October 4, 20187 yr I recently upgraded to 6.6.0, and now in the process of upgrading drives. Both parity drives were done over the weekend and I honestly don't recall if I upgraded the OS just before or just after the parity replacement (but I think it was before). Either way the parity re-build went fine. Tonight I started a single drive replacement. - Stopped the array - Change a single drive assignment, received the message that everything would be destroyed on the drive, also had the note that starting the array would begin the data rebuild. - started the array What happened is the data re-build started, but the array did not come online. I've never seen this happen before. Please see the attached 3 screenshots. They are all visible right now in that form. the bottom of the first shot shows the Array stopped and the parity rebuild running. I considered re-booting the array and seeing what happens but decided it was safer to just let the rebuild run it's course. Has anyone seen this, is it an issue in 6.6.0? I'll add the "fix common problems" note has to do with me having two SSD's in the array, not having SSD Trim installed, and having the Marvell controller. I've done drive replacements before with all of this and without incident or parity error.
October 4, 20187 yr Community Expert Something got the array in a stale config, rebooting will fix it, not sure if the rebuild will finish correctly like this, it likely will but if you're just starting I would reboot now.
October 4, 20187 yr Author 4 hours ago, johnnie.black said: Something got the array in a stale config, rebooting will fix it, not sure if the rebuild will finish correctly like this, it likely will but if you're just starting I would reboot now. I did restart the server when it had been running about 2 hours. Everything came back up fine and the rebuild started again. So what is a "stale config." what would cause it? After stopping the array to do the initial drive swap, the only oddity I noticed was a slight hesitation to return messages and for anything to happen when I tried to start the array. The steps are pretty straight forward so I really don't see I could have done anything wrong. It's The rebuild is still going and should finish mid afternoon. One more swap and all my 7 year old 2 TB drives will be gone I'm hoping that last rebuild time goes faster.
October 4, 20187 yr Community Expert Just now, TODDLT said: So what is a "stale config." what would cause it? After stopping the array to do the initial drive swap, the only oddity I noticed was a slight hesitation to return messages and for anything to happen when I tried to start the array. The steps are pretty straight forward so I really don't see I could have done anything wrong. Something done to an array disk outside Unraid. i.e., running a filesystem check directly on de device, probably also if Unraid detected some configuration change, like disk changing letter, without the diags impossible to tell, and possibly not even with them.
October 4, 20187 yr Author 6 hours ago, johnnie.black said: Something done to an array disk outside Unraid. i.e., running a filesystem check directly on de device, probably also if Unraid detected some configuration change, like disk changing letter, without the diags impossible to tell, and possibly not even with them. The only thing I can think of with the ability to touch one of those drives would be Unassigned Devices, and I really can't think of how that would occur (but conceivably it could). The drives outside the array are/were not mounted. There is a user script that makes sure they stay spundown and thats really it.
Archived
This topic is now archived and is closed to further replies.