| View previous topic :: View next topic |
| Author |
Message |
drdreff
n00b
Joined: 02 Dec 2006
Posts: 17
|
 Posted: Tue Feb 08, 2011 5:24 am Post subject: fsck.xfs /dev/md0 does not exist |
 |
|
First time I've been stumped an not found an answer in the archives...
This just started happening after a long upgrade. My system fails to finish booting ofter fsck fails on my root partition (3 disk raid5). When I give the root password, the root is fully mounted and intact.
I have added TARBALL=yes to rc.conf and re-emerged a dozen packages. I can get up with a liveCD so I can pull logs, but I don't know where to start with this one.
Thanks in advance for the help. |
|
| Back to top |
|
 |
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54821
Location: 56N 3W
|
| Posted: Tue Feb 08, 2011 3:41 pm Post subject: |
|
|
drdreff,
Describe your raid5 setup. Drive and partition names that are donated to the raid.
Post | Code: | | mdadm /dev/<partition> -E |
for one of the raid drives may be useful.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
drdreff
n00b
Joined: 02 Dec 2006
Posts: 17
|
| Posted: Wed Feb 09, 2011 2:59 am Post subject: |
|
|
Well now that is interesting...
| Code: |
livecd / # mdadm /dev/md0 -E
mdadm: No md superblock detected on /dev/md0.
livecd / # cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid5 sda2[0] sdc2[2] sdb2[1]
974566912 blocks level 5, 64k chunk, algorithm 2 [3/3] [UUU]
unused devices: <none>
livecd / # fdisk -l
Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
Device Boot Start End Blocks Id System
/dev/sda1 * 1 13 104391 83 Linux
/dev/sda2 14 60677 487283580 fd Linux raid autodetect
/dev/sda3 60678 60801 996030 82 Linux swap / Solaris
Disk /dev/sdc: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
Device Boot Start End Blocks Id System
/dev/sdc1 * 1 13 104391 83 Linux
/dev/sdc2 14 60677 487283580 fd Linux raid autodetect
/dev/sdc3 60678 60801 996030 82 Linux swap / Solaris
Disk /dev/sdb: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
Device Boot Start End Blocks Id System
/dev/sdb1 * 1 13 104391 83 Linux
/dev/sdb2 14 60677 487283580 fd Linux raid autodetect
/dev/sdb3 60678 60801 996030 82 Linux swap / Solaris
Disk /dev/md0: 998.0 GB, 997956517888 bytes
2 heads, 4 sectors/track, 243641728 cylinders
Units = cylinders of 8 * 512 = 4096 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 65536 bytes / 131072 bytes
Disk identifier: 0x00000000
Disk /dev/md0 doesn't contain a valid partition table
livecd / #
|
Thanks for taking a look... |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54821
Location: 56N 3W
|
| Posted: Wed Feb 09, 2011 11:58 am Post subject: |
|
|
drdreff,
| Code: | | # mdadm /dev/md0 -E |
is not correct. The raid superblock is on the underlying partition.
Try | Code: | | # mdadm /dev/sda2 -E |
for example.
| Code: | Device Boot Start End Blocks Id System
/dev/sda1 * 1 13 104391 83 Linux
/dev/sda2 14 60677 487283580 fd Linux raid autodetect
/dev/sda3 60678 60801 996030 82 Linux swap / Solaris |
is interesting
If /dev/sda1 or /dev/sda3 are part of raid sets, they will not be auto assembled by the kernel as the partition types are not fd.
Thats not wrong, it means you have to use mdadm in an initrd to assemble any raid set you want to use for root.
When your system is running, what do you have in /proc/mdstat ?
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
drdreff
n00b
Joined: 02 Dec 2006
Posts: 17
|
| Posted: Wed Feb 09, 2011 4:04 pm Post subject: |
|
|
Here is a more correct mdadm -E
| Code: |
livecd / # mdadm /dev/sda2 -E
/dev/sda2:
Magic : a92b4efc
Version : 0.90.00
UUID : 8c539f22:a2e6e70f:02bd0424:6f23f4a7
Creation Time : Sat Nov 15 15:11:47 2008
Raid Level : raid5
Used Dev Size : 487283456 (464.71 GiB 498.98 GB)
Array Size : 974566912 (929.42 GiB 997.96 GB)
Raid Devices : 3
Total Devices : 3
Preferred Minor : 0
Update Time : Tue Feb 8 18:57:54 2011
State : clean
Active Devices : 3
Working Devices : 3
Failed Devices : 0
Spare Devices : 0
Checksum : fe89b7ab - correct
Events : 6461940
Layout : left-symmetric
Chunk Size : 64K
Number Major Minor RaidDevice State
this 0 8 2 0 active sync /dev/sda2
0 0 8 2 0 active sync /dev/sda2
1 1 8 18 1 active sync /dev/sdb2
2 2 8 34 2 active sync /dev/sdc2
|
And /proc/mdstat which was above as well:
| Code: |
livecd / # cat /proc/mdstat
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid5 sda2[0] sdc2[2] sdb2[1]
974566912 blocks level 5, 64k chunk, algorithm 2 [3/3] [UUU]
unused devices: <none>
|
The bootable partitions are not part of the raid group. /dev/sda1 is my primary boot, which I periodically mirror to one of the other 3 drives. |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54821
Location: 56N 3W
|
| Posted: Wed Feb 09, 2011 5:09 pm Post subject: |
|
|
drdreff,
| Code: | Version : 0.90.00
...
Preferred Minor : 0 |
shows that this raid set should get /dev/md0, and it supports auto assembly so everything is in order so far.
Try the following on your kernel
| Code: | $ grep AUTODE /usr/src/linux/.config
CONFIG_MD_AUTODETECT=y |
Once upon a time, raid sets were always auto detected. Now you need a kernel option.
Grub can load from a raid1 mirror set as long as its made with a version 0.9 superblock.
It does this by ignoring that its a mirror.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
drdreff
n00b
Joined: 02 Dec 2006
Posts: 17
|
| Posted: Thu Feb 10, 2011 8:24 am Post subject: |
|
|
My kernel already has that set. The RAID is getting reassembled, but fsck can't find the /dev/md0 device to scan. That is the odd part. When I drop to the root shell, the raid is fully assembled and everything is intact, mounted as /.
I thought it was udev, but adding
| Code: |
rc-update add mdadm boot
|
did not help, as was proposed in another place. |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54821
Location: 56N 3W
|
| Posted: Thu Feb 10, 2011 4:45 pm Post subject: |
|
|
drdreff,
Adding mdadm to the boot runlevel can't help as md0 must be assembled before root can be mounted to read any init scripts.
emerge wgetpaste if you don't have it and put your kernel .config on a pastebin as it won't fit in a post.
Tell me the URL.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
drdreff
n00b
Joined: 02 Dec 2006
Posts: 17
|
| Posted: Thu Feb 10, 2011 9:55 pm Post subject: |
|
|
| NeddySeagoon wrote: |
Adding mdadm to the boot runlevel can't help as md0 must be assembled before root can be mounted to read any init scripts.
|
Right, The problem is that my system seems to be assembling the raid and executing init scripts before it fails.
| NeddySeagoon wrote: |
emerge wgetpaste if you don't have it and put your kernel .config on a pastebin as it won't fit in a post.
Tell me the URL. |
I will get that tonight. |
|
| Back to top |
|
|
|
Kernel & Hardware
