* stacked raid devices not autodetected?
@ 2004-12-02 22:32 James Ralston
2004-12-03 15:34 ` Luca Berra
0 siblings, 1 reply; 7+ messages in thread
From: James Ralston @ 2004-12-02 22:32 UTC (permalink / raw)
To: linux-raid
I have several raid devices that I created with the following
commands:
mdadm --create /dev/md0 --level=mirror --raid-devices=2 /dev/sd{b,h}1
mdadm --create /dev/md1 --level=mirror --raid-devices=2 /dev/sd{c,i}1
mdadm --create /dev/md2 --level=mirror --raid-devices=2 /dev/sd{d,j}1
mdadm --create /dev/md3 --level=mirror --raid-devices=2 /dev/sd{e,k}1
mdadm --create /dev/md4 --level=mirror --raid-devices=2 /dev/sd{f,l}1
mdadm --create /dev/md5 --chunk=256 --level=stripe --raid-devices=5 /dev/md{0,1,2,3,4}
On boot, the kernel automatically detects and starts md0 through md4.
However, md5 is *not* autodetected. Using the RAID_AUTORUN ioctl also
has no effect. The only way I've found to reliably start md5 is to
have rc.sysinit run "mdadm -assemble --scan".
Is this a bug? A design limitation? Something else?
(I am currently using kernel-2.6.9-1.675_EL on the RHEL 4AS beta. But
I've never been able to get autodetection of stacked raid devices to
work using any kernel.)
--
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: stacked raid devices not autodetected?
2004-12-02 22:32 stacked raid devices not autodetected? James Ralston
@ 2004-12-03 15:34 ` Luca Berra
[not found] ` <36652.212.158.231.74.1102208997.squirrel@mail.fsck.co.uk>
[not found] ` <20041206113752.GM3858@marowsky-bree.de>
0 siblings, 2 replies; 7+ messages in thread
From: Luca Berra @ 2004-12-03 15:34 UTC (permalink / raw)
To: linux-raid
> On boot, the kernel automatically detects and starts md0 through md4.
> However, md5 is *not* autodetected. Using the RAID_AUTORUN ioctl also
> has no effect. The only way I've found to reliably start md5 is to
> have rc.sysinit run "mdadm -assemble --scan".
>
> Is this a bug? A design limitation? Something else?
this is one of the limitations of the in-kernel autodetection
> (I am currently using kernel-2.6.9-1.675_EL on the RHEL 4AS beta. But
> I've never been able to get autodetection of stacked raid devices to
> work using any kernel.)
if you are not booting from the stacked raid device you can safely have
mdadm called from rc.sysinit, if you are you better use mdassemble in
initrd.
Regards,
L.
--
Luca Berra -- bluca@comedia.it
Communication Media & Services S.r.l.
/"\
\ / ASCII RIBBON CAMPAIGN
X AGAINST HTML MAIL
/ \
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: stacked raid devices not autodetected?
[not found] ` <36652.212.158.231.74.1102208997.squirrel@mail.fsck.co.uk>
@ 2004-12-05 7:26 ` Luca Berra
2004-12-06 6:41 ` Odd md related top output with kernel 2.4.28 Guy
0 siblings, 1 reply; 7+ messages in thread
From: Luca Berra @ 2004-12-05 7:26 UTC (permalink / raw)
To: linux-raid
On Sun, Dec 05, 2004 at 01:09:57AM -0000, A. James Lewis wrote:
>
>Was there not also a kernel command line option you could pass to assemble
>a raid array?... I'm sure I've used this in the past, but I don't remember
>the details...
>
>Anyone confirm or deny?
md=<md device no.>,dev0,dev1,...,devn
L.
(please could you reply to the list only and avoid top-posting)
--
Luca Berra -- bluca@comedia.it
Communication Media & Services S.r.l.
/"\
\ / ASCII RIBBON CAMPAIGN
X AGAINST HTML MAIL
/ \
^ permalink raw reply [flat|nested] 7+ messages in thread
* Odd md related top output with kernel 2.4.28
2004-12-05 7:26 ` Luca Berra
@ 2004-12-06 6:41 ` Guy
2004-12-06 7:15 ` Guy
0 siblings, 1 reply; 7+ messages in thread
From: Guy @ 2004-12-06 6:41 UTC (permalink / raw)
To: linux-raid
Neil, (or anyone)
I just upgraded my system from kernel 2.4.20.8smp (redhat) to 2.4.28
(from kernel.org). I am making apache, and ran top to see how things were
going. 2 processes have strange info. PIDs 8 and 28. Any ideas about what
is going on? The system seems fine.
Thanks,
Guy
From "ps -ef" they look like this:
root 8 1 0 Dec05 ? 00:00:00 [mdrecoveryd]
root 28 1 0 Dec05 ? 00:00:02 [raid5d]
98 processes: 95 sleeping, 3 running, 0 zombie, 0 stopped
CPU0 states: 62.1% user 11.0% system 0.0% nice 0.0% iowait 25.1%
idle
CPU1 states: 25.0% user 12.1% system 0.0% nice 0.0% iowait 62.0%
idle
Mem: 515296k av, 188588k used, 326708k free, 0k shrd, 22800k
buff
36692k active, 84512k inactive
Swap: 524280k av, 0k used, 524280k free 70264k
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU COMMAND
14574 root 13 0 2004 2004 960 R 24.9 0.3 0:00 0 sh
14562 root 16 0 1160 1160 856 R 6.4 0.2 0:00 1 top
14569 root 9 0 752 752 640 S 1.8 0.1 0:00 1 make
14571 root 9 0 1028 1028 888 S 0.9 0.1 0:00 1 sh
1 root 9 0 472 472 424 S 0.0 0.0 0:03 1 init
2 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 keventd
3 root 19 19 0 0 0 SWN 0.0 0.0 0:00 0
ksoftirqd_CPU
4 root 19 19 0 0 0 SWN 0.0 0.0 0:00 1
ksoftirqd_CPU
5 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 kswapd
6 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 bdflush
7 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 kupdated
8 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:00
14 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 ahc_dv_0
15 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 ahc_dv_1
16 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 ahc_dv_2
17 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 ahc_dv_3
18 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 scsi_eh_0
19 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 scsi_eh_1
20 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 scsi_eh_2
21 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 scsi_eh_3
28 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:01
^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: Odd md related top output with kernel 2.4.28
2004-12-06 6:41 ` Odd md related top output with kernel 2.4.28 Guy
@ 2004-12-06 7:15 ` Guy
0 siblings, 0 replies; 7+ messages in thread
From: Guy @ 2004-12-06 7:15 UTC (permalink / raw)
To: linux-raid
I did a reboot, the same thing. Even if the system is idle.
From "top":
8 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:00
28 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:00
29 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:00
30 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:00
From "ps -ef"
root 8 1 0 01:57 ? 00:00:00 [mdrecoveryd]
root 28 1 0 01:57 ? 00:00:00 [raid5d]
root 29 1 0 01:57 ? 00:00:00 [raid1d]
root 30 1 0 01:57 ? 00:00:00 [raid1d]
Guy
-----Original Message-----
From: linux-raid-owner@vger.kernel.org
[mailto:linux-raid-owner@vger.kernel.org] On Behalf Of Guy
Sent: Monday, December 06, 2004 1:42 AM
To: linux-raid@vger.kernel.org
Subject: Odd md related top output with kernel 2.4.28
Neil, (or anyone)
I just upgraded my system from kernel 2.4.20.8smp (redhat) to 2.4.28
(from kernel.org). I am making apache, and ran top to see how things were
going. 2 processes have strange info. PIDs 8 and 28. Any ideas about what
is going on? The system seems fine.
Thanks,
Guy
From "ps -ef" they look like this:
root 8 1 0 Dec05 ? 00:00:00 [mdrecoveryd]
root 28 1 0 Dec05 ? 00:00:02 [raid5d]
98 processes: 95 sleeping, 3 running, 0 zombie, 0 stopped
CPU0 states: 62.1% user 11.0% system 0.0% nice 0.0% iowait 25.1%
idle
CPU1 states: 25.0% user 12.1% system 0.0% nice 0.0% iowait 62.0%
idle
Mem: 515296k av, 188588k used, 326708k free, 0k shrd, 22800k
buff
36692k active, 84512k inactive
Swap: 524280k av, 0k used, 524280k free 70264k
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME CPU COMMAND
14574 root 13 0 2004 2004 960 R 24.9 0.3 0:00 0 sh
14562 root 16 0 1160 1160 856 R 6.4 0.2 0:00 1 top
14569 root 9 0 752 752 640 S 1.8 0.1 0:00 1 make
14571 root 9 0 1028 1028 888 S 0.9 0.1 0:00 1 sh
1 root 9 0 472 472 424 S 0.0 0.0 0:03 1 init
2 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 keventd
3 root 19 19 0 0 0 SWN 0.0 0.0 0:00 0
ksoftirqd_CPU
4 root 19 19 0 0 0 SWN 0.0 0.0 0:00 1
ksoftirqd_CPU
5 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 kswapd
6 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 bdflush
7 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 kupdated
8 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:00
14 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 ahc_dv_0
15 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 ahc_dv_1
16 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 ahc_dv_2
17 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 ahc_dv_3
18 root 9 0 0 0 0 SW 0.0 0.0 0:00 0 scsi_eh_0
19 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 scsi_eh_1
20 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 scsi_eh_2
21 root 9 0 0 0 0 SW 0.0 0.0 0:00 1 scsi_eh_3
28 root 18446744073709551615 -20 0 0 0 SW< 0.0 0.0
0:01
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: stacked raid devices not autodetected?
[not found] ` <20041206113752.GM3858@marowsky-bree.de>
@ 2004-12-06 11:49 ` Luca Berra
2004-12-06 13:16 ` Lars Marowsky-Bree
0 siblings, 1 reply; 7+ messages in thread
From: Luca Berra @ 2004-12-06 11:49 UTC (permalink / raw)
To: linux-raid
On Mon, Dec 06, 2004 at 12:37:52PM +0100, Lars Marowsky-Bree wrote:
>On 2004-12-03T16:34:59, Luca Berra <bluca@comedia.it> wrote:
>
>> > On boot, the kernel automatically detects and starts md0 through md4.
>> > However, md5 is *not* autodetected. Using the RAID_AUTORUN ioctl also
>> > has no effect. The only way I've found to reliably start md5 is to
>> > have rc.sysinit run "mdadm -assemble --scan".
>> >
>> > Is this a bug? A design limitation? Something else?
>> this is one of the limitations of the in-kernel autodetection
>
>You can pull a fix for that one out of the SUSE kernel, where we add md
>devices we just started to the auto detection. It worked pretty well in
>2.4, but my memory is hazed when I try to recall whether it was all
>ported to 2.6...
>
is it worth doing that in-kernel, i really believe that identifying
and starting arrays is a job that is better done in user-space.
regards,
L.
--
Luca Berra -- bluca@comedia.it
Communication Media & Services S.r.l.
/"\
\ / ASCII RIBBON CAMPAIGN
X AGAINST HTML MAIL
/ \
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: stacked raid devices not autodetected?
2004-12-06 11:49 ` stacked raid devices not autodetected? Luca Berra
@ 2004-12-06 13:16 ` Lars Marowsky-Bree
0 siblings, 0 replies; 7+ messages in thread
From: Lars Marowsky-Bree @ 2004-12-06 13:16 UTC (permalink / raw)
To: Luca Berra; +Cc: linux-raid
On 2004-12-06T12:49:24, Luca Berra <bluca@comedia.it> wrote:
> >>> Is this a bug? A design limitation? Something else?
> >>this is one of the limitations of the in-kernel autodetection
> >You can pull a fix for that one out of the SUSE kernel, where we add md
> >devices we just started to the auto detection. It worked pretty well in
> >2.4, but my memory is hazed when I try to recall whether it was all
> >ported to 2.6...
> >
> is it worth doing that in-kernel, i really believe that identifying
> and starting arrays is a job that is better done in user-space.
I'm not disagreeing, I'm only pointing out that it is possible and what
the general fix looks like. We had a customer requirement to get it done
that way, and at the time it was just the fastest way to code it.
I'm a fan of moving discovery to user-land, alas I'm also a fan of frank
words and right now the much hailed user-space discovery of RAIDs,
partitions etc still has a couple of months to go to stabilize ;-)
Sincerely,
Lars Marowsky-Brée <lmb@suse.de>
--
High Availability & Clustering
SUSE Labs, Research and Development
SUSE LINUX Products GmbH - A Novell Business
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2004-12-06 13:16 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-12-02 22:32 stacked raid devices not autodetected? James Ralston
2004-12-03 15:34 ` Luca Berra
[not found] ` <36652.212.158.231.74.1102208997.squirrel@mail.fsck.co.uk>
2004-12-05 7:26 ` Luca Berra
2004-12-06 6:41 ` Odd md related top output with kernel 2.4.28 Guy
2004-12-06 7:15 ` Guy
[not found] ` <20041206113752.GM3858@marowsky-bree.de>
2004-12-06 11:49 ` stacked raid devices not autodetected? Luca Berra
2004-12-06 13:16 ` Lars Marowsky-Bree
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).