Re: RAID 10 with 2 failed drives

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

 



On 20/09/19 17:24, Sarah Newman wrote:
> On 9/20/19 8:59 AM, Wols Lists wrote:
>> On 19/09/19 21:45, Liviu Petcu wrote:
>>> Hello,
>>>
>>> Please let me know if in this situation detailed below, there are
>>> chances of restoring the RAID 10 array and how I can do it safely.
>>> Thank you!
>>
>> This is linux raid 10, not some form of raid 1+0? That's what it looks
>> like to me. I notice it says the array is active! That I think is good
>> news!
> 
> I thought that there should be a flag like 'degraded' if the raid was
> actually running. I can't find the kernel documentation any more.
> 
>>
>> Can you mount it read-only and read it? I would be surprised if you
>> can't, which means the array is running fine in degraded mode. NOT GOOD
>> but not a problem provided nothing further goes wrong. I notice it's
>> also version 0.9 - is it an old array? Have the drives themselves
>> failed? (which I guess is probably the case :-( I guess the drives
>> effectively have just the one partition - 2 - and 1 is something
>> unimportant?
> 
> What you said is definitely true for a near layout for an even number of
> devices and n=2.
> 
> I thought the offset layout meant any two adjacent raid devices failing
> was data loss, assuming this is accurate:
> 
> http://www.ilsistemista.net/index.php/linux-a-unix/35-linux-software-raid-10-layouts-performance-near-far-and-offset-benchmark-analysis.html?start=1
> 
Except you've failed to extrapolate, sorry. We have six drives, not the
four of the example. Although you could still be right. Does "offset=2"
mean 2 copies, offset layout?

The rule with raid-10 is that you can lose AT LEAST n-1 drives where n
is the number of mirrors. So if there are three mirrors of two drives
each, this array is safe. You can lose AT MOST p drives, where p is the
number of drives in a mirror. So this array *could* be safe with 2
mirrors. What you can't do is lose n drives that are mirroring each
other. The fact that the array is active makes me suspect that he is lucky.

NOTE that mdadm has problems with a degraded array if it was working
last time it tried! That doesn't mean it can't get it working, it means
it's trying to draw attention to the problem.

Cheers,
Wol



[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