Operations & troubleshooting

Diagnose health, migrations, queues, sync, storage, PostgreSQL performance, and deployment incidents without damaging production data.

Health checks and smoke tests

Verify the platform after deployment with application, database, cache, queue, storage, and product-level checks.

Recover when a release fails migrations, health checks, or worker startup.

Diagnose stuck Horizon jobs, excessive retries, report timeouts, and worker memory problems.

Fix upload, export, and generated-file failures caused by missing or unwritable application storage.

Investigate high database CPU, large indexes, long transactions, and disk pressure methodically.

Respond to slow or failed field synchronisation without deleting unsynced data.