After many reboots, I've found the issue:
Dec 27 08:34:56 ann-desktop kernel: [ 3.806330] md: bind<sdf5>
* it runs /scripts/local-premount here, BEFORE sda5 is bound to the
multi-disk array, and thus triggers a drop to busybox shell *
Dec 27 08:34:56 ann-desktop kernel: [ 3.857986] md: bind<sda5>
Dec 27 08:34:56 ann-desktop kernel: [ 3.860152] bio: create slab <bio-1>
at 1
Dec 27 08:34:56 ann-desktop kernel: [ 3.860331] md/raid1:md0: active
with 2 out of 2 mirrors
Dec 27 08:34:56 ann-desktop kernel: [ 3.860431] md0: detected capacity
change from 0 to 151846780928
* at this point, md0 is fully set up, but the script ran beforehand so
thinks it isn't *
So all I need now is to find out how to make local-premount wait for
/dev/sda and /dev/sdf. (My guess is if it detects sdf first then it assumes
that sda has already been detected and thus advances too early. That or it
just takes too long sometimes.)
...
After much searching it seems the safest fix is to add rootdelay=2 to
GRUB_CMDLINE_LINUX_DEFAULT in /etc/defaults/grub and then run sudo
update-grub...
Wish me luck!
Benjie.
On 27 December 2011 07:52, Benjie Gillam <benjie@???> wrote:
> Sorry, I meant RAID1 - the 0 of md0 must have confused me in my slightly
> inebriated state last night :-$
>
> This morning the computer booted first time with the second drive detected
> as /dev/sdf so that obviously isn't the issue...
>
> Number Major Minor RaidDevice State
> 0 8 5 0 active sync /dev/sda5
> 1 8 85 1 active sync /dev/sdf5
>
> So I'm none the wiser. :-/
>
> Benjie.
>
> On 26 December 2011 21:33, Benjie Gillam <benjie@???> wrote:
>
>> Hi all,
>>
>> I installed Ubuntu for my parents many many moons ago. I had upgraded it
>> as far as Hardy and then I left it. Recently they've been complaining that
>> certain websites aren't working (due to flash being too old) and their
>> printers drivers not being fully reliable. "No problem," thinks I, I'll
>> just update to 11.10. I have done so, and Unity issues aside (I've
>> installed GNOME Classic for them now) it's gone well...
>>
>> EXCEPT the computer only boots roughly 1 time in 3 (no obvious pattern).
>> The issue, I've found, is due to their RAID0 /home partition not
>> initialising correctly sometimes (sometimes it's "degraded mode", sometimes
>> no message at all, it just locks up). I've had a good poke around and it
>> seems that it doesn't work when /dev/sdb is detected as /dev/sdf instead. I
>> know this in itself isn't a bug, but it seems to be what is causing the
>> issue. I was expecting it to Just Work (TM) since it does for all the other
>> boxes that I have RAID on. I thought the persistent superblock should stop
>> this issue from happening, but autodetection seems to be failing early
>> during boot.
>>
>> I've added relevant data to the bottom of the email to rule out some
>> common issues/confirm what I've done/point out an obvious issue I may have
>> overlooked...
>>
>> For autodetection I require, according to
>> http://tldp.org/HOWTO/Software-RAID-HOWTO.html#toc7.2
>> 1) Autodetection in the kernel [I have stock Ubuntu 11.10 kernel -
>> perhaps this is missing/missing from the initrd]
>> 2) Persistent-superblock [mdadm claims this to be the case - see below]
>> 3) 0xFD partition types [fdisk claims this to be the case]
>>
>> From /var/log/syslog (or /var/log/messages) we get:
>> Dec 26 09:12:15 ann-desktop kernel: [ 3.251290] md: bind<sda5>
>> Dec 26 09:12:15 ann-desktop kernel: [ 3.290692] md: bind<sdb5>
>> Dec 26 09:12:15 ann-desktop kernel: [ 3.293935] bio: create slab
>> <bio-1> at 1
>> Dec 26 09:12:15 ann-desktop kernel: [ 3.294070] md/raid1:md0: active
>> with 2 out of 2 mirrors
>> Dec 26 09:12:15 ann-desktop kernel: [ 3.294097] md0: detected capacity
>> change from 0 to 151846780928
>> Dec 26 09:12:15 ann-desktop kernel: [ 3.296464] md0: unknown
>> partition table
>>
>> This suggests to me that auto-detection is not occurring, so installing
>> something (kernel modules/settings/app/whatever) and then doing "update-initramfs
>> -u" should presumably fix this issue, but I've not much longer to solve
>> it before I leave to go back home tomorrow. Any ideas would be greatly
>> appreciated.
>>
>> Further details follow.
>>
>> I hope you're all having a great Christmas! Happy New Year!
>>
>> Cheers,
>>
>> Benjie.
>>
>>
>> / is not RAID (why bother, just replace)
>> /home is RAID0 (custom set up via mdadm about 4 years ago)
>> Two equal sized drives.
>> The RAID, once it's running, is absolutely fine. It's just detection that
>> fails.
>>
>>
>> $ sudo fdisk -l /dev/sda
>> Disk /dev/sda: 160.0 GB, 160041885696 bytes
>> 255 heads, 63 sectors/track, 19457 cylinders, total 312581808 sectors
>> Units = sectors of 1 * 512 = 512 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 * 63 16000739 8000338+ 83 Linux
>> /dev/sda2 16000740 312576704 148287982+ 5 Extended
>> /dev/sda5 16000803 312576704 148287951 fd Linux RAID
>> autodetect
>>
>>
>> $ sudo fdisk -l /dev/sdb
>> Disk /dev/sdb: 160.0 GB, 160041885696 bytes
>> 255 heads, 63 sectors/track, 19457 cylinders, total 312581808 sectors
>> Units = sectors of 1 * 512 = 512 bytes
>> Sector size (logical/physical): 512 bytes / 512 bytes
>> I/O size (minimum/optimal): 512 bytes / 512 bytes
>> Disk identifier: 0x000aa89b
>>
>> Device Boot Start End Blocks Id System
>> /dev/sdb1 63 4000184 2000061 82 Linux swap /
>> Solaris
>> /dev/sdb2 16000801 312576704 148287952 5 Extended
>> /dev/sdb5 16000803 312576704 148287951 fd Linux RAID
>> autodetect
>>
>>
>> $ cat /proc/mdstat
>> Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5]
>> [raid4] [raid10]
>> md0 : active raid1 sda5[0] sdb5[1]
>> 148287872 blocks [2/2] [UU]
>>
>> unused devices: <none>
>>
>>
>> $ sudo mdadm --detail /dev/md0
>> /dev/md0:
>> Version : 0.90
>> Creation Time : Fri Nov 16 12:01:45 2007
>> Raid Level : raid1
>> Array Size : 148287872 (141.42 GiB 151.85 GB)
>> Used Dev Size : 148287872 (141.42 GiB 151.85 GB)
>> Raid Devices : 2
>> Total Devices : 2
>> Preferred Minor : 0
>> Persistence : Superblock is persistent
>>
>> Update Time : Mon Dec 26 21:06:17 2011
>> State : clean
>> Active Devices : 2
>> Working Devices : 2
>> Failed Devices : 0
>> Spare Devices : 0
>>
>> UUID : f0a0371d:12376ea7:4c4ad349:bc95e7b8
>> Events : 0.436
>>
>> Number Major Minor RaidDevice State
>> 0 8 5 0 active sync /dev/sda5
>> 1 8 21 1 active sync /dev/sdb5
>>
>
>
--
Please post to: Hampshire@???
Web Interface:
https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL:
http://www.hantslug.org.uk
--------------------------------------------------------------
