it should check if all packages (~500k) had their daily downloads done for the previous day. If not, raise a Sentry event.
If not, try to find out where in the pipeline it stopped. Run the queries we do to get the ids to get a count. The check package_download counts and last update dates. Then check rabbitmq if it has messages in it. Check if workers are running. etc. Wrap the whole thing in a Sentry transaction
A script was created to debug locally:
This can be used as the basis of the recurring job.
it should check if all packages (~500k) had their daily downloads done for the previous day. If not, raise a Sentry event.
If not, try to find out where in the pipeline it stopped. Run the queries we do to get the ids to get a count. The check package_download counts and last update dates. Then check rabbitmq if it has messages in it. Check if workers are running. etc. Wrap the whole thing in a Sentry transaction
A script was created to debug locally:
This can be used as the basis of the recurring job.