Hi Sentry Support,
Org: elm-ai
Projects: x-auditsense, s-auditsense, notification-service (and unexpectedly datapipelines)
Setup (project-level Issue Alerts):
Each project has one Issue Alert with filters like
env does not contain LOCAL_DEV, development
level ≥ error
message does not contain several strings (e.g., “Running AI Pipeline”, “Error uploading to AI Pipeline”, “regrade/evidence was u…”)
Action: Send a notification via Webhook (PagerTree).
Sentry PagerTree integration is disabled; no org-wide or catch-all rules configured.
Problems:
Webhooks fire for events that don’t match our filters (e.g., level=info, message contains a blocked string).
Example: https://elm-ai.sentry.io/issues/6762071340/?alert_rule_id=15435648&alert_type=issue¬ification_uuid=b5a216bc-d4d4-4c96-8807-a6fd61df213f
Webhooks appear to originate from datapipelines, a project where we did not configure a PagerTree webhook.
Requests:
Identify which rule fired for the example above (by alert_rule_id / notification_uuid) and why it matched despite filters.
Confirm what field “Message … does not contain” evaluates for Node events (event.message / logentry.formatted?).
Verify whether any org-level/legacy mechanism could send webhooks outside of project Issue Alerts.
Any pointers to rule-evaluation logs we can check ourselves?
Happy to provide additional examples or screenshots.
Thanks,
Eesha Khanna