ruaok: there are a bunch of sentry errors like https://sentry.metabrainz.org/metabrainz/listen... from the mbid_mapping_writer. The stacktraces are getting truncated so we won't be able to know the exact errors. I looked into this and it seems to because of using at a few places `traceback.format_exc`, https://forum.sentry.io/t/some-stack-traces-are... mentions that this can lead to truncating the
stacktrace. alternatively, it is not possible to replace this usage, we can look into increasing the limit of string length reported by sentry.
killme joined the channel
ruaok
_lucifer: oh, interesting. an incorrect cron job ran and borked the underlying DB. fun.
the place where the exception gets reported is here:
getting exceptions from the futures stuff was a pain...
_lucifer
ruaok, oh! i had seen that line but the line numbers in the stack trace didn't match so i was confused. the mismatch is probably because this is async code i guess.
ruaok
background threads, yes.
_lucifer
can we do logger.error(exc) there, does that eat up some info?
*, or
ruaok
I can try that. let me fix the underlying problem.
[bookbrainz-site] dependabot[bot] opened pull request #627 (master…dependabot/npm_and_yarn/hosted-git-info-2.8.9): chore(deps): bump hosted-git-info from 2.7.1 to 2.8.9 https://github.com/bookbrainz/bookbrainz-site/p...