| View previous topic :: View next topic |
| Author |
Message |
bendy
Apprentice
Joined: 18 May 2002
Posts: 274
Location: Gloucester, UK
|
 Posted: Thu Mar 12, 2009 7:21 pm Post subject: shutdown problem with / on raid device |
 |
|
Hi,
I have gentoo installed with /boot on a regular disk partition and everything else on raid partitions. My /etc/fstab:
| Code: | /dev/sda2 /boot ext2 noauto,noatime 1 2
/dev/md1 / ext3 noatime 0 1
/dev/sda3 none swap sw 0 0
/dev/md2 /home ext3 noatime 0 0
/dev/md3 /var reiserfs noatime,notail 0 0
tmpfs /tmp tmpfs defaults,nosuid,size=1024M,mode=1777 0 0
/dev/md5 /work xfs noatime 0 0
/dev/cdroms/cdrom0 /mnt/cdrom iso9660 noauto,ro 0 0
/dev/fd0 /mnt/floppy auto noauto 0 0 |
and my /etc/mdadm.conf:
| Code: | ARRAY /dev/md1 level=raid0 num-devices=2 UUID=bc3bbaf3:6628c7ba:2b02e604:8457fd23
ARRAY /dev/md2 level=raid0 num-devices=2 UUID=b1cea3ab:38398ed5:e0899131:e0e10697
ARRAY /dev/md3 level=raid0 num-devices=2 UUID=54c94f7c:ef16bbe1:19e5e8a0:2413d932
ARRAY /dev/md4 level=raid0 num-devices=2 UUID=d6dbfe2b:fb028dde:2238324a:d30b674d
ARRAY /dev/md5 level=raid0 num-devices=2 UUID=a9989cd6:2d88afe3:fad10696:9af35823 |
This has been working fine for quite some time. However I recently decided to run my system ~arch as I had so many packages added to package.keywords file. This included an upgrade to gentoo-sources-2.6.28-r2, which I compiled with raid auto-detection.
The problem is that my system will no longer shut-down cleanly. I get:
| Code: | * mdraid uses addon code which is deprecated
* and may not be available in the future.
* shutting down RAID devices (mdadm)...
mdadm : stopped /dev/md2
mdadm : stopped /dev/md3
mdadm : stopped /dev/md4
mdadm : stopped /dev/md5
mdadm: fail to stop array /dev/md1: Device or resource busy
Perhaps a running process, mounted filesystem or active volume group?
* ERROR: mdraid failed to stop
.
.
.
* Remounting remaining filesystems read-only...
* Remounting / read only...
INIT: no more processes left in this runlevel. |
Then it freezes, although I can force it with alt-sysreq-u and alt-sysreq-b.
Anyone got any ideas what changed recently to stop my machine shutting down properly?
Thanks! |
|
| Back to top |
|
 |
richard.scott
Veteran
Joined: 19 May 2003
Posts: 1497
Location: Oxfordshire, UK
|
| Posted: Fri Apr 17, 2009 1:40 pm Post subject: |
|
|
Hi,
This is just a warning that / can't be unmounted as its currently in use... I get this too, its nothing to worry about.
To get your system to power off you need to compile your APM or ACPI options as static rather than modules.
Hope this helps,
Rich |
|
| Back to top |
|
|
JanSteen
n00b
Joined: 13 Feb 2005
Posts: 11
Location: Amsterdam/NL
|
| Posted: Sat Apr 18, 2009 6:50 am Post subject: shutdown problem with / on raid device |
|
|
Hi,
Same problem here, started occurring very recently, since my last update I presume. Before, already for some months, I seem to remember that warning you also get,
| Code: | * mdraid uses addon code which is deprecated
* and may not be available in the future. |
It could be that that warning started appearing since I switched to OpenRC? Not sure.
Anyhow, lately my machine also refuses to power off, even though I do have (and already had) all ACPI options compiled into the kernel instead of as modules.
I'll post a reply if I find a solution; anybody else with the same trouble?
CS |
|
| Back to top |
|
|
snailhead
n00b
Joined: 13 Nov 2006
Posts: 5
|
| Posted: Sun May 10, 2009 4:25 am Post subject: Re: shutdown problem with / on raid device |
|
|
| bendy wrote: | Hi,
I have gentoo installed with /boot on a regular disk partition and everything else on raid partitions. My /etc/fstab:
| Code: | /dev/sda2 /boot ext2 noauto,noatime 1 2
/dev/md1 / ext3 noatime 0 1
/dev/sda3 none swap sw 0 0
/dev/md2 /home ext3 noatime 0 0
/dev/md3 /var reiserfs noatime,notail 0 0
tmpfs /tmp tmpfs defaults,nosuid,size=1024M,mode=1777 0 0
/dev/md5 /work xfs noatime 0 0
/dev/cdroms/cdrom0 /mnt/cdrom iso9660 noauto,ro 0 0
/dev/fd0 /mnt/floppy auto noauto 0 0 |
and my /etc/mdadm.conf:
| Code: | ARRAY /dev/md1 level=raid0 num-devices=2 UUID=bc3bbaf3:6628c7ba:2b02e604:8457fd23
ARRAY /dev/md2 level=raid0 num-devices=2 UUID=b1cea3ab:38398ed5:e0899131:e0e10697
ARRAY /dev/md3 level=raid0 num-devices=2 UUID=54c94f7c:ef16bbe1:19e5e8a0:2413d932
ARRAY /dev/md4 level=raid0 num-devices=2 UUID=d6dbfe2b:fb028dde:2238324a:d30b674d
ARRAY /dev/md5 level=raid0 num-devices=2 UUID=a9989cd6:2d88afe3:fad10696:9af35823 |
This has been working fine for quite some time. However I recently decided to run my system ~arch as I had so many packages added to package.keywords file. This included an upgrade to gentoo-sources-2.6.28-r2, which I compiled with raid auto-detection.
The problem is that my system will no longer shut-down cleanly. I get:
| Code: | * mdraid uses addon code which is deprecated
* and may not be available in the future.
* shutting down RAID devices (mdadm)...
mdadm : stopped /dev/md2
mdadm : stopped /dev/md3
mdadm : stopped /dev/md4
mdadm : stopped /dev/md5
mdadm: fail to stop array /dev/md1: Device or resource busy
Perhaps a running process, mounted filesystem or active volume group?
* ERROR: mdraid failed to stop
.
.
.
* Remounting remaining filesystems read-only...
* Remounting / read only...
INIT: no more processes left in this runlevel. |
Then it freezes, although I can force it with alt-sysreq-u and alt-sysreq-b.
Anyone got any ideas what changed recently to stop my machine shutting down properly?
Thanks! |
If I were guess, the script /lib/rcscripts/addons/raid-stop.sh changed, with the key line using mdadm -Ss (scan the config file and shutdown the listed raid devices). I would presume that using mdadm is the "new school"
| Code: | [ -f /proc/mdstat ] || exit 0
# Stop software raid with mdadm (new school)
mdadm_conf="/etc/mdadm/mdadm.conf"
[ -e /etc/mdadm.conf ] && mdadm_conf="/etc/mdadm.conf"
if [ -x /sbin/mdadm -a -f "${mdadm_conf}" ] ; then
ebegin "Shutting down RAID devices (mdadm)"
output=$(mdadm -Ss 2>&1)
ret=$?
[ ${ret} -ne 0 ] && echo "${output}"
eend ${ret}
fi
|
This is a pretty simple script called by halt.sh. I propose the following script instead, but it isn't simpler 
| Code: | [ -f /proc/mdstat ] || exit 0
# Stop software raid with mdadm, but check for md on root
mdadm_conf="/etc/mdadm/mdadm.conf"
[ -e /etc/mdadm.conf ] && mdadm_conf="/etc/mdadm.conf"
if [ -x /sbin/mdadm -a -f "${mdadm_conf}" ] ; then
ebegin "Shutting down RAID devices (mdadm)"
for i in `ls -t /dev/md` ; do
mounted=`mount | egrep "md${i} on / "`
if [ -n "${mounted}" ] ; then
echo /dev/md${i} is mounted on root and should *not* be stopped
ret=0
fi
if [ -z "${mounted}" ] ; then
active=`cat /proc/mdstat | egrep active | egrep md${i}`
ret=0
if [ -n "${active}" ] ; then
output=$(mdadm -S /dev/md${i})
ret=$?
fi
fi
[ ${ret} -ne 0 ] && echo "${output}" && return=$?
done
eend ${return}
fi |
|
|
| Back to top |
|
|
richard.scott
Veteran
Joined: 19 May 2003
Posts: 1497
Location: Oxfordshire, UK
|
| Posted: Sun May 10, 2009 9:45 am Post subject: |
|
|
Hi snailhead,
That new script would make more sense.
I don't list any md devices in my /etc/mdadm.conf file... come to think of it, I've never edited that file on any raid system I manage.
Perhaps you could submit a bug request for the relevant package and see if we can get this changed?
Rich. |
|
| Back to top |
|
|
ClientAlive
n00b
Joined: 07 Jul 2011
Posts: 33
|
| Posted: Mon Nov 28, 2011 11:31 pm Post subject: |
|
|
| Me too but my system is a little more complex in how it's set up. I'm having difficulty finding related info. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
