[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]
Re: [dm-devel] SUSE multipath-tools patch resync
- From: Christophe Varoqui <christophe varoqui gmail com>
- To: Hannes Reinecke <hare suse de>
- Cc: device-mapper development <dm-devel redhat com>
- Subject: Re: [dm-devel] SUSE multipath-tools patch resync
- Date: Wed, 25 May 2011 13:14:39 +0200
On mer., 2011-05-25 at 11:08 +0200, Hannes Reinecke wrote:
> On 05/25/2011 09:51 AM, Christophe Varoqui wrote:
> > On mer., 2011-05-18 at 17:03 +0200, Hannes Reinecke wrote:
> >> git.kernel.org:/pub/scm/linux/kernel/git/hare/multipath-tools.git
> >
> > I started the merge. I'll post comments along the reading. It seems
> > there won't be much : this patchset is clearly a must-have.
> >
> :-)
>
> With some things so bloody obvious that I keep wondering if I'm the
> only one seeing these ...
>
The merge is done. No change wrt your branch. Still not pushed to korg.
I struggle with a hwtable corruption with no /etc/multipath.conf ... can
you reproduce that ?
For example, iterating through the hwtable and printing ->product from
config.c:find_hwe() gives:
...
(StorEdge 3510|T4)
FC2502
RAIGE VOLUME
CSM200_R
LCSM100_[IEFS]
INF-01-00
FLEXLINE 380
hd
dcssblk
The final spurious entries are clearly from the blacklist mem region. I
verified the hwtable is not corrupted after initialization. Corruption
thus occurs between hwtable init and path discovery.
I'll continue tracking it, but if you see something obvious ...
[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]