Hi,
Basically what the title says. We have periodic snapshots (daily) for a number of datasets. Snapshot task on the WebUI always indicates either RUNNING or FINISHED but looking into it, we find no snapshots created. The last run column sometimes correctly mentions the time since last snapshot, other time it gives incorrect information.
Last Run column correctly showing last run as 1 month ago:
Last Run column showing incorrect value:
I suspect my issue may be the same as this topic: Time Machine no longer taking snaphots after Fangtooth upgrade
Running TrueNAS community 25.04.1. For the updates available, I don’t see this fixed in any of the release notes. Please share if this is known issue and any updates on a fix?
Thanks,
Just wondering, how big are the datasets as you are backing up every hour. Can it that it never completes?
My other snapshots work fine, only the time machine never gets created. But they are system managed and not manually entered tasks.
the datasets are significant in size (range from 350GBs to 350TBs) but they aren’t being used heavily for quite some time so max snapshot sizes would be 50GBs on seldom cases. I’m not sure what you mean by system managed and not manually entered. Possibly because I’m not familiar with time machine snapshots. As for my snapshots, I have periodic tasks configured that (are supposed to) take hourly or daily snapshots depending on the dataset.
By system managed I mean that when you create a dataset share with Time Machine profile, it automatically sets up snapshot tasks (but this is broken as of Fangtooth).
All other snapshots are manual tasks created by the user.
What if you set the time schedule not to every hour but longer or disable a few? You run probably more tasks and all at the same time on each hour and on your whole pool?
When it says “running” you can click on that I think and see some info?
I also have staggered snapshop tasks set up on various datasets and these work fine
(staggered: hourly to be removed daily/weekly; daily to keep 3 months, monthly to keep for x years)
ah I see. yes, I’ve manually setup snapshots tasks on truenas.
I tried looking for any zfs snapshot processes via command-line but didn’t find any. also, tried using midclt command but got a Failed connection handshake error.
Also, clicking on “RUNNING” snapshot tasks doesn’t output anything and while I see some tasks running in the “Running Jobs” section at the top right in the menu bar, none of the tasks are snapshots tasks.
That being said, I also have a replication task among the tasks in “Running Jobs” section in main menu that has been running since 25th of August. This is the time when snapshots stopped being created to this appears to be part of the same problem.
I want to reboot the server to see if it makes a difference but cannot as it’s production.
If you have a replication task running at the same time, I guess a conflict might arise with snapshots, especially with multiple ones. (I am not a developer though).
Most tasks can be cancelled when you click on them, or else try to turn them off in the Data Protection section if you can’t boot the server.
I would turn off replication and then see if the snapshots work and then turn them on one-by-one or change schedules.
I also tried midctl commands as requested for the bug report of my Time Machine snapshots, but never got them work (due to permissions).
@EtienneB thanks very much for your replies. Unfortunately, nothing worked and I couldn’t figure out the issue. Had to reboot the server in the end which seems to have resumed the snapshots and replications.
1 Like
