Jump to content

Dual parity - 1 failed drive. Is this the correct procedure?

Featured Replies

Posted

Hello.

 

I have a 5 disk array with 2 parity drives. The parity drives are 4gb and the data drives are 4gb, 3gb, and 3gb for a total of 10gb + dual parity.

 

One of the parity drives is failing. I got a notification that there's a lot of read errors. 408 of them. I suppose I have no idea if that's a lot of not or if I can keep using the drive, but I'd just as soon retire it. I have a 4gb replacement drive ready to install and use once I preclear it.

 

Do I use the same procedure as described here in Replacing a Data Drive? I see there's another document that talks about the Parity Swap Procedure but that doesn't seem like the right one. I just wanted to double check before I get started because I'm not sure if a parity drive is considered a "data drive" or if there's some entirely other process I should do.

 

Thanks!

  • Community Expert

The official online documentation on replacing disks is here in the online documentation accessible via the ‘Manual’ link at the bottom of the Unraid GUI.

 

you are correct in that the Parity Swap procedure is not the one to use in your case.

  • Community Expert

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

  • Author
22 hours ago, trurl said:

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

 

Done!

 

Edited by neal_is_king
no need to keep diagnostics posted here

  • Author
1 hour ago, itimpi said:

The official online documentation on replacing disks is here in the online documentation accessible via the ‘Manual’ link at the bottom of the Unraid GUI.

 

you are correct in that the Parity Swap procedure is not the one to use in your case.

 

Thanks itimpi. Am I correct in thinking that the next "Upgrading parity disk(s)" section is pretty close to what I'll be doing? So basically 

 

1. add new disk and preclear
2. stop array 
3. pick new disk from "Parity 2" dropdown
4. start array

 

and then it will rebuild that half of the array and I'm good to go?

 

  • Community Expert

None of your disks are disabled so nothing needs to be replaced or rebuilt.

 

SMART for parity looks OK, you can run an extended SMART test on it if you want.

 

 

  • Author

Hmmm, interesting. I'm running the extended test now, stuck on 10%, stays there for a few hours IIRC. So a positive value for "UDMA CRC error count" is meaningless and the disk is OK?

  • Community Expert
48 minutes ago, neal_is_king said:

 

Thanks itimpi. Am I correct in thinking that the next "Upgrading parity disk(s)" section is pretty close to what I'll be doing? So basically 

 

1. add new disk and preclear
2. stop array 
3. pick new disk from "Parity 2" dropdown
4. start array

 

and then it will rebuild that half of the array and I'm good to go?

 

Just worth pointing out that step 1) is unnecessary (unless you want to stress test the new drive)  as building parity writes to every sector on the disk anyway.

  • Author

I did some more googling and apparently this is most likely a cabling issue, and I am safe to acknowledge the errors for now and do nothing, but if it keeps popping up I may need a new cable.

  • Community Expert
2 minutes ago, neal_is_king said:

So a positive value for "UDMA CRC error count" is meaningless and the disk is OK?

It means you have connection issues of some sort, and although it can be the drive the vast majority of the time it tends to be the cabling to the drive (power or SATA).

  • Author
24 minutes ago, itimpi said:

It means you have connection issues of some sort, and although it can be the drive the vast majority of the time it tends to be the cabling to the drive (power or SATA).

 

This drive is 1 of 4 on a breakaway cable to a LSI 9207-8i controller. If it is a cable issue, is it common for just one of the drives to experience that?

  • Community Expert
8 hours ago, neal_is_king said:

 

This drive is 1 of 4 on a breakaway cable to a LSI 9207-8i controller. If it is a cable issue, is it common for just one of the drives to experience that?

It can certainly happen :( 

i have also experience of a single port on the HBA becoming unreliable whatever 4 way cable is used.

  • Author

Well I am hoping for the best. Thanks a lot itimpi and trurl !!

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...