Re: openafs bug 886768

[Benjamin Kaduk <kaduk@mit.edu>]

On Tue, Feb 20, 2018 at 08:51:16PM +0100, Salvatore Bonaccorso wrote:
> Hi Thorsten,
> 
> On Tue, Feb 20, 2018 at 02:45:48PM +0100, Thorsten Alteholz wrote:
> > Hi everybody,
> > 
> > the latest security update of the kernel to version 3.2.0-5 in Jessie
> > resulted in #886768 [1] for openafs.
> > 
> > Wouldn't it be better to do the openafs upload via security as well?
> > At the moment openafs in Jessie is just broken until the next point release.
> 
> Whilest one arguably can say that the issue was introduced/uncovered
> by a security update, the package has already been accepted by the SRM
> (thanks for that to Julien and Adam!).
> 
> So affected persons could already install the fixed packages via
> proposed-updates, but maybe Julien and Adam can be conviced that an
> update is important enought to schedule an update earlier via a SUA?

It's probably also worth noting that this is not the first time that
a linux security update caused an openafs regression, and thus not
the first time that this question has been raised.  I don't have a
great recollection of the specific arguments presented from previous
instances, but do remember that the conclusion was to not treat the
openafs change as a security update.  So presumably some compelling
new argument would be desired in order to actually reopen the
discussion.

-Ben