External-Mirrors/awx
2
0
Fork 0
mirror of https://github.com/ansible/awx.git synced 2026-06-24 08:07:47 -02:30
Code Issues Packages Projects Releases Wiki Activity

[AAP-74497] Reset orphaned waiting jobs when controller node is deprovisioned (#16467)

Browse Source 
Reset orphaned waiting jobs when controller node is deprovisioned

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Seth Foster
2026-06-02 10:46:52 -04:00
committed by GitHub
parent b14b9e1771
commit 5cc467d4cf
2 changed files with 46 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
awx/main/scheduler/task_manager.py
View File

@@ -688,6 +688,17 @@ class TaskManager(TaskBase):
logger.error(f'{j.execution_node} is not a registered instance; reaping {j.log_format}')
reap_job(j, 'failed')
# Reset waiting jobs whose controller_node was deprovisioned (e.g. K8s pod replaced).
# These jobs will never be picked up because no live node is listening for them.
registered_control_nodes = Instance.objects.filter(node_type__in=('control', 'hybrid')).values_list('hostname', flat=True)
orphaned_waiting = UnifiedJob.objects.filter(status='waiting').exclude(controller_node__in=registered_control_nodes)
for j in orphaned_waiting:
logger.warning(f'{j.controller_node} is not a registered instance; resetting {j.log_format} to pending')
j.status = 'pending'
j.controller_node = ''
j.execution_node = ''
j.save(update_fields=['status', 'controller_node', 'execution_node'])
def process_tasks(self):
# maintain a list of jobs that went to an early failure state,
# meaning the dispatcher never got these jobs,