Re: persistent removed MD referrence on reboot

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

 



Thank you Adam and linbloke.

I ran "sudo mdadm --manage --stop /dev/md127" to clear out /proc/mdstat

Then "sudo mdadm --zero-superblock /dev/sdc1" twice; no return info on
the first one,
but the second had a response similar to "No superblock found".

Rebooted and the phantom raid device is gone.

Skip

On Wed, Aug 1, 2012 at 1:10 AM, linbloke <linbloke@xxxxxxxxxxx> wrote:
>
> On 1/08/12 11:43 AM, Skip Coombe wrote:
>>
>> Thanks for the quick response.
>>
>> I did "dd -if=/dev/zero -of=/dev/sdc" and modified the the partition to
>> 83.
>>
>> Today I followed your advice to investigate initrd. I deleted all the
>> ARRAY specs from /etc/mdadm/mdadm.conf (they were
>> from the time of installing 12-4, including the md2 ARRAY), wrote the
>> new initrd, and rebooted.
>
>
> I think Adam led you to the right path. Your reference to 12-4 leads me to
> guess that you're running Ubuntu 12.04? For me on these systems, I would do
> the following:
>
> - zero superblocks:
>
> sudo mdadm --zero-superblock /dev/sdc1
>
> running that commands multiple times until they return "No superblock
> found" or similar message. Once you get that response, there are no mdadm
> superblocks remaining.
>
> - update /etc/mdadm/mdadm.conf to remove references to old md arrays.
> - update initramfs:
>
> update-initramfs -u
>
>
>
> From there you should be able to reboot and find only those arrays
> specified in mdadm.conf (also arrays auto-built by the kernel from devices
> with partitions of type fd). If you're still not getting the expected
> outcome, pls send the output of "parted -l" and /etc/mdadm/mdadm.conf, uname
> -a, mdadm -V
>
> HTH
>
> j
>
>
>> The reference to md2 went away, but now I see
>> ======================
>> $ cat /proc/mdstat
>> Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5]
>> [raid4] [raid10]
>> md127 : inactive sdc[1](S)
>>        488385560 blocks super 1.2
>>
>> unused devices:<none>
>> ======================
>>
>> I will repeat after "rm -rf /etc/mdadm" from the initrd dir tree. If
>> that does not work I will probably just reinstall 12-4. I hate to do
>> all the config again, but I've burned too many hours on this already.
>> There are no superblocks left so I hope to get a clean start.
>>
>> n.b. md127 seems to be the default array, but still sdc is identified.
>> There must be more info buried in the initrd.
>>
>> Thanks for the good advice, but I am still missing something.
>>
>> Skip
>>
>> On Mon, Jul 30, 2012 at 9:26 PM, Adam Goryachev
>> <adam@xxxxxxxxxxxxxxxxxxxxxx>  wrote:
>>>
>>> On 07/31/2012 11:02 AM, Skip Coombe wrote:
>>>>
>>>> I have an md entry referring to a badly removed RAID1 device (by me)
>>>> after
>>>> the associated device crashed (can't afford a replacement drive at this
>>>> time).
>>>> Despite attempts to remove the reference, after a reboot I see:
>>>>
>>>> ================================
>>>> $ cat /proc/mdstat
>>>> Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5]
>>>> [raid4]
>>>> [raid10]
>>>> md2 : inactive sdc[1](S)
>>>>         488385560 blocks super 1.2
>>>>
>>>> unused devices:<none>
>>>> ================================
>>>>
>>>> I am trying to repurpose the good drive as a not-RAID device. I have
>>>> done
>>>> sudo mdadm --zero-superblock /dev/sdc
>>>> and
>>>> sudo mdadm --zero-superblock --force /dev/sdc
>>>> and even low level formatted /dev/sdc
>>>
>>> What do you mean "low level formatted"? Have you done a
>>> dd if=/dev/zero of=/dev/sdc
>>>
>>>> ================================
>>>> $ ls /etc/mdadm/mdadm.conf
>>>> ls: cannot access /etc/mdadm/mdadm.conf: No such file or directory
>>>> ================================
>>>
>>> Did you update the initrd file? This file probably contains a copy of
>>> the
>>> mdadm.conf
>>>
>>> Also, did you modify the partition type, from fd to 83
>>> I would suggest running the above dd command, which will erase ALL
>>> traces of
>>> anything, including any data on the drive (so back it up elsewhere
>>> first).
>>>
>>> Most likely, the md code looks at a couple of different places on the
>>> drive
>>> for the md information, so when you run the zero-superblock, it is only
>>> erasing one of the possible locations. See man mdadm, especially the
>>> --metadata section.
>>>
>>> Regards,
>>> Adam
>>
>>
>>



--
Skip Coombe
skip.coombe@xxxxxxxxx
919.442.VLSI
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux RAID Wiki]     [ATA RAID]     [Linux SCSI Target Infrastructure]     [Linux Block]     [Linux IDE]     [Linux SCSI]     [Linux Hams]     [Device Mapper]     [Device Mapper Cryptographics]     [Kernel]     [Linux Admin]     [Linux Net]     [GFS]     [RPM]     [git]     [Yosemite Forum]


  Powered by Linux