[PATCH 0/1] [Jaunty] LP#334994 -- degraded raid boot failures

[Stefan Bader]

Andy Whitcroft wrote:
> On Mon, Mar 30, 2009 at 10:10:08AM +0200, Stefan Bader wrote:
>> Andy Whitcroft wrote:
>>> After much debugging we have found the bug causing the degraded raid
>>> boot failures the server team reported.  The patch following this email
>>> adjusts the teardown of raids to ensure they are completely removed before
>>> returning to userspace.  This prevents a later race between the delayed
>>> destructors and the recreation of the raid in degraded mode.
>>>
>>> Proposing for Jaunty.
>>>
>>> Andy Whitcroft (1):
>>>   UBUNTU: SAUCE: md: wait for possible pending deletes after stopping
>>>     an array
>>>
>>>  drivers/md/md.c |    4 ++--
>>>  1 files changed, 2 insertions(+), 2 deletions(-)
>>>
>>>
>> Looks like a clear, "crap, why hasn't that done before". ACK Is there 
>> potential for this getting backported to Intrepid and Hardy or has this 
>> been freshly introduced?
> 
> It appears the code is the same in Intrepid and similar in Hardy.
> It is odd that noone has ever hit this on those releases but it is pretty
> reproducible on Jaunty.  I am a little wary of changing things on releases
> where this has not reproduced.  Perhaps we should submit this upstream as
> well and get their acceptance before pushing it back to previous releases.
> 
> -apw
Yeah, sounds like a plan. IIRC Neil Brown is maintaining md.

Stefan

-- 

When all other means of communication fail, try words!