Hi All,

 

My parity and disk 1 are disabled for some reason.  Parity check ran on 6/1 with zero errors.  Fix common problems says check the main tab.  The SMART shows no errors.  The unraid log shows no errors.  I've tried rebooting it, and it temporarily fixed the issue before going back to being disabled.  I'd love some help on what I should do to test it.  I'm running a Ryzen 2700 with ECC memory.  I know Ryzens can have some issues as I have every once in a while where all disks report errors.  But a reboot solves that.  I'm running a fix common problems extended test.  What else should I do?

If you didn't reboot, your diagnostics could provide information about what happened and help to give you a course of action.

That error happened after the reboot.  Here are the diagnostics.  Fix common problems extended scan seems to freeze.  

server-diagnostics-20210606-1800.zip

5 hours ago, Mlatx said:

I've tried rebooting it, and it temporarily fixed the issue before going back to being disabled. 

Not clear what you mean here. Reboot can't possibly fix disabled disks, even temporarily.

 

Since you did reboot after the disks became disabled, we can't see what happened to disable the disks. SMART for both disks looks OK, and emulated disk1 is mounted, so you can rebuild to the same disks if you want. You should at least check connections before attempting rebuild.

 

https://wiki.unraid.net/Manual/Storage_Management#Rebuilding_a_drive_onto_itself

 

Do you have backups of everything important and irreplaceable?

 

 

Just noticed this:

Jun  6 11:31:20 server emhttpd: unclean shutdown detected

Not sure if that could explain why it "forgot" the disks were disabled. Did you do the reboot from the webUI?

3 hours ago, trurl said:

Just noticed this:

Jun  6 11:31:20 server emhttpd: unclean shutdown detected

Not sure if that could explain why it "forgot" the disks were disabled. Did you do the reboot from the webUI?

Yes, I rebooted from the webUI.  Should I reboot a different way?

9 hours ago, Mlatx said:

Yes, I rebooted from the webUI.  Should I reboot a different way?

That is the correct way but you shouldn't get an unclean shutdown after rebooting from webUI. Next time try stopping the array before shutting down or rebooting just to make sure it is able to cleanly stop the array and something isn't holding things up until timeout forces reboot.

 

Do you need any further advice on rebuilding those disks?

Thanks Constructor.  No I've rebuilt disks before.  Thanks for helping out.  If it happens again, I'll make sure to download diagnostics before restarting.

4 hours ago, Mlatx said:

Thanks Constructor.

Your welcome, Newbie.😉

So here I am again back in the same situation.  This seems to happen every 6-8 weeks.  Yesterday I had errors show up on the same parity disk and data disk.  The parity has already been reconstructed, and the data disk is in progress.  Here is the diagnostics report after the errors showed up.  Is there anything here that is suspicious?

server-diagnostics-20210727-0741.zip

Problems with multiple, maybe even all, disks. Suggests a power or controller issue.

It's not a power issue.  It probably is a controller issue.  The last time I had this prior to unraid 6.9 it happened across all disks.  Here is a post where a member described the issue being common for Ryzen boards, 

I need IOMMU, so what is my other option, a separate controller card?

1 minute ago, Mlatx said:

a separate controller card?

Other then using a different board that's your best bet.

ok thanks

 

52 minutes ago, Mlatx said:

ok thanks

 

Advices to choose a controller : https://forums.unraid.net/topic/102010-recommended-controllers-for-unraid

 

Awesome.  I was going to buy a new motherboard but found an 8 port card with the LSI chip.  Thanks.