Snapshot Retention Not Deleting Historical Snapshot

Hey Team Awesome.

I Have a Question…

So i upgraded to Scale on my System from 13.1 up to 24 using ISO install. The Pools are up and working, i had a little Struggle with my Replication Task that didnt want to Replicate from TrueNas Core - Scale. so we instead setup Scale to Pull from Core.

But now i have a Snapshot Problem on the Scale that just does not want to remove old Snapshots Automatically based on Policy, Policy was set as 1 week retention on Source and as such this is now created the New Snapshot with the Retention Policy in place and will remove the Snapshots a week Later, However… The old Snapshot (Dated a Month old) just dont want to Remove automatically, I had attempted to Force the Autoremove with a Periodic Snapshot task on the Pool and recursive down to the last dataset…(Came to my attention that Both task cant run at the same time.) This worked for the first two Dataset. However The Last Dataset and its dataset(Childs) dont want to remove their older Snapshots.

Any Possible Idea where i can go looking further?

Did the snapshot name change between Core & Scale?

We used the Automatic Naming scheme within core, if the upgrade changes the naming scheme that would have been implemented by Scale and not us as Sysadmin

Which doesn’t answer the question.

However IF the naming convention changed in any way then historical CORE snapshots would not be deleted by the Scale snapshot task as they depend absolutely on the naming of the snapshot

Sorry for not answering your question Directly. I provided the best information i have that relates to your question, How else can i answer your statement as i do require assistance in this matter and do not wish to manually delete the snapshots. Bigger headaches from what im reading on the Forums.

All Names correlates with this Scheme
auto-2024-12-02_05-00
Obviously Time adjusted as per Snapshot Taken, but that the Naming Convention used on the System from Oldest auto-2024-09-06_08-00 to the latest (Last) auto-2024-12-10_13-00 Which does not appear to be any different in the naming scheme of Truenas

Manually deleting snapshots isn’t an issue.

You can even use the command line and delete all snapshots according to a later than “date”.

As for your issue - then I have no idea

Thank you @NugentS ill Mention this the team, Was Hoping someone tells me ZFS has a Command for this lol. Other than Manual Removal.

If i cannot Get it on this Community, will i be able to get it at the IXsystems Forum
https://ixsystems.atlassian.net/ or is this for Enterprise Support.?

Again thank you for the Assistance @NugentS

That’s for bug reporting. Not sure this is a bug.

I had a similar issue when upgrading from FreeNAS 11 to TrueNAS CORE 12, in that snapshots created in 11 started accumulating under 12. Naming scheme was exactly the same, with a 2-week lifetime and snapped every 3 hr. It occurred despite reimporting the previous config file after the upgrade.

My interpretation was that it was a glitch in the upgrade process, and once satisfied that the snapshots created under 12 were being correctly auto-deleted after 2 weeks, I just manually deleted the unneeded and outdated 11 ones via the GUI. Did not experience any issues afterwards that might be associated with this.

What could be the anticipated problem with manually deleting the legacy snapshots, using the process within TrueNAS?

Ahhh Thank you for that clearification @NugentS
Truly a Awesome Community

1 Like

The Fear of Manual Removal with the Snapshots is purely caution, i had read that certain users Experienced Problems With their Applications that they run directly from Truenas, after the manual removal of snapshots, they are unable to run there apps and had to re-install… In our case we only use if as a Storage location for our Server Clusters.

Being that it is housing most of my Clients Data extra Caution is used to ensure the integrity of the System as well as the Data set are maintained.

Using a CLI script to do so ensure the process is completed to the standard that is set by Truenas for there ZFS systems, is just the dream at this point from what im reading and seeing.

Are you running applications, and it is those snapshots you want to delete?

No luckily im not, and i had resolved my issue with the most simple and dumbstruck solution.

On Scale there is an option in your replication to either keep as Source, or Custom on Snapshot retention, I just adjusted that value to 1 week and applied the Custom Retention plan and it cleaned everything out as it should with no hassle no fuzz.

Sometimes looking too deep into the Matter can blind you from simple solutions.

OK excellent - glad you got it sorted! :ok_hand:

1 Like

Auto snapshots; mainly those auto taken during an app update, or other update are not deleted. They are maintained by the system so apps can be rolled back in event of an issue. These auto snapshots will build up and have to be manually deleted. This can be done from the GUI, but is a manual operation. Data Protection >> Periodic Snapshot Tasks >> Snapshots button. This will list all snapshots on the system.

1 Like