Description of problem: Currently some tasks like paunch uses async mechanism in ansible. These tasks have the first task to run a long running job asynchronously and the second task to run periodically check status of the first task. Currently the second polling task is retried every 3 seconds but this interval is not guaranteed and it can be delayed for example there are multiple nodes, and more frequent interval causes more additional overhead. This makes the timeout detection happen very late. In the real field we saw that the paunch task timed out in a controller node after 1 hour but ansible detected that after 2 hours. Version-Release number of selected component (if applicable): How reproducible: Sometimes, when timeout is detected Steps to Reproduce: 1. 2. 3. Actual results: Timeout is detected a while after the task times out Expected results: Timeout is detected immediately after the task times out Additional info:
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Release of components for Red Hat OpenStack Platform 16.2.4), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2022:8794