[SATLUG] First question - RAID5 quit working

Don Wright satlug at sbcglobal.net
Sun Aug 16 16:53:49 CDT 2009


On Sat, 15 Aug 2009 17:43:54 -0700 (PDT), David Labens
<adlabens at swbell.net> wrote:

>root at RCH-SERVER:/home/admiral# mdadm --assemble --scan
>mdadm: /dev/md/0 assembled from 2 drives - not enough to start the array.
>			
>I KNOW that two brand new Western Digital HDDs did not go bad simultaneously, and really doubt that either of them went bad.


Actually, having parts from the same batch fail simultaneously is a
rather frequent problem in RAID. It could also be a common event like a
power glitch that damaged components on both drives and was just barely
under the threshold to kill them all.

Unlike everyone else so far, I'm going to tell you to run diagnostics on
all the drives before you destroy the data on the array by forcing it to
build. Start with smartctl from smartmontools. It should be readily
available in Ubuntu. Use the short selftest to get an idea how *all* the
drives are working - it's non-destructive and only takes a couple of
minutes.
http://smartmontools.sourceforge.net/

Then check the two drives that aren't getting along with mdadm using the
long test. This takes a few hours but you can do both at the same time.
If this passes you can be fairly certain the drives will function once
rebuilt into the array.

Now switch to the RAID tools and check the individual drives. I'll cover
that in the next message.  --Don

-- 
A computer is a hole in your desk you pour time into.


More information about the SATLUG mailing list