Re: mdadm 3.2.4 problem

[NeilBrown <neilb@suse.de>]

[-- Attachment #1: Type: text/plain, Size: 3448 bytes --]

On Sat, 12 May 2012 17:30:16 +0400 nowhere <nowhere@hakkenden.ath.cx> wrote:

> Hello,
> 
> I have a RAID5 array, which was created with mdadm 3.2.3
> Here is one of its member's superblock:
> 
> /dev/sdl:
>           Magic : a92b4efc
>         Version : 1.2
>     Feature Map : 0x0
>      Array UUID : cc0fbdd2:0feade15:963fe546:13a072fa
>            Name : heimdall:0  (local to host heimdall)
>   Creation Time : Sun May  6 10:58:30 2012
>      Raid Level : raid5
>    Raid Devices : 8
> 
>  Avail Dev Size : 3907027120 (1863.02 GiB 2000.40 GB)
>      Array Size : 13674590720 (13041.11 GiB 14002.78 GB)
>   Used Dev Size : 3907025920 (1863.02 GiB 2000.40 GB)
>     Data Offset : 2048 sectors
>    Super Offset : 8 sectors
>           State : clean
>     Device UUID : 64962294:3699d869:510bef35:9d2fd78d
> 
>     Update Time : Sat May 12 17:03:28 2012
>        Checksum : df1714ca - correct
>          Events : 96
> 
>          Layout : left-symmetric
>      Chunk Size : 512K
> 
>    Device Role : Active device 6
>    Array State : AAAAAAA. ('A' == active, '.' == missing)
> 
> Some time ago mdadm have been upgraded to 3.2.4

That was probably about 1 day ago :-)  3.2.4 was only released on Wednesday.

> 
> So, today I have replaced failed drive, but could not add it:
> mdadm --manage /dev/md0 --add /dev/sdm 
> mdadm: add new device failed for /dev/sdm as 8: Invalid argument
> 
> And there are lines in dmesg:
> md: sdm does not have a valid v1.2 superblock, not importing!
> md: md_import_device returned -22

That's bad.  Looks like I'll need to do a 3.2.5 fairly quickly as that will
affect lots of people.
I've pushed out a change that fixes it for me:
http://neil.brown.name/git?p=mdadm;a=commitdiff;h=d9751e06a601b5576b1b9e2c8126584083110ca5

Thanks for the report.

NeilBrown



> 
> mdadm creates new superblock on sdm:
> /dev/sdm:
>           Magic : a92b4efc
>         Version : 1.2
>     Feature Map : 0x0
>      Array UUID : cc0fbdd2:0feade15:963fe546:13a072fa
>            Name : heimdall:0  (local to host heimdall)
>   Creation Time : Sun May  6 10:58:30 2012
>      Raid Level : raid5
>    Raid Devices : 8
> 
>  Avail Dev Size : 3906767024 (1862.89 GiB 2000.26 GB)
>      Array Size : 13674590720 (13041.11 GiB 14002.78 GB)
>   Used Dev Size : 3907025920 (1863.02 GiB 2000.40 GB)
>     Data Offset : 262144 sectors
>    Super Offset : 8 sectors
>           State : clean
>     Device UUID : ecaaf8af:f41c6ca1:8190b6f8:37b53a06
> 
>     Update Time : Sat May 12 16:46:20 2012
>        Checksum : 6dabb37a - correct
>          Events : 0
> 
>          Layout : left-symmetric
>      Chunk Size : 512K
> 
>    Device Role : spare
>    Array State : AAAAAAA. ('A' == active, '.' == missing)
> 
> New direve's size is equals to other drive's sizes (checked with
> blockdev --getsz)
> 
> The only difference between superblocks created by diferent mdadm
> versions is data offset. Old one has 2048 sector offset, new one -
> 262144. I think, this leads to available device size to be smaller with
> new superblock and inability to add drive to the array.
> 
> Downgrading mdadm allowed me to add device into array.
> 
> --
> 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


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]