[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: mdadm raid5 giving me heartburn.....



On Sun, Sep 14, 2008 at 9:43 AM, Michael Habashy <mjh2000@gmail.com> wrote:
> further, when i try to execute any of these mdadm commands on the
> /dev/mdX devices, I get errors like:
> mdadm: fail to stop array /dev/md0: Device or resource busy
>
> How can i get past them ?? did i miss a step ???
>
> I keep on seeing debian LiveCD , what is that ? how will this help ??
> is the is the regular debian cd, but in rescue mode ???
>
> the layout of my drives is such:
>
> 500g /dev/sda1,/dev/sda5,/dev/sda6
> 500g /dev/sdb1,/dev/sdb5,/dev/sdb6
> 400g /dev/sdc5
> 400g /dev/sdd5
>
> md0 = 300meg = /boot
> md1 = 10gig = swap
> md2 = 400gig =/
>
> Can you give a good faq..that is really helpfull...i googled a lot and
> came up empty
> If you give me a step by step - i would also appriciate it..because i
> think i am missing a step.
> thanks for your help.
>
>
>
> On Sun, Sep 14, 2008 at 8:48 AM, Michael Habashy <mjh2000@gmail.com> wrote:
>> On Sun, Sep 14, 2008 at 8:10 AM, Alex Samad <alex@samad.com.au> wrote:
>>> On Sat, Sep 13, 2008 at 11:39:47PM -0400, Michael Habashy wrote:
>>>> the current state of my system -- it is up and running but i do not
>>>> belive for long:
>>>>
>>>> root@rider:~# cat /proc/mdstat
>>>> Personalities : [raid1] [raid6] [raid5] [raid4]
>>>> md1 : active raid5 sda5[3](F) sdb5[4](F) sdd5[2] sdc5[1]
>>>>       781417344 blocks level 5, 64k chunk, algorithm 2 [3/2] [_UU]
>>>>
>>>> md2 : active raid1 sda6[0]
>>>>       9767424 blocks [2/1] [U_]
>>>>
>>>> md0 : active raid1 sda1[0]
>>>>       289024 blocks [2/1] [U_]
>>>
>>>
>>> looks like you lost some of your mirror sets, what did you mirror sda6
>>> with ?  just add it back to the raid set (find out why it dropped out),
>>> mdadm -A /dev/md2 /dev/<what ever the device was>
>>>
>>> Plus sdb seems to have failed in the raid5.
>>>
>>> My guess is sdb has gone faulty, try running some smart tests against
>>> it. You will want to be carefull, if you loose another disk you will
>>> loose your raid5 (and if its sda your raid1's.
>>>
>>> Find out whats wrong with sdb and fix it, either replace the drive,
>>> check the cables or power or ....
>>>
>>> It should survive a reboot, presuming sda is hd0
>>>
>>>>
>>>>
>>>>
>>>> It looks like i got some kind of hardware/software issue --this is a
>>>> newly built raid5 system.
>>>>
>>>> Device md1 is my main / root ssystem.
>>>>
>>>> How can i fix this ?? I do not want to reboot becasue i am afraid that
>>>> it will not come up...
>>>>
>>>> I need magic or lots of prays...please try to provide both...
>>>>
>>>> thanks
>>>> mjh
>>>>
>>>>
>>>> --
>>>> To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
>>>> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
>>>>
>>>>
>>>
>>> --
>>> "I don't want to win? If that were the case, why the heck am I on the bus sixteen hours a day, shaking thousands of hands, giving hundreds of speeches, getting pillared in the press and cartoons and still staying on message to win?"
>>>
>>>        - George W. Bush
>>> 02/28/2000
>>> as quoted in Newsweek
>>>
>>> -----BEGIN PGP SIGNATURE-----
>>> Version: GnuPG v1.4.9 (GNU/Linux)
>>>
>>> iEYEARECAAYFAkjM/0oACgkQkZz88chpJ2PZ3ACgpbS0ZRGk5AQTMsUAZKS//1Q/
>>> oJoAoPM+f5DlADAbmMtrKxQhzhn43bhQ
>>> =Jj15
>>> -----END PGP SIGNATURE-----
>>>
>>>
>> Thanks for the quick response.
>>
>> I think sda and sdb are my 500gig harddrives that have swap and /boot
>> on them mirrored.
>> The raid5 which is / root is going against 4 drives.  The 2 (500 gig)
>> drives and 2 400 gig sata drives.
>>
>> Can you tell me what specific utilities to run against these drives ??
>> I have  googled alot for recovery of mdadm raid drives...i could not
>> find anything..except telling me how to remove and add drives.
>>
>> do not know how to use smartmonitor utilies..do you know of a good faq
>> that i can look at??
>>
>> thanks
>> mjh
>>
>

Just for your information, the current state of my drives are :

root@rider:~# cat /proc/mdstat
Personalities : [raid1] [raid6] [raid5] [raid4]
md1 : active raid5 sdc5[1] sdd5[2]
      781417344 blocks level 5, 64k chunk, algorithm 2 [3/2] [_UU]

md2 : active raid1 sda6[0]
      9767424 blocks [2/1] [U_]

md0 : active raid1 sda1[0]
      289024 blocks [2/1] [U_]

unused devices: <none>


Reply to: