Incident Arena

The carrier turned off a legacy SOAP binding and you need to cut traffic to the regenerated client without creating a second outage.

Different worker pools are sending different SOAP auth headers because one cache shard still serves an expired token.

A carrier changed its SOAP contract and your generated client is now out of sync.

A batch shipment job restarted in the middle of a partial run and now risks replaying writes without item-level checkpoints.

A carrier created the label, but your internal save failed before the workflow completed.

A quiet carrier maintenance window changed the SOAP contract and now scheduled jobs are failing before business hours.

Permanent carrier failures have accumulated in the DLQ and now need structured triage instead of blind replay.

The external label exists and is billable, but your internal workflow lost the write before persistence completed.

The same integration path passes in sandbox but fails in production because the environments do not actually share one contract.

Handle a live carrier incident when the one identifier support needs was never logged.