Torque bugzilla (was Re: [torquedev] pbs_server segfault in
glen.beane at gmail.com
Tue Dec 30 19:10:47 MST 2008
On Tue, Dec 30, 2008 at 7:23 PM, Chris Samuel <csamuel at vpac.org> wrote:
> First off I'd say that we're not alone with this issue,
> the Linux kernel folks have a bugzilla system which is,
> shall we say, not always updated as it should be. ;-)
> ----- "Glen Beane" <glen.beane at gmail.com> wrote:
>> I think if TORQUE really wants community participation, then the
>> torque bugs should be logged in a public place and not CRIs internal
>> tracking system. Log support requests and maybe even feature
>> requests internally, but bugs should be available to the community.
> So maybe what should happen there is that support requests that
> come in to the Cluster Resources RT system as the result of a
> Torque bug should cause the CR person who owns that RT request
> to open a Bugzilla bug referencing the RT ticket but anonymising
> details that the customer may not want disclosed.
> But for everything else I suspect that we want people to open
> their own bugs in bugzilla directly.
> Perhaps this list should get all bugzilla emails too ?
>> We should also actively promote this whenever someone sends
>> a bug report to the user list.
> I think that's a good plan, but we should also point out
> existing tickets that should be contributed/subscribed too
> if we are aware of them.
>> If the developers at CRI and the community developers
>> use it then we will avoid the problems that occurred
>> last time. It is too easy for a bug report sent to the
>> user list to get lost or overlooked.
> Indeed, and the same for feature requests..
> I think people also need to understand that the number
> of developers is *really* small, pretty much just Josh,
> Glen and Garrick from my understanding. I do what I can
> but I wouldn't trust me with commit access to the Torque
> SVN.. ;-)
Alan Taufer at CRI does a lot of work on TORQUE, but most of it as a
result from customer bug reports and feature requests I think.
More information about the torquedev