Databricks Error:
TIMEDOUT
What does this error mean?
A Databricks job run was stopped because it exceeded the configured maximum run duration. No error was raised by the code — the run was killed by the timeout policy.
Common causes
- 1A query or transformation is processing more data than expected
- 2Cluster startup took too long (cold start on a new cluster)
- 3An infinite loop or deadlock in notebook code
- 4Network bottleneck slowing down data reads from cloud storage
- 5The job timeout is set too low for the actual workload
How to fix it
- 1Check the run duration history to see if this is a one-off or a growing trend.
- 2Review the task logs to identify which step was running when the timeout occurred.
- 3Optimise the slow query: add partition filters, check for cartesian joins, or reduce the data volume being processed.
- 4Increase the cluster size or use auto-scaling if the workload has genuinely grown.
- 5Increase the job timeout in the job configuration if the current value is too conservative.