Description of problem: Trying to transition from 3.6 beta 7 to 1st release candidate. Run engine-setup to pull the new packages and reconfigure. Engine-setup stops the engine then complains about 2 async tasks [ INFO ] Cleaning async tasks and compensations The following system tasks have been found running in the system: The following commands have been found running in the system: Command ID: ccbf659f-0124-4563-ac15-7329c1bc8ad0 Command Type: 3203 Created at: 30 Status: ACTIVE Command ID: fe98cc6d-edf8-487b-852e-30bbcd4b2be4 Command Type: 3201 Created at: 30 Status: ACTIVE The following compensations have been found running in the system: Would you like to try to wait for that? Yes waits forever, No exits with the engine stopped. There is no row in the async_tasks table and vdsClient getTaskStatus show nothing. There are no outstanding tasks in the GUI. I set the zombie task timeout to 30 a couple of days ago and attempted the engine-setup upgrade _many_ times and rebooted the engine after each attempt. I'm confident these tasks are rubbish. How do I get rid of them to perform the upgrade? Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Can you please attach the logs and maybe the db dump as well?
I tried it regularly for a week or so with the same result. After the 2nd release candidate was released, I did a yum update on the engine host and tried again. It was successful. It found the 2 stuck tasks, cleared them and continued with the upgrade. I can see the logs from when it was unsuccessful but the database dump would likely be useless. Still want them?
(In reply to Darryl Bond from comment #2) > Still want them? Probably the Db will be useless now that the upgrade succeeded. Can you please attach the report generated by: engine-log-collector --no-hypervisors collect
Created attachment 1085332 [details] Collected sos file without var log
Created attachment 1085333 [details] sos file - var log Had to split the file up as it was slightly too big. Perhaps the collector script should do that.