[lsb-discuss] Making progress on LSB bugs

Wichmann, Mats D mats.d.wichmann at intel.com
Fri Jul 21 07:03:53 PDT 2006



As the (unofficial) LSB Bugmaster, I'm getting concerned as
the LSB open bug count again creeps towards 300.  Granted we
use the bugzilla for a lot of stuff besides critial bugs
(minor bugs, nice-to-have's, feture requests, work items and
trackers for future releases, etc.) but 300 is still a big
number for a project this size.

It's pretty clear that one problem is there are not enough
people with enough time, but I think we can do a bit better.
To that end, I'm requesting some actions that should not take
more than a few hours from the people who own bugs - as well as
possibly some future actions from ones who don't own any bugs.

1. If you own bugs, please do a search on bugs.linuxbase.org
for your bugs, and take a look at them.

2. If you have bugs that are actually fixed, please mark
them resolved.

3. If you have bugs that may be fixed, but you're not sure, try
to either verify this, or let me know and I'll try to pursue
it. Of course, return to step 2 if you find out they're fixed.

4. If you have bugs that would be quick to fix, but have not
gotten around to it, let's get those knocked off and out of
the way

5. If you have bugs that you still intend to work on, but
can't finish in the next few days (maybe because they're
big and you need to schedule the time; maybe because they're
something that lead to a future release), make sure they're
marked Accepted, not New.

6. If you have bugs in either Accepted or New state that you
don't expect to get around to, please give them away.  Do this
by assiging them to someone else that has agreed to work on
them, or barring that, give them to lsb-bugs at freestandards.org.
I am not letting bugs sit owned by lsb-bugs for very long,
so I'll see these and know I need to find a new owner.

7. If you have bugs which could be fixed relatively easily if
we had only decided what to do, please indicat this by setting
the Keyword "helpwanted" on it.  Also use this keyword if
you've fixed a bug bug can't close it because you need someone
else to test it, maybe on hardware you don't have access to.

I will try to find new owners for the bugs that were given
away, will try to schedule the work to resolve issues and
check status, and will also review the list for stale bugs
that should be closed.

After a pass or two at this (there may have to be a couple
of passes if many bugs end up reassigned), we should end up
with a much cleaner bug list, dead wood pruned out and the
live bugs assigned to the right people and Accepted so it's
clear those people expect to work on them as some point.

Yeah, I realize in saying this that I own the majority of
the bugs - this morning's count is 92 of the 291 open bugs,
and should probably take most of the ones with Nick's name
on them until such time as he is able to work on bugs again.
But a bunch of that is by default - me taking bugs, esp. spec
bugs, so they'll have an owner, I know it is not realistic
for me to find the time to fix all of the ones on my plate.

The bugs which need group work are a main topic for the
LSB "virtual sprints".  I tried to hold one of those a
week ago, but it was not a success, I got no feedback
going and nobody else brought up any bugs to discuss. We
need to do better at this.  This is where hopefully some
of you who do not have assigned bugs can help - we
gather on the IRC channel (and can have a phone bridge
open as well, if desired) and kick thoughts around on
bugs where the answer is not immediately clear.  This
does not have to be limited to long-time "core"
developers, either, this is a place where more eyes on
the problem helps, and it's a good way to learn more about
the internals.




More information about the lsb-discuss mailing list