Skip to content

Reject unscoped requests to task cluster endpoint - #1284

Merged
jake-low merged 1 commit into
mainfrom
jlow/task-cluster-perf-fixes
Oct 1, 2026
Merged

jake-low merged 1 commit into
mainfrom
jlow/task-cluster-perf-fixes

Conversation

@jake-low

@jake-low jake-low commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor

Previously, PUT /api/v2/taskCluster with no challenge id or bbox would run k-means clustering over every task in the database, which would take minutes (the reverse proxy would reply with a 504 long before the query returned) and potentially spill a large amount of data to temp files on disk, which managed to cause the DB autoscaler to kick in twice recently (on Sept 25 and Sept 30).

Previously, PUT /api/v2/taskCluster with noe challenge id or bbox would
run k-means clustering over every task in the database, which would take
minutes (the reverse proxy would reply with a 504 long before the query
returned) and potentially spill a large amount of data to temp files on
disk, which managed to cause the DB autoscaler to kick in twice recently
(on Sept 25 and Sept 30).
@sonarqubecloud

sonarqubecloud Bot commented Oct 1, 2026

Copy link
Copy Markdown

@jake-low
jake-low merged commit b9b2e69 into main Oct 1, 2026
9 checks passed
@jake-low
jake-low deleted the jlow/task-cluster-perf-fixes branch October 1, 2026 01:39
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant