Currently if we extract the DialInfo from a Request Context during an active health check, then the Upstream in the DialInfo is nil.
This PR attempts to set the Upstream to a sensible value, based on wether or not the Upstream has been overriden in the active health check's config.
* reverse_proxy: re-add healthy upstreams metric
Signed-off-by: Mohammed Al Sahaf <msaa1990@gmail.com>
* lint
Signed-off-by: Mohammed Al Sahaf <msaa1990@gmail.com>
---------
Signed-off-by: Mohammed Al Sahaf <msaa1990@gmail.com>
* buffer requests for fastcgi by default
* fix import cycle
* fix the return value of bufferedBody
* more comments about fastcgi buffering
---------
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
* reverse proxy: rewrite requests and responses for websocket over http2
* delete protocol pseudo-header
* modify cloned requests
* set request variable to track if it's a h2 websocket
* use request bodu
* rewrite request body
* use WebSocket instead of Websocket in the headers
* use logger check for zap loggers
* fix lint
i.e. Revert commit f5dce84a70
Two years ago, the patch in #4952 was a seemingly necessary way to fix an issue (sort of an edge case), but it broke other more common use cases (see #6666).
Now, as of #6669, it seems like the original issue can no longer be replicated, so we are reverting that patch, because it was incorrect anyway.
If it turns out the original issue returns, a more proper patch may be in #6669 (even if used as a baseline for a future fix). A potential future fix could be an opt-in setting.
* caddyhttp: Add `MatchWithError` to replace SetVar hack
* Error in IP matchers on TLS handshake not complete
* Use MatchWithError everywhere possible
* Move implementations to MatchWithError versions
* Looser interface checking to allow fallback
* CEL factories can return RequestMatcherWithError
* Clarifying comment since it's subtle that an err is returned
* Return 425 Too Early status in IP matchers
* Keep AnyMatch signature the same for now
* Apparently Deprecated can't be all-uppercase to get IDE linting
* Linter
Most of the errors that can be seen here are write errors due to clients
aborting the request from their side. Often seen ones include:
* writing: ... write: broken pipe
* writing: ... connection timed out
* writing: http2: stream closed
* writing: timeout...
* writing: h3 error...
Most of these errors are beyond of the control of caddy on the client side,
probably nothing can be done on the server side. It still warrants
researching when these errors occur very often, so a change in level from
error to warn is better here to not polute the logs with errors in the
normal case.
* caddyhttp: Reject 0-RTT early data in IP matchers and set Early-Data header when proxying
See RFC 8470: https://httpwg.org/specs/rfc8470.html
Thanks to Michael Wedl (@MWedl) at the University of Applied Sciences St. Poelten for reporting this.
* Don't return value for {remote} placeholder in early data
* Add Caddyfile support
Before this change, a read of size (let's say) < 10, into a buffer of size 10, will return EOF because we're using CopyN to limit to the size of the buffer. That resulted in the body being read from later, which should only happen if it couldn't fit in the buffer.
With this change, the body is properly NOT set when it can all fit in the buffer.
* reverseproxy: Add more debug logs
This makes debug logging very noisy when reverse proxying, but I guess
that's the point.
This has shown to be useful in troubleshooting infrastructure issues.
* Update modules/caddyhttp/reverseproxy/streaming.go
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
* Update modules/caddyhttp/reverseproxy/streaming.go
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
* Add opt-in `trace_logs` option
* Rename to VerboseLogs
---------
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
* use gofmput to format code
* use gci to format imports
* reconfigure gci
* linter autofixes
* rearrange imports a little
* export GOOS=windows golangci-lint run ./... --fix
* caddyhttp: Make use of http.ResponseController
Also syncs the reverseproxy implementation with stdlib's which now uses ResponseController as well 2449bbb5e6
* Enable full-duplex for HTTP/1.1
* Appease linter
* Add warning for builds with Go 1.20, so it's less surprising to users
* Improved godoc for EnableFullDuplex, copied text from stdlib
* Only wrap in encode if not already wrapped
* update quic-go to v0.37.0
* Bump to Go 1.20
* Bump golangci-lint version, yml syntax consistency
* Use skip-pkg-cache workaround
* Workaround needed for both?
* Seeding weakrand is no longer necessary
---------
Co-authored-by: Matt Holt <mholt@users.noreply.github.com>
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
Ideally I'd just remove the parameter to caddy.Context.Logger(), but
this would break most Caddy plugins.
Instead, I'm making it variadic and marking it as partially deprecated.
In the future, I might completely remove the parameter once most
plugins have updated.
* reverseproxy: Close hijacked conns on reload/quit
We also send a Close control message to both ends of
WebSocket connections. I have tested this many times in
my dev environment with consistent success, although
the variety of scenarios was limited.
* Oops... actually call Close() this time
* CloseMessage --> closeMessage
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
* Use httpguts, duh
* Use map instead of sync.Map
Co-authored-by: Francis Lavoie <lavofr@gmail.com>
Jesper Brix Rosenkilde
WeidiDeng
Mohammed Al Sahaf
Matthew Holt
bt90
Francis Lavoie
Kévin Dunglas
Steffen Busch
Jens-Uwe Mager
Yolan Romailler
Fred Cox
Patrick Koenig
Jacob Gadikian
mmm444
Marten Seemann
Corin Langosch
Trea Hauet
Y.Horie