16:02:26 <poelcat> #startmeeting Fedora 14 Alpha Blocker Meeting
16:02:26 <zodbot> Meeting started Fri Jul 16 16:02:26 2010 UTC.  The chair is poelcat. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:02:26 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:03:18 <adamw> https://bugzilla.redhat.com/showdependencytree.cgi?id=611990&hide_resolved=1
16:04:09 <adamw> jlaska always comes up with a nicer list sorted by component but I can never remember how...
16:05:20 <poelcat> okay, i'll call them out, might need help w/ some of the meetbot foo
16:05:35 <adamw> oh hey, here ya go
16:05:35 <poelcat> #topic https://bugzilla.redhat.com/show_bug.cgi?id=596985
16:05:36 <buggbot> Bug 596985: high, low, ---, jglisse, NEW, hang at start of X11 on fresh install from DVD; no debug info
16:05:49 <adamw> http://tinyurl.com/2apxuo7
16:06:24 <poelcat> looks like no feedback yet from developer
16:07:14 <poelcat> has anyone else looked at this bug yet and should it stay on the list?
16:07:34 <adamw> this is one we usually wouldn't accept as a blocker, as so far it's a single-system bug
16:08:18 <poelcat> adamw: so bump from list with that as the reason?
16:08:22 <adamw> john calls out criterion 6 but really that criterion is intended to cover a case where there's a bug in *the installer itself*, not a hardware-related X bug which, as a consequence, means anaconda can't run
16:08:54 <adamw> it's something of a messy bug though, and it's interesting that he apparently first filed a problem with an NVIDIA card in the system, which would seem to imply there's at _least_ two bugs there
16:09:51 <poelcat> adamw: what should we do w/ it for the purpose of our meeting now?
16:10:12 <adamw> for now i would vote to bump it with a comment that it can be re-considered if more people seem to suffer the same bug, or the developer thinks it'll affect a lot of systems
16:11:18 <poelcat> #action https://bugzilla.redhat.com/show_bug.cgi?id=596985 remove from F14Alpha blocker--will reconsider if this bug impact turns out to be more widespread
16:11:20 <buggbot> Bug 596985: high, low, ---, jglisse, NEW, hang at start of X11 on fresh install from DVD; no debug info
16:11:34 <poelcat> any comments from other people?
16:11:41 <adamw> i'll update the bug
16:11:45 * poelcat realizes forgot to do rollcall
16:11:53 <poelcat> adamw: thanks
16:12:04 <poelcat> #topic https://bugzilla.redhat.com/show_bug.cgi?id=613695
16:12:05 <buggbot> Bug 613695: medium, medium, ---, mgracik, ASSIGNED, ImportError: cannot import name GObject
16:12:48 <poelcat> looks like there is a patch, but no indication on timeline for applying it
16:13:27 <poelcat> any other info for this bug or discussion?
16:14:18 <adamw> looking
16:16:23 <adamw> this came from the rawhide acceptance testing
16:16:26 <adamw> https://fedoraproject.org/wiki/Test_Results:Fedora_14_Pre-Alpha_Rawhide_Acceptance_Test_1
16:17:08 <poelcat> anyone know martin gracik's irc nick?
16:17:16 * poelcat can't find a people page for him
16:17:34 <adamw> that has an extended description: "when attempting to active advanced storage devices "
16:18:27 <adamw> arguably that makes this not an alpha blocker, since alpha is only supposed to be able to install "using IDE, SATA and SCSI storage devices, with the default file system and LVM"
16:18:33 <adamw> none of which come under the 'advanced storage devices' bit, iirc
16:18:38 <adamw> (possibly scsi?)
16:18:51 <poelcat> adamw: what should our actions on this bug be then?
16:19:14 <adamw> it'd be nice to be sure about the impact
16:19:16 <adamw> jlaska: ping?
16:20:17 <adamw> if jlaska's not about i'd vote to ask for more info on the bug before making a decision
16:20:45 <adamw> i'd say we need to know if the bug breaks that criterion or not, then when we have that info we can make a decision straight away in the bug
16:21:32 <poelcat> adamw: okay, can you put question in the bug?
16:22:05 <poelcat> #action https://bugzilla.redhat.com/show_bug.cgi?id=613695 request more information to determine if this bug breaks criterion or not
16:22:06 <buggbot> Bug 613695: medium, medium, ---, mgracik, ASSIGNED, ImportError: cannot import name GObject
16:22:10 <adamw> will do
16:22:20 <poelcat> #topic https://bugzilla.redhat.com/show_bug.cgi?id=613817
16:22:22 <buggbot> Bug 613817: medium, low, ---, skvidal, NEW, API break: .ts.run() exception has changed
16:25:24 <adamw> i think this qualifies; it looks like the impact is unpredictable depending on the package set to be installed, but it certainly has the potential ability to break installation
16:26:01 <adamw> it may be worth asking whether installation would actually be able to proceed in the failure case here even with this bug resolved - it seems like you need an ultimately invalid package install operation to trigger it, so i don't know what anaconda would do after that - it sounds like it might fail anyway
16:27:53 <adamw> seeing if anyone's available frm anaconda team
16:29:09 <adamw> ...seems not
16:29:14 <adamw> usually we get more input for these meetings, heh
16:29:29 <poelcat> adamw: yeah, i didn't do a good job of pinging people at the start
16:29:54 <adamw> i'd say again we need more info on this. if it can break an install that would otherwise work, i'd say that's a blocker...if not, it's not
16:29:56 <adamw> wdyt?
16:30:31 <poelcat> adamw: using the info we have i say call it a blocker
16:30:51 <poelcat> and let devel or someone else come back and argue it isn't
16:31:01 <adamw> fair enough
16:31:45 <poelcat> #action https://bugzilla.redhat.com/show_bug.cgi?id=613817 keep as blocker unless feedback from devel is otherwise
16:31:46 <buggbot> Bug 613817: medium, low, ---, skvidal, NEW, API break: .ts.run() exception has changed
16:32:02 <poelcat> adamw: can you update bug w/ whiteboard field, etc.?
16:32:12 <poelcat> #topic https://bugzilla.redhat.com/show_bug.cgi?id=614488
16:32:13 <buggbot> Bug 614488: medium, low, ---, anaconda-maint-list, NEW, Dropped vesamenu.c32 from rawhide boot.iso
16:32:57 <adamw> done
16:34:14 <adamw> again an interesting one. i wish jlaska was more clear about the impact =)
16:34:54 <adamw> i'm not sure any of the criteria actually cover this, if the installer does go ahead and boot.
16:40:27 <adamw> wdyt?
16:40:30 <poelcat> adamw: w/o knowing much it seems like a blocker to me
16:41:43 <adamw> under which criterion?
16:43:02 <poelcat> hm
16:43:13 <adamw> you could argue for "The rescue mode of the installer must start successfully and be able to detect and mount an existing default installation " if it makes it difficult/impossible to get into rescue mode, i guess
16:44:01 <Oxf13> oh this is today, forgot to put it on my calendar
16:44:45 <adamw> hello oxf13
16:47:23 <adamw> Oxf13: we're on https://bugzilla.redhat.com/show_bug.cgi?id=614488 in case you hadn't seen
16:47:24 <buggbot> Bug 614488: medium, low, ---, anaconda-maint-list, NEW, Dropped vesamenu.c32 from rawhide boot.iso
16:47:32 <adamw> the impact's not very clear from jlaska's report
16:47:54 * poelcat brb
16:49:51 <adamw> well, my vote's we can't take this as a blocker without more info on exactly what it breaks
16:50:59 <poelcat> okay, any other votes to the contrary?
16:51:37 <poelcat> keep until next week or drop and request more info to regain blocker status?
16:52:04 <adamw> well i'd just leave it as f14alpha but not add the acceptedblocker keyword, and post a comment asking for info
16:52:09 <Oxf13> so I think what happens is you don't even get a text menu, you might just get a bare prompt, which will be somewhat confusing for people
16:52:11 <adamw> we can review when we get further info, and/or next week
16:52:14 <Oxf13> I'd say it's a blocker
16:52:28 <adamw> Oxf13: under which alpha criterion?
16:52:29 <Oxf13> particularly since it should be an easy fix
16:52:57 <Oxf13> under the "boy it would suck if we shipped alpha like that" criteria (which is kind of hard to enumerate)
16:53:13 <adamw> sorry, that criterion doesn't exist. =)
16:53:20 <adamw> the criteria are https://fedoraproject.org/wiki/Fedora_14_Alpha_Release_Criteria . it's a release blocking bug if it breaks one of those.
16:53:39 <Oxf13> surely we have some loophole that allows us to make judgment calls on issues that don't neatly fall into any of our existing buckets...
16:54:02 <poelcat> what about ammending the criteria to say there must be graphical isolinux/grub
16:54:05 <poelcat> screen.
16:54:22 <adamw> well, the 'Alpha Blocker Bugs' section, underneath. but for that the bug needs to be 'high' or 'critical' severity, and i'd find it hard to call this bug one of those.
16:54:39 <adamw> poelcat: yeah, someone could certainly propose such a criterion on the list.
16:54:50 <Oxf13> if it's going to generate a crapton of bugs filed by people, I would consider that a high.
16:55:08 <adamw> we have definitions of severity, too, y'know =)
16:55:18 <adamw> "High: the bug makes the program in question unusable, or a major packaging guideline violation (license problem, bundled library, etc) "
16:55:29 <adamw> "Urgent: the bug makes whole system unusable (or it is a security bug, which is per definition urgent) "
16:55:31 <Oxf13> yes, and it's all too easy to get bound up in protocol, and let a bullshit bug slip through because it didn't neatly fit into a bucket.
16:55:52 <adamw> we created the protocol because things were too arbitrary when we did everything by feel.
16:56:04 <Oxf13> so lets not create a second problem
16:56:08 <adamw> if we decide to just start throwing bugs in as blockers on feel, it renders the protocol useless.
16:56:19 <Oxf13> which is inability to look at a bug and make a decision.
16:56:32 <Oxf13> no, the protocol is a guideline, to help with those decisions
16:56:38 <poelcat> how about if we start a separate mail thread on this
16:56:38 <Oxf13> but we're going to run into things that don't neatly fit.
16:56:41 <adamw> i don't believe it was ever suggested that that was the case.
16:56:50 <poelcat> we have one more bug and I'd like to go soon :)
16:56:56 <Oxf13> so we can either say "boohoo, it doesn't fit" or we can say "hey this is a blocker, we should add it, and then make sure the next time this comes up it fits"
16:57:12 <adamw> poelcat's already given the correct procedure if you feel such bugs should be blocking: propose a criterion to cover it
16:57:23 <Oxf13> wow, really?
16:57:37 <Oxf13> you can't look at this bug and say "yep, it's a blocker"?
16:57:46 <Oxf13> gotta get all the paperwork filed, a few acks, another meeting or two?
16:57:50 <adamw> yes. that way, we don't have to keep making judgment calls forever, and try and remember which ones we made before, and make the same ones again when similar bugs come up.
16:58:03 <Oxf13> these things can happen in parallel
16:58:14 <adamw> no, i can't. and frankly not just because of the protocol. given the info in the bug, even on feel i'm not sure it's a blocker. but that's irrelevant, really. there's a reason we came up with the process.
16:58:29 <Oxf13> whatever.  You guys have fun with that.
16:58:33 <Oxf13> I'll be over here, doing real work.
16:58:46 <adamw> it doesn't need any meetings. you send an email to test list saying 'hey, i'd like to add this release criterion'. then we argue about it. it's classic fedora process. =)
16:59:38 <poelcat> back to the bug at hand
16:59:44 <poelcat> what is our #action?
17:00:14 <adamw> i'd say same as i said earlier: leave it as blocking f14alpha but don't add the keyword yet, ask for further info, review when we get it
17:00:42 <poelcat> did someone want to propose and amendment to the criteria?
17:00:46 <poelcat> on the list
17:01:24 <adamw> maybe. frankly, personally, i'd propose it as a beta or even final criterion. to me, whether we display a graphical boot menu for the installer is mostly a polish issue.
17:01:25 <poelcat> #action https://bugzilla.redhat.com/show_bug.cgi?id=614488 leave as blocking f14alpha, need additional information
17:01:26 <buggbot> Bug 614488: medium, low, ---, anaconda-maint-list, NEW, Dropped vesamenu.c32 from rawhide boot.iso
17:01:58 <poelcat> adamw: do i need to adjust the "action" ?
17:02:08 <adamw> nope, that's fine
17:02:31 <poelcat> #topic https://bugzilla.redhat.com/show_bug.cgi?id=614538
17:02:33 <buggbot> Bug 614538: medium, low, ---, mgracik, ASSIGNED, firstboot does not start after manual install
17:02:35 <poelcat> last one
17:04:31 <poelcat> maybe no longer a problem?
17:04:40 <zodbot> Announcement from my owner (stickster): Fedora Board open IRC meeting in #fedora-board-meeting at 1800 UTC today (~1 hour from now).
17:04:47 <poelcat> set to needinfo reporter?
17:05:22 <adamw> no, i don't think so - that last comment is only a side note
17:05:48 <adamw> from what i can tell, martin knows that firstboot doesn't run, knows why, and has a fix, but doesn't have permissions to push a new firstboot
17:07:02 <adamw> i think we've considered this a break of "In most cases, the installed system must boot to a functional graphical environment (see Blocker_Bug_FAQ) " in the past
17:07:28 <adamw> because you do user account creation in firstboot; if firstboot is skipped you have no user accounts when you hit gdm, and you can't login as root by default, so the desktop really isn't functional
17:07:43 <adamw> so i'd consider it a blocker under that
17:10:04 <poelcat> okay, that makes sense
17:10:58 <adamw> so, yeah, i'd be +1 for this being a blocker
17:11:26 <poelcat> #action https://bugzilla.redhat.com/show_bug.cgi?id=614538 accepted as a blocker
17:11:28 <buggbot> Bug 614538: medium, low, ---, mgracik, ASSIGNED, firstboot does not start after manual install
17:11:33 <poelcat> anything else before we wrap?
17:13:20 <poelcat> thanks for your time everyone
17:13:23 <poelcat> #endmeeting