ruanwenjun opened a new issue, #10854: URL: https://github.com/apache/dolphinscheduler/issues/10854
### Search before asking - [X] I had searched in the [issues](https://github.com/apache/dolphinscheduler/issues?q=is%3Aissue) and found no similar issues. ### What happened I find when there is workflowInstance running, and the database restart. The workflow instance's status will be incorrect, it will be running all the time. This is due to when we receive the worker's response, we will update the taskInstance's meta, and update the database, but when we update database failed, we will not rollback the meta in memory, so the worker send response again, the master will skip the state, due to it think the state has already been handled. ### What you expected to happen The master can recover, when database restart. ### How to reproduce 1. Running 1000 workflowIntsance 2. Restart database 3. Find some workflowInstance will always be running ### Anything else _No response_ ### Version dev ### Are you willing to submit PR? - [ ] Yes I am willing to submit a PR! ### Code of Conduct - [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
