Guh, this is complicated.
Fixes#4640
This also follows up on #4398 (reverting it) which made a change that technically worked, but was incorrect. It changed the condition in `hostsFromKeysNotHTTP` from `&&` to `||`, but then the function no longer did what its name said it would do, and it would return hosts even if they were marked with `http://`, if they used a non-HTTP port. That wasn't the intent of it. The test added in there was kept though, because it is a valid usecase.
The actual fix is to check _earlier_ whether all the addresses explicitly have `http://`, and if so we can short circuit and skip considering the rest.
Huge thank-you to Tailscale (https://tailscale.com) for making this change possible!
This is a great feature for Caddy and Tailscale is a great fit for a standard implementation.
* caddytls: GetCertificate modules; Tailscale
* Caddyfile support for get_certificate
Also fix AP provisioning in case of empty subject list (persist loaded
module on struct, much like Issuers, to surive reprovisioning).
And implement start of HTTP cert getter, still WIP.
* Update modules/caddytls/automation.go
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
* Use tsclient package, check status for name
* Implement HTTP cert getter
And use reuse CertMagic's PEM functions for private keys.
* Remove cache option from Tailscale getter
Tailscale does its own caching and we don't need the added complexity...
for now, at least.
* Several updates
- Option to disable cert automation in auto HTTPS
- Support multiple cert managers
- Remove cache feature from cert manager modules
- Minor improvements to auto HTTPS logging
* Run go mod tidy
* Try to get certificates from Tailscale implicitly
Only for domains ending in .ts.net.
I think this is really cool!
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
Some new users mistakenly try to define two sites without braces around each. Doing this can yield a confusing error message saying that their site address is an "unknown directive".
We can do better by keeping track of whether the current site block was parsed with or without a brace, then changing the error message later based on that.
For example, now this invalid config:
```
foo.example.com
respond "foo"
bar.example.com
respond "bar"
```
Will yield this error message:
```
$ caddy adapt
2021/08/22 19:21:31.028 INFO using adjacent Caddyfile
adapt: Caddyfile:4: unrecognized directive: bar.example.com
Did you mean to define a second site? If so, you must use curly braces around each site to separate their configurations.
```
* httpcaddyfile: Add shortcut for proxy hostport placeholder
I've noticed that it's a pretty common pattern to write a proxy like this, when needing to proxy over HTTPS:
```
reverse_proxy https://example.com {
header_up Host {http.reverse_proxy.upstream.hostport}
}
```
I find it pretty hard to remember the exact placeholder to use for this, and I continually need to refer to the docs when I need it. I think a simple fix for this is to add another Caddyfile placeholder for this one to shorten it:
```
reverse_proxy https://example.com {
header_up Host {proxy_hostport}
}
```
* Switch the shortcut name
* httpcaddyfile: ensure hosts to skip can always be collected
Previously, some hosts that should be skipped in logging would
be missed as the current logic would only collect them after
encountering the first server that would log. This change makes sure
the ServerLogConfig is initialized before iterating over the server
blocks.
* httpcaddyfile: add test case for skip hosts behavior
This change is aimed at enhancing the logging module within the
Caddyfile directive to allow users to configure logs other than the HTTP
access log stream, which is the current capability of the Caddyfile [1].
The intent here is to leverage the same syntax as the server log
directive at a global level, so that similar customizations can be added
without needing to resort to a JSON-based configuration.
Discussion for this approach happened in the referenced issue.
Closes https://github.com/caddyserver/caddy/issues/3958
[1] https://caddyserver.com/docs/caddyfile/directives/log
The HTTP Caddyfile adapter can now configure the PKI app, and the acme_server directive can now be used to specify a custom CA used for issuing certificates. More customization options can follow later as needed.
This is probably an invasive change, but existing tests continue to pass.
It seems to make sense this way. There is likely an edge case I haven't
considered.
This changes the signature of UnmarshalGlobalFunc but this is probably OK since it's only used by this repo as far as we know.
We need this change in order to "remember" the previous value in case a global option appears more than once, which is now a possibility with the cert_issuer option since Caddy now supports multiple issuers in the order defined by the user.
Bonus: the issuer subdirective of tls now supports one-liner for "acme" when all you need to set is the directory:
issuer acme <dir>
* httpcaddyfile: First pass at implementing server options
* httpcaddyfile: Add listener wrapper support
* httpcaddyfile: Sort sbaddrs to make adapt output more deterministic
* httpcaddyfile: Add server options adapt tests
* httpcaddyfile: Windows line endings lol
* caddytest: More windows line endings lol (sorry Matt)
* Update caddyconfig/httpcaddyfile/serveroptions.go
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
* httpcaddyfile: Reword listener address "matcher"
* Apply suggestions from code review
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
* httpcaddyfile: Deprecate experimental_http3 option (moved to servers)
* httpcaddyfile: Remove validation step, no longer needed
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
* httpcaddyfile: Ensure handle_path is sorted as equal to handle
* httpcaddyfile: Make mutual exclusivity grouping deterministic (I hope)
* httpcaddyfile: Add comment linking to the issue being fixed
* httpcaddyfile: Typo fix, comment clarity
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
* Update caddyconfig/httpcaddyfile/httptype.go
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
We recently introduced `if !cp.SettingsEmpty()` which conditionally
adds the connection policy to the list. If the condition evaluates to
false, the policy wouldn't actually be added, even if
hasCatchAllTLSConnPolicy was set to true on the previous line.
Now we set that variable in accordance with whether we actually add
the policy.
While debugging this I noticed that catch-all policies added early in
that loop (i.e. not at the end if we later determine we need one) are
not always at the end of the list. They should be, though, since they
are selected by which one matches first, and having a catch-all first
would nullify any more specific ones later in the list. So I added a
sort in consolidateConnPolicies to take care of that.
Should fix#3670 and
https://caddy.community/t/combining-on-demand-tls-with-custom-ssl-certs-doesnt-seem-to-work-in-2-1-1/9719
but I won't know for sure until somebody verifies it, since at least in
the GitHub issue there is not yet enough information (the configs are
redacted).
* Bring `ensure_origin` and `origins` to caddyfile admin config
* Add unit test for caddyfile admin config update
* Add caddyfile adapt test for typical admin setup
* httpcaddyfile: Replace admin config error message when there's more arguments than needed
Replace d.Err() to d.ArgErr() since the latter provides similarly informative error message
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
Catch-alls should always go last. Normally this is the case, but we have
a special case for comparing one wildcard-host site block to another
non-wildcard host site block; and a catch-all site block is also a
non-wildcard host site block, so now we have to special-case the
catch-all site block. Sigh.
This could be reproduced with a Caddyfile that has two site blocks:
":80" and "*.example.com", in that order.
* httpcaddyfile: Add shorthands for parameterized placeholders
httpcaddyfile: Now with regexp instead
httpcaddyfile: Allow dashes, gofmt
httpcaddyfile: Compile regexp only once
httpcaddyfile: Cleanup struct
httpcaddyfile: Optimize the replacers, pull out of the loop
httpcaddyfile: Add `{port}` shorthand
* httpcaddyfile: Switch `r.` to `re.`
* httpcaddyfile: Make global options pluggable
* httpcaddyfile: Add a global options adapt test
* httpcaddyfile: Wrap err
Co-Authored-By: Dave Henderson <dhenderson@gmail.com>
* httpcaddyfile: Revert wrap err
Co-authored-by: Dave Henderson <dhenderson@gmail.com>
This can lead to nicer, smaller JSON output for Caddyfiles like this:
a {
tls internal
}
b {
tls foo@bar.com
}
i.e. where the tls directive only configures automation policies, and
is merely meant to enable TLS on a server block (if it wasn't implied).
This helps keeps implicit config implicit.
Needs a little more testing to ensure it doesn't break anything
important.
* httpcaddyfile: Exclude access logs written to files from default log
Even though any logs can just be ignored, most users don't seem to like
configuring an access log to go to a file only to have it doubly appear
in the default log.
Related to:
- #3294
- https://caddy.community/t/v2-logging-format/7642/4?u=matt
- https://caddy.community/t/caddyfile-questions/7651/3?u=matt
* caddyhttp: General improvements to access log controls (fixes#3310)
* caddyhttp: Move log config nil check higher
* Rename LoggerName -> DefaultLoggerName
Panic would happen if an automation policy was specified in a singular
server block that had no hostnames in its address. Definitely an edge
case.
Fixed a bug related to checking for server blocks with a host-less key
that tried to make an automation policy. Previously if you had only two
server blocks like ":443" and another one at ":80", the one at ":443"
could not create a TLS automation policy because it thought it would
interfere with TLS automation for the block at ":80", but obviously that
key doesn't enable TLS because it is on the HTTP port. So now we are a
little smarter and count only non-HTTP-empty-hostname keys.
Also fixed a bug so that a key like "https://:1234" is sure to have TLS
enabled by giving it a TLS connection policy. (Relaxed conditions
slightly; the previous conditions were too strict, requiring there to be
a TLS conn policy already or a default SNI to be non-empty.)
Also clarified a comment thanks to feedback from @Mohammed90
- Create two default automation policies; if the TLS app is used in
isolation with the 'automate' certificate loader, it will now use
an internal issuer for internal-only names, and an ACME issuer for
all other names by default.
- If the HTTP Caddyfile adds an 'automate' loader, it now also adds an
automation policy for any names in that loader that do not qualify
for public certificates so that they will be issued internally. (It
might be nice if this wasn't necessary, but the alternative is to
either make auto-HTTPS logic way more complex by scanning the names in
the 'automate' loader, or to have an automation policy without an
issuer switch between default issuer based on the name being issued
a certificate - I think I like the latter option better, right now we
do something kind of like that but at a level above each individual
automation policies, we do that switch only when no automation
policies match, rather than when a policy without an issuer does
match.)
- Set the default LoggerName rather than a LoggerNames with an empty
host value, which is now taken literally rather than as a catch-all.
- hostsFromKeys, the function that gets a list of hosts from server
block keys, no longer returns an empty string in its resulting slice,
ever.
We now store the parsed site/server block keys with the server block,
rather than parsing the addresses every time we read them.
Also detect conflicting schemes, i.e. TLS and non-TLS cannot be served
from the same server (natively -- modules could be built for it).
Also do not add site subroutes (subroutes generated specifically from
site blocks in the Caddyfile) that are empty.
Certificate selection used to be a module, but this seems unnecessary,
especially since the built-in CustomSelectionPolicy allows quite complex
selection logic on a number of fields in certs. If we need to extend
that logic, we can, but I don't think there are SO many possibilities
that we need modules.
This update also allows certificate selection to choose between multiple
matching certs based on client compatibility and makes a number of other
improvements in the default cert selection logic, both here and in the
latest CertMagic.
The hardest part of this was the conn policy consolidation logic
(Caddyfile only, of course). We have to merge connection policies that
we can easily combine, because if two certs are manually loaded in a
Caddyfile site block, that produces two connection policies, and each
cert is tagged with a different tag, meaning only the first would ever
be selected. So given the same matchers, we can merge the two, but this
required improving the Tag selection logic to support multiple tags to
choose from, hence "tags" changed to "any_tag" or "all_tags" (but we
use any_tag in our Caddyfile logic).
Combining conn policies with conflicting settings is impossible, so
that should return an error if two policies with the exact same matchers
have non-empty settings that are not the same (the one exception being
any_tag which we can merge because the logic for them is to OR them).
It was a bit complicated. It seems to work in numerous tests I've
conducted, but we'll see how it pans out in the release candidates.
If a site block has a key like "http://localhost:2016", then the log for
that site must be mapped to "localhost:2016" and not just "localhost"
because "localhost:2016" will be the value of the Host header of requests.
But a key like "localhost:80" does not include the port since the Host
header will not include ":80" because it is a standard port.
Fixes https://caddy.community/t/v2-common-log-format-not-working/7352?u=matt
It's hard to say whether this was actually a bug, but the linked issue
shows why the old behavior was confusing. Basically, we infer that a
rewrite handler is supposed to act as an internal redirect, which likely
means it will no longer match the matcher(s) it did before the rewrite.
So if the rewrite directive shares a matcher with any adjacent route or
directive, it can be confusing/misleading if we consolidate the rewrite
into the same route as the next handler, which shouldn't (probably) match
after the rewrite is complete.
This is kiiiind of a hacky workaround to a quirky problem.
For edge cases like these, it is probably "cleaner" to just use handle
blocks instead, to group handlers under the same matcher, nginx-style.
* added sni tests
* set the default sni when there is no host to match
* removed invalid sni test. Disabled tests that rely on host headers.
* readded SNI tests. Added logging of config load times
