working pool in a new case is now degraded

Cabby

Cadet
Joined
Sep 15, 2013
Messages
9
Hello,

when the scale beta was released, I created a NAS for testing out of a Node 804 case, older 1tb disk and a cross-flashed HBA from Dell.

This worked fine, although some of the older disk (raidz1) failed over time, which I replaced with newly bought Seagate 1TB disk (CMR).

Now the MB failed, so I had to move everything to a different case, a Define R5 and a different MB.

The import of the pool worked like a charm, but almost immediately the pool was shown as degraded because of read and write failures.

I recreated the pool, but the same happened. The curious thing was, the two new Seagate drives were shown as faulty, not the older ones.

I created another pool with other disks, that works flawlessly, so I the hba and the cables are appear to be fine.

The only difference is mounting in the case, from hanging down in the node to being in the traditional horizontal position in the Define.

I haven't smart checked the drives yet, but I doubt that both failed after such a short usage time.

Any ideas?

TIA Carsten
 

Ericloewe

Server Wrangler
Moderator
Joined
Feb 15, 2014
Messages
20,194
Well, what's the SMART data like for these disks?
 

NugentS

MVP
Joined
Apr 16, 2020
Messages
2,947
The issue is obvious - you have no hardware. Please read forum rules and post your hardware. In particular post the make and models of your disks
 

sretalla

Powered by Neutrality
Moderator
Joined
Jan 1, 2016
Messages
9,700

Cabby

Cadet
Joined
Sep 15, 2013
Messages
9
Thanks for the replies, have been away the whole week.

So, the disks causing the raidz1 pool to be degraded were two Seagate Barracuda ST1000DM010, the other two drives being a ST31000528AS and a ST1000Dl004.

The ST1000DM010 are still CMR, and put in a different PC, the Seatools Smart tests do not return any errors.

I created a different pool with 4x WD Disk on the HBA, and that works fine.

The other disk in the pool were rather old, so maybe there is a incompability there, but then one would expect these drives to throw up errors, not the new ones? And the pool was fine, when it was still in the other case, on the same HBA.

Well a curios edge case maybe?
 
Top