- Introduce StrictHostMatching mode for sites that require clientauth
- Error if QUIC is enabled whilst TLS clientauth is configured
(Our QUIC implementation does not yet support TLS clientauth, but
maybe it will in the future - fixes#2095)
- Error if one but not all TLS configs for the same hostname have a
different ClientAuth CA pool
Also add SSL_PROTOCOL and SSL_CIPHER env vars for fastcgi.
* Implement placeholders for ssl_protocol and ssl_cipher
* gofmt
* goimports
* Housekeeping and implement as {tls_protocol} and {tls_cipher}
* tls: Fall back to certificate keyed by empty name (fixes#2035)
This should only happen for sites defined with an empty hostname (like
":8080") and which are using self-signed certificates or some other
funky self-managed certificate. But that certificate should arguably
be used for all incoming SNI names.
* tls: Revert to serving any certificate if no match, regardless of SNI
Also fix self-signed certs to include IP addresses in their name
if they are configured to serve an IP address
* Remove tests which are now irrelevant (behavior reverted)
It would be good to revisit this in the future.
Caddy can now obtain certificates when behind load balancers and/or in
fleet/cluster configurations, without needing any extra configuration.
The only requirement is sharing the same $CADDYPATH/acme folder.
This works with the HTTP challenge, whereas before the DNS challenge
was required. This commit allows one Caddy instance to initiate the
HTTP challenge and another to complete it.
When sharing that folder, certificate management is synchronized and
coordinated, without the Caddy instances needing to know about each
other. No load balancer reconfiguration should be required, either.
Currently, this is only supported when using FileStorage for TLS
storage (which is ~99.999% of users).
- Using xenolf/lego's likely-temporary acmev2 branch
- Cleaned up vendor folder a little bit (probably more to do)
- Temporarily set default CA URL to v2 staging endpoint
- Refactored user management a bit; updated tests (biggest change is
how we get the email address, which now requires being able to make
an ACME client with a User with a private key so that we can get the
current ToS URL)
- Automatic HTTPS now allows specific wildcard pattern hostnames
- Commented out (but kept) the TLS-SNI code, as the challenge type
may return in the future in a similar form
See discussion on #2015; the initial change had removed this check, and
I can't remember why I removed it or if it was accidental. Anyway, it's
back now.
Also introduce caddy.OnProcessExit which is a list of functions that
run before exiting the process cleanly; these do not count as shutdown
callbacks, so they do not return errors and must execute quickly.
- Expose the list of Caddy instances through caddy.Instances()
- Added arbitrary storage to caddy.Instance
- The cache of loaded certificates is no longer global; now scoped
per-instance, meaning upon reload (like SIGUSR1) the old cert cache
will be discarded entirely, whereas before, aggressively reloading
config that added and removed lots of sites would cause unnecessary
build-up in the cache over time.
- Key certificates in the cache by their SHA-256 hash instead of
by their names. This means certificates will not be duplicated in
memory (within each instance), making Caddy much more memory-efficient
for large-scale deployments with thousands of sites sharing certs.
- Perform name-to-certificate lookups scoped per caddytls.Config instead
of a single global lookup. This prevents certificates from stepping on
each other when they overlap in their names.
- Do not allow TLS configurations keyed by the same hostname to be
different; this now throws an error.
- Updated relevant tests, with a stark awareness that more tests are
needed.
- Change the NewContext function signature to include an *Instance.
- Strongly recommend (basically require) use of caddytls.NewConfig()
to create a new *caddytls.Config, to ensure pointers to the instance
certificate cache are initialized properly.
- Update the TLS-SNI challenge solver (even though TLS-SNI is disabled
currently on the CA side). Store temporary challenge cert in instance
cache, but do so directly by the ACME challenge name, not the hash.
Modified the getCertificate function to check the cache directly for
a name match if one isn't found otherwise. This will allow any
caddytls.Config to be able to help solve a TLS-SNI challenge, with one
extra side-effect that might actually be kind of interesting (and
useless): clients could send a certificate's hash as the SNI and
Caddy would be able to serve that certificate for the handshake.
- Do not attempt to match a "default" (random) certificate when SNI
is present but unrecognized; return no certificate so a TLS alert
happens instead.
- Store an Instance in the list of instances even while the instance
is still starting up (this allows access to the cert cache for
performing renewals at startup, etc). Will be removed from list again
if instance startup fails.
- Laid groundwork for ACMEv2 and Let's Encrypt wildcard support.
Server type plugins will need to be updated slightly to accommodate
minor adjustments to their API (like passing in an Instance). This
commit includes the changes for the HTTP server.
Certain Caddyfile configurations might error out with this change, if
they configured different TLS settings for the same hostname.
This change trades some complexity for other complexity, but ultimately
this new complexity is more correct and robust than earlier logic.
Fixes#1991Fixes#1994Fixes#1303
This adds the ask sub-directive to tls that defines the URL of a backend HTTP service to be queried during the TLS handshake to determine if an on-demand TLS certificate should be acquired for incoming hostnames. When the ask sub-directive is defined, Caddy will query the URL for permission to acquire a cert by making a HTTP GET request to the URL including the requested domain in the query string. If the backend service returns a 2xx response Caddy will acquire a cert. Any other response code (including 3xx redirects) are be considered a rejection and the certificate will not be acquired.
Renewed certificates would not be reloaded into the cache because their
names conflict with names of certificates already in the cache; this
was intentional when loading new certs to avoid confusion, but is
problematic when renewing, since the old certificate doesn't get
evicted from the cache. (Oops.)
Here, I remedy this situation by explicitly deleting the old cert from
the cache before adding the renewed one back in.
https://caddy.community/t/random-ocsp-response-errors-for-random-clients/2473?u=matt
Certificates are keyed by name in the cache, optimized for fast lookups
during TLS handshakes using SNI. A more "correct" way that is truly a
1:1 would be to cache certificates by a hash of the leaf's DER bytes,
but this involves an extra index to maintain. So instead of that, we
simply choose to prevent overlap when keying certificates by server
name. This avoids the ambiguity when updating OCSP staples, for instance.
* introduced own ChallengeProvider type, based on acme.ChallengeProvider to avoid vendoring/version mismatches in Caddy plugins; see Caddy issue #1697
* fixed up comments for ChallengeProvider
* moved ChallengeProvider to caddytls/tls.go
