multiple arrays and unmapped partitions
mauelshagen at redhat.com
Tue Apr 29 09:07:22 UTC 2008
kpartx workaround works for you or not ?
In order to be able to test this further, I'd need the partition
table sector(s) in addition to the "dmraid -rD" output you provided.
On Sun, Apr 27, 2008 at 03:44:16PM +0200, Lucas B. Cohen wrote:
> Have there been significant changes or advances regarding the issue of
> partitions not being mapped when more than one array is present in a system,
> since Liviu Balan's message on the subject  last year?
> With rc13 and rc14, depending on kernel version, either the first (2.6.18)
> or second array (2.6.22 and up) get "missed" in my system (Intel Software
> Raid on Intel P35 with ICH9R).
> Aside from Heinz's reply to  (suggesting a kpartx-based workaround), I've
> noticed that deactivating the non-mapped array *only* before activating it
> allows the missing nodes to be created in /dev/mapper. (Something like
> dmraid -an isw_di && dmraid -ay isw_di)
> I imagine the "use kpartx instead of my own DOS metadata format handler"
> item in rc14's TODO is relevant to this.
> In case it can be of use, the output of "dmraid -rD" for my system is
> accessible at .
> Any insight is appreciated, as I'm having a hard time finding relevant
> information on this problem elsewhere.
> Lucas Cohen
>  http://www.redhat.com/archives/ataraid-list/2007-April/msg00009.html
>  http://elmo.xzibition.com/~lbc/dmraid-metadata.tar.gz
> Ataraid-list mailing list
> Ataraid-list at redhat.com
Heinz Mauelshagen Red Hat GmbH
Consulting Development Engineer Am Sonnenhang 11
Storage Development 56242 Marienrachdorf
Mauelshagen at RedHat.com PHONE +49 171 7803392
FAX +49 2626 924446
More information about the Ataraid-list