perf(memory): run under jemalloc to bound RSS - #275
Merged
Conversation
Preload libjemalloc and set PYTHONMALLOC=malloc so freed per-session pages are returned to the OS via madvise decay, instead of glibc/pymalloc ratcheting RSS to the peak concurrent working set. Replaces the earlier MALLOC_ARENA_MAX mitigation.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Run the container process under jemalloc so freed per-session memory is returned to the OS.
Problem: glibc/pymalloc never release the arenas fragmented by per-session widget churn, so RSS ratchets to the peak concurrent working set and stays there until restart — the production OOM.
Fix: preload libjemalloc (
PYTHONMALLOC=malloc, decay-based page purge) so memory follows users back down. Replaces the earlierMALLOC_ARENA_MAXmitigation.The ENV is placed after the build layers so image builds don't run under the preload. Verify after any image change: