SEO

2.6.23-gentoo-r3
built with non mdadm tools
ie raidtools
has been working for some time
Occasionally in upgrades I might have to create devices or raidstart or
something.
This is a data only non lvm ext3 raid5 of roughly 144g (3x73g)
So I recently and perhaps too casually bumped the kernel up to 23 from 21
worried about some other problem I was having with udev. Yes I'm bleeding
all over, I hate it when that happens.
So at boot the volume would normally be mounted as in fstab. Now it stops in
a shell.
zoomzoom ~ # raidstart /dev/md0
cannot determine md version: no MD device file in /dev.
zoomzoom ~ #
zoomzoom ~ # cd /dev
zoomzoom dev # MAKEDEV md
zoomzoom dev # modprobe md
zoomzoom dev # raidstart /dev/md0
/dev/md0: No such device
zoomzoom dev #

I did see an option in the new 23 kernel not sure if turning it on caused
this or was possibly fallout from an nfs mount left up at reboot.
--- Multiple devices driver support (RAID and LVM) ?
?
? ? <M> RAID support
? ?
? ? <M> Linear (append) mode
? ?
? ? < > RAID-0 (striping) mode
? ?
? ? <M> RAID-1 (mirroring) mode
? ?
? ? < > RAID-10 (mirrored striping) mode (EXPERIMENTAL)
? ?
? ? <M> RAID-4/RAID-5/RAID-6 mode
? ?
? ? [ ] Support adding drives to a raid-5 array
? ?
? ? <M> Multipath I/O support
? ?
? ? <M> Faulty test module for MD
? ?
? ? <M> Device mapper support
? ?
? ?[*] Device mapper debugging support
? ?
? ? <M> Crypt target support
? ?
? ? <M> Snapshot target (EXPERIMENTAL)
? ?
? ? <M> Mirror target (EXPERIMENTAL)
? ?
? ? <M> Zero target (EXPERIMENTAL)

see the option for adding drives after raid5? That had been enabled '*' and
removing it did not change the kernel image.
If anyone has some safe ideas I'd appreciate it.
If you read the help on that add drive option it talks about moving data,
I'd sure hate to think that happened. I think this just showwed in the 23
kernel.

Thanks.
-Walt

Walt Shekrota wrote:

> 2.6.23-gentoo-r3
> built with non mdadm tools
> ie raidtools
> has been working for some time
> Occasionally in upgrades I might have to create devices or raidstart or
> something.
> This is a data only non lvm ext3 raid5 of roughly 144g (3x73g)
> So I recently and perhaps too casually bumped the kernel up to 23 from 21
> worried about some other problem I was having with udev. Yes I'm bleeding
> all over, I hate it when that happens.
> So at boot the volume would normally be mounted as in fstab. Now it stops
> in a shell.
> zoomzoom ~ # raidstart /dev/md0
> cannot determine md version: no MD device file in /dev.
> zoomzoom ~ #
> zoomzoom ~ # cd /dev
> zoomzoom dev # MAKEDEV md
> zoomzoom dev # modprobe md
> zoomzoom dev # raidstart /dev/md0
> /dev/md0: No such device
> zoomzoom dev #
>
> I did see an option in the new 23 kernel not sure if turning it on caused
> this or was possibly fallout from an nfs mount left up at reboot.
> --- Multiple devices driver support (RAID and LVM)
> ? ?
> ? ? <M> RAID support
> ? ?
> ? ? <M> Linear (append) mode
> ? ?
> ? ? < > RAID-0 (striping) mode
> ? ?
> ? ? <M> RAID-1 (mirroring) mode
> ? ?
> ? ? < > RAID-10 (mirrored striping) mode (EXPERIMENTAL)
> ? ?
> ? ? <M> RAID-4/RAID-5/RAID-6 mode
> ? ?
> ? ? [ ] Support adding drives to a raid-5 array
> ? ?
> ? ? <M> Multipath I/O support
> ? ?
> ? ? <M> Faulty test module for MD
> ? ?
> ? ? <M> Device mapper support
> ? ?
> ? ?[*] Device mapper debugging support
> ? ?
> ? ? <M> Crypt target support
> ? ?
> ? ? <M> Snapshot target (EXPERIMENTAL)
> ? ?
> ? ? <M> Mirror target (EXPERIMENTAL)
> ? ?
> ? ? <M> Zero target (EXPERIMENTAL)
>
> see the option for adding drives after raid5? That had been enabled '*'
> and removing it did not change the kernel image.
> If anyone has some safe ideas I'd appreciate it.
> If you read the help on that add drive option it talks about moving data,
> I'd sure hate to think that happened. I think this just showwed in the 23
> kernel.
>


An odd thing is fdisk -l shows the 3 raid5 disks at the end and something
extra
Disk /dev/sdd: 73.4 GB, 73407865856 bytes
255 heads, 63 sectors/track, 8924 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/sdd1 1 8924 71681998+ fd Linux raid
autodetect

Disk /dev/sde: 73.4 GB, 73407865856 bytes
255 heads, 63 sectors/track, 8924 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/sde1 1 8924 71681998+ fd Linux raid
autodetect

Disk /dev/sdf: 73.4 GB, 73407865856 bytes
255 heads, 63 sectors/track, 8924 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/sdf1 1 8924 71681998+ fd Linux raid
autodetect

Disk /dev/dm-0: 1073 MB, 1073741824 bytes
255 heads, 63 sectors/track, 130 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000000

Disk /dev/dm-0 doesn't contain a valid partition table

What is dm-0?
Thanks.
-Walt

Walt Shekrota wrote:

> Walt Shekrota wrote:
>
>> 2.6.23-gentoo-r3
>> built with non mdadm tools
>> ie raidtools
>> has been working for some time
>> Occasionally in upgrades I might have to create devices or raidstart or
>> something.
>> This is a data only non lvm ext3 raid5 of roughly 144g (3x73g)
>> So I recently and perhaps too casually bumped the kernel up to 23 from 21
>> worried about some other problem I was having with udev. Yes I'm bleeding
>> all over, I hate it when that happens.
>> So at boot the volume would normally be mounted as in fstab. Now it stops
>> in a shell.


I cleared this all up by using mdadm and doing an assemble.
I'm thinking the old raidtools are having an issue with devices possibly by
my udev issues.
Not sure. I might be missing some subtle option in the kernel build causing
this cascade of issues.
-Walt