Re: trouble adding spare

[Max Waterman <davidmaxwaterman@fastmail.co.uk>]

Someone else's post prompted me to try :

$ sudo /sbin/mdadm --detail --scan
ARRAY /dev/md2 level=raid5 num-devices=6 metadata=00.90 spares=2 
UUID=15bfec75:595ac793:0914f8ee:862effd8

and use that in mdadm.conf instead of the output of :

~$ sudo /sbin/mdadm --examine --scan
ARRAY /dev/md2 level=raid5 num-devices=6 
UUID=15bfec75:595ac793:0914f8ee:862effd8
   spares=2

This fixes my problem of the second spare drive not being added. I'm not 
sure why.

However, mdadm now gives me a 'message' :

mdadm: metadata format 00.90 unknown, ignored.

What's this about? Anything to worry about?

Max.
Max Waterman wrote:
> Can no one provide any enlightenment on this issue?
>
> Max.
>
> Max Waterman wrote:
>> Justin Piszcz wrote:
>>>
>>>
>>> On Wed, 14 Jan 2009, Max Waterman wrote:
>>>
>>>> Max Waterman wrote:
>>>>>>
>>>>>> If that does not work, clean the drive out manually after 
>>>>>> removing it as a spare from the array:
>>>>>> dd if=/dev/zero of=/dev/sdi bs=1M
>>>>> This is taking forever
>>>> I stopped it after about half an hour, hoping that it had made 
>>>> garbage of whatever was the problem.
>>>>
>>>> I then added it into the array and it worked as before.
>>>>
>>>> However, it still wasn't added when I rebooted.
>>>>
>>>> ...but, there is something different. Now, when I --examine 
>>>> /dev/sdi, it gives output for the whole array including 2 spares, 
>>>> which is the end result I expect :
>>>>
>>>> Active Devices : 6
>>>> Working Devices : 8
>>>> Failed Devices : 0
>>>> Spare Devices : 2
>>>>
>>>> However, --detail /dev/md2 still denies /dev/sdi exists :
>>>>
>>>> Active Devices : 6
>>>> Working Devices : 7
>>>> Failed Devices : 0
>>>> Spare Devices : 1
>>>>
>>>> What's up?
>>>>
>>>> Max.
>>>>
>>>> -- 
>>>> 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
>>>>
>>>
>>> The superblock is at the end of the disk for 0.90.00 superblocks, so 
>>> you would need to write to end of the disk or just let dd clear out 
>>> the entire disk.  Honestly though that may not fix the root cause of 
>>> the problem, we may need to wait for others on the list and/or Neil 
>>> to check this one out. It seems very strange, can you post the 
>>> regular output that people ask for so once they see your mail, they 
>>> will have (hopefully) most of what they need to diagnose the problem.
>>>
>>> 1. mdadm -D /dev/mdX
>>> 2. cat /proc/mdstat
>>> 3. mdadm -Esvvv
>>> 4. uname -a
>> 5. mdadm -E /dev/sdi
>>
>> OK, I uploaded the output of the above commands to here :
>>
>> <http://reality.sgiweb.org/maxw/tmp/md2Details/>
>>
>> I hope someone can shed some light on this.
>>
>> Thanks,
>>
>> Max.
>> -- 
>> 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
>
> -- 
> 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