"failed-disk" ?

"failed-disk" ?

[Robert L. Harris]

[-- Attachment #1: Type: text/plain, Size: 1119 bytes --]



I'm trying to "mkraid /dev/md1" with the following:

raiddev /dev/md1
  raid-level 1
  nr-raid-disks 2
  nr-spare-disks 0
  chunk-size 16
  persistent-superblock 1
  device failed-disk
  raid-disk 0
  device /dev/sdb1
  raid-disk 1

I'm trying to convert this machine to raid1 on root.  Don't ask about
the md1, long story.  At any rate when I do the mkraid I get this:

root@rharris-build2:~# mkraid /dev/md1
handling MD device /dev/md1
analyzing super-block
couldn't call stat() on device failed-disk -- No such file or directory
mkraid: aborted, see the syslog and /proc/mdstat for potential clues.

root@rharris-build2.acs:~# mkraid -V
mkraid version 0.90.0

Thoughts?

:wq!
---------------------------------------------------------------------------
Robert L. Harris                     | PGP Key ID: E344DA3B
                                         @ x-hkp://pgp.mit.edu 
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.

Diagnosis: witzelsucht  	

IPv6 = robert@ipv6.rdlg.net	http://ipv6.rdlg.net
IPv4 = robert@mail.rdlg.net	http://www.rdlg.net

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: "failed-disk" ?

[Lars Marowsky-Bree]

On 2003-03-20T16:22:50,
   "Robert L. Harris" <Robert.L.Harris@rdlg.net> said:

> I'm trying to "mkraid /dev/md1" with the following:
> 
> raiddev /dev/md1
>   raid-level 1
>   nr-raid-disks 2
>   nr-spare-disks 0
>   chunk-size 16
>   persistent-superblock 1
>   device failed-disk
>   raid-disk 0
>   device /dev/sdb1
>   raid-disk 1
> 
> I'm trying to convert this machine to raid1 on root.  Don't ask about
> the md1, long story.  At any rate when I do the mkraid I get this:
> 
> root@rharris-build2:~# mkraid /dev/md1
> handling MD device /dev/md1
> analyzing super-block
> couldn't call stat() on device failed-disk -- No such file or directory
> mkraid: aborted, see the syslog and /proc/mdstat for potential clues.

Does your system have a device named 'failed-disk' ? I would assume not ;-)


Sincerely,
    Lars Marowsky-Brée <lmb@suse.de>

-- 
SuSE Labs - Research & Development, SuSE Linux AG
  
"If anything can go wrong, it will." "Chance favors the prepared (mind)."
  -- Capt. Edward A. Murphy            -- Louis Pasteur
-
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: "failed-disk" ?

[Wolfram Schlich]

[-- Attachment #1: Type: text/plain, Size: 1076 bytes --]

* Robert L. Harris <Robert.L.Harris@rdlg.net> [2003-03-20 22:25]:
> 
> 
> I'm trying to "mkraid /dev/md1" with the following:
> 
> raiddev /dev/md1
>   raid-level 1
>   nr-raid-disks 2
>   nr-spare-disks 0
>   chunk-size 16
>   persistent-superblock 1
>   device failed-disk
>   raid-disk 0
>   device /dev/sdb1
>   raid-disk 1
> 
> I'm trying to convert this machine to raid1 on root.  Don't ask about
> the md1, long story.  At any rate when I do the mkraid I get this:
> 
> root@rharris-build2:~# mkraid /dev/md1
> handling MD device /dev/md1
> analyzing super-block
> couldn't call stat() on device failed-disk -- No such file or directory
> mkraid: aborted, see the syslog and /proc/mdstat for potential clues.
> 
> root@rharris-build2.acs:~# mkraid -V
> mkraid version 0.90.0
> 
> Thoughts?

You need to use "failed-disk" in place of "raid-disk", not anywhere
else :-)

E.g.:
--8<--
device /some/dev1
raid-disk 0

device /some/dev2
failed-fisk 1
--8<--
-- 
Wolfram Schlich; Friedhofstr. 8, D-88069 Tettnang; +49-(0)178-SCHLICH

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: "failed-disk" ?

[Robert L. Harris]

[-- Attachment #1: Type: text/plain, Size: 1811 bytes --]



No but according to the RAID HOWTO it says it's a valid keyword for when
you need to build a mirror with a failed/missing disk (i.e. converting
root to raid and can't kill your existing /)


Thus spake Lars Marowsky-Bree (lmb@suse.de):

> On 2003-03-20T16:22:50,
>    "Robert L. Harris" <Robert.L.Harris@rdlg.net> said:
> 
> > I'm trying to "mkraid /dev/md1" with the following:
> > 
> > raiddev /dev/md1
> >   raid-level 1
> >   nr-raid-disks 2
> >   nr-spare-disks 0
> >   chunk-size 16
> >   persistent-superblock 1
> >   device failed-disk
> >   raid-disk 0
> >   device /dev/sdb1
> >   raid-disk 1
> > 
> > I'm trying to convert this machine to raid1 on root.  Don't ask about
> > the md1, long story.  At any rate when I do the mkraid I get this:
> > 
> > root@rharris-build2:~# mkraid /dev/md1
> > handling MD device /dev/md1
> > analyzing super-block
> > couldn't call stat() on device failed-disk -- No such file or directory
> > mkraid: aborted, see the syslog and /proc/mdstat for potential clues.
> 
> Does your system have a device named 'failed-disk' ? I would assume not ;-)
> 
> 
> Sincerely,
>     Lars Marowsky-Br?e <lmb@suse.de>
> 
> -- 
> SuSE Labs - Research & Development, SuSE Linux AG
>   
> "If anything can go wrong, it will." "Chance favors the prepared (mind)."
>   -- Capt. Edward A. Murphy            -- Louis Pasteur

:wq!
---------------------------------------------------------------------------
Robert L. Harris                     | PGP Key ID: E344DA3B
                                         @ x-hkp://pgp.mit.edu 
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.

Diagnosis: witzelsucht  	

IPv6 = robert@ipv6.rdlg.net	http://ipv6.rdlg.net
IPv4 = robert@mail.rdlg.net	http://www.rdlg.net

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: "failed-disk" ?

[danci]

On Thu, 20 Mar 2003, Robert L. Harris wrote:

>
>
> No but according to the RAID HOWTO it says it's a valid keyword for when
> you need to build a mirror with a failed/missing disk (i.e. converting
> root to raid and can't kill your existing /)

It is, but:

'failed-disk' replaces 'raid-disk' to inform the md driver, that a device
'/dev/sda1' should not be automatically synchronized.

This gives you an operative array in degraded mode, so you can copy files
from your current root to your new degraded array.

You have to change the file /etc/fstab (on the NEW filesystem) to use
/dev/md0 for root, arrange the boot manager (lilo, grub, whatever) to boot
with the new root (/dev/md0), reboot and 'un-fail' :) (or add) /dev/sda1
to the array (using mdadm or raidhotadd).


  D.


>
>
> Thus spake Lars Marowsky-Bree (lmb@suse.de):
>
> > On 2003-03-20T16:22:50,
> >    "Robert L. Harris" <Robert.L.Harris@rdlg.net> said:
> >
> > > I'm trying to "mkraid /dev/md1" with the following:
> > >
> > > raiddev /dev/md1
> > >   raid-level 1
> > >   nr-raid-disks 2
> > >   nr-spare-disks 0
> > >   chunk-size 16
> > >   persistent-superblock 1
> > >   device failed-disk
> > >   raid-disk 0
> > >   device /dev/sdb1
> > >   raid-disk 1
> > >
> > > I'm trying to convert this machine to raid1 on root.  Don't ask about
> > > the md1, long story.  At any rate when I do the mkraid I get this:
> > >
> > > root@rharris-build2:~# mkraid /dev/md1
> > > handling MD device /dev/md1
> > > analyzing super-block
> > > couldn't call stat() on device failed-disk -- No such file or directory
> > > mkraid: aborted, see the syslog and /proc/mdstat for potential clues.
> >
> > Does your system have a device named 'failed-disk' ? I would assume not ;-)
> >
> >
> > Sincerely,
> >     Lars Marowsky-Br?e <lmb@suse.de>
> >
> > --
> > SuSE Labs - Research & Development, SuSE Linux AG
> >
> > "If anything can go wrong, it will." "Chance favors the prepared (mind)."
> >   -- Capt. Edward A. Murphy            -- Louis Pasteur
>
> :wq!
> ---------------------------------------------------------------------------
> Robert L. Harris                     | PGP Key ID: E344DA3B
>                                          @ x-hkp://pgp.mit.edu
> DISCLAIMER:
>       These are MY OPINIONS ALONE.  I speak for no-one else.
>
> Diagnosis: witzelsucht
>
> IPv6 = robert@ipv6.rdlg.net	http://ipv6.rdlg.net
> IPv4 = robert@mail.rdlg.net	http://www.rdlg.net
>

-- 
___________________________________________________________________
|    Danilo Godec    |     Agenda d.o.o.    |   ISP for business  |
|  jr. Syst. Admin   |    Gosposvetska 84   |     WAN networks    |
|  danci@agenda.si   |    si-2000 Maribor   |  Internet/Intranet  |
| tel:+386.2.2340860 |       Slovenija      | Application servers |
| fax:+386.2.2340854 | http://www.agenda.si |  Caldera OpenLinux  |

Re: "failed-disk" ?

[Robert L. Harris]

[-- Attachment #1: Type: text/plain, Size: 4652 bytes --]



  I'm trying to rebuild a system with a non-mirrored root.  I
intentionally killed /dev/sda by yanking it (hotswap disk) and let the
machine croak.  I put the replacement disk in and rebooted, rebuilt (we
have a system similar to jumpstart that makes this very fast and clean).

  I'm at the point of trying to restart /dev/md0 with one mirror dead.
My raidtab looks like this:

raiddev /dev/md0
  raid-level 1
  nr-raid-disks 2
  nr-spare-disks 0
  chunk-size 16
  persistent-superblock 1
  device /dev/lvm-vg1/lvm1
  raid-disk 0
  device /dev/lvm-vg2/lvm2
  raid-disk 1

I changed the first raid-disk line to be:
  failed-disk 0

and then tried:

root@rharris-build2.acs:~# raidstart /dev/md0
/dev/md0: Invalid argument
md: could not lock [dev 9e:0e], zero-size? Marking faulty.
md: could not import [dev 9e:0e]!
md: autostart [dev 9e:0e] failed!


At this point /dev/lvm-vg1/lvm1 doesn't exist and will need to be
rebuilt but I'd like to get the machine up and running with /dev/md0 
intact.

Any idea what's wrong here?  

Robert



Thus spake danci@agenda.si (danci@agenda.si):

> On Thu, 20 Mar 2003, Robert L. Harris wrote:
> 
> >
> >
> > No but according to the RAID HOWTO it says it's a valid keyword for when
> > you need to build a mirror with a failed/missing disk (i.e. converting
> > root to raid and can't kill your existing /)
> 
> It is, but:
> 
> 'failed-disk' replaces 'raid-disk' to inform the md driver, that a device
> '/dev/sda1' should not be automatically synchronized.
> 
> This gives you an operative array in degraded mode, so you can copy files
> from your current root to your new degraded array.
> 
> You have to change the file /etc/fstab (on the NEW filesystem) to use
> /dev/md0 for root, arrange the boot manager (lilo, grub, whatever) to boot
> with the new root (/dev/md0), reboot and 'un-fail' :) (or add) /dev/sda1
> to the array (using mdadm or raidhotadd).
> 
> 
>   D.
> 
> 
> >
> >
> > Thus spake Lars Marowsky-Bree (lmb@suse.de):
> >
> > > On 2003-03-20T16:22:50,
> > >    "Robert L. Harris" <Robert.L.Harris@rdlg.net> said:
> > >
> > > > I'm trying to "mkraid /dev/md1" with the following:
> > > >
> > > > raiddev /dev/md1
> > > >   raid-level 1
> > > >   nr-raid-disks 2
> > > >   nr-spare-disks 0
> > > >   chunk-size 16
> > > >   persistent-superblock 1
> > > >   device failed-disk
> > > >   raid-disk 0
> > > >   device /dev/sdb1
> > > >   raid-disk 1
> > > >
> > > > I'm trying to convert this machine to raid1 on root.  Don't ask about
> > > > the md1, long story.  At any rate when I do the mkraid I get this:
> > > >
> > > > root@rharris-build2:~# mkraid /dev/md1
> > > > handling MD device /dev/md1
> > > > analyzing super-block
> > > > couldn't call stat() on device failed-disk -- No such file or directory
> > > > mkraid: aborted, see the syslog and /proc/mdstat for potential clues.
> > >
> > > Does your system have a device named 'failed-disk' ? I would assume not ;-)
> > >
> > >
> > > Sincerely,
> > >     Lars Marowsky-Br?e <lmb@suse.de>
> > >
> > > --
> > > SuSE Labs - Research & Development, SuSE Linux AG
> > >
> > > "If anything can go wrong, it will." "Chance favors the prepared (mind)."
> > >   -- Capt. Edward A. Murphy            -- Louis Pasteur
> >
> > :wq!
> > ---------------------------------------------------------------------------
> > Robert L. Harris                     | PGP Key ID: E344DA3B
> >                                          @ x-hkp://pgp.mit.edu
> > DISCLAIMER:
> >       These are MY OPINIONS ALONE.  I speak for no-one else.
> >
> > Diagnosis: witzelsucht
> >
> > IPv6 = robert@ipv6.rdlg.net	http://ipv6.rdlg.net
> > IPv4 = robert@mail.rdlg.net	http://www.rdlg.net
> >
> 
> -- 
> ___________________________________________________________________
> |    Danilo Godec    |     Agenda d.o.o.    |   ISP for business  |
> |  jr. Syst. Admin   |    Gosposvetska 84   |     WAN networks    |
> |  danci@agenda.si   |    si-2000 Maribor   |  Internet/Intranet  |
> | tel:+386.2.2340860 |       Slovenija      | Application servers |
> | fax:+386.2.2340854 | http://www.agenda.si |  Caldera OpenLinux  |
> 

:wq!
---------------------------------------------------------------------------
Robert L. Harris                     | PGP Key ID: E344DA3B
                                         @ x-hkp://pgp.mit.edu 
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.

Diagnosis: witzelsucht  	

IPv6 = robert@ipv6.rdlg.net	http://ipv6.rdlg.net
IPv4 = robert@mail.rdlg.net	http://www.rdlg.net

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: "failed-disk" ?

[danci]

On Mon, 24 Mar 2003, Robert L. Harris wrote:

>   I'm at the point of trying to restart /dev/md0 with one mirror dead.
> My raidtab looks like this:
>
> raiddev /dev/md0
>   raid-level 1
>   nr-raid-disks 2
>   nr-spare-disks 0
>   chunk-size 16
>   persistent-superblock 1
>   device /dev/lvm-vg1/lvm1
>   raid-disk 0
>   device /dev/lvm-vg2/lvm2
>   raid-disk 1
>
> I changed the first raid-disk line to be:
>   failed-disk 0

I'm not sure if this is your problem, but the Software RAID How-To says:

"Don't put the failed-disk as the first disk in the raidtab, that will
give you problems with starting the RAID. Create the RAID, and put a
filesystem on it."

It may be that SW RAID cannot really use LVMs. It is usually the
other way around (LVM using RAID arrays).

  D.

Re: "failed-disk" ?

[Robert L. Harris]

[-- Attachment #1: Type: text/plain, Size: 1609 bytes --]



Actually I thought that meant only when converting a root to raid, but
as a random guess tried this.  I had to physically swap the lines so it
looked like:

device /dev/lvm-vg2/lvm2
raid-disk 1
device /dev/lvm-vg1/lvm1
failed-disk 0

It came right up after that.


Thus spake danci@agenda.si (danci@agenda.si):

> On Mon, 24 Mar 2003, Robert L. Harris wrote:
> 
> >   I'm at the point of trying to restart /dev/md0 with one mirror dead.
> > My raidtab looks like this:
> >
> > raiddev /dev/md0
> >   raid-level 1
> >   nr-raid-disks 2
> >   nr-spare-disks 0
> >   chunk-size 16
> >   persistent-superblock 1
> >   device /dev/lvm-vg1/lvm1
> >   raid-disk 0
> >   device /dev/lvm-vg2/lvm2
> >   raid-disk 1
> >
> > I changed the first raid-disk line to be:
> >   failed-disk 0
> 
> I'm not sure if this is your problem, but the Software RAID How-To says:
> 
> "Don't put the failed-disk as the first disk in the raidtab, that will
> give you problems with starting the RAID. Create the RAID, and put a
> filesystem on it."
> 
> It may be that SW RAID cannot really use LVMs. It is usually the
> other way around (LVM using RAID arrays).
> 
>   D.

:wq!
---------------------------------------------------------------------------
Robert L. Harris                     | PGP Key ID: E344DA3B
                                         @ x-hkp://pgp.mit.edu 
DISCLAIMER:
      These are MY OPINIONS ALONE.  I speak for no-one else.

Diagnosis: witzelsucht  	

IPv6 = robert@ipv6.rdlg.net	http://ipv6.rdlg.net
IPv4 = robert@mail.rdlg.net	http://www.rdlg.net

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]