You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
operations-per-run is handy to not run up against the github api rate-limiter, however, every time the action runs: it tries to process the same (default 30) issues that it did on the prior runs. This has the affect that the stale-bot is not making progress on large backlogs over time.
The solution
Prefer to randomize the ordering, so that operations-per-run randomly picks which issues to process. This will have the effect of eventually (randomly) seeing every issue, instead of always only looking at the first (or last) 30 issues by default provided by the operations-per-run and ascending config settings. Thanks.
The text was updated successfully, but these errors were encountered:
It seems that 90 operations gives 23 issues, but then we'll end up repeatedly checking those same issues per actions/stale#692
So this ups the number to 200 which might be fast enough to at least tread water on our backlog.
Hopefully if we can eventually get a randomized order, we can stale more quickly with fewer operations.
The problem
operations-per-run is handy to not run up against the github api rate-limiter, however, every time the action runs: it tries to process the same (default
30
) issues that it did on the prior runs. This has the affect that the stale-bot is not making progress on large backlogs over time.The solution
Prefer to randomize the ordering, so that operations-per-run randomly picks which issues to process. This will have the effect of eventually (randomly) seeing every issue, instead of always only looking at the first (or last) 30 issues by default provided by the operations-per-run and ascending config settings. Thanks.
The text was updated successfully, but these errors were encountered: