28/08/2026
Most production incidents resolved by a workaround and a cause is not defined. We pay for the unresolved errors every month even the workaround closes the ticket.
In our support and modernisation work it often happens when customer wants fast workaround instead of spending more time and resolve root cause. This is consistent. An error occurs so rarely that restarting the service requires less effort than investigating the cause, so restarting the service is the solution. Two years later, the same error occurs again, either because an audit question has not been passed or because a migration cannot be continued.
Works on resolution of root cause is cumbersome and usually is not easily reproducible. We have spent days on faults that a restart would have hidden by lunchtime. Sometimes the workaround could be correct step.
Our standard way of working is to set a named investigator for a recurring fault and a written finding. The client sees that finding whether it flatters us or not.
