[Bug] [workflow] version 3.3.1 condition task node not run when parent node failed

[sanfeng-lhh]

Search before asking

• I had searched in the issues and found no similar issues.

What happened

I build a pipline like below. when node1 run success, condition node that is named check_status run, and success node run. But when node1 run failed, condition node that is named check_status not run.

workflow_1761270973885.json

What you expected to happen

when node1 run failed, condition node that is named check_status should run and shell node that is named failed should run.

How to reproduce

workflow_1761270973885.json The json file is about the workflow pipline, you can build a pipline like this to reproduce it

Anything else

No response

Version

3.3.1

Are you willing to submit PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[Mrhs121]

I used the workflow you provided to reproduce this problem. Why can't the execution of the condition task be triggered when the upstream task fails(For example, node1 in the workflow you provided failed)? The main reasons are as follows：

When node1 fails, an event as shown below will be sent

You can also see similar logs in the log

[WI-16][TI-0] - 2025-10-24 16:10:45.873 INFO  [ds-workflow-eventbus-worker-2] o.a.d.s.m.e.WorkflowEventBus:[41] - Publish event: WorkflowTopologyLogicalTransitionWithTaskFinishLifecycleEvent{task=node1taskState=FAILURE}

Next, the running status of this event will be analyzed in the tryToTriggerSuccessorsAfterTaskFinish method, due to satisfy the conditions for isEndOfTaskChain, returned directly, so there is no trigger downstream tasks scheduling

[sanfeng-lhh]

I have fixed this issue in our project and I would like to submit this fix to the community.