it in practice, there
// is a good chance the MX records can still change, at least on initial delivery
// failures.
// todo: possibly detect that future deliveries will fail due to long ttl's of cached records that are preventing delivery.
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// If we failed due to requiretls not being satisfied, make the delivery permanent.
// It is unlikely the recipient domain will implement requiretls during our retry
// period. Best to let the sender know immediately.
if ! permanent && nmissingRequireTLS > 0 && nmissingRequireTLS == len ( hosts ) {
qlog . Info ( "marking delivery as permanently failed because recipient domain does not implement requiretls" )
permanent = true
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
fail ( qlog , m , backoff , permanent , remoteMTA , secodeOpt , errmsg )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// deliverHost attempts to deliver m to host. Depending on tlsMode we'll do
// opportunistic or required STARTTLS or skip TLS entirely. Based on tlsPKIX we do
// PKIX/WebPKI verification (for MTA-STS). If we encounter DANE records, we verify
// those. If the message has a message header "TLS-Required: No", we ignore TLS
// verification errors.
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
//
// deliverHost updates m.DialedIPs, which must be saved in case of failure to
// deliver.
//
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// The haveMX and next-hop-authentic fields are used to determine if DANE is
// applicable. The next-hop fields themselves are used to determine valid names
// during DANE TLS certificate verification.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
//
// The returned hostResult holds TLSRPT reporting results for the connection
// attempt. Its policy type can be the zero value, indicating there was no finding
// (e.g. internal error).
func deliverHost ( log * mlog . Log , resolver dns . Resolver , dialer smtpclient . Dialer , cid int64 , ourHostname dns . Domain , transportName string , host dns . IPDomain , enforceMTASTS , haveMX , origNextHopAuthentic bool , origNextHop dns . Domain , expandedNextHopAuthentic bool , expandedNextHop dns . Domain , m * Msg , tlsMode smtpclient . TLSMode , tlsPKIX bool , recipientDomainResult * tlsrpt . Result ) ( permanent , tlsDANE , badTLS bool , secodeOpt string , remoteIP net . IP , errmsg string , hostResult tlsrpt . Result , ok bool ) {
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
// About attempting delivery to multiple addresses of a host: ../rfc/5321:3898
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
tlsRequiredNo := m . RequireTLS != nil && ! * m . RequireTLS
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
start := time . Now ( )
var deliveryResult string
defer func ( ) {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
mode := string ( tlsMode )
if tlsPKIX {
mode += "+mtasts"
}
if tlsDANE {
mode += "+dane"
}
metricDelivery . WithLabelValues ( fmt . Sprintf ( "%d" , m . Attempts ) , transportName , mode , deliveryResult ) . Observe ( float64 ( time . Since ( start ) ) / float64 ( time . Second ) )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
log . Debug ( "queue deliverhost result" ,
mlog . Field ( "host" , host ) ,
mlog . Field ( "attempt" , m . Attempts ) ,
mlog . Field ( "tlsmode" , tlsMode ) ,
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
mlog . Field ( "tlspkix" , tlsPKIX ) ,
mlog . Field ( "tlsdane" , tlsDANE ) ,
mlog . Field ( "tlsrequiredno" , tlsRequiredNo ) ,
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
mlog . Field ( "permanent" , permanent ) ,
mlog . Field ( "badtls" , badTLS ) ,
mlog . Field ( "secodeopt" , secodeOpt ) ,
mlog . Field ( "errmsg" , errmsg ) ,
mlog . Field ( "ok" , ok ) ,
mlog . Field ( "duration" , time . Since ( start ) ) )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
} ( )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Open message to deliver.
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
f , err := os . Open ( m . MessagePath ( ) )
if err != nil {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return false , false , false , "" , nil , fmt . Sprintf ( "open message file: %s" , err ) , hostResult , false
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
msgr := store . FileMsgReader ( m . MsgPrefix , f )
defer func ( ) {
err := msgr . Close ( )
log . Check ( err , "closing message after delivery attempt" )
} ( )
cidctx := context . WithValue ( mox . Context , mlog . CidKey , cid )
ctx , cancel := context . WithTimeout ( cidctx , 30 * time . Second )
defer cancel ( )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// We must lookup the IPs for the host name before checking DANE TLSA records. And
// only check TLSA records for secure responses. This prevents problems with old
// name servers returning an error for TLSA requests or letting it timeout (not
// sending a response). ../rfc/7672:879
var daneRecords [ ] adns . TLSA
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
var tlsHostnames [ ] dns . Domain
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
if host . IsDomain ( ) {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
tlsHostnames = [ ] dns . Domain { host . Domain }
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
if m . DialedIPs == nil {
m . DialedIPs = map [ string ] [ ] net . IP { }
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
countResultFailure := func ( ) {
recipientDomainResult . Summary . TotalFailureSessionCount ++
hostResult . Summary . TotalFailureSessionCount ++
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
metricDestinations . Inc ( )
authentic , expandedAuthentic , expandedHost , ips , dualstack , err := smtpclient . GatherIPs ( ctx , log , resolver , host , m . DialedIPs )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
destAuthentic := err == nil && authentic && origNextHopAuthentic && ( ! haveMX || expandedNextHopAuthentic ) && host . IsDomain ( )
if ! destAuthentic {
log . Debugx ( "not attempting verification with dane" , err , mlog . Field ( "authentic" , authentic ) , mlog . Field ( "expandedauthentic" , expandedAuthentic ) )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// Track a DNSSEC error if found.
var errCode adns . ErrorCode
if err != nil {
if errors . As ( err , & errCode ) && errCode . IsAuthentication ( ) {
// Result: ../rfc/8460:567
reasonCode := fmt . Sprintf ( "dns-extended-error-%d-%s" , errCode , strings . ReplaceAll ( errCode . String ( ) , " " , "-" ) )
fd := tlsrpt . Details ( tlsrpt . ResultValidationFailure , reasonCode )
hostResult = tlsrpt . MakeResult ( tlsrpt . TLSA , host . Domain , fd )
countResultFailure ( )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
} else {
// todo: we could lookup tlsa records, and log an error when they are not dnssec-signed. this should be interpreted simply as "not doing dane", but it could be useful to warn domain owners about, they may be under the impression they are dane-protected.
hostResult = tlsrpt . MakeResult ( tlsrpt . NoPolicyFound , host . Domain )
}
} else if tlsMode == smtpclient . TLSSkip {
metricDestinationsAuthentic . Inc ( )
// TLSSkip is used to fallback to plaintext, which is used with a TLS-Required: No
// header to ignore the recipient domain's DANE policy.
// possible err is propagated to below.
} else {
metricDestinationsAuthentic . Inc ( )
// Look for TLSA records in either the expandedHost, or otherwise the original
// host. ../rfc/7672:912
var tlsaBaseDomain dns . Domain
tlsDANE , daneRecords , tlsaBaseDomain , err = smtpclient . GatherTLSA ( ctx , log , resolver , host . Domain , expandedNextHopAuthentic && expandedAuthentic , expandedHost )
if tlsDANE {
metricDestinationDANERequired . Inc ( )
}
if err != nil {
metricDestinationDANEGatherTLSAErrors . Inc ( )
}
if err == nil && tlsDANE {
tlsMode = smtpclient . TLSRequiredStartTLS
hostResult = tlsrpt . Result { Policy : tlsrpt . TLSAPolicy ( daneRecords , tlsaBaseDomain ) }
if len ( daneRecords ) == 0 {
// If there are no usable DANE records, we still have to use TLS, but without
// verifying its certificate. At least when there is no MTA-STS. Why? Perhaps to
// prevent ossification? The SMTP TLSA specification has different behaviour than
// the generic TLSA. "Usable" means different things in different places.
// ../rfc/7672:718 ../rfc/6698:1845 ../rfc/6698:660
log . Debug ( "no usable dane records, requiring starttls but not verifying with dane" )
metricDestinationDANESTARTTLSUnverified . Inc ( )
daneRecords = nil
// Result: ../rfc/8460:576 (this isn't technicall invalid, only all-unusable...)
hostResult . FailureDetails = [ ] tlsrpt . FailureDetails {
{
ResultType : tlsrpt . ResultTLSAInvalid ,
ReceivingMXHostname : host . XString ( false ) ,
FailureReasonCode : "all-unusable-records+ignored" ,
} ,
}
} else {
log . Debug ( "delivery with required starttls with dane verification" , mlog . Field ( "allowedtlshostnames" , tlsHostnames ) )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// Based on CNAMEs followed and DNSSEC-secure status, we must allow up to 4 host
// names.
tlsHostnames = smtpclient . GatherTLSANames ( haveMX , expandedNextHopAuthentic , expandedAuthentic , origNextHop , expandedNextHop , host . Domain , tlsaBaseDomain )
} else if ! tlsDANE {
log . Debugx ( "not doing opportunistic dane after gathering tlsa records" , err )
err = nil
hostResult = tlsrpt . MakeResult ( tlsrpt . NoPolicyFound , tlsaBaseDomain )
} else if err != nil {
fd := tlsrpt . Details ( tlsrpt . ResultTLSAInvalid , "" )
var errCode adns . ErrorCode
if errors . As ( err , & errCode ) {
fd . FailureReasonCode = fmt . Sprintf ( "extended-dns-error-%d-%s" , errCode , strings . ReplaceAll ( errCode . String ( ) , " " , "-" ) )
if errCode . IsAuthentication ( ) {
// Result: ../rfc/8460:580
fd . ResultType = tlsrpt . ResultDNSSECInvalid
countResultFailure ( )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
}
hostResult = tlsrpt . Result {
Policy : tlsrpt . TLSAPolicy ( daneRecords , tlsaBaseDomain ) ,
FailureDetails : [ ] tlsrpt . FailureDetails { fd } ,
}
if tlsRequiredNo {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
log . Debugx ( "error gathering dane tlsa records with dane required, but continuing without validation due to tls-required-no message header" , err )
err = nil
metricTLSRequiredNoIgnored . WithLabelValues ( "badtlsa" ) . Inc ( )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
// else, err is propagated below.
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// todo: for requiretls, should an MTA-STS policy in mode testing be treated as good enough for requiretls? let's be strict and assume not.
// todo: ../rfc/8689:276 seems to specify stricter requirements on name in certificate than DANE (which allows original recipient domain name and cname-expanded name, and hints at following CNAME for MX targets as well, allowing both their original and expanded names too). perhaps the intent was just to say the name must be validated according to the relevant specifications?
// todo: for requiretls, should we allow no usable dane records with requiretls? dane allows it, but doesn't seem in spirit of requiretls, so not allowing it.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
if err == nil && m . RequireTLS != nil && * m . RequireTLS && ! ( tlsDANE && len ( daneRecords ) > 0 ) && ! enforceMTASTS {
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
log . Info ( "verified tls is required, but destination has no usable dane records and no mta-sts policy, canceling delivery attempt to host" )
metricRequireTLSUnsupported . WithLabelValues ( "nopolicy" ) . Inc ( )
// Resond with proper enhanced status code. ../rfc/8689:301
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return false , tlsDANE , false , smtp . SePol7MissingReqTLS , remoteIP , "missing required tls verification mechanism" , hostResult , false
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
}
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Dial the remote host given the IPs if no error yet.
var conn net . Conn
if err == nil {
if m . DialedIPs == nil {
m . DialedIPs = map [ string ] [ ] net . IP { }
}
conn , remoteIP , err = smtpclient . Dial ( ctx , log , dialer , host , ips , 25 , m . DialedIPs )
}
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
cancel ( )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Set error for metrics.
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
var result string
switch {
case err == nil :
result = "ok"
case errors . Is ( err , os . ErrDeadlineExceeded ) , errors . Is ( err , context . DeadlineExceeded ) :
result = "timeout"
case errors . Is ( err , context . Canceled ) :
result = "canceled"
default :
result = "error"
}
metricConnection . WithLabelValues ( result ) . Inc ( )
if err != nil {
log . Debugx ( "connecting to remote smtp" , err , mlog . Field ( "host" , host ) )
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return false , tlsDANE , false , "" , remoteIP , fmt . Sprintf ( "dialing smtp server: %v" , err ) , hostResult , false
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
var mailFrom string
if m . SenderLocalpart != "" || ! m . SenderDomain . IsZero ( ) {
mailFrom = m . Sender ( ) . XString ( m . SMTPUTF8 )
}
rcptTo := m . Recipient ( ) . XString ( m . SMTPUTF8 )
// todo future: get closer to timeouts specified in rfc? ../rfc/5321:3610
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
log = log . Fields ( mlog . Field ( "remoteip" , remoteIP ) )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
ctx , cancel = context . WithTimeout ( cidctx , 30 * time . Minute )
defer cancel ( )
mox . Connections . Register ( conn , "smtpclient" , "queue" )
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// Initialize SMTP session, sending EHLO/HELO and STARTTLS with specified tls mode.
var firstHost dns . Domain
var moreHosts [ ] dns . Domain
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
if len ( tlsHostnames ) > 0 {
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// For use with DANE-TA.
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
firstHost = tlsHostnames [ 0 ]
moreHosts = tlsHostnames [ 1 : ]
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
}
var verifiedRecord adns . TLSA
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
opts := smtpclient . Opts {
IgnoreTLSVerifyErrors : tlsRequiredNo ,
RootCAs : mox . Conf . Static . TLS . CertPool ,
DANERecords : daneRecords ,
DANEMoreHostnames : moreHosts ,
DANEVerifiedRecord : & verifiedRecord ,
RecipientDomainResult : recipientDomainResult ,
HostResult : & hostResult ,
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
sc , err := smtpclient . New ( ctx , log , conn , tlsMode , tlsPKIX , ourHostname , firstHost , opts )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
defer func ( ) {
if sc == nil {
conn . Close ( )
} else {
sc . Close ( )
}
mox . Connections . Unregister ( conn )
} ( )
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
if err == nil && m . SenderAccount != "" {
// Remember the STARTTLS and REQUIRETLS support for this recipient domain.
// It is used in the webmail client, to show the recipient domain security mechanisms.
// We always save only the last connection we actually encountered. There may be
// multiple MX hosts, perhaps only some support STARTTLS and REQUIRETLS. We may not
// be accurate for the whole domain, but we're only storing a hint.
rdt := store . RecipientDomainTLS {
Domain : m . RecipientDomain . Domain . Name ( ) ,
STARTTLS : sc . TLSEnabled ( ) ,
RequireTLS : sc . SupportsRequireTLS ( ) ,
}
if err = updateRecipientDomainTLS ( ctx , m . SenderAccount , rdt ) ; err != nil {
err = fmt . Errorf ( "storing recipient domain tls status: %w" , err )
}
}
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
if err == nil {
implement dnssec-awareness throughout code, and dane for incoming/outgoing mail delivery
the vendored dns resolver code is a copy of the go stdlib dns resolver, with
awareness of the "authentic data" (i.e. dnssec secure) added, as well as support
for enhanced dns errors, and looking up tlsa records (for dane). ideally it
would be upstreamed, but the chances seem slim.
dnssec-awareness is added to all packages, e.g. spf, dkim, dmarc, iprev. their
dnssec status is added to the Received message headers for incoming email.
but the main reason to add dnssec was for implementing dane. with dane, the
verification of tls certificates can be done through certificates/public keys
published in dns (in the tlsa records). this only makes sense (is trustworthy)
if those dns records can be verified to be authentic.
mox now applies dane to delivering messages over smtp. mox already implemented
mta-sts for webpki/pkix-verification of certificates against the (large) pool
of CA's, and still enforces those policies when present. but it now also checks
for dane records, and will verify those if present. if dane and mta-sts are
both absent, the regular opportunistic tls with starttls is still done. and the
fallback to plaintext is also still done.
mox also makes it easy to setup dane for incoming deliveries, so other servers
can deliver with dane tls certificate verification. the quickstart now
generates private keys that are used when requesting certificates with acme.
the private keys are pre-generated because they must be static and known during
setup, because their public keys must be published in tlsa records in dns.
autocert would generate private keys on its own, so had to be forked to add the
option to provide the private key when requesting a new certificate. hopefully
upstream will accept the change and we can drop the fork.
with this change, using the quickstart to setup a new mox instance, the checks
at internet.nl result in a 100% score, provided the domain is dnssec-signed and
the network doesn't have any issues.
2023-10-10 13:09:35 +03:00
// SMTP session is ready. Finally try to actually deliver.
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
has8bit := m . Has8bit
smtputf8 := m . SMTPUTF8
var msg io . Reader = msgr
size := m . Size
if m . DSNUTF8 != nil && sc . Supports8BITMIME ( ) && sc . SupportsSMTPUTF8 ( ) {
has8bit = true
smtputf8 = true
size = int64 ( len ( m . DSNUTF8 ) )
msg = bytes . NewReader ( m . DSNUTF8 )
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
err = sc . Deliver ( ctx , mailFrom , rcptTo , size , msg , has8bit , smtputf8 , m . RequireTLS != nil && * m . RequireTLS )
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
if err != nil {
log . Infox ( "delivery failed" , err )
}
var cerr smtpclient . Error
switch {
case err == nil :
deliveryResult = "ok"
case errors . Is ( err , os . ErrDeadlineExceeded ) , errors . Is ( err , context . DeadlineExceeded ) :
deliveryResult = "timeout"
case errors . Is ( err , context . Canceled ) :
deliveryResult = "canceled"
case errors . As ( err , & cerr ) :
deliveryResult = "temperror"
if cerr . Permanent {
deliveryResult = "permerror"
}
default :
deliveryResult = "error"
}
if err == nil {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return false , tlsDANE , false , "" , remoteIP , "" , hostResult , true
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
} else if cerr , ok := err . ( smtpclient . Error ) ; ok {
// If we are being rejected due to policy reasons on the first
// attempt and remote has both IPv4 and IPv6, we'll give it
// another try. Our first IP may be in a block list, the address for
// the other family perhaps is not.
permanent := cerr . Permanent
if permanent && m . Attempts == 1 && dualstack && strings . HasPrefix ( cerr . Secode , "7." ) {
permanent = false
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// If server does not implement requiretls, respond with that code. ../rfc/8689:301
secode := cerr . Secode
if errors . Is ( cerr . Err , smtpclient . ErrRequireTLSUnsupported ) {
secode = smtp . SePol7MissingReqTLS
metricRequireTLSUnsupported . WithLabelValues ( "norequiretls" ) . Inc ( )
}
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return permanent , tlsDANE , errors . Is ( cerr , smtpclient . ErrTLS ) , secode , remoteIP , cerr . Error ( ) , hostResult , false
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
} else {
implement outgoing tls reports
we were already accepting, processing and displaying incoming tls reports. now
we start tracking TLS connection and security-policy-related errors for
outgoing message deliveries as well. we send reports once a day, to the
reporting addresses specified in TLSRPT records (rua) of a policy domain. these
reports are about MTA-STS policies and/or DANE policies, and about
STARTTLS-related failures.
sending reports is enabled by default, but can be disabled through setting
NoOutgoingTLSReports in mox.conf.
only at the end of the implementation process came the realization that the
TLSRPT policy domain for DANE (MX) hosts are separate from the TLSRPT policy
for the recipient domain, and that MTA-STS and DANE TLS/policy results are
typically delivered in separate reports. so MX hosts need their own TLSRPT
policies.
config for the per-host TLSRPT policy should be added to mox.conf for existing
installs, in field HostTLSRPT. it is automatically configured by quickstart for
new installs. with a HostTLSRPT config, the "dns records" and "dns check" admin
pages now suggest the per-host TLSRPT record. by creating that record, you're
requesting TLS reports about your MX host.
gathering all the TLS/policy results is somewhat tricky. the tentacles go
throughout the code. the positive result is that the TLS/policy-related code
had to be cleaned up a bit. for example, the smtpclient TLS modes now reflect
reality better, with independent settings about whether PKIX and/or DANE
verification has to be done, and/or whether verification errors have to be
ignored (e.g. for tls-required: no header). also, cached mtasts policies of
mode "none" are now cleaned up once the MTA-STS DNS record goes away.
2023-11-09 19:40:46 +03:00
return false , tlsDANE , errors . Is ( cerr , smtpclient . ErrTLS ) , "" , remoteIP , err . Error ( ) , hostResult , false
new feature: when delivering messages from the queue, make it possible to use a "transport"
the default transport is still just "direct delivery", where we connect to the
destination domain's MX servers.
other transports are:
- regular smtp without authentication, this is relaying to a smarthost.
- submission with authentication, e.g. to a third party email sending service.
- direct delivery, but with with connections going through a socks proxy. this
can be helpful if your ip is blocked, you need to get email out, and you have
another IP that isn't blocked.
keep in mind that for all of the above, appropriate SPF/DKIM settings have to
be configured. the "dnscheck" for a domain does a check for any SOCKS IP in the
SPF record. SPF for smtp/submission (ranges? includes?) and any DKIM
requirements cannot really be checked.
which transport is used can be configured through routes. routes can be set on
an account, a domain, or globally. the routes are evaluated in that order, with
the first match selecting the transport. these routes are evaluated for each
delivery attempt. common selection criteria are recipient domain and sender
domain, but also which delivery attempt this is. you could configured mox to
attempt sending through a 3rd party from the 4th attempt onwards.
routes and transports are optional. if no route matches, or an empty/zero
transport is selected, normal direct delivery is done.
we could already "submit" emails with 3rd party accounts with "sendmail". but
we now support more SASL authentication mechanisms with SMTP (not only PLAIN,
but also SCRAM-SHA-256, SCRAM-SHA-1 and CRAM-MD5), which sendmail now also
supports. sendmail will use the most secure mechanism supported by the server,
or the explicitly configured mechanism.
for issue #36 by dmikushin. also based on earlier discussion on hackernews.
2023-06-16 19:38:28 +03:00
}
}
implement "requiretls", rfc 8689
with requiretls, the tls verification mode/rules for email deliveries can be
changed by the sender/submitter. in two ways:
1. "requiretls" smtp extension to always enforce verified tls (with mta-sts or
dnssec+dane), along the entire delivery path until delivery into the final
destination mailbox (so entire transport is verified-tls-protected).
2. "tls-required: no" message header, to ignore any tls and tls verification
errors even if the recipient domain has a policy that requires tls verification
(mta-sts and/or dnssec+dane), allowing delivery of non-sensitive messages in
case of misconfiguration/interoperability issues (at least useful for sending
tls reports).
we enable requiretls by default (only when tls is active), for smtp and
submission. it can be disabled through the config.
for each delivery attempt, we now store (per recipient domain, in the account
of the sender) whether the smtp server supports starttls and requiretls. this
support is shown (after having sent a first message) in the webmail when
sending a message (the previous 3 bars under the address input field are now 5
bars, the first for starttls support, the last for requiretls support). when
all recipient domains for a message are known to implement requiretls,
requiretls is automatically selected for sending (instead of "default" tls
behaviour). users can also select the "fallback to insecure" to add the
"tls-required: no" header.
new metrics are added for insight into requiretls errors and (some, not yet
all) cases where tls-required-no ignored a tls/verification error.
the admin can change the requiretls status for messages in the queue. so with
default delivery attempts, when verified tls is required by failing, an admin
could potentially change the field to "tls-required: no"-behaviour.
messages received (over smtp) with the requiretls option, get a comment added
to their Received header line, just before "id", after "with".
2023-10-24 11:06:16 +03:00
// Update (overwite) last known starttls/requiretls support for recipient domain.
func updateRecipientDomainTLS ( ctx context . Context , senderAccount string , rdt store . RecipientDomainTLS ) error {
acc , err := store . OpenAccount ( senderAccount )
if err != nil {
return fmt . Errorf ( "open account: %w" , err )
}
err = acc . DB . Write ( ctx , func ( tx * bstore . Tx ) error {
// First delete any existing record.
if err := tx . Delete ( & store . RecipientDomainTLS { Domain : rdt . Domain } ) ; err != nil && err != bstore . ErrAbsent {
return fmt . Errorf ( "removing previous recipient domain tls status: %w" , err )
}
// Insert new record.
return tx . Insert ( & rdt )
} )
if err != nil {
return fmt . Errorf ( "adding recipient domain tls status to account database: %w" , err )
}
return nil
}