Koozali.org: home of the SME Server

RAID - Drive replacement

Offline girkers

  • *
  • 296
  • +0/-0
    • gk computer services
RAID - Drive replacement
« on: April 19, 2010, 02:05:50 AM »
I was using SME's inbuilt RAID support with a 3 + 1 (RAID5 with hotspare) configuration. The problem is I have had two drives fail at once and I am now currently without a server until I receive the warranty replacement drives. I was wondering when I put the two (2) replacement drives in, is there anything that I would need to do with the new drives to get the system up and running again?

Thanks

Offline CharlieBrady

  • *
  • 6,918
  • +3/-0
Re: RAID - Drive replacement
« Reply #1 on: April 19, 2010, 04:57:53 AM »
I was using SME's inbuilt RAID support with a 3 + 1 (RAID5 with hotspare) configuration. The problem is I have had two drives fail at once and I am now currently without a server until I receive the warranty replacement drives. I was wondering when I put the two (2) replacement drives in, is there anything that I would need to do with the new drives to get the system up and running again?

That depends on whether the hotspare was one of the drives which failed. With RAID5, if two drives fail at once (which sounds unlikely to me) then you will need to do a fresh install and restore from backup.

Offline girkers

  • *
  • 296
  • +0/-0
    • gk computer services
Re: RAID - Drive replacement
« Reply #2 on: April 19, 2010, 08:33:43 AM »
From what I could gain at the time of testing one had failed in January but I had not noticed for whatever reason so it was just running on the three (3) drives. Then when it finally started having more issues at the start of this month and I turned on SMART, found that two (2) of the drives where faulty.

After I posted my original message I went looking for information on the what happens with RAID5 when it drops a drive and from what I have read it can still work with only two (2) drives, just no redundancy and slower performance.

Has anyone else got anything experience or information on my situation that they care to share?

Thanks :)

Offline Stefano

  • *
  • 10,894
  • +3/-0
Re: RAID - Drive replacement
« Reply #3 on: April 19, 2010, 09:35:50 AM »
From what I could gain at the time of testing one had failed in January but I had not noticed for whatever reason

do you read admin's mail?

Quote
so it was just running on the three (3) drives. Then when it finally started having more issues at the start of this month and I turned on SMART, found that two (2) of the drives where faulty.

After I posted my original message I went looking for information on the what happens with RAID5 when it drops a drive and from what I have read it can still work with only two (2) drives, just no redundancy and slower performance.

Has anyone else got anything experience or information on my situation that they care to share?

Thanks :)

so, you are running on a degraded raid 5 array (2 hd) ?

if so, you can safely insert a new hd and add it via console..

Offline girkers

  • *
  • 296
  • +0/-0
    • gk computer services
Re: RAID - Drive replacement
« Reply #4 on: April 20, 2010, 02:28:28 AM »
do you read admin's mail?
I had gotten an email about the array rebuilding but my ignorant assumption was that there was simply a glitch and it was rebuilding.

so, you are running on a degraded raid 5 array (2 hd) ?

if so, you can safely insert a new hd and add it via console..

I don't have the server on at the moment so I may turn it on today and see if it boots up.

Thanks for the feedback

Offline girkers

  • *
  • 296
  • +0/-0
    • gk computer services
Re: RAID - Drive replacement
« Reply #5 on: April 20, 2010, 08:29:00 AM »
No boot  :( it goes like this:
Code: [Select]
md: md1 stop.
md: bind<sda1>
md: bind<sdb1>
raid1: raid set md1 active with 2 out of 3 mirrors
mdadm: /dev/md1 has been started with 2 drives (out of 3)
md: md2 stopped.
md: bind <sda2>
md: bind <sdb2>
mdadm: /dev/md2 assembled from 1 drive and 1 spare - not enough to start the array.
Making device-mapper control node
Scanning logical volumes
   Reading all physical volumes. This may take a while...
Activating logical volumes
   Volume group "main" not found
ERROR: /bin/lvm exited abnormally! (pid 482)
Creating root device
Mounting root filesystem
mount: error 6 mounting ext3
mount: error 2 mount none
Switching to new root
switchroot: mount failed:22
umount /initrd/dev failed: 2
Kernel panic - not syncing: Attempted to kill init!

So I am thinking that I have to reinstall when I get my drives back and restore from a backup, thoughts?
« Last Edit: April 20, 2010, 08:31:07 AM by girkers »