will mdadm work with a raid created using raidtools

will mdadm work with a raid created using raidtools

[Andrew Nelson]

I may be a bit late in asking this because I already gave it a try but will
mdadm work with a raid created using raidtools?

//andy

Re: will mdadm work with a raid created using raidtools

[Neil Brown]

On Wednesday February 15, freeandy@gmail.com wrote:
> I may be a bit late in asking this because I already gave it a try but will
> mdadm work with a raid created using raidtools?

Yes.  There is no difference in the arrays created by raidtools or
mdadm.


NeilBrown

Re: will mdadm work with a raid created using raidtools

[Andrew Nelson]

Neil Brown wrote:
> On Wednesday February 15, freeandy@gmail.com wrote:
> 
>>I may be a bit late in asking this because I already gave it a try but will
>>mdadm work with a raid created using raidtools?
> 
> 
> Yes.  There is no difference in the arrays created by raidtools or
> mdadm.
> 
> 

Cool.  Now I just gotta figure out why I can't create the raid.  Sadly when this
system crashed I lost the old /etc/raidtab config so I really have no idea how
the raid was set up, but here is what I've done so far.


alfalfa:~# mdadm --examine /dev/hdg
/dev/hdg:
          Magic : a92b4efc
        Version : 00.90.00
           UUID : 04407130:1c19b2b9:2c573158:ee8ac467
  Creation Time : Sun Dec 21 16:19:13 2003
     Raid Level : raid1
   Raid Devices : 2
  Total Devices : 2
Preferred Minor : 0

    Update Time : Tue Jan 13 23:12:57 2004
          State : active
 Active Devices : 1
Working Devices : 2
 Failed Devices : 0
  Spare Devices : 1
       Checksum : 6ff770df - correct
         Events : 0.29


      Number   Major   Minor   RaidDevice State
this     2      34        0        2      spare   /dev/hdg

   0     0      33        1        0      active sync   /dev/hde1
   1     1       0        0        1      faulty removed
   2     2      34        0        2      spare   /dev/hdg
alfalfa:~# mdadm --examin /dev/hde
mdadm: No super block found on /dev/hde (Expected magic a92b4efc, got ffffffff)
alfalfa:~# mdadm --examin /dev/hde1
/dev/hde1:
          Magic : a92b4efc
        Version : 00.90.00
           UUID : 04407130:1c19b2b9:2c573158:ee8ac467
  Creation Time : Sun Dec 21 16:19:13 2003
     Raid Level : raid1
   Raid Devices : 2
  Total Devices : 2
Preferred Minor : 0

    Update Time : Tue Feb 14 21:44:46 2006
          State : active
 Active Devices : 2
Working Devices : 2
 Failed Devices : 0
  Spare Devices : 0
       Checksum : 73e5445d - correct
         Events : 0.42


      Number   Major   Minor   RaidDevice State
this     0      33        1        0      active sync   /dev/hde1

   0     0      33        1        0      active sync   /dev/hde1
   1     1      34        1        1      active sync   /dev/hdg1
alfalfa:~# mdadm --examin /dev/hdg1
/dev/hdg1:
          Magic : a92b4efc
        Version : 00.90.00
           UUID : 04407130:1c19b2b9:2c573158:ee8ac467
  Creation Time : Sun Dec 21 16:19:13 2003
     Raid Level : raid1
   Raid Devices : 2
  Total Devices : 2
Preferred Minor : 0

    Update Time : Tue Feb 14 21:44:46 2006
          State : active
 Active Devices : 2
Working Devices : 2
 Failed Devices : 0
  Spare Devices : 0
       Checksum : 73e54460 - correct
         Events : 0.42


      Number   Major   Minor   RaidDevice State
this     1      34        1        1      active sync   /dev/hdg1

   0     0      33        1        0      active sync   /dev/hde1
   1     1      34        1        1      active sync   /dev/hdg1



I then try to assemble the raid with

mdadm --assemble /dev/md0 /dev/hdg1 /dev/hde1


This appears to work fine but then the computer locks up.  With this info in syslog.


Feb 14 21:57:19 localhost kernel: md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DI
SKS=27
Feb 14 21:57:19 localhost kernel: md: md0 stopped.
Feb 14 21:57:51 localhost kernel: md: md0 stopped.
Feb 14 21:57:52 localhost kernel: md: bind<hdg1>
Feb 14 21:57:52 localhost kernel: md: bind<hde1>
Feb 14 21:57:52 localhost kernel: md: md0: raid array is not clean -- starting b
ackground reconstruction
Feb 14 21:57:52 localhost kernel: md: raid1 personality registered as nr 3
Feb 14 21:57:52 localhost kernel: raid1: raid set md0 active with 2 out of 2 mir
rors
Feb 14 21:57:52 localhost kernel: md: syncing RAID array md0
Feb 14 21:57:52 localhost kernel: md: minimum _guaranteed_ reconstruction speed:
 1000 KB/sec/disc.
Feb 14 21:57:52 localhost kernel: md: using maximum available idle IO bandwith (
but not more than 200000 KB/sec) for reconstruction.
Feb 14 21:57:52 localhost kernel: md: using 128k window, over a total of 1953583
36 blocks.
Feb 14 21:58:36 localhost kernel: hde: dma_timer_expiry: dma status == 0x21
Feb 14 21:58:46 localhost kernel: hde: DMA timeout error
Feb 14 21:58:46 localhost kernel: hde: dma timeout error: status=0x51 { DriveRea
dy SeekComplete Error }
Feb 14 21:58:46 localhost kernel: hde: dma timeout error: error=0x40 { Uncorrect
ableError }, LBAsect=1594001, high=0, low=1594001, sector=1593535
Feb 14 21:58:46 localhost kernel: end_request: I/O error, dev hde, sector 159353
5

Does anyone have any idea what I'm doing wrong?

Thanks much

//andy

Re: will mdadm work with a raid created using raidtools

[Frank Blendinger]

[-- Attachment #1: Type: text/plain, Size: 949 bytes --]

On Thu, Feb 16, 2006 at 09:15:27AM -0600, Andrew Nelson wrote:
> Feb 14 21:58:36 localhost kernel: hde: dma_timer_expiry: dma status == 0x21
> Feb 14 21:58:46 localhost kernel: hde: DMA timeout error
> Feb 14 21:58:46 localhost kernel: hde: dma timeout error: status=0x51 { DriveRea
> dy SeekComplete Error }
> Feb 14 21:58:46 localhost kernel: hde: dma timeout error: error=0x40 { Uncorrect
> ableError }, LBAsect=1594001, high=0, low=1594001, sector=1593535
> Feb 14 21:58:46 localhost kernel: end_request: I/O error, dev hde, sector 159353
> 5
> 
> Does anyone have any idea what I'm doing wrong?

It's probably not your fault - blame /dev/hde! This sounds like a bad
error on the disk - you should really get a new one, and try to copy
/dev/hde to the new disk (with dd_rescue for example). This _might_ save
the data.

Then you can try to create the array with the new disk and hope that
it will work.


Greetings,
Frank

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

Re: will mdadm work with a raid created using raidtools

[Andrew Nelson]

Frank Blendinger wrote:
> On Thu, Feb 16, 2006 at 09:15:27AM -0600, Andrew Nelson wrote:
> 
>>Feb 14 21:58:36 localhost kernel: hde: dma_timer_expiry: dma status == 0x21
>>Feb 14 21:58:46 localhost kernel: hde: DMA timeout error
>>Feb 14 21:58:46 localhost kernel: hde: dma timeout error: status=0x51 { DriveRea
>>dy SeekComplete Error }
>>Feb 14 21:58:46 localhost kernel: hde: dma timeout error: error=0x40 { Uncorrect
>>ableError }, LBAsect=1594001, high=0, low=1594001, sector=1593535
>>Feb 14 21:58:46 localhost kernel: end_request: I/O error, dev hde, sector 159353
>>5
>>
>>Does anyone have any idea what I'm doing wrong?
> 
> 
> It's probably not your fault - blame /dev/hde! This sounds like a bad
> error on the disk - you should really get a new one, and try to copy
> /dev/hde to the new disk (with dd_rescue for example). This _might_ save
> the data.
> 
> Then you can try to create the array with the new disk and hope that
> it will work.
> 


I thought the whole idea of a raid 1 was that if one drive went bad I could just
plug a new drive in and the raid would rebuild without problems.

What have I done wrong so that this isn't possible?

//andy

Re: will mdadm work with a raid created using raidtools

[Frank Blendinger]

[-- Attachment #1: Type: text/plain, Size: 1020 bytes --]

On Thu, Feb 16, 2006 at 11:31:06AM -0600, Andrew Nelson wrote:
> > It's probably not your fault - blame /dev/hde! This sounds like a bad
> > error on the disk - you should really get a new one, and try to copy
> > /dev/hde to the new disk (with dd_rescue for example). This _might_ save
> > the data.
> > 
> > Then you can try to create the array with the new disk and hope that
> > it will work.
> 
> 
> I thought the whole idea of a raid 1 was that if one drive went bad I could just
> plug a new drive in and the raid would rebuild without problems.

That is certainly right. I just wanted to tell you, that your /dev/hde
probably has a serious hardware error, and that you should replace it!

Of course you can just throw it out, put in a new drive, rebuild the
array with your other drive and the new one, and then resync. This
should work just fine.

I guess my first answer was quite confusing, sorry. It's absolutely not
necessary to copy the old hde with dd_rescue.


Greetings,
Frank

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

Re: will mdadm work with a raid created using raidtools

[Andrew Nelson]

Frank Blendinger wrote:
> On Thu, Feb 16, 2006 at 11:31:06AM -0600, Andrew Nelson wrote:
> 
>>>It's probably not your fault - blame /dev/hde! This sounds like a bad
>>>error on the disk - you should really get a new one, and try to copy
>>>/dev/hde to the new disk (with dd_rescue for example). This _might_ save
>>>the data.
>>>
>>>Then you can try to create the array with the new disk and hope that
>>>it will work.
>>
>>
>>I thought the whole idea of a raid 1 was that if one drive went bad I could just
>>plug a new drive in and the raid would rebuild without problems.
> 
> 
> That is certainly right. I just wanted to tell you, that your /dev/hde
> probably has a serious hardware error, and that you should replace it!
> 
> Of course you can just throw it out, put in a new drive, rebuild the
> array with your other drive and the new one, and then resync. This
> should work just fine.
> 
> I guess my first answer was quite confusing, sorry. It's absolutely not
> necessary to copy the old hde with dd_rescue.
> 
> 
> Greetings,
> Frank

Is it possible to assemble the array with just /dev/hdg?  I swear I saw an
option somewhere that allowed just such a thing but now I can't find it anywhere.

//andy

Re: will mdadm work with a raid created using raidtools

[Andrew Nelson]

Andrew Nelson wrote:
> Frank Blendinger wrote:
> 
>>On Thu, Feb 16, 2006 at 11:31:06AM -0600, Andrew Nelson wrote:
>>
>>
>>>>It's probably not your fault - blame /dev/hde! This sounds like a bad
>>>>error on the disk - you should really get a new one, and try to copy
>>>>/dev/hde to the new disk (with dd_rescue for example). This _might_ save
>>>>the data.
>>>>
>>>>Then you can try to create the array with the new disk and hope that
>>>>it will work.
>>>
>>>
>>>I thought the whole idea of a raid 1 was that if one drive went bad I could just
>>>plug a new drive in and the raid would rebuild without problems.
>>
>>
>>That is certainly right. I just wanted to tell you, that your /dev/hde
>>probably has a serious hardware error, and that you should replace it!
>>
>>Of course you can just throw it out, put in a new drive, rebuild the
>>array with your other drive and the new one, and then resync. This
>>should work just fine.
>>
>>I guess my first answer was quite confusing, sorry. It's absolutely not
>>necessary to copy the old hde with dd_rescue.
>>
>>
>>Greetings,
>>Frank
> 
> 
> Is it possible to assemble the array with just /dev/hdg?  I swear I saw an
> option somewhere that allowed just such a thing but now I can't find it anywhere.
> 
> //andy
> 

Never mind I figured it out.

mdadm -AR /dev/md0 /dev/hdg1

So far this seems to work.

Thanks for all the help.  I'll replace hde ASAP :).


//andy