Misaligned raid0

Heinz Mauelshagen mauelshagen at redhat.com
Wed Dec 21 14:15:24 UTC 2005


On Wed, Dec 07, 2005 at 08:01:54PM -0500, Phillip Susi wrote:
> I have been using dmraid for a few months now to access my two disk 
> raid0 on my VIA SATA Raid controller.  Today I noticed some IO errors 
> beyond the end of the disk in my syslog and when I investigated, it 
> looks to me like my raid setup is broken.
> 
> The dm table that dmraid creates for the main raid volume device looks 
> like this:
> 
> 0 144607678 striped 2 128 8:0 0 8:16 0
> 
> Now what I noticed is that 144607678 is NOT an even multiple of the 
> stripe width of 256 sectors ( 128 sector stripe factor x 2 disks ). 
> Shouldn't the total length of a striped mapper device allways be an even 
> multiple of the stripe width?  I think that the last fractional stripe 
> at the end of the disk is the problem, because there aren't a full 128 
> sectors left on the disk to access before rolling over to the next disk 
> in the stripe.

Hrm, partial strides should be ok with device-mapper.
Seems like reiser decided to allocate past EOD.

Heinz

> 
> So does this mean that my bios created a broken stripe?  And why does 
> dmraid and the kernel device mapper accept such broken values?
> 
> Wish me luck with using resizefs.reiserfs to shrink down the volume a 
> bit to avoid that broken tail end.
> 
> _______________________________________________
> Ataraid-list mailing list
> Ataraid-list at redhat.com
> https://www.redhat.com/mailman/listinfo/ataraid-list

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Red Hat GmbH
Consulting Development Engineer                   Am Sonnenhang 11
Cluster and Storage Development                   56242 Marienrachdorf
                                                  Germany
Mauelshagen at RedHat.com                            +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-




More information about the Ataraid-list mailing list