Fix pycache not working on linux local dev boxes - #13342
Draft
cdrini wants to merge 1 commit into
Draft
Conversation
…tup cost Neither the bind-mounted app source nor site-packages are writable by the container user on Linux, so every process start recompiled the entire import tree from scratch. Redirects caching to a dedicated, pre-owned volume and pre-warms it at image build time. Co-Authored-By: Drini Cami <cdrini@gmail.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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Neither the bind-mounted app source nor site-packages are writable by the container user on Linux, so every process start recompiled the entire import tree from scratch. Redirects caching to a dedicated, pre-owned volume and pre-warms it at image build time.
Technical
Total memory over time (non-solr containers, summed)
xychart-beta title "Total memory across non-solr containers (MB)" x-axis "Elapsed (s)" [0, 7, 15, 22, 29, 36, 44, 51, 58, 66, 73, 80, 87, 95, 102, 109, 116] y-axis "Memory (MB)" 0 --> 1400 line [604 "before", 679, 671, 755, 1085, 955, 995, 1300, 918, 920, 923, 926, 921, 923, 922, 926, 920] line [633 "after", 616, 619, 628, 812, 885, 894, 797, 797, 797, 801, 804, 798, 804, 800, 803, 798]The
beforeline spikes harder mid-startup (peaking ~1.3GB around the reindex burst at 51s) thanafter(~895MB peak) — less to compile means less transient memory pressure during the busiest part of startup, not just a lower resting value.Testing
Screenshot
Stakeholders