So as a test I tried to update a stack using the exact same name for the vmdk. Surprisingly, it let me put the request through, but now the task is perpetually stuck in an "Error" or "Failure" state (under the Pending Actions tab). Is there a way to clear/cancel those pending actions?
Little bit of bad news. If it does not remove itself, you are stuck with it for now. No remove tool yet..
Maybe wait to late but i know that in older version the pending operations would be removed after a set period of time (my understanding was that it was 24 hours. not 100% sure though).
Well, one of the pending actions cleared, but this one from 1/30 just seems persistent:
"Create AppStack at "[PXPRDCLVMW33_DS05] cloudvolumes/apps/" - Failed 5 times"
The only option I have is to Retry. I have no idea what AppStack it is that we were trying to create.
It does a check every 4 hours. So it is not even showing the name of the AppStack it is unable to create?
Right - no name on the failed stack.
Little bit of bad news. If it does not remove itself, you are stuck with it for now. No remove tool yet..
I'm guessing it's just in the db somewhere? I may go digging around.
Yes. Backups are your friend if you go digging
We ran into the same issue, using AppVolumes 2.10 now. Any solution one year later?
No, nothing. Development seems to have been focused on making the necessary changes to fit into the unified Astro UI future, to the detriment of every App Volumes 2.x admin with a production deployment. VMware overall seems to be missing the boat when it comes to operationalizing all of these great technologies. Most of my issues these days either have NO solution, or I have to dig into command lines and databases that aren't officially supported, or I just have to wait for Dev to come through in some future release. Why App Volumes had to be the first technology in the Horizon stack to be released for the future model when it SO CLEARLY IS NOT READY is beyond me. So now we have an unusable 3.0 release (for prod environments, and if you can get it to fully work in test, more power to you), and a lapsed 2.x branch that isn't fully compatible with the Horizon 7 release.
I get no sense of urgency or duty to the customer (us) from VMware. Just a few sympathetic shrugs and nothing else.
Hi. I am currently experiencing this issue as well; is there any update on this, or a supported way to purge "stuck" pending actions? I have two that have been stuck for about 7 months and we would like to purge them if there's a way. I have checked and do not see a way to do this in the GUI. We are on AppVol Manager 2.16.
Thank you.
I've found a simple way to resolve this issue without having to go mess with the DB directly.
Simply launch the Ruby console from the manager machine as detailed in VMware Knowledge Base
Once the console loads up, simply run the following command:
Delayed::Job.destroy_all
Wait a couple minutes until the command is finished, and rejoice, your Pending Actions are all cleaned up
In response to MentalWorX ..
Genius! this worked for me.