Bad drive, phantom drive after replacement

[mguebert]

Hello,

I am running FreeNas 11.1 U7 I have array that had a drive go bad, it was DA2 and was a 3tb seagate. I marked the drive offline. Put a new drive in a different slot and it assigned the new drive da2 again. I clicked on the old drive and clicked replace and selected da2 which was the new drive. The array resilvered, but it still shows degraded and now has a phantom entry.

Zpool Status output

glabel status

Note da12 is the old failed 3tb drive

At this point I am not sure what to do and not lose any data. I have another drive available to me. Would it make any sense to install it and do a replace on 2765334936214370224?

I can't figure why it assigned the new drive DA2 and the old drive DA12, DA2 does show as part of the volume and DA12 does not.

 

[Chris Moore]

At this point I am not sure what to do and not lose any data.

Did you try rebooting the system? Is the defective drive out of the system now?
You should not be at risk of data loss because you are using RAIDz2 and that gives you two drives of redundancy, there is only one drive failed.

Would it make any sense to install it and do a replace on 2765334936214370224?

No. You already did a replace.

I can't figure why it assigned the new drive DA2 and the old drive DA12, DA2 does show as part of the volume and DA12 does not.

Get the defective drive out of the system.

 

[mguebert]

Did you try rebooting the system? Is the defective drive out of the system now?
You should not be at risk of data loss because you are using RAIDz2 and that gives you two drives of redundancy, there is only one drive failed.

Yes I did reboot. With and without the old drive present. I tried again and got the same results, other than DA12 not showing in the view disk tab.

This is the output after removing the bad disk again.

 

[mguebert]

Hello everyone, I could really use some guidance, I would appreciate any that can be given. I have googled and each situation I can find is similar but not the same. That being said, because the resilver completed normally without errors replacing the bad drive, is it likely that data loss would occur if I installed another new disk and did a replace on the phantom disk?

 

[mguebert]

I went ahead and did the replace on the phantom drive. Everything resilvered and the pool showed no errors for two days. Now I have another weird issue. Da9 is showing this error

But when I look at my disk and volume status and zpool status everything looks normal. A reboot clears this for about 6-8 hrs then back again.

I am assuming whatever problem here is a intermittent issue and I am ordering a new drive. Damn 3tb seagates have been nothing but problems.

 

[mguebert]

I am wondering if I am approaching asking for help the wrong way?

Too much info?
Not enough?

 

[Ericloewe]

A reboot clears this

That's what reboots do, the error counts are not persistent.

da9 needs to be replaced. In fact, go through all drives and make sure they're healthy. If that's beyond your current skill level, paste the output of smartctl -x /dev/daX for all disks into a file and attach it here and we'll have a look.

Once we're sure about the state of your disks, we can figure out what, if anything else is needed.

 

[mguebert]

Thanks for your reply, I have smart checks setup for 4 times per month for all drives, 2 long and 2 short tests. I checked all of these after replacing the phantom drive and did not find any alarming errors. The strange thing for me is the flashing critical error in the dashboard, but when I check the disks or run a zpool status it doesn't show any errors with DA9. I will be replacing that disk tomorrow night after I return from a business trip. I just hope I don't lose any more disks, there is still 2 more seagates in there. So far they are the only disks I have had fail. This will be the 4th one, where as the WD and toshiba drives so far have been good.