fixing / warning with sentry error-reporting?

[syphar]

I just stumbled on a small problem (in my config?) question is if we should fix this in django-pq, or at least document the behavior.

• default sentry DSNs (in docs, heroku, ...) are with scheme https
• since https://github.com/getsentry/raven-python/commit/7a693286b6304d31a7743ffcc35dfea4e02b8a43 , seems to be in raven 3.6.0, the threaded http transport for the http or https scheme is default.
• pqworker forks a process to do the job-work
• the threaded raven transport uses a background thread to send the events to sentry (see the source)
• our Worker class kills the forked process when the work is done.

So the result is, failed jobs are not sent to sentry.

What should be the best solution to fix this?

• wait somehow until the background thread is finished?
• on registering, raise an error when an async transport is used?
• document the behavior?

edit: changed raven-version where this appeared.

[syphar]

any idea on this? In reality, sentry integration is broken for many configurations atm.

I'll be happy to provide a PR if it's clear which solution to follow.

[bretth]

I haven't had a chance to fix this, but a quick review suggests to me that changing the way pqworker loads sentry is the best course of action. Instead of using the SENTRY_DSN the pqworker management command should use a new setting PQ_SENTRY_DSN, and contrib.sentry should provide warnings or exception if a threaded transport is used. The rational for sentry to switch to async methods makes sense for the request/response cycle but has no real benefit for background tasks anyway.