Expire idle pooled connections when the peer sends no Keep-Alive - #2683
Open
DavidTavoularis wants to merge 1 commit into
Open
Expire idle pooled connections when the peer sends no Keep-Alive#2683DavidTavoularis wants to merge 1 commit into
DavidTavoularis wants to merge 1 commit into
Conversation
Copy link
Copy Markdown
Sorry, something went wrong.
Podman closes an idle API connection after twice its service_timeout, 10s by default, and advertises no Keep-Alive header. DefaultConnectionKeepAliveStrategy therefore falls back to RequestConfig's 3-minute default, so the pool keeps handing out connections the daemon dropped long ago. Revalidation on lease is disabled (c61da29, for docker-java#1726) and HttpClient5 will not retry a POST, so /containers/create and /containers/{id}/exec surface the dead connection to the caller as "NoHttpResponseException: localhost:2375 failed to respond". Set a connection keep-alive fallback so an idle connection is expired before any known daemon's window can close it. It is a fallback rather than a ceiling: a peer that advertises its own Keep-Alive timeout is still honoured. Tests come in two layers. IdleConnectionReuseTest drives a fake daemon over tcp:// and needs no docker at all; IdleConnectionReuseIT runs against whatever DOCKER_HOST points at and skips unless a raw-socket probe shows the daemon actually hangs up on idle connections, which it does not on docker. Fixes docker-java#2682 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WCQ8B2RchzdWf7xbhkaXQr Signed-off-by: David Tavoularis <david.tavoularis@mycom-osi.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Podman closes an idle API connection after twice its service_timeout, 10s by default, and advertises no Keep-Alive header. DefaultConnectionKeepAliveStrategy therefore falls back to RequestConfig's 3-minute default, so the pool keeps handing out connections the daemon dropped long ago. Revalidation on lease is disabled (c61da29, for #1726) and HttpClient5 will not retry a POST, so /containers/create and /containers/{id}/exec surface the dead connection to the caller as "NoHttpResponseException: localhost:2375 failed to respond".
Set a connection keep-alive fallback so an idle connection is expired before any known daemon's window can close it. It is a fallback rather than a ceiling: a peer that advertises its own Keep-Alive timeout is still honoured.
Tests come in two layers. IdleConnectionReuseTest drives a fake daemon over tcp:// and needs no docker at all; IdleConnectionReuseIT runs against whatever DOCKER_HOST points at and skips unless a raw-socket probe shows the daemon actually hangs up on idle connections, which it does not on docker.
Fixes #2682