varnish: remove thread pool overrides, use built-in defaults - #1350
Closed
gquintard wants to merge 5 commits into
Closed
varnish: remove thread pool overrides, use built-in defaults#1350gquintard wants to merge 5 commits into
gquintard wants to merge 5 commits into
Conversation
thread_pool_max=10000/min=5000 (x2 pools = up to ~20k threads) is wildly oversized for a 64-logical-CPU container running CPU-bound synthetic responses. Strip it to get a clean varnishd-defaults baseline before sweeping custom pool sizes. Also fixes a missing line-continuation after vsl_mask=none that made -s malloc,256m dead code, never reaching varnishd. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
gquintard
requested review from
Kaliumhexacyanoferrat and
MDA2AV
as code owners
August 27, 2026 21:42
Contributor
|
👋 Heads up! This PR modifies the following frameworks:
|
Contributor
Author
|
/benchmark -f varnish --save |
Contributor
|
👋 Benchmark request received. A collaborator will review and approve the run. |
2 tasks
Contributor
Benchmark ResultsFramework:
Full log |
gquintard
marked this pull request as draft
August 28, 2026 05:10
Explicit config replacing bare defaults: more pools to reduce cross-CCX lock/cache-line contention on the Threadripper's multi-CCX layout, sized well below the original 5000/10000 min/max but above varnishd's built-in defaults. Also fixes a missing line-continuation after vsl_mask=none that made -s malloc,256m dead code, never reaching varnishd. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Contributor
Author
|
/benchmark -f varnish --save |
Contributor
|
👋 Benchmark request received. A collaborator will review and approve the run. |
Contributor
|
|
Owner
|
/benchmark -f varnish --save |
Contributor
|
👋 Benchmark request received. A collaborator will review and approve the run. |
Contributor
Benchmark ResultsFramework:
Full log |
Contributor
Author
|
this confirms that tweaking the thread pools doesn't do much, and is even harmful. Closing |
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
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.
Summary
thread_pool_max=10000/thread_pool_min=5000(x2 pools = up to ~20k threads) is wildly oversized for the 64-logical-CPU cpuset the benchmark container runs on for CPU-bound synthetic responses. This strips the override to get a clean varnishd-defaults baseline before sweeping custom pool sizes.vsl_mask=nonethat made-s malloc,256mdead code, never reaching varnishd.Test plan
varnishstatthread/queue counters (MAIN.sess_queued,MAIN.threads_limited) against current main