Wrong Pool State cache - invalid config

[Angora]

Hey unraiders,

 

I already searched the forum, but didnt find any solution that would help me currently.

I have following setup:

Cache: 1 nvme 1TB (to get it somehow to work, i added another nvme to that pool, but that wasnt installed at the start)

Array: 1 ssd 256GB (just to make unraid happy to have a disk there, not really used)

Data: 2 nvme 2TB 

 

I stopped the array to exchange the 256GB ssd, but since that wasnt that easy either, i just wanted to go back to my "normal" and start the array again. When i did that i got the error message you can see as the topic name.

 

I already unassinged the cache drive, rebooted and reassinged it. Didnt help. Does anyone have a good idea, why this is happening and how i could solve that problem?

 

I just found following message under the cache itself:

  pool: cache
 state: DEGRADED
status: One or more devices could not be used because the label is missing or
	invalid.  Sufficient replicas exist for the pool to continue
	functioning in a degraded state.
action: Replace the device using 'zpool replace'.
   see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-4J
  scan: scrub repaired 0B in 00:00:11 with 0 errors on Mon Feb 12 10:21:24 2024
config:

	NAME                     STATE     READ WRITE CKSUM
	cache                    DEGRADED     0     0     0
	  mirror-0               DEGRADED     0     0     0
	    8514342029840584511  FAULTED      0     0     0  was /dev/nvme0n1p1
	    /dev/nvme3n1p1       ONLINE       0     0     0

errors: No known data errors

How can this happen with a restart? And how can i solve this?

 

Thanks.

 

 

homestation-diagnostics-20240406-0958.zip

Edited April 6 by Angora
Add diagnostics. Add another error msg.

[JorgeB]

3 hours ago, Angora said:

How can this happen with a restart?

Unlikely that this was the result of a restart, at some point there was another device assigned to this pool, try this:

 

unassign the cache device

start array

stop array

set cache pool slots to 2

re-assign the nvme device

start array

post new diags.

 

[Angora]

Thanks for your response. 

 

When i tried to unassign the device i couldnt start the array (was grayed out). Only when i set the number of devices for the cache to 0 i could start the array, and so i did. Then i stopped again. 

Then i reacreated the pool with 2 devices, added the original one and startet the array once more.

 

It starts but with "Device is disabled, contents emulated". 

 

Quote

Unlikely that this was the result of a restart, at some point there was another device assigned to this pool, try this:

As i said, i added another 1TB nvme in the meantime to be able to start the array again. But i dont want that to be the normal case.

 

Any new ideas?

 

 

 

------------------------------

I just noticed that i have an historical device shown at the bottom of the page:

dev1 WD_Red_SN700_1000GB_23465W800450

Does this lead to anything?

 

Thanks again!

homestation-diagnostics-20240406-1427.zip

Edited April 6 by Angora
Typos, Historical Device

[JorgeB]

Try this:

 

with the array running type

zfs detach cache 8514342029840584511

then stop the array

unassign the cache pool device

start array

stop array

set cache pool slots to 1

re-assign the nvme device

start array

post new diags.

[Angora]

Quote

unrecognized command 'detach'

Is what i get for the first command. Probably you meant zpool?

 

 

--------------------------

Done with zpool detach [...]. See Diag.

 

It looks green at this time beeing. So it might be solved. But how could i prevent this from happening?

homestation-diagnostics-20240406-1637.zip

Edited April 6 by Angora
Added diag and executed command with zpool

[JorgeB]

18 hours ago, Angora said:

Is what i get for the first command. Probably you meant zpool?

Yep, sorry.

 

18 hours ago, Angora said:

It looks green at this time beeing. So it might be solved. But how could i prevent this from happening?

Looks OK now, difficult to say how it got that way without the diags from when it happen, don't expect it will happen gain, but if it does, save the diags before rebooting.

[Angora]

Alright, thanks for your help. I will have an eye on it and make a diag, if that should happen again.

 

Thank you very much!