linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* 3-disk RAID1 - only two disks active
@ 2005-11-15 13:26 vherva
  2005-11-15 23:01 ` Neil Brown
  0 siblings, 1 reply; 3+ messages in thread
From: vherva @ 2005-11-15 13:26 UTC (permalink / raw)
  To: linux-raid

I had two disks taken off from the three-disk raid set (hda,hdb,hdc - hdc
failed by itself, hdb was marked bad with raidsetfaulty). When the new disks
were added back to the set (with raidhotadd), only two of them became active
and one remained as a hot spare.

I there a way to reconfigure all three to become active again?

This is kernel 2.4.32-rc1.

> cat /proc/mdstat
Personalities : [raid1]
read_ahead 1024 sectors
md1 : active raid1 hdb1[2] hdc1[1] hda1[0]
      102720 blocks [2/2] [UU]

md2 : active raid1 hdb2[2] hdc2[1] hda2[0]
      1024064 blocks [2/2] [UU]

md3 : active raid1 hdb3[2] hdc3[1] hda3[0]
      37955648 blocks [2/2] [UU]

unused devices: <none>

> mdadm --detail /dev/md1
/dev/md1:
        Version : 00.90.00
  Creation Time : Fri May 10 13:42:55 2002
     Raid Level : raid1
     Array Size : 102720 (100.33 MiB 105.19 MB)
    Device Size : 102720 (100.33 MiB 105.19 MB)
   Raid Devices : 2
  Total Devices : 3
Preferred Minor : 1
    Persistence : Superblock is persistent

    Update Time : Mon Nov 14 23:17:03 2005
          State : active
 Active Devices : 2
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 1
           
           UUID : d3f91d5b:5587896d:982609b1:c6a3e584
         Events : 0.191
    
    Number   Major   Minor   RaidDevice State
       0       3        1        0      active sync   /dev/hda1
       1      22        1        1      active sync   /dev/hdc1
       
       2       3       65        2      spare   /dev/hdb1


> cat /etc/raidtab
raiddev             /dev/md3
raid-level                  1
nr-raid-disks               3
persistent-superblock       1
nr-spare-disks              0
    device          /dev/hda3
    raid-disk     0
    device          /dev/hdb3
    raid-disk     1
    device          /dev/hdc3
    raid-disk     2
raiddev             /dev/md2
raid-level                  1
nr-raid-disks               3
persistent-superblock       1
nr-spare-disks              0
    device          /dev/hda2
    raid-disk     0
    device          /dev/hdb2
    raid-disk     1
    device          /dev/hdc2
    raid-disk     2
raiddev             /dev/md1
raid-level                  1
nr-raid-disks               3
persistent-superblock       1
nr-spare-disks              0
    device          /dev/hda1
    raid-disk     0
    device          /dev/hdb1
    raid-disk     1
    device          /dev/hdc1
    raid-disk     2

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: 3-disk RAID1 - only two disks active
  2005-11-15 13:26 3-disk RAID1 - only two disks active vherva
@ 2005-11-15 23:01 ` Neil Brown
  2005-11-16  3:23   ` Ville Herva
  0 siblings, 1 reply; 3+ messages in thread
From: Neil Brown @ 2005-11-15 23:01 UTC (permalink / raw)
  To: linux-raid, vherva

On Tuesday November 15, vherva@turing.netspan.fi wrote:
> I had two disks taken off from the three-disk raid set (hda,hdb,hdc - hdc
> failed by itself, hdb was marked bad with raidsetfaulty). When the new disks
> were added back to the set (with raidhotadd), only two of them became active
> and one remained as a hot spare.
> 
> I there a way to reconfigure all three to become active again?

It looks from the details you gave that it never was a three-disk
set.  It was only ever two-disk, possibly with a spare.

> 
> This is kernel 2.4.32-rc1.

With a recent 2.6 kernel, you could
   mdadm --grow --raid-disk=3 /dev/md1
but with a 2.4 kernel, the only option is to re-create the array:

  mdadm --stop /dev/md1
  mdadm --create /dev/md1 --level=1 --raid-disks=3 /dev/hda1 /dev/hdc1 missing
  mdadm /dev/md1 --add /dev/hdb1

NeilBrown

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: 3-disk RAID1 - only two disks active
  2005-11-15 23:01 ` Neil Brown
@ 2005-11-16  3:23   ` Ville Herva
  0 siblings, 0 replies; 3+ messages in thread
From: Ville Herva @ 2005-11-16  3:23 UTC (permalink / raw)
  To: Neil Brown; +Cc: linux-raid, Samuli Kärkkäinen

On Wed, Nov 16, 2005 at 10:01:10AM +1100, you [Neil Brown] wrote:
> On Tuesday November 15, vherva@turing.netspan.fi wrote:
> > I had two disks taken off from the three-disk raid set (hda,hdb,hdc - hdc
> > failed by itself, hdb was marked bad with raidsetfaulty). When the new disks
> > were added back to the set (with raidhotadd), only two of them became active
> > and one remained as a hot spare.
> > 
> > I there a way to reconfigure all three to become active again?
> 
> It looks from the details you gave that it never was a three-disk
> set.  It was only ever two-disk, possibly with a spare.

I can't dig up any hard evidence, but my impression is that it _was_ [UUU]
back when I had three disks in it the last time. That was in 2003, though.
It was created with the mentioned raidtab and 0.90 raidttools - shouldn't
that result into three active disks? 

Later one disk was removed, and later one broken. Now two have been added
back.

Anyway, you may well be right, perhaps I misremember and it was 2 active + 1
one spare.

> > This is kernel 2.4.32-rc1.
> 
> With a recent 2.6 kernel, you could
>    mdadm --grow --raid-disk=3 /dev/md1
> but with a 2.4 kernel, the only option is to re-create the array:
> 
>   mdadm --stop /dev/md1
>   mdadm --create /dev/md1 --level=1 --raid-disks=3 /dev/hda1 /dev/hdc1 missing
>   mdadm /dev/md1 --add /dev/hdb1

Ok, thanks. One of the md's is rootfs, so I'll have to wait for a reboot to
try that.



-- v -- 

v@iki.fi


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2005-11-16  3:23 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-11-15 13:26 3-disk RAID1 - only two disks active vherva
2005-11-15 23:01 ` Neil Brown
2005-11-16  3:23   ` Ville Herva

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).