bad /proc/mdstat message

bad /proc/mdstat message

[Sharif Islam]

I am getting the following message:
mdadm: bad /proc/mdstat line starts : Event:


I wasn't able to figure out the reason. My array is acting normal and
/proc/mdstat looks ok to me.

# cat /proc/mdstat
Personalities : [linear] [raid0]
read_ahead 1024 sectors
Event: 1
md0 : active linear hdl[5] hdk[4] hdh[3] hdg[2] hdf[1] hde[0]
      703324416 blocks 64k rounding

unused devices: <none>

Any hints?

RE: bad /proc/mdstat message

[Guy]

My /proc/mdstat does not contain "Event: 1", or "Event".
So I would guess it is related to an event! :)

Look in "/var/log/messages", and dmesg.

I can't find any info that explains the contents of /proc/mdstat.
It also seems like mdadm has a bug!  It should support the "Event" line.
Maybe just ignore it.  Maybe /proc/mdstat should give details about the
events.  Just my 3 cents.

-----Original Message-----
From: linux-raid-owner@vger.kernel.org
[mailto:linux-raid-owner@vger.kernel.org] On Behalf Of Sharif Islam
Sent: Monday, January 12, 2004 1:29 PM
To: linux-raid@vger.kernel.org
Subject: bad /proc/mdstat message 

I am getting the following message:
mdadm: bad /proc/mdstat line starts : Event:


I wasn't able to figure out the reason. My array is acting normal and
/proc/mdstat looks ok to me.

# cat /proc/mdstat
Personalities : [linear] [raid0]
read_ahead 1024 sectors
Event: 1
md0 : active linear hdl[5] hdk[4] hdh[3] hdg[2] hdf[1] hde[0]
      703324416 blocks 64k rounding

unused devices: <none>

Any hints?

-
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

Re: bad /proc/mdstat message

[Mike Fedyk]

On Mon, Jan 12, 2004 at 12:28:37PM -0600, Sharif Islam wrote:
> I am getting the following message:
> mdadm: bad /proc/mdstat line starts : Event:
> 
> 
> I wasn't able to figure out the reason. My array is acting normal and
> /proc/mdstat looks ok to me.
> 
> # cat /proc/mdstat
> Personalities : [linear] [raid0]
> read_ahead 1024 sectors
> Event: 1
> md0 : active linear hdl[5] hdk[4] hdh[3] hdg[2] hdf[1] hde[0]
>       703324416 blocks 64k rounding
> 
> unused devices: <none>
> 
> Any hints?

Did you see anything in your logs?

Re: bad /proc/mdstat message

[Sharif Islam]

On Mon, 12 Jan 2004, Mike Fedyk wrote:

> On Mon, Jan 12, 2004 at 12:28:37PM -0600, Sharif Islam wrote:
> > I am getting the following message:
> > mdadm: bad /proc/mdstat line starts : Event:
> >
> >
> > I wasn't able to figure out the reason. My array is acting normal and
> > /proc/mdstat looks ok to me.
> >
> > # cat /proc/mdstat
> > Personalities : [linear] [raid0]
> > read_ahead 1024 sectors
> > Event: 1
> > md0 : active linear hdl[5] hdk[4] hdh[3] hdg[2] hdf[1] hde[0]
> >       703324416 blocks 64k rounding
> >
> > unused devices: <none>
> >
> > Any hints?
>
> Did you see anything in your logs?
Nope, no errors and such.
-----
kjournald starting.  Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on md(9,1), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
----
I think this is something linear raid specific. I have another machine
with RAID 5 and I don't have the Event: 1 line there either.

Re: bad /proc/mdstat message

[Luca Berra]

On Mon, Jan 12, 2004 at 12:28:37PM -0600, Sharif Islam wrote:
>I am getting the following message:
>mdadm: bad /proc/mdstat line starts : Event:
>
>I wasn't able to figure out the reason. My array is acting normal and
>/proc/mdstat looks ok to me.
>
># cat /proc/mdstat
>Personalities : [linear] [raid0]
>read_ahead 1024 sectors
>Event: 1
where does this line come from?
are you running a vanilla or patched kernel? which version?

-- 
Luca Berra -- bluca@comedia.it
        Communication Media & Services S.r.l.
 /"\
 \ /     ASCII RIBBON CAMPAIGN
  X        AGAINST HTML MAIL
 / \

Linear RAID (was Re: bad /proc/mdstat message)

[Sharif Islam]

On Mon, 12 Jan 2004, Luca Berra wrote:

> On Mon, Jan 12, 2004 at 12:28:37PM -0600, Sharif Islam wrote:
[...]

> ># cat /proc/mdstat
> >Personalities : [linear] [raid0]
> >read_ahead 1024 sectors
> >Event: 1

> where does this line come from?
> are you running a vanilla or patched kernel? which version?

No patching.

# uname -r
2.4.21-4.ELsmp
Red Hat Enterprise Linux AS release 3


Few other things I noticed while playing with linear raid. I haven't
seen much documentation online about it, so bare with me. It is a test
machine so I am not worried about data loss or things like that.

1. The raid creation process was very quick, unlike RAID 1 and RAID 5.
2. Every time I reboot the raid partition doesn't get mounted. However,
if I run
mdadm -Cv  -llinear -n6 /dev/md0 /dev/hd{e,f,g,h,k,l}
and
mount /dev/md0 /raid
it comes back again with all my data.

I included the partition in /etc/fstab, but that gives error. Also my
rc.local has mdadm -As + mount command. That doesn't have any effect
either.

and

3. I get the following warning message when I run
the mdadm -Cv command, for all the disks.

mdadm: /dev/hdi appears to contain an ext2fs file system
    size=120103040K  mtime=Mon Jan 12 12:42:37 2004
...
Continue creating array?

I hit yes, and my array is back. Here mtime is the last time my array
was up.

Did I screw up somewhere or all these make some sense?

-Sharif

RE: Linear RAID (was Re: bad /proc/mdstat message)

[Christian Kivalo]

> 2. Every time I reboot the raid partition doesn't get
> mounted.
hi!

are the partitions of type "Linux Raid Autodetect" (type fd)? i don't
know for sure if that apply's for linear also, but for all other raid
types, having the partition type 'fd', the array get's assembled at boot
time and properly mountet through fstab.

christian

Re: Linear RAID (was Re: bad /proc/mdstat message)

[Mike Fedyk]

On Mon, Jan 12, 2004 at 09:19:48PM +0100, Christian Kivalo wrote:
> > 2. Every time I reboot the raid partition doesn't get
> > mounted.
> hi!
> 
> are the partitions of type "Linux Raid Autodetect" (type fd)? i don't
> know for sure if that apply's for linear also, but for all other raid

Yes, it works with linear...

Re: Linear RAID (was Re: bad /proc/mdstat message)

[Luca Berra]

On Mon, Jan 12, 2004 at 02:01:19PM -0600, Sharif Islam wrote:
>> ># cat /proc/mdstat
>> >Personalities : [linear] [raid0]
>> >read_ahead 1024 sectors
>> >Event: 1
>
>> where does this line come from?
>> are you running a vanilla or patched kernel? which version?
>
>No patching.
>
># uname -r
>2.4.21-4.ELsmp
>Red Hat Enterprise Linux AS release 3
heavily patched i would say.
complain to redhat since they seem to patch kernels and not userland
tools :)

>Few other things I noticed while playing with linear raid. I haven't
>seen much documentation online about it, so bare with me. It is a test
>machine so I am not worried about data loss or things like that.
>
>1. The raid creation process was very quick, unlike RAID 1 and RAID 5.
>2. Every time I reboot the raid partition doesn't get mounted. However,
>if I run
>mdadm -Cv  -llinear -n6 /dev/md0 /dev/hd{e,f,g,h,k,l}
>and
>mount /dev/md0 /raid
>it comes back again with all my data.

redhat by default used raidtab and raidtools to start raid devices, i
don't know if they changed this for AS3

>I included the partition in /etc/fstab, but that gives error. Also my
>rc.local has mdadm -As + mount command. That doesn't have any effect
>either.
i think you don't have a correct mdadm.conf file
to create one from scratch use
echo "DEVICE partitions" > /etc/mdadm.conf
mdadm  --examine  --scan --config=partitions | grep "^ARRAY" >> /etc/mdadm.conf

>3. I get the following warning message when I run
>the mdadm -Cv command, for all the disks.
>
>mdadm: /dev/hdi appears to contain an ext2fs file system
>    size=120103040K  mtime=Mon Jan 12 12:42:37 2004
>...
>Continue creating array?
>
>I hit yes, and my array is back. Here mtime is the last time my array
>was up.
you should mdadm -A to start arrays, not mdadm -C


-- 
Luca Berra -- bluca@comedia.it
        Communication Media & Services S.r.l.
 /"\
 \ /     ASCII RIBBON CAMPAIGN
  X        AGAINST HTML MAIL
 / \