What should I check when Sentry capture calls succeed but no Bicho event appears? #68
|
How can delivery problems be narrowed down after changing a Sentry DSN? Which checks distinguish a transport problem from a stored event hidden by search filters? |
Replies: 1 comment
|
Start with the actual network destination. An old Check authentication next. All supplied keys must match the route project; secret-enabled backend projects also need the generated transport headers. Never add an admin token to the SDK as a fix. For browser requests, inspect the preflight and POST separately. Bicho's ingest routes support cross-origin ingestion with wildcard origin and Sentry authentication headers. Read/admin CORS is separate. A proxy, CSP, or edge security rule can still block delivery; interactive challenges are unsuitable for SDK routes. Send one clearly labeled exception and await delivery before shutdown: Sentry.captureException(new Error("Bicho setup check"));
const flushed = await Sentry.flush(5000);An event ID or Documentation - Move from Sentry Source - Bicho: Worker routes |
Start with the actual network destination. An old
tunnel, custom transport, or environment override can keep sending to Sentry after the visible DSN changes. Confirm requests reach your Bicho instance's envelope or store route and that any envelope DSN names the receiving origin and project.Check authentication next. All supplied keys must match the route project; secret-enabled backend projects also need the generated transport headers. Never add an admin token to the SDK as a fix.
For browser requests, inspect the preflight and POST separately. Bicho's ingest routes support cross-origin ingestion with wildcard origin and Sentry authentication headers. Read/admin CORS is separate. A pro…