The command "acrocmd list tasks" shows "not_run_yet"
We use acrocmd to check when the last backup was run, so we get alerts when Acronis decides to either not run, or error.
However I am seeing an increasing case where an Acronis activity (show in the Acronis logs as "deleting backup plan") keeps running between the time the backup completes (about 19:00 till 23:00) and our checking script (which runs at 6am)
This seems to either wipe out the last run time of the plan (but it still shows it in the logs) or deletes the plan and redeploys it, causing it to show as never having run.
It is rather bad for us as it means I'm getting errors on the Acronis management console saying backups have never completed (when they have) and our script alerting us to failed backups.
Is this expected behaviour ?
Is there a away of disabling it ?
when was this change brought in ? (so I can downgrade all of our servers to the release before)
We are now running 11.5.39029, but have tried a few version up to this.
Many thanks
Update, one of the servers that has this, has passed the check today.
In the logs it doesn't show the deleting backup plan and redeploying it, as the last attached logs show.
I will keep checking in case this is a one off.
Hello Thomas,
The "list tasks" command doesn't always work as you would expect (showing just the regular sub-tasks and their status) and doesn't fit very well as a monitoring tool.
For this purpose, I would suggest using "list plans" instead.
You might also want to check out a reporting script we published in our KB.
We use GFI agent, so I will pass this information on to them. See which option they use to check, though I suspect they use "list tasks".
Though this doesn't explain why Acronis management console also reports back as never competed, but when connecting to the local machine it reports correctly.
many thanks for the response.
GFI have confirmed they use the call "list tasks"
attached is last nights log, which shows the plan being deleted and redeployed, and reason for this action ?
The screen shot shows the result of the acrocmd "list tasks" and "list plans"
Many thanks
Also the screen shot shows the list plans has no start/finish date.....
Just testing the "list plans" command and I can't see how it can tell me if the backup ran correct, unless I'm missing a switch
# acrocmd list plans
On the boxes that have this issue above it show the name of the backup task, whether it's running or idle the status ok/warning/error and the last start date. However on the boxes that randomly run this redeploy the list shows status "ok" and blank for the last start time......
on a normal working box this shows status "ok" and has the last start date filled out.
see screen shots.
I have attached log from the server that redeployed, with a screen shot of the output for "list plans"
and a screen shot of "list plans" from a server that didn't redeploy
I'm back to the same point before, even the management console is saying that the backups have not run. GFI are looking into poling the "list plans" command but this won't fix my issue of Acronis running a delete on my plans and the redeploying them.
Hello Thomas,
Thank you for the clarifications.
This is a rather strange situation because centralized plans are normally revoked (automatically deleted) only in certain specific cases. For example, if the backup task is edited and an agent is removed from the list of "what to backup", the task will be automatically revoked from the agent as well.
Another case where this happens is when the agent is deleted from the management server and re-added.
Finally, for Virtual Appliances, plans can be redeployed if virtual machines are moved between ESX hosts and the agent bindings are changed.
Is there any possibility of anything like this (through scripts or otherwise) happening on the affected systems?
Thank you.
I have checked on the Management server and on the server and there are no tasks like this running.
Is there any debugging/monitoring I can run to catch what is causing this ?
You can try enabling debug logs by following the instructions in this article:
https://kb.acronis.com/content/43370
Please don't forget to turn off the debug logging after catching the problem, this will cause problems if left running.
Perfect, I shall enable this Monday as I don't want to full up the disk with logs over the weekend.
Thanks
I have enabled this on the server in question and on the PC that the management server is install on.
I will post this up this tomorrow.
I have attached the log from the server with the agent, but I can't create the zip file from the management server. It runs for 20 minutes and then errors with "Not enough memory"
I have the log output from the management server, if that's any help (with the debug option selected)
Hello Thomas,
The logs show some possible problems with the internal service database, but this is not a very likely cause for the behavior.
My suggestion would be to open a support ticket for a full investigation on why this happens.
If the agent doesn't have any local plans or vaults, one possible solution you could try would be to reset the agent database I mentioned. But I want to stress again that this is just resolving the internal error I mentioned above, not a guaranteed solution to your problem.
The way to do this would be:
- Remove the agent from the AMS server;
- Connect directly to this agent and check to make sure that all the plans, vault etc have been removed;
- Stop the agent MMS service;
- Rename the C:\ProgramData\Acronis\BackupAndRecovery\MMSData folder
- Start the service;
- Add the machine to the AMS server.
I can give that a go in the next few days. When I removed the agent I did delete the whole Acronis folder from ProgramData.
At the moment I've deployed a plan from the local machine to it's self to see if that also experiences the same issue.
many thanks for the help.