chore: pin celeris main (be12a98) in the refapps, debugvars and the bench server - #323
Merged
Merged
Conversation
…ench server
Moves all ten modules from c9f7dad to be12a98 so the next validation run covers
what the release will actually ship. Since the previous pin, celeris gained:
#572 one shared bounded close-time recv drain for both loop engines (#571)
#574 clear() on four reuse queues whose [:0] truncation pinned every slot
up to cap -- measured to be the fix for the soak's only failure
#531 x/sys v0.47.0 -> v0.48.0, klauspost/compress v1.19.2 -> v1.20.0
#575 docs: go.mod is the single source of truth for the Go floor (#417)
observability's middleware/metrics and middleware/otel move too, from the
v1.5.11 tags to the same pseudo-version, so every celeris module in the tree
is one commit rather than a mix of tags and main.
The point of this repin is #574. The 24h soak (34368601903) failed on exactly
two cells of forty-eight -- auth_session_ratelimit/io_uring on both arches,
I-MEM-1 only -- and a local A/B of two trees differing by that one line took
the failing cell from 1651 violations to zero, with the trough slope going
1071 -> 52 B/s (epoll, for scale, is 131). That result is amd64-under-docker
on one machine; this repin is what lets the cluster confirm it on both arches.
All ten modules build for linux/amd64 against the new pin.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Moves all ten modules from
c9f7dadtobe12a98so the next validation run covers what the release will actually ship.Since the previous pin, celeris gained:
clear()on four reuse queues whose[:0]truncation pinned every slot up tocapobservability'smiddleware/metricsandmiddleware/otelmove too, from the v1.5.11 tags to the same pseudo-version, so every celeris module in the tree is one commit rather than a mix of tags and main.Why now
The point of this repin is goceleris/celeris#574. The 24h soak (run 34368601903) failed on exactly two cells of forty-eight —
auth_session_ratelimit/io_uring on both architectures, I-MEM-1 only. A local A/B of two trees differing by that single line took the failing cell from 1,651 violations to zero, with the trough slope going 1,071 → 52 B/s (epoll, for scale, sits at 131).That result is one machine, amd64 under Docker. This repin is what lets the cluster confirm it on both architectures, which the release bar requires and I cannot claim without.
All ten modules build for linux/amd64 against the new pin.