[Pkg-linaro-lava-devel] Bug#763319: Bug#763319: Fails unless python-django-longerusername in installed

[Sjoerd Simons]

On Mon, 2014-09-29 at 11:26 +0100, Neil Williams wrote:
> tag 763319 + upstream, pending
> thanks
> 
> On Mon, 29 Sep 2014 11:49:00 +0200
> Sjoerd Simons <sjoerd at debian.org> wrote:
> 
> > Source: lava-server
> > Version: 2014.08.2-1
> > Severity: grave
> 
> Please take care with bug severity, grave needs a justification from
> Policy. 

Actually serious needs a justification from policy, grave however does
not.. As a reminder grave is "makes the package in question unusable or
mostly so" (among other things). You could argue that severity serious
would have been better (with policy justfication 3.5: Every package must
specify the dependency information about other packages that are
required for the first to work correctly.).

> There is a fix for this issue already applied upstream and the
> next release is due this week. RC status will have no effect on how
> quickly this gets fixed and it's arguable whether this is severity
> important as existing installs continue to operate unless the
> longerusername package is removed. 

I don't set severities on bugs to get fixes for the packages any
quicker. I set them because they are, in my opinion, the correct
severities for the respective bugs. 

For RC status all that counts is the package as is, not what will be in
the next release so arguing that severity is incorrect due to fixes that
have been done upstream which are meant to trickle down to debian makes
little sense. 

> I'm keeping the severity as-is but
> please just bear in mind that not every HTTP500 affecting lava-server is
> of serious or grave severity.

Sure, and as you could see from the bugreport i actually did investigate
what the issue was here.

> The Django1.7 changes had to be applied quickly as a hotfix release and
> there are issues which are being integrated into the next full release.
> 
> For now, installing the longerusername package fixes this problem.

That's all fine, and really, i don't mind hitting these kind of issues
while our boxes are running testing (Actually i'm very happy we can run
from packages these days instead of random release deployments).. But
still, the bug severity is orthogonal to all this :)


-- 
Sjoerd Simons <sjoerd at debian.org>