Re: mdadm: making a spare actie

[Jon Buckingham <jbuckingham@blueyonder.co.uk>]

[-- Attachment #1: Type: text/plain, Size: 5153 bytes --]

Neil Brown wrote:
> On Thursday June 19, jbuckingham@blueyonder.co.uk wrote:
>> Neil Brown wrote:
>>> On Thursday June 19, jbuckingham@blueyonder.co.uk wrote:
>>>> I have also done
>>>> mdadm /dev/md0 -a /dev/sdb5
>>>> and this results in a recovery...
>>>>
>>>> nas:~ # cat /proc/mdstat
>>>> Personalities : [raid6] [raid5] [raid4]
>>>> md0 : active raid5 sdb5[4] sda5[0] sdd5[3] sdc5[2]
>>>>        733142016 blocks level 5, 64k chunk, algorithm 2 [4/3] [U_UU]
>>>>        [=>...................]  recovery =  7.3% (17900780/244380672) finish=174.1min speed=21666K/sec
>>>>
>>>> unused devices: <none>
>>>>
>>>> Which I've been through before, but still ends up as a spare.
>>> That suggests that it hits some IO error during recovery and aborts.
>>>
>>> Are there any kernel log messages during the time that it is
>>> recovering?
>>>
>>> NeilBrown
>>>
>>>
>> No.
>> After the "add" completed, and a reboot it seems it is still a
>> "spare".
>> Strange.
> 
> What would be interesting to see is the --examine output and the dmesg
> just as the recovery after the add has completed.  i.e. just before
> the reboot.
> 
> The dmesg you have included is after the reboot.  It confirms that
> sdb5 is non-refresh, presumably the event count is behind for some
> reason (as can be seen from the --examine output you send in the first
> email).  However it doesn't contain any hint as to why.
> 
> NeilBrown
> 
> 
>> Then from dmesg:
>>
>> device-mapper: ioctl: 4.11.0-ioctl (2006-10-12) initialised: dm-devel@redhat.com
>> md: md0 stopped.
>> md: bind<sdc5>
>> md: bind<sdd5>
>> md: bind<sdb5>
>> md: bind<sda5>
>> md: kicking non-fresh sdb5 from array!
>> md: unbind<sdb5>
>> md: export_rdev(sdb5)
>> raid5: automatically using best checksumming function: pIII_sse
>>     pIII_sse  :  5640.000 MB/sec
>> raid5: using function: pIII_sse (5640.000 MB/sec)
>>
>> raid5: device sda5 operational as raid disk 0
>> raid5: device sdd5 operational as raid disk 3
>> raid5: device sdc5 operational as raid disk 2
>> raid5: allocated 4204kB for md0
>> raid5: raid level 5 set md0 active with 3 out of 4 devices, algorithm 2
>> RAID5 conf printout:
>>   --- rd:4 wd:3
>>   disk 0, o:1, dev:sda5
>>   disk 2, o:1, dev:sdc5
>>   disk 3, o:1, dev:sdd5
>>
>> I am tempted to rebuild the whole thing now, since I have tried
>> quite a few variations and not solved it. There must be some deeper rooted problem that
>> is causing this issue on the disk.
>>
>> Thanks again,
>>
>> Jon B
> 
> 
It is currently rebuilding (I had shutdown before it completed yesterday,
so it is continuing after booting now),
and here is the information requested (I'll forward
on the results after the partitioning when it completes
in 3 hours time, or when I get home again!).

nas: # mdadm -E /dev/sda5 (a "good" partition)
----------------------------------------------
/dev/sda5:
           Magic : a92b4efc
         Version : 00.90.03
            UUID : b54e46e1:b6a6e6ea:3ae5a5a5:04e207e4
   Creation Time : Fri Aug  4 22:42:14 2006
      Raid Level : raid5
   Used Dev Size : 244380672 (233.06 GiB 250.25 GB)
      Array Size : 733142016 (699.18 GiB 750.74 GB)
    Raid Devices : 4
   Total Devices : 4
Preferred Minor : 0

     Update Time : Fri Jun 20 09:32:25 2008
           State : clean
  Active Devices : 3
Working Devices : 4
  Failed Devices : 1
   Spare Devices : 1
        Checksum : f11d23b5 - correct
          Events : 0.3796196

          Layout : left-symmetric
      Chunk Size : 64K

       Number   Major   Minor   RaidDevice State
this     0       8        5        0      active sync   /dev/sda5

    0     0       8        5        0      active sync   /dev/sda5
    1     1       0        0        1      faulty removed
    2     2       8       37        2      active sync   /dev/sdc5
    3     3       8       53        3      active sync   /dev/sdd5
    4     4       8       21        4      spare   /dev/sdb5

nas:# mdadm -E /dev/sdb5 (the "bad/spare" partition)
----------------------------------------------------
/dev/sdb5:
           Magic : a92b4efc
         Version : 00.90.03
            UUID : b54e46e1:b6a6e6ea:3ae5a5a5:04e207e4
   Creation Time : Fri Aug  4 22:42:14 2006
      Raid Level : raid5
   Used Dev Size : 244380672 (233.06 GiB 250.25 GB)
      Array Size : 733142016 (699.18 GiB 750.74 GB)
    Raid Devices : 4
   Total Devices : 4
Preferred Minor : 0

     Update Time : Fri Jun 20 09:32:25 2008
           State : clean
  Active Devices : 3
Working Devices : 4
  Failed Devices : 1
   Spare Devices : 1
        Checksum : f11d23c7 - correct
          Events : 0.3796196

          Layout : left-symmetric
      Chunk Size : 64K

       Number   Major   Minor   RaidDevice State
this     4       8       21        4      spare   /dev/sdb5

    0     0       8        5        0      active sync   /dev/sda5
    1     1       0        0        1      faulty removed
    2     2       8       37        2      active sync   /dev/sdc5
    3     3       8       53        3      active sync   /dev/sdd5
    4     4       8       21        4      spare   /dev/sdb5


There is nothing in /var/log/messages since the reboot.

Cheers

Jon B

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/x-pkcs7-signature, Size: 3281 bytes --]