BugZappers/Meetings/Minutes-2008-Oct-07

Bug Triage Meeting :: 2008-10-07

Attendees

• jds2001
• John5342
• ke4qqq
• poelcat
• jlaska
• mcepl
• comphappy

NEW->ASSIGNED

• Should we change the definition of transition to ASSIGNED?
  • BugZappers/BugStatusWorkFlow#ASSIGNED
• Adding a keyword of Triaged seems to be the best alternative
• Most attendees believe that this problem is not pervasive enough to change the established process
• RESOLVED: revisit in a few months if this is really an issue

Greasemonkey Bugzilla Assist Scripts

• try out latest version from comphappy
  • http://bashton.fedorapeople.org/bugzilla_buttons_for_fed.user.js
• mcepl requests testing by @redhat.com users
  • http://mcepl.fedorapeople.org/scripts/greasemonkey/redhat/bugzillaCopyOwnerToCC.user.js

FTBFS (Fails To Build From Source)

• Auto-filed for packages that fail to build from source
• Ask owner to automatically set new bugs in the following way so they do not need to be triaged
  1. Change state to ASSIGNED
  2. Revew comment text to make sure it is descriptive and explains what is happening
• OWNER: jds2001 to close with mdomsch

FEver Bugs

• Auto-filed for packages that have new versions upstream at SourceForge
• http://fedoraproject.org/wiki/PackageMaintainers/FEver
• Ask owner to automatically set new bugs in the following way so they do not need to be triaged
  1. Change state to ASSIGNED
  2. Revew comment text to make sure it is descriptive and explains what is happening
  3. Add FutureFeature keyword
• OWNER: poelcat

Topics for Next Week

1. okay to triage your own bugs?
2. How to create an organized effort to tackle growing list of new rawhide bugs: http://tinyurl.com/6llac8

IRC Transcript

poelcat	who is here?	07:01
poelcat	or did you already do that? :)	07:01
* jds2001 , John5342 ke4qqq		07:01
John5342	jds2001: beat me to it	07:02
poelcat	should we wait for anyone else or get started?	07:02
jds2001	how bout comphappy?	07:02
jds2001	is he gonna be able to make it?	07:02
jds2001	ahh, there's mcepl :)	07:02
mcepl	here I am	07:02
poelcat	mcepl: hi!	07:02
mcepl	hi	07:02
poelcat	so first off, I reorganized and trimmed down this page	07:03
poelcat	BugZappers/Meetings	07:03
* John5342 likes it		07:03
poelcat	my personal goal is to do better meeting recaps for the next couple of months and see how that goes	07:03
poelcat	including sending them to fedora-test-list after each meeting	07:04
* mcepl *really* likes it		07:04
poelcat	after the new year I'd hope that someone else can take this over or help with them	07:04
jds2001	poelcat: sorry for giving you about a year's worth of raw irc logs from here :)	07:04
poelcat	it doesn't take that long to do, but combined with everything else	07:04
poelcat	it can be hard to be consistent	07:04
poelcat	so i'm hoping that having another feedback loop the mailing list about what we are doing	07:05
jds2001	yeah, I think that the new arrangement is easier.	07:05
poelcat	helps raise visibility, interest, etc.	07:05
John5342	poelcat: and as for the implied request for help in the long term i might be able to help out if nobody else steps up	07:05
poelcat	John5342: i'll keep you in mind :)	07:05
mcepl	just a note on NEW->ASSIGNED -- I don't think we should take it too seriously -- there is only one guy bitching, other people who really need it, can use ON_DEV or something.	07:05
jds2001	and the good news is that the list is about to be flooded with XO stuff :)	07:05
jds2001	so probably new folks coming in that may not have known about this before.	07:06
poelcat	is that why #fedora-qa seems busier than usual?	07:06
comphappy	my hope is that I can be ore active again once march rolls arround	07:06
jds2001	yeah, I think that's part of it.	07:06
poelcat	mcepl: let's talk about that topic as it blocks a few things	07:06
poelcat	comphappy: hi!	07:06
mcepl	poelcat: and because people found out that it is better place to find out what's going on then be swamped by newbies on #fedora ?	07:06
poelcat	lol	07:07
* John5342 googles for XO		07:07
mcepl	poelcat: and of course, Fedora10Beta helps ... ;-)	07:07
poelcat	John5342: OLPC	07:07
mcepl	John5342: OLPC -- one laptop per child is called XO	07:07
John5342	thanks	07:07
comphappy	poelcat: did you give that gs script a whirl	07:08
poelcat	so how shall we resolve the NEW->ASSIGNED controversy	07:08
poelcat	comphappy: let's cover that in a little bit	07:08
comphappy	ok I got to take off in about 5	07:08
* jlaska arrives late		07:08
poelcat	i'm willing to consider "Triaged" keyword	07:08
mcepl	poelcat: my suggestion is stonewall it -- doing nothing.	07:08
mcepl	-1 for a new keyword	07:08
poelcat	mcepl: why is that?	07:09
mcepl	see above -- too much work for nothing	07:09
* poelcat wants to make sure other teams feel like we want to help/work with them vs. get in the way		07:09
* comphappy agrees with mcepl		07:09
comphappy	never seen a complaint about it until it was talked about on the ML	07:09
mcepl	if there is anybody who has legitimate problems (like I was told by dm & al. and KDE people) there are possible solutions	07:10
poelcat	mcepl: so are you suggesting continuing to triage everyone's bugs the same way?	07:10
mcepl	yup	07:10
poelcat	jlaska: hi	07:10
* John5342 believes that Triaged keyword is non-intrusive and can also be automated using the gs stuff but should still solve the issues. no matter how minor they are.		07:11
jds2001	hmm, i dont really want to give the sense of inflexibility	07:11
poelcat	so i see a couple of roads....	07:11
mcepl	jds2001: there are just few trolls bitching ... don't feed them	07:11
poelcat	1) do nothing for 2 more months	07:11
poelcat	2) take a vote now	07:11
poelcat	3) ?	07:11
comphappy	jds2001, but it is also just_another_thing	07:11
* mcepl votes for 1)		07:11
* comphappy 1		07:12
* poelcat +1 to 1)		07:12
poelcat	ke4qqq: jlaska jds2001 ?	07:12
jds2001	+1 to 1. (but aren't we doing 2? :) )	07:12
poelcat	if you're at the meeting you have a "vote" :)	07:12
poelcat	lol	07:12
* ke4qqq wonders if our minds will change in 2 months		07:12
poelcat	ke4qqq: well we are kind of deciding not to decide	07:13
mcepl	ke4qqq: we'll see whether there is a real problem, or whether it just settles down.	07:13
* poelcat waits for someone to quote the Rush song		07:13
* John5342 refuses to quote songs		07:13
comphappy	poelcat, not to push the issue, but did you have any comments on that	07:13
* mcepl don't that many American songs ;-)		07:13
* poelcat meant voting as in vote on a different way to do it		07:13
mcepl	*know	07:13
poelcat	comphappy: i didn't know there was a new version	07:14
* mcepl understands, but maybe being a lawyer he has it too easy to distinguish between substantial and procedural votes ;-)		07:14
poelcat	comphappy: link?	07:14
* comphappy got a warning for showing up late because of linux stuff yesterday		07:14
comphappy	http://bashton.fedorapeople.org/bugzilla_buttons_for_fed.user.js	07:14
poelcat	comphappy: its okay if you have to go... thanks for all your help!	07:15
comphappy	code is not very clean right now, but I will fix that in the next week or so	07:15
poelcat	comphappy: okay	07:15
comphappy	but it seems to work	07:15
poelcat	anymore thoughts/comments/votes on the current topic of NEW/ASSIGNED ?	07:16
comphappy	i will keep my irc up while i get ready so just put your comments in	07:16
jlaska	poelcat: so the consensus is to hold out until post F10?	07:17
poelcat	jlaska: the only +1s are for keeping things they way they are	07:18
poelcat	so i'll consider that concensus for now :)	07:18
jds2001	sounds good :)	07:19
mcepl	comphappy: consider http://mcepl.fedorapeople.org/scripts/greasemonkey/redhat/bugzillaCopyOwnerToCC.user.js -- I seem to have some improvements in.	07:19
poelcat	next topic... FTBFS	07:19
poelcat	did we want to recommend/decide for sure that they auto-set those to ASSIGNED?	07:19
jds2001	I thought I told mdomsch to throw those in ASSIGNED, but come to think of it, there may have been some python-bugzilla thing that prevented him from doing that, not sure.	07:20
poelcat	jds2001: i think it would have to be a followup action	07:20
mcepl	poelcat: yes, the description should be more thorough and more emphasize that this is a Fedora action not some random whiner (which some maintainer tends to think), but otherwise, there is not much to triage.	07:20
poelcat	because you can't start in ASSIGNED	07:20
mcepl	poelcat: well, I don't care if they are filed as NEW and then automagically immediately switched to ASSIGNed.	07:21
jds2001	i think that's what I told him to do, and mentioned that there's a way to suppress the mail.	07:21
poelcat	mcepl: i agree	07:22
jds2001	but this was many moons ago :)	07:22
jds2001	when he first started auto-filing bugs	07:22
poelcat	does anyone else think we could be missing something by not triaging?	07:22
poelcat	jds2001: could you followup with him?	07:22
jds2001	matt already doess something of a sanity check to make sure that rawhide wasn't busted that day, etc.	07:23
jds2001	sure.	07:23
mcepl	jds2001: well, if he screws up, then bug triagers won't fix it.	07:23
poelcat	any objections to a resolution of "FTBFS bugs should be auto-changed to ASSIGNED by Matt's script" ?	07:24
jds2001	(and mail suppressed on that action to save him from being endlessly flamed)	07:25
poelcat	jds2001: good point	07:25
John5342	+1 for that resolution together with jds2001 comment.	07:25
jds2001	but I think that's really easy to do in python-bugzilla	07:26
poelcat	any objections to said resolution?	07:26
mcepl	+1 to take a second look at the text of the bug	07:26
mcepl	s/\+1/& +/	07:26
poelcat	mcepl: lol that is not what the resolution is :)	07:26
poelcat	mcepl: so you're voting -1 ?	07:26
mcepl	no +1	07:27
jds2001	mcepl: you're a lawyere, you're out of order! :)	07:27
* poelcat brings the gavel down		07:27
* comphappy + 1		07:27
mcepl	with comment "please, take a second look at the text of the FTBS bug"	07:27
poelcat	mcepl: who?	07:27
mcepl	Matt	07:27
poelcat	what is he lookin at? that isn't clear to me	07:27
jds2001	just to make sure that the text is clear that this isn't some random Joe complaining	07:28
jds2001	and what will happen if it's not resolved, etc.	07:28
* poelcat is still confused		07:29
mcepl	poelcat: don't worry	07:29
poelcat	can someone state from the beginning what needs to happen?	07:29
poelcat	otherwise we can't have clear minutes :)	07:29
* mcepl is quite certain that he shouldn't say anything		07:29
jds2001	sure, two things:	07:29
jds2001	1) Move to ASSIGNED automatically, supppressing mail	07:29
jds2001	2) Review the text of the bug that's filed for clarity to the maintainer.	07:30
poelcat	shouldn't we be advisin Matt on #2 ?	07:30
poelcat	or we just want to ask him to "be sure"  ?	07:30
poelcat	that sounds funny to me, but oh well :)	07:30
jds2001	I can offer some advice, yeah	07:30
poelcat	jds2001: thanks	07:31
poelcat	okay next topic... a related one: FEver	07:31
poelcat	do we want to vote on a similiar resolution for default bug filing behavior?	07:32
John5342	+1 to the same resolutions from me.	07:32
mcepl	poelcat: "be sure" is better and mention my experience with Firefox guys who were ignoring FTBFS on seamonkey as "don't feed troll who doesn'ŧ know how to build"	07:32
poelcat	e.g. ASSIGNED, FutureFeature, etc. ?	07:32
* poelcat was +1 for past resolution too		07:33
mcepl	I am lost, what am I voting about now?	07:33
John5342	mcepl: FEver. at least thats what i just voted on	07:33
jds2001	FEver, the automated thing that files bugs about new releases.	07:34
jds2001	new upstream releases of packages, that is.	07:34
poelcat	we talked about it here BugZappers/Meetings/Minutes-2008-Sep-30#FEver_Bugs	07:34
poelcat	wanted to close out the discussion	07:34
poelcat	is there anything we want to add beyond asking them to autochange to ASSIGNED and add FutureFeature?	07:37
John5342	poelcat: presumably also supress mail from that one in the same way as the last issue	07:37
poelcat	John5342: good point	07:38
mcepl	+1	07:38
poelcat	how about the text they are using?	07:38
jds2001	+1 here	07:38
jds2001	mcepl: i just looked at a FTBFS bug and it directs folks to a rather well-written wiki page.	07:38
mcepl	do we have an example of the text?	07:38
mcepl	jds2001: OK, so Martin is lazy. Oh well :(	07:39
John5342	mcepl: https://bugzilla.redhat.com/show_bug.cgi?id=464491	07:41
buggbot	Bug 464491: medium, medium, ---, wart@kobold.org, ASSIGNED, manaworld-0.0.26 is available	07:41
poelcat	okay... any disagreement with recommending: 1) auto set to ASSIGNED 2) FutureFeature keyword, 3) suppress email?	07:42
poelcat	otherwise i'll assume everyone is +1 to moving forward	07:42
jds2001	yep	07:42
John5342	+1 here	07:42
jds2001	+1 to moving forward, that is.	07:42
* poelcat +1		07:42
mcepl	+1	07:43
poelcat	the beast is growing again ... http://tinyurl.com/6llac8	07:43
poelcat	(list of NEW rawhide bugs)	07:43
ke4qqq	+1	07:43
poelcat	should we have a "Bug day" or "Bug Week" or _________ ?	07:43
mcepl	poelcat: yes, fighting with X & Gecko bugs and it is tough	07:43
poelcat	i'd also encourage people to post triage questions with bz url to fedora-test-list	07:44
poelcat	i got a lot of questions answered that way	07:44
poelcat	best to group a bunch of them thought, rather than one email at a time :)	07:44
John5342	autosend all rawhideASSIGNED	07:46
John5342	oops	07:46
John5342	ignore me	07:46
poelcat	should we have a "Bug day" or "Bug Week" or _________ ? Need some ideas and input here :)	07:50
poelcat	beyond "yes"	07:50
jds2001	yes :)	07:51
mcepl	yes (and thinking about the rest)	07:51
* poelcat hits jds2001 w/ the gavel		07:51
jds2001	So the problem that I see with em is what you and I had talked about.	07:51
jds2001	that a "go forth and triage" day isn't really productive of anyone's time.	07:51
poelcat	mcepl: have you had any past success with them or other ideas we should try?	07:51
jds2001	I think that jlaska's test days have been quite the success, though.	07:51
jds2001	but that's focused and there are folks right there to help.	07:52
poelcat	maybe he could come as a guest speaker and give a motivational talk? ;-)	07:52
mcepl	poelcat: no (I haven't organized any yet) and I have no clue. Still thinking ...	07:52
mcepl	I think that being focused might help -- "let's go and squash this beast" is much better than everybody wandering around for anybody to kill.	07:53
John5342	i agree with mcepl. easy to get a bulk session going with a target. not so much for in general.	07:54
mcepl	and the beast must be smaller than "all NEW Rawhide bugs" ;-)	07:54
jlaska	poelcat: you want a motivational talk from me? :)	07:54
poelcat	shall we think more about this and discuss again next week?	07:55
poelcat	jlaska: we aren't sure we can afford your rate, but you would get the free wireless headset mic	07:55
poelcat	;-)	07:55
jlaska	ha!	07:55
poelcat	anything else we want to cover this week or add to next week's agenda?	07:56
mcepl	please, all RH folks, bomb me with bugs in http://mcepl.fedorapeople.org/scripts/greasemonkey/redhat/bugzillaCopyOwnerToCC.user.js	07:57
ke4qqq	?	07:58
jds2001	apparently that's a greasemonkey script that only works with RH employees	07:58
jds2001	(they see a slightly different view)	07:58
ke4qqq	hey quick "process" question - is it ok to triage bugs you created, or should you not?	07:59
ke4qqq	is there supposed to be separation?	07:59
mcepl	ke4qqq: I rather don't do it.	07:59
* poelcat tends to triage my own... i never make mistakes ;-)		08:00
ke4qqq	that tends to be my thought as well mcepl - but I hate the idea of someone else having to do it.	08:00
mcepl	ke4qqq: triage two more additinoal bugs for every bug you file ;-)	08:01
poelcat	let's revisit next week as we are at the hour mark	08:01
mcepl	OK	08:01
ke4qqq	k	08:02
poelcat	thanks everyone for coming!	08:02
John5342	bye all. thanks	08:02
* poelcat will send out recap by tomorrow, earlier if weather permits		08:02
mcepl	bye	08:03

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!