ubuntu-bugcontrol team mailing list archive

["Christopher M. Penalver" <christopher.m.penalver@xxxxxxxxx>]

Phillip Susi, thanks for advising on this.

On Thu, Jun 12, 2014 at 7:58 PM, Phillip Susi <psusi@xxxxxxxxxx> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA512
>
> On 06/12/2014 08:15 PM, Christopher M. Penalver wrote:
> | Daniel Letzeisen, thanks for your e-mail, and your trying to help with
> | triaging linux (Ubuntu) bugs. Regarding your comments:
> |
> |> " About a week ago, Ubuntu released kernel 2.6.32-61 for Lucid/10.04
> Server as a security fix to this bug:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1326367 Unfortunately,
> the fix caused a regression, which is correctly documented and tagged by an
> Ubuntu dev here:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1327300 The regression
> caused a variety of symptoms such as hanging and loss of sound (there are a
> couple other bugs, but I don't have the link at the moment):
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1327220
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1327014
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1328360 Christopher
> Penalver quickly marked those bugs as "Won't Fix" and said Lucid Desktop
> wasn't supported anymore (even though the kernel is part of the Server
> package and an update had just been pushed)."
> |
> | This is because all the bugs you are marking duplicates of 1327220
> | aren't using Lucid Server, but Lucid Desktop, which is EoL as of
> | https://wiki.ubuntu.com/Releases . As a member of Ubuntu Bug Control
> | you should have already been keenly aware of this link. However, if
> | there has been a change in the End of Life for Lucid Desktop I've not
> | been made aware of, I'll be more than happy to reverse the Status and
> | triage appropriately.
>
> Sounds like a rather sticky situation.  I would say that the other bugs
> should not be marked wontfix out of hand simply because they are no longer
> supported.

I disagree with you here on this, as that is the exact reason to mark
a bug report as Won't Fix, is because they aren't supported. There is
a long precedence on marking these situations as Won't Fix via the
Ubuntu Kernel Team's triaging of past linux (Ubuntu) bug reports.
Those using EoL platforms having their bugs marked Won't Fix is not a
new, or unusually controversial situation on Launchpad.

>  A reasonable attempt should be made to address the issues in the
> still supported package, without causing unintended breakage, especially if
> the same root cause also causes breakage in still supported packages.

I agree, and currently open LP#1327220 will address this. However, by
marking Lucid Desktop bug reports a duplicate of LP#1327220, Daniel is
encouraging those with an EoL platform experiencing breakages in their
systems to continue using them, in stark contrast to the fact it's
EoL. This is giving them the false impression that their platform is
still supported (which it's not), and safe to continue using until the
EoL date of Lucid Server (also not).

>  Only if there is no reasonable way to accomplish that would marking the fallout
> in the unsupported packages as wontfix be justified.
>
> To put that another way, as long as there is an open bug against a still
> supported package caused by this regression that the others can rightly be
> duped against, then they should be duped rather than severed and marked
> wontfix.

That's where I disagree. If the current duplicates of LP#1327300 are
Lucid Server, then yes, marking them Invalid/Won't Fix would be
improper, and I have some Status reversal+apologizing to do to a
number of bug reporters. However, they are Lucid Desktop, and
shouldn't be marked duplicates, and should be marked as Won't Fix /
Invalid.

Thanks again Phillip.

Christopher M. Penalver
E-Mail: christopher.m.penalver@xxxxxxxxx