Made a mistake when replacing a disk and I'm at a loss now

[Starkiller]

Running Unraid 6.8.3

 

So yesterday I wanted to replace a 2TB data disk by a new shucked WD 10TB disk.

 

Although this isn't the first time that I replaced a disk with a bigger one, I forgot to unassign the 2TB disk before shutting down and switching the disks. And then I didn't found any better than choosing for a new config.

 

After that, I made sure that my parity disks and data disks were assigned as they were before removing the 2TB disk. Then I started the parity build, with the 10TB disk assigned to the same slot where the previos 2TB disk was.

 

Unraid told me that this disk was unformatted.

 

With still 9 hours of rebuilding left, I went to sleep and found out in the morning that the server was shutdown.

 

I started the server again for a parity build but later today, it was again shutdown.

 

Will I have a chance when I rebuild everything again with the 2TB disk and after everything has been rebuild I replace the 2TB disk with the 10TB disk as explained in the wiki?

 

Or might there be a hardware problem? Never had any problems untill now. This server has been running for almost a decade now.

Edited April 9, 2021 by Starkiller

[trurl]

1 minute ago, Starkiller said:

forgot to unassign the 2TB disk

Does not matter. All you had to do was assign the new disk to that slot and start to begin rebuild. 

 

But, after that everything you did was wrong. 

 

Hang on to that original disk it is your only hope of getting those files.

 

On mobile now will check back later. 

 

In the meantime 

Go to Tools-Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread. 

[Starkiller]

1 hour ago, trurl said:

Does not matter. All you had to do was assign the new disk to that slot and start to begin rebuild. 

 

But, after that everything you did was wrong. 

 

Hang on to that original disk it is your only hope of getting those files.

 

On mobile now will check back later. 

 

In the meantime 

Go to Tools-Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread. 

Problem was that Unraid didn't recognize the new shucked WD 10TB disk. I found out afterwards that this had something to do with the 3.3v pin but at that time it was already too late as I had already opted for a new config.

In attachment the requested diagnostics.

toschestation-diagnostics-20210409-1949.zip

[JorgeB]

You need to do a new config with the old 2TB disk and re-sync parity, but this:

 

1 hour ago, Starkiller said:

I went to sleep and found out in the morning that the server was shutdown

suggests a hardware problem, server should never shutdown own it's own, start by checking if the CPU cooler needs cleaning, an overheating CPU can cause a shutdown, and more likely to happen during a sync since it can be CPU intensive.

[trurl]

54 minutes ago, Starkiller said:

I had already opted for a new config

New Config rebuilds parity so it agrees with the currently assigned disks. Since that original disk wasn't assigned, after the parity rebuild it wouldn't be possible to rebuild that disk to a replacement. You really should have asked for advice.

 

I have seen advice that you need to unassign a disk before replacing it. I have often responded to such advice saying it isn't necessary. Shouldn't matter if you do or you don't, so why do it? In this case, since you thought it was a mistake to not do it, you ran off the rails.

 

The only thing Unraid needs to see is a different disk assigned to that slot and no changes to any other disk assignments, and it will rebuild to that different disk. Doesn't matter if the original disk has been removed or not or unassigned or not.

 

Since you did New Config, as far as Unraid was concerned, all disks were different, so obviously the requirements for rebuilding to the new disk couldn't apply.

[Starkiller]

On 4/9/2021 at 8:29 PM, JorgeB said:

You need to do a new config with the old 2TB disk and re-sync parity, but this:

 

suggests a hardware problem, server should never shutdown own it's own, start by checking if the CPU cooler needs cleaning, an overheating CPU can cause a shutdown, and more likely to happen during a sync since it can be CPU intensive.

Server is  up and running again. Cleaned the CPU cooler and the shutdowns have disappeared since then. 

Thank you @JorgeB and @trurl! Lesson learned for the next time :)