Add procstat telegraf plugins for gunicorn web services#69
Open
mvdbeek wants to merge 1 commit into
Open
Conversation
The web nodes only collected host-level memory, so the 2026-06-30 memory surge on main1/main2 could not be reconstructed per-process (the handler group already had procstat). Add per-PID procstat for both gunicorn services (:8080/:8081) so worker RSS/CoW growth is captured going forward.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Adds per-PID
procstattelegraf plugins for both gunicorn web services (:8080/:8081) ongalaxy_main_web_servers.Why
During the 2026-06-30 memory + load surge on main1/main2, the web nodes only collected host-level memory (
mem.used), so the incident could not be reconstructed per-process — we had to fall back to livesmaps/psand log forensics. The handler group already shipsprocstat; this brings the web nodes to parity so worker RSS / CoW growth is captured historically and the next occurrence is a Grafana lookup.Notes
-b 0.0.0.0:8080/:8081) — the only per-service discriminator in the cmdline — which preserves the @0/@1 split (they behaved differently in the incident: @1 throttled, @0 didn't) and excludes the backup-staging gunicorn.--max-requests/ memory limits / swap; those mitigations are intentionally left for a separate change.