Error with cpu-user.rrd

[Nordlicht-13]

I'm running TrueNAS 13.1 and got this one error, that I can't remove
with a scrub of the pool.

errors: Permanent errors have been detected in the following files:

diskpool-1/.system/rrd-639741fdc4834a14a00ba13ff2decee1:/localhost/aggregation-cpu-average/cpu-user.rrd

Is there something that can be done to get rid of this error?
Tried to set System Dataset Pool from diskpool-1 to boot-pool.

Thank for any help.

 

[Nordlicht-13]

If anybody has the the same problem, here is what fixed it on my TrueNAS.
Just start a scrub, let it run for a couple of minutes and then stop it with

Code:

zpool scrub -s zstorage

That worked for me clearing permanent errors for files.

 

[sretalla]

I don't think that repaired the file, it just prevented the system from scrubbing the area of the disks that file resides on, so the list at the "end" of the scrub is now empty.

You could delete that file and reboot to properly clear it, but also look into why you have pool corruption in general.

 

[Nordlicht-13]

Right, the error is still there, but the file is not there.
There is no directory .system under /mnt/diskbool-1/

 

[sretalla]

There is no directory .system under /mnt/diskbool-1/

That's not where it's mounted.

Look at the output from zfs mount

Then look for it in the right place.

 

[Nordlicht-13]

Thanks
mount is to /var/db/system/
I stopped syslog and set the system-data-set to boot-pool
now

Code:

zpool status -xv

give out

Code:

errors: Permanent errors have been detected in the following files:

        <0x2c2>:<0x54>

 

[sretalla]

Well that's metadata...

As I said earlier in this thread, you need to look into why you're getting this corruption. Maybe sharing hardware details might help you to get advice on what might be the cause.

Metadata corruption can only really be corrected with a pool rebuild... Backup, destroy pool, build pool again (correcting hardware issues in the process), restore.

 

[Nordlicht-13]

Okay, I now have a second pool in my NAS.
Is the right way to replicate the pool-1 to pool-2 and than rename
the pool-2 to pool-1 to have everything work again as before?

 

[sretalla]

Is the right way to replicate the pool-1 to pool-2 and than rename
the pool-2 to pool-1 to have everything work again as before?

Stop VMs, jails and services, replicate, rename and start your things again, all should be fine.

 

[Nordlicht-13]

Thanks a lot for the help. I appreciate the quick reply from you.