raid 5 out of sync won't mount

raid 5 out of sync won't mount

[Monty Charlton]

Hello all,

I've got a helluva problem and I'm hoping someone here can possibly help 
out.

I've got raid5 (sdb1,sbc1,sdd1) on a partition, and one of the drives 
(sdd1) went bad.  So I (stupidly) hit the reset button on the computer, 
but now the partition won't mount.  Raid complains that one drive is bad 
(which it is), and that the other is out of sync.  See part of dmesg 
below...

Am I hosed?  Can I fake out raid by commenting out some of the code that 
tries to determine whether or not the drive is fresh?

Thanks,
Monty


part of dmesg:

md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
raid5: measuring checksumming speed
    8regs     :   247.600 MB/sec
    32regs    :   248.000 MB/sec
    pII_mmx   :   337.200 MB/sec
    p5_mmx    :   409.200 MB/sec
raid5: using function: p5_mmx (409.200 MB/sec)
md: raid5 personality registered as nr 4
  [events: 00000003]
  [events: 00000004]
  [events: 00000002]
md: autorun ...
md: considering sdd1 ...
md:  adding sdd1 ...
md:  adding sdc1 ...
md:  adding sdb1 ...
md: created md0
md: bind<sdb1,1>
md: bind<sdc1,2>
md: bind<sdd1,3>
md: running: <sdd1><sdc1><sdb1>
md: sdd1's event counter: 00000002
md: sdc1's event counter: 00000004
md: sdb1's event counter: 00000003
md: superblock update time inconsistency -- using the most recent one
md: freshest: sdc1
md: kicking non-fresh sdd1 from array!
md: unbind<sdd1,2>
md: export_rdev(sdd1)
md0: kicking faulty sdb1!
md: unbind<sdb1,1>
md: export_rdev(sdb1)
md0: removing former faulty sdd1!
md: md0: raid array is not clean -- starting background reconstruction
md0: max total readahead window set to 496k
md0: 2 data-disks, max readahead per data-disk: 248k
raid5: device sdc1 operational as raid disk 1
raid5: not enough operational devices for md0 (2/3 failed)
RAID5 conf printout:
  --- rd:3 wd:1 fd:2
  disk 0, s:0, o:0, n:0 rd:0 us:1 dev:[dev 00:00]
  disk 1, s:0, o:1, n:1 rd:1 us:1 dev:sdc1
  disk 2, s:0, o:0, n:2 rd:2 us:1 dev:[dev 00:00]
raid5: failed to run raid set md0
md: pers->run() failed ...
md :do_md_run() returned -22
md: md0 stopped.
md: unbind<sdc1,0>
md: export_rdev(sdc1)
md: ... autorun DONE.

Re: raid 5 out of sync won't mount

[Luca Berra]

On Tue, Jun 18, 2002 at 08:45:45AM -0600, Monty Charlton wrote:
> Hello all,
> 
> I've got a helluva problem and I'm hoping someone here can possibly help 
> out.
> 
> I've got raid5 (sdb1,sbc1,sdd1) on a partition, and one of the drives 
> (sdd1) went bad.  So I (stupidly) hit the reset button on the computer, 
> but now the partition won't mount.  Raid complains that one drive is bad 
> (which it is), and that the other is out of sync.  See part of dmesg 
> below...
> 
> Am I hosed?  Can I fake out raid by commenting out some of the code that 
> tries to determine whether or not the drive is fresh?
edit /etc/raidtab
mark the drive you know as bad as
failed-disk instead of raid-disk
run mkraid --force on the array
read the message you get and do as instructed.

L.

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

Re: raid 5 out of sync won't mount

[Monty Charlton]

Luca Berra wrote:
> On Tue, Jun 18, 2002 at 08:45:45AM -0600, Monty Charlton wrote:
> 
>>Hello all,
>>
>>I've got a helluva problem and I'm hoping someone here can possibly help 
>>out.
>>
>>I've got raid5 (sdb1,sbc1,sdd1) on a partition, and one of the drives 
>>(sdd1) went bad.  So I (stupidly) hit the reset button on the computer, 
>>but now the partition won't mount.  Raid complains that one drive is bad 
>>(which it is), and that the other is out of sync.  See part of dmesg 
>>below...
>>
>>Am I hosed?  Can I fake out raid by commenting out some of the code that 
>>tries to determine whether or not the drive is fresh?
> 
> edit /etc/raidtab
> mark the drive you know as bad as
> failed-disk instead of raid-disk
> run mkraid --force on the array
> read the message you get and do as instructed.

Thank you for the reply.

I read that in the howto, but feared it would destroy my data.  As the 
mkraid manpage says, "... destroys all of the data on the consituent 
devices."

I read the message, and am about to use the secret flag, but I want to 
be sure.  This will not simply wipe out my data?  This will only resync 
my two good drives?  Sorry to be such a coward...just trying to be sure 
of what I'm doing.

Thanks again,
Monty

Re: raid 5 out of sync won't mount

[Jakob Oestergaard]

On Tue, Jun 18, 2002 at 06:15:49PM +0200, Luca Berra wrote:
> On Tue, Jun 18, 2002 at 08:45:45AM -0600, Monty Charlton wrote:
> > Hello all,
> > 
> > I've got a helluva problem and I'm hoping someone here can possibly help 
> > out.
> > 
> > I've got raid5 (sdb1,sbc1,sdd1) on a partition, and one of the drives 
> > (sdd1) went bad.  So I (stupidly) hit the reset button on the computer, 
> > but now the partition won't mount.  Raid complains that one drive is bad 
> > (which it is), and that the other is out of sync.  See part of dmesg 
> > below...
> > 
> > Am I hosed?  Can I fake out raid by commenting out some of the code that 
> > tries to determine whether or not the drive is fresh?
> edit /etc/raidtab
> mark the drive you know as bad as
> failed-disk instead of raid-disk
> run mkraid --force on the array
> read the message you get and do as instructed.

Yes, that is exactly the way to do it.

If you mark the right disk as failed, everything will succeed and you
will not lose your data even though mkraid claims that you will.

If you, however, mark the wrong disk as failed, or forget to mark a
disk as failed, no amount of magic is going to bring those data back.

If it is really important data, it might be advisable to do a raw copy
of the individual partitions to some other machine - just in case. But
again, if you do it right, there's no problem  ;)

-- 
................................................................
:   jakob@unthought.net   : And I see the elder races,         :
:.........................: putrid forms of man                :
:   Jakob Østergaard      : See him rise and claim the earth,  :
:        OZ9ABN           : his downfall is at hand.           :
:.........................:............{Konkhra}...............:
-
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: raid 5 out of sync won't mount

[Jakob Oestergaard]

On Tue, Jun 18, 2002 at 10:52:48AM -0600, Monty Charlton wrote:
> Luca Berra wrote:
> >On Tue, Jun 18, 2002 at 08:45:45AM -0600, Monty Charlton wrote:
...
> 
> Thank you for the reply.
> 
> I read that in the howto, but feared it would destroy my data.  As the 
> mkraid manpage says, "... destroys all of the data on the consituent 
> devices."
> 
> I read the message, and am about to use the secret flag, but I want to 
> be sure.  This will not simply wipe out my data?  This will only resync 
> my two good drives?  Sorry to be such a coward...just trying to be sure 
> of what I'm doing.

I think our mails crossed on the wire   ;)    Please see the other mail
I just sent to the list.

Especially the part about perhaps dd'ing images of your partitions to
some other storage...

-- 
................................................................
:   jakob@unthought.net   : And I see the elder races,         :
:.........................: putrid forms of man                :
:   Jakob Østergaard      : See him rise and claim the earth,  :
:        OZ9ABN           : his downfall is at hand.           :
:.........................:............{Konkhra}...............:
-
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: raid 5 out of sync won't mount

[Monty Charlton]

On Tue, Jun 18, 2002 at 06:56:38PM +0200, Jakob Oestergaard wrote:
> Yes, that is exactly the way to do it.
> 
> If you mark the right disk as failed, everything will succeed and you
> will not lose your data even though mkraid claims that you will.
> 
> If you, however, mark the wrong disk as failed, or forget to mark a
> disk as failed, no amount of magic is going to bring those data back.
> 
> If it is really important data, it might be advisable to do a raw copy
> of the individual partitions to some other machine - just in case. But
> again, if you do it right, there's no problem  ;)

Thanks so much!  That worked great.  fullmonty.org lives again thanks to you guys. :)

Monty


--
Monty Charlton
http://fullMonty.org