Comment # 6 on bug 1162479 from Coly Li

(In reply to Neil Brown from comment #5)
> The problem here related to the fact that the RAID0 has two components of
> different sizes.
> A bug was introduced in Linux 3.14 which change the way data was laid out on
> those drives.
> So if you create such RAID0 with 3.13 (or earlier) and access it on 3.14 (or
> later) the data might get corrupted.
> We added code in Linux 5.4 so that array like that will be rejected (with
> error 524) unless they are explicitly marked to say which layout to use.
> 
> mdadm was also change to explicitly set a layout on newer kernels.
> 
> It seems that the new mdadm isn't being used here.
> 
> Coly: where are we up to with the new mdadm?  Is it in Tumbleweed?  If so,
> when did it land?

I just add the mdadm part to openSUSE:Base:System, it should be in Tumbleweed
soon. Here is the changelog,
- Create: add support for RAID0 layouts. (bsc#1162479)
  0015-Create-add-support-for-RAID0-layouts.patch
- Assemble: add support for RAID0 layouts. (bsc#1162479)
  0016-Assemble-add-support-for-RAID0-layouts.patch
- mdadm.8: add note information for raid0 growing operation
  (bsc#1129900)
  1002-mdadm.8-add-note-information-for-raid0-growing-opera.patch


Coly Li