mdadm bug - inconsistent output in -D mode

DISCUSSION Peter Rabbitson 2 participants 2 messages Jan 15, 2009 – Jan 18, 2009 MD RAID

lore

Share & tools

lore

Peter Rabbitson

author

2009-01-15 11:50 UTC #0 lore raw

Hi,

I suppose this is a remnant of the effort to convert all sector counts to bytes. Consider this output:

root@Thesaurus:~# mdadm -E /dev/sda1
/dev/sda1:
          Magic : a92b4efc
        Version : 1.1
    Feature Map : 0x0
     Array UUID : 6b11b1ba:78985745:b320fc1a:1db68bcf
           Name : Thesaurus:Crypta  (local to host Thesaurus)
  Creation Time : Sat Mar  8 16:33:41 2008
     Raid Level : raid5
   Raid Devices : 4

Avail Dev Size : 578130828 (275.67 GiB 296.00 GB) <--- Correct:

     Array Size : 1730162688 (825.01 GiB 885.84 GB)  <--- sectors(bytes)
  Used Dev Size : 576720896 (275.00 GiB 295.28 GB)
    Data Offset : 264 sectors
   Super Offset : 0 sectors
          State : active
    Device UUID : 85e0e62b:56129e5f:b0d82e78:2ace7d37

    Update Time : Thu Jan 15 12:42:53 2009
       Checksum : 61a32529 - correct
         Events : 764390

         Layout : left-symmetric
     Chunk Size : 2048K

    Array Slot : 5 (failed, failed, 2, 3, 0, 1)  <--- by the way: wtf?

Array State : uUuu 2 failed <--- ditto

root@Thesaurus:~# mdadm -D /dev/md5
/dev/md5:
        Version : 01.01
  Creation Time : Sat Mar  8 16:33:41 2008
     Raid Level : raid5
     Array Size : 865081344 (825.01 GiB 885.84 GB) <--- bytes(bytes)
  Used Dev Size : 576720896 (550.00 GiB 590.56 GB) <--- sectors(2xbytes)
   Raid Devices : 4
  Total Devices : 4
Preferred Minor : 5
    Persistence : Superblock is persistent

Update Time : Thu Jan 15 12:42:29 2009
State : active
Active Devices : 4
Working Devices : 4
Failed Devices : 0
Spare Devices : 0

Layout : left-symmetric
Chunk Size : 2048K

Name : Thesaurus:Crypta (local to host Thesaurus) UUID : 6b11b1ba:78985745:b320fc1a:1db68bcf
Events : 764391

Number Major Minor RaidDevice State
4 8 67 0 active sync /dev/sde3
5 8 1 1 active sync /dev/sda1
2 8 19 2 active sync /dev/sdb3
3 8 35 3 active sync /dev/sdc3

root@Thesaurus:~# mdadm -V
mdadm - v2.6.7.1 - 15th October 2008
root@Thesaurus:~# uname -a
Linux Thesaurus 2.6.24.7.th1 #1 PREEMPT Sun May 11 20:18:05 CEST 2008
i686 GNU/Linux

P.S. I know my kernel is old, but I suspect it's a mdadm problem.

Michał Przyłuski

2009-01-18 20:04 UTC #1 lore raw

2009/1/15 Peter Rabbitson <[email protected]>:

Quoted 4 lines

Hi,
I suppose this is a remnant of the effort to convert all sector counts to bytes. Consider this output:

Hello,
I have the same issue (not a biggie anyway) with a 6*750G raid6. I guess it might be
related to 1.x superblocks though, I don't recall that happening with my old 0.9 stuff.

[root@kylie kotek]# ~kotek/mdadm-2.6.4/mdadm --detail /dev/md1 /dev/md1:

        Version : 01.01.03

Creation Time : Mon Nov 10 19:59:48 2008
Raid Level : raid6
Array Size : 2930294784 (2794.55 GiB 3000.62 GB)
Used Dev Size : 1465147392 (698.64 GiB 750.16 GB)
Raid Devices : 6
Total Devices : 6
Preferred Minor : 1
Persistence : Superblock is persistent

Update Time : Sun Jan 18 20:44:35 2009
State : clean
Active Devices : 6
Working Devices : 6
Failed Devices : 0
Spare Devices : 0

Chunk Size : 1024K

Name : 1
UUID : 4eaba5a9:cb767b93:c73450fe:c1dc27c9
Events : 487870

Number Major Minor RaidDevice State
0 8 48 0 active sync /dev/sdd
1 8 64 1 active sync /dev/sde
4 8 96 2 active sync /dev/sdg
5 8 80 3 active sync /dev/sdf
7 8 32 4 active sync /dev/sdc
6 8 16 5 active sync /dev/sdb
[root@kylie kotek]# ~kotek/mdadm-2.6.7/mdadm --detail /dev/md1 /dev/md1:

        Version : 01.01

Creation Time : Mon Nov 10 19:59:48 2008
Raid Level : raid6
Array Size : 2930294784 (2794.55 GiB 3000.62 GB)
Used Dev Size : 1465147392 (1397.27 GiB 1500.31 GB)
Raid Devices : 6
Total Devices : 6
Preferred Minor : 1
Persistence : Superblock is persistent

Update Time : Sun Jan 18 20:44:39 2009
State : clean
Active Devices : 6
Working Devices : 6
Failed Devices : 0
Spare Devices : 0

Chunk Size : 1024K

Name : 1
UUID : 4eaba5a9:cb767b93:c73450fe:c1dc27c9
Events : 487870

and --examine in all cases gives (what seems to be) correct data, that is

Avail Dev Size : 1465148904 (698.64 GiB 750.16 GB)

     Array Size : 5860589568 (2794.55 GiB 3000.62 GB)

Used Dev Size : 1465147392 (698.64 GiB 750.16 GB)

Quoted 1 line

root@Thesaurus:~# mdadm -E /dev/sda1

[snip]

Quoted 3 lines

   Array Slot : 5 (failed, failed, 2, 3, 0, 1)  <--- by the way: wtf?
Array State : uUuu 2 failed <--- ditto

    Array Slot : 6 (0, 1, failed, failed, 2, 3, 5, 4)

Array State : uuuuuU 2 failed

That's how it looks here, and I guess that's perfectly OK. There was a post on the list
by Neil regarding that matter some weeks ago. I believe those 2*fails are due to a way that raid5/6 is created - as a degraded array, which will rebuild to the nth disk.

However, in my case, it might be caused by the fact the array was created with 1 really
missing drive (i.e. it contained 3, then 4, then 6 drives).

Oh, got it:
date 20 December 2008 02:16
subject Re: can you help explain some --examine output to me?

Quoted 2 lines

root@Thesaurus:~# mdadm -V
mdadm - v2.6.7.1 - 15th October 2008

here 2.6.2 and 2.6.4 are correct, and 2.6.7 not quite.

Well ok, correct in terms of "human readable output", hard to say if raw data is
correct, all depends if it's sectors or bytes, as Peter mentioned.

[kotek@kylie ~]$ uname -a
Linux kylie 2.6.23-0.214.rc8.git2.fc8 #1 SMP Fri Sep 28 17:10:49 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux

Quoted 5 lines

root@Thesaurus:~# uname -a
Linux Thesaurus 2.6.24.7.th1 #1 PREEMPT Sun May 11 20:18:05 CEST 2008
i686 GNU/Linux
P.S. I know my kernel is old, but I suspect it's a mdadm problem.

Mine is older! ;-)

Greets,
Mike