Re: Updated MDADM, array dies.
From: Dan Williams <hidden>
Date: 2008-11-26 22:13:53
On Wed, Nov 26, 2008 at 9:40 AM, Greg Cormier [off-list ref] wrote:
I have a RAID5 array of 4 disks. Was working well. I did a "yum
update" and noticed it updated mdadm to mdadm 2.6.7.1, amongst other
things (new kernel 2.2.27 instead of .26) , and now my array won't
mount.
mdadm --assemble /dev/md0 /dev/sdb1 /dev/sdc1 /dev/sdd1 /dev/sde1
mdadm: cannot open device /dev/sdb1: Device or resource busy
mdadm: /dev/sdb1 has no superblock - assembly aborted
[root@zeus ~]# cat /proc/mdstat
Personalities :
md_d0 : inactive sdb1[0](S) sdd1[2](S)
976766976 blocks
unused devices: <none>Looks like you have some partially assembled array there. perhaps the initrd tried to assemble the array before all devices were available...? It shows up in your dmesg at this point:
NVRM: loading NVIDIA UNIX x86 Kernel Module 177.80 Wed Oct 1 14:38:10 PDT 2008 md: bind<sdd1> md: bind<sdb1> NET: Registered protocol family 10
You can clean that up with "mdadm -Ss" and then try to assemble the array again, but I suspect that at the next reboot you may be in the same boat. What are the contents of /etc/mdadm/mdadm.conf?