Jump to content

2 of 4 drives stuck at "mounting" after a reboot (Solved?)


skidz7

Recommended Posts

After a recent power loss my unraid shares failed to come back online.  I got into the web utility and found that 2 of the drives were stuck in a "mounting" state.  (Disk 1 and 2)

 

I've made no recent changes.  I do a parity check every 30 days, so far never found any errors.

 

Unraid version: 4.7 Pro

 

Parity drive:  Western Digital 2GB on Mobo Sata

Disk1:  WD 2GB on Mobo Sata  - Stuck at Mounting

Disk2:  Hitachi 2GB on PCI sata - Stuck at Mounting

Disk3:  Hitachi 2GB on PCI Sata

 

Looked at another thread which recommended doing a reiserfsck -- check

Unmounted samba and then the drives

Started reiserfsck -- check on one drive but after a few minutes appeared to hang so I stopped it. 

I looked at the admin webpage after that and it showed my drives as unformatted.

I then rebooted the box the box everything came back up normally, however it showed 1 parity error.  I'm currently running a parity check which has (so far) found 3 errors (got another 7+ hours to go).

 

All of my shares have come back up and everything looks normal aside from the parity errors.  I'm happy that it's working again but the geek in me wants to know what happened.

Any recommendations, any ideas as to what might have been the issue?

Syslog captured during the mounting problem is attached.

ur_syslog.txt

Link to comment

ReiserFS is a journaled file system. During initial boot the file system was replaying transactions that were not completed because of the sudden power loss.

 

Oct 13 07:23:46 Tower kernel: REISERFS (device md3): replayed 3 transactions in 2 seconds

Oct 13 07:23:46 Tower kernel: REISERFS (device md3): Using r5 hash to sort names

Oct 13 07:23:52 Tower ntpd[1196]: synchronized to 69.164.206.65, stratum 3

Oct 13 07:26:21 Tower login[1252]: ROOT LOGIN  on `tty1'

Oct 13 07:29:32 Tower kernel: REISERFS (device md1): replayed 781 transactions in 348 seconds

Oct 13 07:29:32 Tower kernel: REISERFS (device md1): Using r5 hash to sort names

 

It appears that REISERFS was still replaying transactions for your other two disks. How long did you give it to complete mounting? Here is a link to when I had the same issue and it took several times as long to mount all of my disks.

 

http://lime-technology.com/forum/index.php?topic=15643.msg145883#msg145883

Link to comment

It appears that REISERFS was still replaying transactions for your other two disks. How long did you give it to complete mounting? Here is a link to when I had the same issue and it took several times as long to mount all of my disks.

 

http://lime-technology.com/forum/index.php?topic=15643.msg145883#msg145883

 

Thanks for that info.  I guess I was paying too much attention to the web interface and not enough attention to the log.  I mistakenly assumed that it was stuck or that nothing was happening in the background.  I probably gave it 4-5 minutes as I frantically searched the forums here for possible solutions.

 

Sounds like the thing to do if there's an unclean shutdown is to look at the syslog to see if it's replaying transactions.....and of course check my UPS batteries.

Link to comment

Taking a closer look it seems that only disk2 did not finish replaying transactions. I think when you rebooted the transactions were still there to replay on disk2. When you performed a parity sync it corrected the parity drive - my guess is that you are all good.

I've seen disks take 15 minutes or more to replay their transactions after a non-clean shutdown.  You just needed to be a bit more patient.
Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...