16:07:39 <roshi> #startmeeting F21-blocker-review
16:07:39 <zodbot> Meeting started Wed Sep 10 16:07:39 2014 UTC.  The chair is roshi. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:07:39 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:07:39 <roshi> #meetingname F21-blocker-review
16:07:39 <zodbot> The meeting name has been set to 'f21-blocker-review'
16:07:40 <roshi> #topic Roll Call
16:07:53 <danofsatx> again, I'm kinda-sorta here
16:08:12 * roshi is here
16:08:55 <roshi> sorry for getting started late, I got distracted with brokenness :)
16:09:08 <roshi> anybody else here?
16:09:21 * tflink can be here
16:09:28 <danofsatx> just use chickens.
16:09:38 <danofsatx> erm, s/use/us
16:09:54 <roshi> adamw, sgallagh? You guys around?
16:09:56 * sgallagh waves
16:10:05 <tflink> danofsatx: I suspect that chickens might not appreciate having a DVD or usb drive stuck into them
16:10:19 <danofsatx> welll......
16:10:23 <sgallagh> I'm juggling a fair number of things (including lunch), so apologies if I'm slow to respond
16:10:39 <roshi> no problem
16:11:03 * mattdm gets a plate of food from the indian buffet
16:11:40 <amita> mattdm, indian?
16:11:44 * pwhalen is here
16:12:02 <mattdm> amita yes it is deeelicious :)
16:12:10 <roshi> I think we have enough to get started with actual reviews :)
16:12:15 <amita> nice mattdm :)
16:12:22 <amita> roshi, yeah
16:12:55 <roshi> #chair tflink pwhalen danofsatx amita mattdm
16:12:55 <zodbot> Current chairs: amita danofsatx mattdm pwhalen roshi tflink
16:13:17 <roshi> any volunteers for secretary duty?
16:13:32 <roshi> if you haven't done it, but want to - I can walk someone through it after the meeting
16:13:48 * danofsatx can't today
16:14:14 * satellit listening  sorry I am late (Workstation meeting)
16:14:17 <adamw> man, i keep forgetting what time this is
16:14:20 * adamw here
16:14:25 <adamw> i can secretary
16:14:29 <roshi> no worries satellit, we were late too
16:14:31 <adamw> once i have breakfast and coffee
16:14:40 <roshi> thanks adamw
16:14:48 <amita> roshi, I woule like to
16:14:53 <roshi> #topic Introduction
16:14:53 <roshi> Why are we here?
16:14:53 <roshi> #info Our purpose in this meeting is to review proposed blocker and nice-to-have bugs and decide whether to accept them, and to monitor the progress of fixing existing accepted blocker and nice-to-have bugs.
16:14:57 <roshi> #info We'll be following the process outlined at:
16:14:59 <roshi> #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting
16:15:02 <roshi> #info The bugs up for review today are available at:
16:15:04 <roshi> #link http://qa.fedoraproject.org/blockerbugs/current
16:15:06 <danofsatx> if you weren't up so late tracking down sssd errors, you'd have breakfast at a reasonable hour adamw
16:15:07 <roshi> #info The criteria for release blocking bugs can be found at:
16:15:09 <roshi> #link https://fedoraproject.org/wiki/Fedora_21_Alpha_Release_Criteria
16:15:13 <roshi> #link https://fedoraproject.org/wiki/Fedora_21_Beta_Release_Criteria
16:15:16 <roshi> #link https://fedoraproject.org/wiki/Fedora_21_Final_Release_Criteria
16:15:19 <roshi> secretary amita?
16:15:34 <amita> no for walk through :P
16:15:42 <amita> its adamw
16:16:07 <roshi> part of the walkthrough I was going to take people through was actually doing it, just after the meeting
16:16:50 <roshi> adamw or I can walk you through it after the meeting - it's pretty simple
16:16:56 <roshi> ok, onto the first blocker!
16:16:57 <roshi> #topic (1139015) F21 Workstation Alpha TC6 network install fails with crash, hang, reboot or "pane is dead" - caused by anaconda segfault
16:17:00 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1139015
16:17:03 <roshi> #info Proposed Blocker, distribution, NEW
16:17:09 * amita clicks
16:17:11 <adamw> +1, obviously
16:17:19 <pwhalen> +1
16:17:37 <sgallagh> +1
16:17:39 <roshi> +1
16:17:46 <kalev> +1
16:17:50 <sgallagh> Though it sounds like this was a bad compose more than anything else
16:17:56 <tflink> +1
16:18:24 <amita> +1
16:18:59 <roshi> proposed #agreed - 1139015 - AcceptedBlocker - This is a clear violation of the Alpha Criteria "Installer must run"
16:19:00 <danofsatx> +1
16:19:06 <danofsatx> ack 'n' all
16:19:26 <sgallagh> Ack
16:19:39 <pwhalen> ack
16:19:41 <kalev> ack
16:19:44 <amita> ack
16:19:49 <roshi> #agreed - 1139015 - AcceptedBlocker - This is a clear violation of the Alpha Criteria "Installer must run"
16:20:03 <roshi> #topic (1135516) freeipa-client needs python-backports-ssl_match_hostname
16:20:05 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1135516
16:20:08 <roshi> #info Proposed Blocker, freeipa, VERIFIED
16:20:39 * amita clicks
16:20:53 <kalev> I'd be rather conservative what we mark as blockers, don't want to end up delaying the Alpha for small issues
16:20:56 <adamw> achievement unlocked: coffee
16:20:56 <kalev> but FE, sure
16:21:04 <adamw> amita: we can probably take the 'clicks' as understood
16:21:09 <roshi> looks like we're just waiting on the fix to land?
16:21:20 <danofsatx> +1 FE
16:21:29 <adamw> it matches the criterion precisely.
16:21:30 <amita> but its says it is already VERIFIED
16:21:40 <adamw> "It must be possible to join the system to a FreeIPA or Active Directory domain at install time and post-install, and the system must respect the identity, authentication and access control configuration provided by the domain."
16:21:57 <sgallagh> +1 blocker
16:21:57 <danofsatx> ok then, -1 FE, +1 block
16:22:01 <roshi> it'd be a blocker for the freeipa criteria for sure, and a fix is pending, so I'm fine with a +1 blocker
16:22:07 <adamw> amita: it's verified fixed in the update, but the update isn't in stable. actually maybe i'm not supposed to use VERIFIED for that, i forget.
16:22:21 <amita> adamw, hmm
16:22:30 <amita> so you may want to change the state?
16:22:34 <adamw> kalev: this makes it impossible to do an unattended/scripted realm join at install time
16:22:55 <adamw> amita: possibly, but i have to go remember what the rules about VERIFIED are first. it's not super important
16:23:03 <pwhalen> +1 blocker
16:23:09 <amita> adamw, ok
16:23:27 <pwhalen> adamw, please let me know too, as i flip that as well once i have confirmed it as working
16:23:36 <roshi> proposed #agreed - 1135516 - AcceptedBlocker - This clearly violates the Alpha Criteria: "It must be possible to join the system to a FreeIPA or Active Directory domain at install time and post-install, and the system must respect the identity, authentication and access control configuration provided by the domain."
16:23:44 <adamw> ack
16:23:48 <amita> ack
16:23:49 <pwhalen> ack
16:23:56 <sgallagh> Ack
16:23:56 <roshi> #agreed - 1135516 - AcceptedBlocker - This clearly violates the Alpha Criteria: "It must be possible to join the system to a FreeIPA or Active Directory domain at install time and post-install, and the system must respect the identity, authentication and access control configuration provided by the domain."
16:24:01 <roshi> #topic (1138746) virt-manager needs to split the version filed of the .treeinfo file on _ and - not just -
16:24:04 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1138746
16:24:07 <roshi> #info Proposed Blocker, virt-manager, ON_QA
16:24:36 <sgallagh> +1 FE, but as this bug is really about the host OS, I don't think it's a true blocker for F21 itself
16:25:13 * jreznik_ is here, again a bit later
16:26:23 <roshi> yeah, I'm not sure this qualifies as alpha blocker
16:26:55 <amita> https://bugzilla.redhat.com/show_bug.cgi?id=1138746#c1 - virt is generally Beta
16:27:16 <roshi> seems like a host issue and a fix is already pending
16:27:29 <roshi> wouldn't hurt to pull it in for the next TC though
16:27:31 <pwhalen> +1 FE
16:28:10 <adamw> well, i mean, i guess someone might be using virt-inst on live images?
16:28:36 <roshi> perhaps?
16:28:38 <adamw> or virt-manager with its 'guided install' thing, i guess this is
16:28:57 <roshi> I mean, an FE would fix the issue and get it in before alpha release
16:29:04 <amita> bug definition says virt-manager
16:29:15 <kalev> well, the fix is already there, so let's pull it in as FE?
16:29:25 <roshi> I'd lean towards FE
16:29:27 <danofsatx> +1 FE
16:29:27 <adamw> yeah, i guess this is the thing where you create a new machine and it tries to do something clever with the target distro?
16:29:30 <kalev> +1 FE
16:29:44 * adamw isn't really convinced, but doesn't mind...+/-0
16:29:59 <roshi> convinced of what? blocker or fe?
16:30:10 <adamw> fe
16:30:14 <adamw> it's obviously not a blocker
16:30:39 <roshi> well, we've got votes for FE
16:31:01 <danofsatx> and the local fix is simply mv F21_blah_blah.iso F21-blah-blah.iso
16:31:13 <amita> hmm :)
16:31:18 <roshi> I don't see any issues with it being pulled in, though I would question people running virt-manager from the batch of alpha lives we have in the first place
16:31:29 <roshi> for anything important anyways
16:31:53 <amita> but it means if anyone is creating VM with - in its name will face trace back
16:32:15 <amita> I mean _
16:32:19 <danofsatx> heh...the Redmine server I just built was from the Server TC5 release ;)
16:32:20 <sgallagh> danofsatx: That doesn't help at all with the netinstall case
16:32:52 <danofsatx> sgallagh: true, hadn't considered that.
16:32:55 <roshi> proposed #agreed - RejectedBlocker AcceptedFreezeException - This doesn't violate any of the alpha criteria so is not a blocker. However it was accepted as a FreezeException to be included in the next TC.
16:32:59 <adamw> amita: it's not the VM name, it's the ISO image name
16:33:12 <sgallagh> roshi: Well, I suspect that there's a lot of intersection between "people testing Alpha TCs" and "people running F21 prerelease on their client machine"
16:33:19 <adamw> roshi: that's nice and vague ;)
16:33:26 <kalev> I think we've historically been too strict with FE's. If the fix is there and seems sane to everybody, it should be fine to pull it in.
16:33:37 <adamw> sgallagh: we mean specifically running VMs off the live image, as that's the only case where FE actually makes any difference
16:33:37 * roshi shrugs towards adamw
16:33:50 <adamw> kalev: we have historically been too liberal with them and they keep busting our composes
16:33:52 <sgallagh> kalev: Well, there are cases where the fallout isn't worth it
16:34:07 <roshi> true sgallagh
16:34:11 <adamw> sgallagh: if you run a VM from an *installed* F21 Alpha system you can just do 'yum update libvirt'
16:34:16 <sgallagh> e.g. A freeze exception for a soname bump of libc? Not gonna happen
16:34:21 <adamw> hell, you can do that in the damn live image if you want to run a VM, we have that working.
16:34:40 <roshi> the workarounds are there - this doesn't really *stop* you from doing anything
16:34:53 <adamw> sgallagh: it's a wrinkle people keep forgetting about, but it's always vital to consider what breaking the freeze actually *achieves* when you're considering FEs
16:35:07 <sgallagh> Yes
16:35:30 <adamw> i mean, i don't really mind this one, i don't think it's gonna explode anything, but i'm just not sure strictly considered it's actually worth breaking a freeze for.
16:36:16 <roshi> yeah
16:36:51 <roshi> well, how about we just say rejected blocker and be done with it
16:37:15 <roshi> far as I can tell, this doesn't break anything a simple workaround doesn't fix quickly and easily
16:37:28 <sgallagh> Fine by me
16:37:54 <danofsatx> so, -1 blocker?
16:38:28 <roshi> proposed #agreed - 1138746 - RejectedBlocker - This doesn't violate any of the alpha criteria and workarounds exist.
16:38:36 <danofsatx> ack
16:38:54 <adamw> ack
16:39:08 <pwhalen> ack
16:39:12 * roshi supposes he's an implicit ack since he wrote it
16:39:20 <roshi> #agreed - 1138746 - RejectedBlocker - This doesn't violate any of the alpha criteria and workarounds exist.
16:39:51 <roshi> #topic (1138746) virt-manager needs to split the version filed of the .treeinfo file on _ and - not just -
16:39:54 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1138746
16:39:56 <roshi> #info Proposed Blocker, virt-manager, ON_QA
16:40:08 <sgallagh> -1 blocker
16:40:27 <roshi> bah
16:40:31 <roshi> that's the one we just did
16:40:34 <roshi> #undo
16:40:34 <zodbot> Removing item from minutes: INFO by roshi at 16:39:56 : Proposed Blocker, virt-manager, ON_QA
16:40:39 <roshi> #undo
16:40:39 <zodbot> Removing item from minutes: <MeetBot.items.Link object at 0xf333850>
16:40:45 <roshi> #undo
16:40:45 <zodbot> Removing item from minutes: <MeetBot.items.Topic object at 0x10033150>
16:40:55 <roshi> ok, that's all the blockers
16:40:57 <pwhalen> i have a late addition, just added
16:41:07 <roshi> link pwhalen ?
16:41:17 <pwhalen> https://bugzilla.redhat.com/show_bug.cgi?id=1136990
16:41:43 <pwhalen> this is actually a dup of https://bugzilla.redhat.com/show_bug.cgi?id=1124987
16:41:55 <roshi> #topic (1136990) - TypeError: must be an interface
16:42:08 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1136990
16:42:09 <pwhalen> but prevents net installs on arm, i confirmed locally that the new glibc fixes it
16:42:45 <roshi> #info Proposed Blocker, glibc, NEW
16:43:35 <kalev> I would definitely be +1 FE if it gets fixed in time
16:43:43 <roshi> are netinsts blocking on arm?
16:44:12 <jreznik_> well, it's primary arch
16:44:13 <roshi> I thought the typical arm install was more of a local thing
16:44:26 <pwhalen> glibc-2.19.90-36.fc21 fixed it, regenerated the install tree locally
16:44:56 <roshi> well, we don't block on ws netinst (as far as I tracked that conversation anyways) - so I didn't know if arm cared about the netinst is all
16:45:00 <sgallagh> ARM is blocking for Server
16:45:18 <roshi> ah
16:45:24 <roshi> which does care about netinst :)
16:45:40 <adamw> we did talk about arm server netinst and concluded we cared, IIRC
16:45:47 <roshi> +1 blocker
16:45:47 <adamw> (we = server SIG, he said, switching hats rapidly)
16:46:08 <roshi> works for me, I just wasn't sure where the cards fell on that one
16:46:09 <pwhalen> adamw, right, it was discussed yesterday
16:46:10 <roshi> if that makes sense
16:46:11 <sgallagh> adamw: You remember correctly
16:46:14 <pwhalen> +1
16:46:22 <sgallagh> +1 blocker
16:46:41 <adamw> +1
16:46:42 <kalev> +1 blocker, if we make the image it should actually work
16:47:01 <jreznik_> +1 blocker
16:47:03 <roshi> proposed #agreed - 1136990 - AcceptedBlocker - This clearly violates the following Alpha Criteria: When using a dedicated installer image, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces.
16:47:22 <sgallagh> Ack
16:47:24 <pwhalen> ack
16:47:28 <jreznik> ack
16:47:38 <kalev> ack
16:47:42 <roshi> #agreed - 1136990 - AcceptedBlocker - This clearly violates the following Alpha Criteria: When using a dedicated installer image, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces.
16:47:50 <jreznik> not sure how long it will be, maybe we can add for what product we block
16:47:58 <roshi> any more ninja blockers?
16:48:00 <pwhalen> thanks, sorry about the late addition
16:48:03 <pwhalen> ;)
16:48:07 <roshi> no worries pwhalen :)
16:48:30 <roshi> onto the FE's then?
16:48:36 <roshi> we've got 9 of them...
16:48:42 <sgallagh> 10
16:49:03 <roshi> huh, only 9 showing on my page
16:49:03 <adamw> whew, for aminute there i thought we were going to have another short blocker meeting
16:49:08 * adamw has ten
16:49:10 <adamw> hit refresh
16:49:10 <sgallagh> roshi: Refresh
16:49:12 * roshi refreshes
16:49:24 <roshi> for the record, mine was easier
16:49:28 <sgallagh> hahaha
16:49:29 <roshi> you guys asked for this :p
16:49:43 <sgallagh> (Unfortunately, I *did* ask for the tenth one)
16:50:00 <roshi> but no, you guys were all like "Hey, let's be thorough and stuff"
16:50:08 <roshi> ok, onto the first one
16:50:08 <roshi> #topic (1133394) [abrt] cheese: g_type_check_instance_is_fundamentally_a(): cheese killed by SIGSEGV
16:50:11 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1133394
16:50:14 <roshi> #info Proposed Freeze Exceptions, cheese, ON_QA
16:50:17 <adamw> such fools we are
16:51:30 <adamw> sure, non-crashing cheese is a good thing on lives
16:51:31 <adamw> +1
16:51:49 <roshi> yup
16:51:50 <roshi> +1
16:51:53 <amita> +1
16:52:44 <roshi> proposed #agreed - 1133394 - AcceptedFreezeException - This change can be pulled into the next compose.
16:53:15 <sgallagh> +0, for the record
16:53:43 <sgallagh> (I don't think it's necessarily worth breaking freeze for, but it's unlikely to have any wider impact, so meh)
16:54:07 <adamw> sgallagh: leaves are usually a bit less worrying :)
16:54:10 <roshi> it's a default app people are likely to test with on lives though
16:54:29 <sgallagh> Which is why I'm only +0 and not -1 :)
16:54:37 <roshi> fair enough, lol
16:54:44 <roshi> any ack/nack/patch?
16:54:50 <sgallagh> Ack
16:55:54 <roshi> ?
16:55:58 <adamw> ack
16:56:01 <adamw> well
16:56:09 <adamw> your agreeds are rather vague this morning ;)
16:56:26 <adamw> maybe #agreed - 1133394 - AcceptedFreezeException - cheese is a highly visible and commonly tested app for live images, worth fixing
16:56:36 <roshi> ack
16:56:41 * roshi was just doing the edit
16:57:07 <sgallagh> ack
16:57:32 <amita> ack
16:57:33 <roshi> the next ones will be more informative adamw :)
16:57:52 <adamw> ack
16:58:07 <roshi> #agreed - 1133394 - AcceptedFreezeException - cheese is a highly visible and commonly tested app for live images, worth fixing
16:58:18 <roshi> #topic (1140165) Cockpit 0.21 is broken with Docker 1.2.0
16:58:18 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1140165
16:58:18 <roshi> #info Proposed Freeze Exceptions, cockpit, MODIFIE
16:59:39 <sgallagh> Docker broke ABI underneath Cockpit and is already in F21. This is a fix so Cockpit (which is acting as our user-friendly interface to Docker) isn't a huge embarrassment at Alpha
17:00:09 <roshi> so this is when the host running cockpit can't manage the containers? or something wrong in our docker container image with cockpit installed?
17:00:19 * roshi is a bit fuzzy with cockpit
17:01:03 <kalev> +1 FE, the fix is available so might just as well pull it in
17:01:27 <roshi> I can see it being worth breaking freeze since cockpit is a key feature of F21
17:01:37 <sgallagh> roshi: The host cannot manage docker containers running on it through cockpit
17:01:44 <roshi> ok
17:01:47 <roshi> thanks sgallagh
17:01:52 <sgallagh> All docker actions result in an error
17:02:16 <roshi> +1 FE, since this touches a lot of the shiny new F21 shiny-ness
17:02:24 <kalev> it's not like we have an release to ship to mirrors today, so might just as well spend the extra time to polish up things and fix embarrasing bugs
17:02:29 <kalev> especially if the fix is already available
17:03:02 <roshi> yeah
17:03:03 <kalev> if we were releasing tomorrow, then I'd be more conservative :)
17:03:24 <sgallagh> Well, it's an FE, not a blocker for much that reason
17:03:37 * kalev nods.
17:03:49 <adamw> seems like you can get the fix through a yum update, but i guess the polish argument kinda sways me on this one as it would look unfortunate if you just deployed straight out of the DVD ISO and tried to use it without updating...
17:04:02 <roshi> that was my thought
17:04:20 <roshi> cockpit and running docker containers are two the the things we're touting with f21
17:04:23 <roshi> votes?
17:04:36 <roshi> I count 2 +1s
17:04:36 * sgallagh abstains for obvious reasons
17:06:01 <roshi> is that a +1 from you adamw ?
17:06:13 <adamw> sure
17:06:27 <adamw> sgallagh: we don't really have conflict-of-interest rules for blocker review
17:06:37 <adamw> the highly advanced heuristic vote counting algorithm takes it into account
17:06:44 <adamw> (i've been known to vote against my own proposals)
17:07:18 <roshi> proposed #agreed - 1140165 - AcceptedFreezeException - This bug touches cockpit and docker interoperability which are both key features for F21. It's worth breaking freeze to get these changes into a more polished image.
17:07:29 <adamw> ack
17:07:39 <roshi> less vague? :)
17:08:01 <amita> roshi, you are just fine, I guess :)
17:08:20 <danofsatx> ok, I'm out. have fun, folks....
17:08:25 <roshi> later danofsatx
17:08:43 <roshi> ack/nack/patch?
17:09:01 <amita> ack
17:09:10 <sgallagh> ack
17:09:16 <roshi> #agreed - 1140165 - AcceptedFreezeException - This bug touches cockpit and docker interoperability which are both key features for F21. It's worth breaking freeze to get these changes into a more polished image.
17:09:30 <roshi> #topic (1116316) gnome-initial-setup not coming up after Live Workstation install
17:09:33 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1116316
17:09:36 <roshi> #info Proposed Freeze Exceptions, gnome-initial-setup, NEW
17:10:17 * amita brb
17:10:47 <adamw> +1, it's not crucial but it'd be good to fix it
17:10:50 <roshi> +1 FE
17:10:56 * roshi thought this looked familiar
17:11:02 <adamw> the g-i-s workflow is kinda useful, and obviously we want folks to test it
17:11:03 <adamw> roshi: :)
17:11:59 <roshi> anyone else?
17:12:33 <pwhalen> +1 fe
17:13:35 <kalev> I don't see a lot of value in theoretical voting here. I mean, if we say +1 FE, it usually means +1 to this _particualar build_ since it looks sane and safe.
17:13:58 <roshi> proposed #agreed - 1116316 - AcceptedFreezeException - g-i-s provides a useful workflow that we want to have tested before release.
17:14:31 <kalev> I'd be happy to say +1 FE if there was an actual build available, but this voting for an unknown thing :)
17:14:32 <satellit> I get it in VirtualBox but in russian...
17:14:35 <roshi> kalev: I see it as a mix between the two, if the package is sane and good isn't enough reason to break freeze - it also has to make sense for the image itself to get that update
17:14:39 <roshi> AIUI
17:15:23 <roshi> well, the +1 FE is "when a fix is available we've alread said we want it pulled in because it helps the image itself"
17:15:40 <roshi> if we were just pulling in good packages, then we wouldn't really need a freeze process anyways
17:15:44 <roshi> far as I can tell
17:15:47 <adamw> kalev: in fact the conception is just the other way around
17:15:54 <adamw> though in practice it's kind of a fudge
17:16:09 <roshi> I could be worng though
17:16:13 <adamw> the idea is that we vote on whether the issue potentially merits a freeze break, and do the evaluation of the actual fix when we decide whether to pull it in
17:16:23 <adamw> in practice, when we can actually see the complexity of the fix at vote time we tend to consider it
17:16:26 <adamw> eh, humans :P
17:16:32 <satellit> g-i-s worked for me in VB root but no user workstation 21 Alpha TC6 x86_64
17:16:40 <adamw> i thought we actually already had the fix for this one, though?
17:16:47 * satellit just now in Vb
17:17:07 <roshi> I thought we did too
17:17:26 <roshi> guess not - at least noted in the bug
17:17:43 <adamw> hum, oh yeah, i guess mclasen just identified the problem, not the fix
17:18:37 * satellit logged in to getting started
17:18:59 <roshi> so I see 3 +1s
17:19:02 <roshi> any other votes?
17:19:37 <roshi> ack/nack/patch?
17:20:08 <adamw> ack
17:20:13 <pwhalen_> ack
17:20:45 <roshi> #agreed - 1116316 - AcceptedFreezeException - g-i-s provides a useful workflow that we want to have tested before release.
17:21:06 <roshi> #topic (1139475) cannot select from full list of keyboard layouts
17:21:06 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1139475
17:21:07 <roshi> #info Proposed Freeze Exceptions, gnome-initial-setup, NEW
17:22:25 <roshi> +1, being able to get your keyboard layout shouldn't be something you can't do
17:22:44 <kalev> +1 here, nice and small fix and already available
17:23:30 <pwhalen> +1
17:23:54 <adamw> +1
17:24:06 <jreznik> +1
17:24:27 <roshi> proposed #agreed - 1139475 - AcceptedFreezeException - Being able to configure the proper keyboard layout is key functionality that should be present in the image and a fix is already available.
17:26:03 <kalev> ack
17:26:17 <pwhalen> ack
17:26:51 <sgallagh> Ack (but the "key" pun was a bit obvious, don't you think?)
17:27:09 <roshi> nah - no one reads these things anyways, right :p
17:27:23 <roshi> I could patch it to say "pressing"
17:27:37 <sgallagh> Nah, I enjoy a good pun.
17:27:40 <roshi> #agreed - 1139475 - AcceptedFreezeException - Being able to configure the proper keyboard layout is key functionality that should be present in the image and a fix is already available.
17:27:43 <adamw> ack
17:27:45 <adamw> d'oh
17:28:09 <roshi> #topic (811967) libvirt in a VM often brings up 'default' network when it shouldn't, kills vm networking
17:28:12 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=811967
17:28:15 <roshi> #info Proposed Freeze Exceptions, libvirt, NEW
17:30:06 * jsmith looks up from his day-job to look at the details of this one
17:30:09 <adamw> i'm definitely +1 FE to this in theory, but with a definite chance for not taking a fix if it's too messy
17:30:13 <adamw> it's a high impact issue, though
17:30:18 <kalev> for me, this one too needs an actual fix I could look at and vote if it's safe to pull in, hard to say now
17:30:57 <roshi> this could be a messy fix
17:31:28 <kalev> like, taking out Boxes would be one possible fix, but I wouldn't be +1 to that without Workstation WG-s agreement
17:31:43 <kalev> we can't just blanket approve any fixes that might come up
17:31:56 <sgallagh> I'm -1 FE on this all-around.
17:32:03 <jsmith> I'm torn...
17:32:07 <sgallagh> Yes, it's a long-standing issue, but the risk it poses is very large.
17:32:15 <sgallagh> I'd rather it get sorted out between Alpha and Beta
17:32:20 <adamw> kalev: this review process is definitely not about code review, it's just about whether the issue is potentially important enough to merit freeze breaking
17:32:27 <kalev> I just don't see value in voting on this until an actual fix is available.
17:32:33 <adamw> kalev: we kind of have short-cycle / timing problems with doing formal review of the actual fix
17:32:50 <adamw> kalev: we tend to have very short windows and it's not practical to try and convene a meeting every time an FE fix lands, unfortunately
17:32:50 <kalev> pff :)
17:33:23 <adamw> we've taken out Boxes before, but that was when it was still pretty early
17:35:22 <roshi> I dunno about this one
17:35:59 <adamw> maybe we should punt it till we see an actual fix?
17:36:02 <adamw> reserve judgment?
17:36:11 <roshi> that works for me
17:36:13 <kalev> sure, I would be +1 to that
17:36:24 <roshi> +1 punt
17:36:57 <pwhalen> +1 punt
17:37:03 <sgallagh> I'm sticking with -1 FE
17:37:15 <sgallagh> Even an actual fix has the potential to be destabilizing
17:37:41 <roshi> everything has that potential though :)
17:37:50 <roshi> so that's +3 to punt
17:38:21 <sgallagh> roshi: Some more than most
17:39:04 <roshi> proposed #agreed - 811967 - Punt - We'll wait to decide FE status until an actual fix is available to look at, since this issue has the potential to touch or break other things.
17:39:10 <roshi> true :)
17:39:10 <adamw> ack
17:39:14 <pwhalen> ack
17:40:01 <kalev> ack
17:40:05 <roshi> #agreed - 811967 - Punt - We'll wait to decide FE status until an actual fix is available to look at, since this issue has the potential to touch or break other things.
17:40:36 <roshi> #topic (1136994) GTK+3 themes must be updated for upstream checkbox 'checked' state
17:40:39 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1136994
17:40:41 <roshi> #info Proposed Freeze Exceptions, mate-themes, ON_QA
17:41:13 <roshi> +1 for this
17:41:29 <roshi> makes user creation in anaconda unintuitive
17:41:58 <kalev> well, the fix is already done, could just as well close the ticket
17:42:07 <kalev> not much to vote on here ...
17:42:22 <kalev> spin-kickstarts doesn't go through any QA process and the fix has already landed there.
17:42:44 <sgallagh> *rubber stamp*
17:43:11 <roshi> works for me
17:43:12 <kalev> sure, +1 FE to that fix that already landed then :)
17:43:31 <roshi> I say we just close the bug if the fix has landed
17:43:44 <roshi> no reason to keep it in the "Accepted FEs" list for us to go over again
17:43:49 <pwhalen> roshi, +1
17:43:55 <satellit> in yum install @KDE?
17:45:01 <roshi> there's another similar FE proposed right now that is that satellit
17:45:04 <roshi> iirc
17:45:07 <adamw> kalev: eh?
17:45:11 <adamw> how is this a spin-kickstarts fix?
17:45:16 <adamw> oh right
17:45:19 <adamw> mate just fudged it
17:45:34 <adamw> i was hoping to find ten minutes to actually fix their damn themes but i didn';t
17:45:50 <adamw> the bug shouldn't be closed because the themes are still broken
17:46:06 <roshi> I htought the fix landed already?
17:46:20 <kalev> they landed a hacky workaround
17:46:36 <adamw> their 'fix' was 'use a different theme when running live'
17:46:48 <amita> ha
17:46:53 <roshi> works for me :p
17:47:02 * roshi misunderstood it
17:47:20 <kalev> it's the mate teams choice how they fix their stuff
17:47:30 <kalev> if they choose to use a different theme, good for them
17:48:08 <adamw> the bug report wasn't 'checkboxes don't show up in anaconda'
17:48:12 <adamw> the bug report was 'fix your themes'
17:48:21 <adamw> kalev: oh, but they don't even change the theme permanently - only during installation
17:48:37 <adamw> kalev: post-install it still uses the MATE theme, which will presumably still have invisible checkboxes everywhere
17:48:44 <adamw> er, only during live boot*
17:48:50 <jreznik> yep
17:48:57 <roshi> cinnamon had a similar issue
17:49:07 <adamw> roshi: the other desktops all went and actually fixed their themes, i think
17:49:17 <jreznik> it's not a good example of what should land in .x.y.z release :(
17:49:26 <adamw> anyway, there's not really any action for us here, i'm just venting :)
17:49:49 <kalev> yes, I very much agree this fix isn't in the right place and done the right way, but it's not really up to the meeting here to be the worlds police
17:50:07 <kalev> it's not one of the primary products either, and not a blocking spin either
17:50:17 <roshi> so we need an FE until all themes are updated? or just one?
17:50:24 <roshi> how many themes are we talking about?
17:50:36 <adamw> roshi: mate ships a truckload
17:50:43 <adamw> the FE would just be for the one the installer uses
17:51:00 <roshi> the installer uses and the one it installs by default?
17:51:03 <adamw> i might still find time to go fix it, iunno. i guess i don't care hugely.
17:51:12 <adamw> roshi: they're the same, i think, this Bluementa thing
17:51:45 <amita> editing the kickstart did not help as he claims
17:51:54 <roshi> ok, but then the hack fix that's there doesn't fix the issue - so we still need an FE to get the *actual* in
17:52:32 <roshi> +1 FE to fix the right theme
17:53:21 <adamw> i think the hack will 'fix' the very specific case of checkboxes in the live installer, it just doesn't fix anything else.
17:53:35 <roshi> right
17:53:42 <roshi> but the theme still remains the issue
17:53:45 <adamw> i guess I'm +1 FE so i can ninja in a sane fix if i get time to make one
17:53:53 <adamw> it can only affect MATE anyway, it's not gonna break Workstation
17:54:05 <satellit__e> +1  FE it is not blocking
17:54:26 <kalev> +1 FE if the fix is in mate theme, -1 if someone submits a build that patches gtk3 :-)
17:54:57 <adamw> kalev: :P
17:55:37 <adamw> $ git branch -d rewrite-gtk3-for-mate
17:55:48 <adamw> well, i guess i need a new plan.
17:55:52 <roshi> proposed #agreed - 1136994 - AcceptedFreezeException - A fix to the theme used in MATE during both installation and the default theme post install (shold be the same theme for both) will be accepted during freeze.
17:56:24 <adamw> ack
17:56:28 <kalev> ack
17:56:36 <jsmith> ACK
17:56:37 <roshi> #agreed - 1136994 - AcceptedFreezeException - A fix to the theme used in MATE during both installation and the default theme post install (shold be the same theme for both) will be accepted during freeze.
17:56:38 <jreznik> ack
17:56:58 <roshi> next is similar
17:56:59 <roshi> #topic (1136985) theme must be updated for GTK+ 'checked' state (not 'active') for checkboxes
17:57:02 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1136985
17:57:04 <roshi> #info Proposed Freeze Exceptions, oxygen-gtk3, ON_QA
17:57:24 <kalev> I would be +1 blocker to that one, actually. KDE is a blocking spin.
17:57:46 <kalev> but since the fix is available, +1 FE
17:57:51 <roshi> it needs pulled in, that update is already there and working
17:57:55 <roshi> +1 FE
17:58:45 <satellit__e> +1 FE
17:59:46 <roshi> proposed #agreed - AcceptedFreezeException - The existing fix should be pulled in during freeze as it fixes anaconda usability issues for KDE.
18:00:11 <jreznik> ack
18:00:17 <kalev> ack
18:00:19 <satellit__e> ack
18:00:23 <adamw> ack
18:00:26 <roshi> #agreed - AcceptedFreezeException - The existing fix should be pulled in during freeze as it fixes anaconda usability issues for KDE.
18:00:43 <roshi> #topic (1134861) sanlock missing on i668 and armv7 architectures
18:00:43 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1134861
18:00:43 <roshi> #info Proposed Freeze Exceptions, sanlock, ON_QA
18:01:44 <jsmith> +1 from me
18:02:32 <pwhalen> +1
18:02:58 * roshi doesn't know what sanlock is
18:03:05 * adamw either
18:03:19 <adamw> roshi: can you add https://bugzilla.redhat.com/show_bug.cgi?id=1134566 to the FE review list?
18:03:45 <roshi> sure
18:04:05 <kalev> +1 from me, looks safe and shouldn't affect any produced media and fixes an annoyance (broken dep mails going out to the libvirt list every day)
18:04:22 <jreznik> +1, as kalev said
18:04:26 <adamw> yeah, small thing but i don't mind taking it for that
18:04:35 <roshi> that's enough +1s for me
18:05:49 <roshi> proposed #agreed - 1134861 - AcceptedFreezeException - The small fix has already been pushed and shouldn't break any other media.
18:06:17 <adamw> ack
18:06:24 <adamw> wait, what do you mean 'already been pushed'?
18:06:36 <roshi> to updates-testing
18:06:41 * roshi changes the words
18:06:42 <adamw> oh, i'd say 'is already available'
18:07:01 <roshi> proposed #agreed - 1134861 - AcceptedFreezeException - The small fix is available and shouldn't break any other media.
18:07:06 <kalev> ack
18:07:14 <jreznik> ack
18:07:23 <pwhalen> ack
18:07:29 <roshi> #agreed - 1134861 - AcceptedFreezeException - The small fix is available and shouldn't break any other media.
18:07:53 <roshi> #topic (1134128) [abrt] shotwell hangs during any import operation
18:07:56 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1134128
18:07:59 <roshi> #info Proposed Freeze Exceptions, shotwell, ON_QA
18:08:09 <kalev> +1
18:09:08 <jreznik> +1
18:09:17 <roshi> +1 for the same reason as the cheese FE
18:09:28 <adamw> yeah, again this is something people are likely to try on a live or immediately post-install
18:09:30 <adamw> +1
18:09:40 <pwhalen> +1
18:10:20 <roshi> proposed #agreed - AcceptedFreezeException - Shotwell is a very visible default application that should work on the live images and get plenty of testing.
18:10:36 <kalev> ack
18:10:45 <pwhalen> ack
18:10:55 <jreznik> ack
18:11:03 <roshi> it could do w/o the testing bit I think, but it works
18:11:33 <roshi> #agreed - AcceptedFreezeException - Shotwell is a very visible default application that should work on the live images and get plenty of testing.
18:11:50 <roshi> #topic (1139962) Fedora 21, FreeIPA 4.0.2: sssd does not find user private group from server
18:11:53 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1139962
18:11:56 <roshi> #info Proposed Freeze Exceptions, sssd, NEW
18:13:44 <roshi> +1 because FreeIPA is a key F21 feature
18:14:17 <adamw> we don't have the fix yet, but it should at least be restricted to sssd and hence reasonably easy to scope for testing
18:14:43 <roshi> and if you have a reproducer that's easy to do
18:14:51 <roshi> testing should be quick on any fixes down the line
18:15:00 * roshi keeps meaning to mess with freeIPA, but hasn't yet
18:17:21 <adamw> freeipa! freeipa!
18:17:25 <adamw> any other votes? + or -?
18:17:41 <adamw> i wish i had done a bit more research into the actual consequences of the bug, but i was too busy working out the cause
18:17:50 <roshi> just mine I think
18:18:23 * sgallagh returns
18:18:55 <adamw> just in time
18:20:23 <roshi> I presume you're +1 adamw
18:20:32 <roshi> pwhalen, kalev, jreznik ?
18:20:39 <adamw> i'm about +0.8 (see what i mean about conflict of interest)
18:20:49 <roshi> haha
18:20:54 * roshi rounds up
18:20:57 <adamw> sometimes Blocker/FE-Nominating-AdamW and Blocker/FE-Voting-AdamW have arguments
18:21:04 * sgallagh studies the bugs quickly
18:21:12 <adamw> (yeah, that nominating-adamw guy is a douche!)
18:21:43 <roshi> lol
18:21:51 <adamw> sgallagh: so if you have a user account that's only a member of a user private group, not any public groups, sssd fails to find that group membership
18:21:51 <kalev> sorry, wandered off to the kitchen
18:21:59 <adamw> at least i'm pretty sure that's the extent of the bug
18:22:10 <sgallagh> Geez, that came up again?
18:22:16 <kalev> let's just punt it and vote in the ticket when the fix is available?
18:22:25 <sgallagh> I fixed almost that exact bug back in 2011 :-/
18:22:26 <adamw> i believe the client system falls back to just knowing it has a gid but it doesn't know what the hell group it is, i don't know what the practical effects of that are
18:22:50 <sgallagh> Punting until a fix is known is probably reasonable
18:23:04 * roshi is fine with punting
18:23:14 <roshi> I don't grok what all this could touch
18:24:08 <sgallagh> roshi: For anyone enrolled in a domain, any change to how SSSD processes group memberships can have significant consequences
18:24:28 <sgallagh> It could change how authorization decisions turn out or just generally break POSIX permissions.
18:24:45 <sgallagh> So I'd be -0.5 right now without a patch I myself could test :)
18:24:47 <adamw> roshi: so far the consequence I know is 'when I do ls /home, /home/user001 has group (somelongnumericstring) not user001'
18:24:49 <roshi> that's no good
18:25:04 <adamw> i am fairly sure this implies rather worse consequences are lurking around the place, but i can't *prove* it :)
18:25:11 <adamw> oh, and you get an error message every time you log in
18:25:35 <adamw> sgallagh: https://fedorahosted.org/sssd/ticket/2436#comment:5
18:25:55 <roshi> punt +1 for me
18:26:01 <adamw> i'm fine with punt till we have a fix to look at
18:26:09 <kalev> punt +1
18:26:33 <adamw> i can quite easily test that a fix fixes the bug, but testing whether it has unexpected consequences in more complex environments is a bit more work (though i can at least also test any fix on my production domain)
18:26:33 <pwhalen> +1 punt
18:27:20 <roshi> proposed #agreed - 1139962 - Punt - We'll wait to decide on FE status until a fix is available because the full extent of what this bug represents isn't clear at this time.
18:28:23 <kalev> +1
18:28:23 <sgallagh> BTW, one late-addition to the FEs: BZ #1117432
18:28:29 <sgallagh> Ack
18:28:32 <kalev> ack
18:28:41 <pwhalen> ack
18:28:52 <roshi> that's two additional FE after this one we need to go over
18:28:57 <roshi> #agreed - 1139962 - Punt - We'll wait to decide on FE status until a fix is available because the full extent of what this bug represents isn't clear at this time.
18:29:14 <adamw> ack
18:29:15 <adamw> doh
18:29:27 <roshi> #topic (1134566) - pungi/lorax: failed to open magic file: could not find any valid magic files
18:29:35 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1134566
18:29:49 <roshi> #info Proposed FreezeException, lorax, NEW
18:30:42 <kalev> sure, the patch only touches ppc codepaths
18:31:30 <kalev> +1 FE
18:31:37 <adamw> i believe there's a matching pungi bug too
18:31:38 <gustavold> notice that bz includes a patch for pungi and a patch for lorax
18:31:40 <adamw> lemme find that one
18:31:48 <roshi> +1
18:32:00 <adamw> https://bugzilla.redhat.com/show_bug.cgi?id=1136535 is the pungi bug
18:32:03 <gustavold> the pungi bz is https://bugzilla.redhat.com/show_bug.cgi?id=1136535 (though it is just a clone)
18:32:05 <gustavold> :)
18:32:10 <adamw> i'm +1 FE to both as long as they don't touch non-PPC code
18:32:21 <adamw> it'd be nice to unblock the compose process for PPC folks so they can work on their alpha
18:32:25 <pwhalen> +1 fe
18:32:56 <roshi> so, everyone fine with applying their +1's to both bugs?
18:33:16 <adamw> +1 to that. +1s all around! +1+1+1
18:33:23 <pwhalen> heh
18:33:27 <roshi> I see your +1, and i +1 it
18:33:28 <adamw> it's like one of those 'gift' super mario levels around here
18:33:35 <adamw> 1-ups everywhere
18:33:45 <sgallagh> +1
18:33:47 <kalev> +1 to both, yes
18:35:22 <roshi> proposed #agreed - 1134566 - AcceptedFreezeException - Pulling in this fix to all the PPC folks to get their Alpha out, provided the fix doesn't touch code for any other arch.
18:35:44 <roshi> use the same reasoning for 1136535?
18:35:56 <adamw> i'd just put both bugs in the agreement
18:36:01 <adamw> efficiency!
18:36:06 <roshi> works for me
18:36:20 * adamw already marked them both as acceptedFE, so...:P
18:36:53 <roshi> proposed #agreed - 1134566, 1136535 - AcceptedFreezeException - Pulling in these fixes to allow the PPC folks to get their Alpha out, provided the fixes don't touch code for any other arch.
18:37:18 <roshi> well, if that's the case I should skip the proposed step
18:37:19 <sgallagh> Ack
18:37:31 <kalev> Ack
18:37:32 <pwhalen> ack
18:37:39 <roshi> #agreed - 1134566, 1136535 - AcceptedFreezeException - Pulling in these fixes to allow the PPC folks to get their Alpha out, provided the fixes don't touch code for any other arch.
18:37:43 <roshi> last one
18:38:02 <roshi> #topic (1117432) - split qrcode funnctionality out to a optional package
18:38:11 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1117432
18:38:33 <roshi> #info Proposed FreezeException, python-qrcode, ASSIGNED
18:38:37 <sgallagh> Short version: FreeIPA recently grew support for one-time-passwords and uses a QRCode to set up Google Authenticator/FreeOTP
18:38:52 <sgallagh> The dependency chain for python-qrcode is enormous and is pulling in way too many packages
18:39:23 <sgallagh> It's being split to reduce the dependency set for the core functionality.
18:39:42 <adamw> +1
18:39:43 <sgallagh> We asked for this as an FE on the grounds that it will shrink the size of the install media as well as the installed system
18:39:53 <roshi> is anything broken now? or just pulling in more than we'd like for minimal?
18:39:56 <sgallagh> Which is difficult to accomplish post-install
18:40:03 <sgallagh> roshi: Pulling in a lot more than we want.
18:40:05 <adamw> i noticed yesterday that when i do a minimal install it's ~260 packages, when I do a minimal install with freeipa-client it's ~350
18:40:24 <sgallagh> adamw: IIRC, about 150 of those are due to this dep.
18:40:28 <roshi> is that something we usually break freeze for?
18:40:31 <pwhalen> ouch
18:40:41 <adamw> roshi: for making the install images not needlessly bloated? i'd say sure
18:40:45 <roshi> (I'm not against it - just wondering what the history is for this kinda thing)
18:40:48 <adamw> it's not something that's come up much before, but seems reasonable
18:41:11 <pwhalen> +1
18:41:17 <roshi> yeah, I just didn't know what the precedence was - it makes sense to me to split it out
18:41:24 <roshi> +1
18:41:33 <adamw> i can't actually recall us having something like this as an FE before, oddly enough
18:41:43 <adamw> (it might have happened, but it doesn't stick out in my memory)
18:42:22 <kalev> we used to have a desktop spin blocker for getting it under 1000 MB and pulled in a lot of similar fixes there
18:42:29 <roshi> proposed #agreed - 1117432 - AcceptedFreezeException - To keep the minimal package set down it makes sense to break freeze for this package to be split.
18:42:42 <jsmith> ACK
18:43:06 <sgallagh> Ack
18:43:07 <kalev> ack, althought for the record I'll say I was +0 earlier since there's no actual fix to look at
18:43:17 <roshi> #agreed - 1117432 - AcceptedFreezeException - To keep the minimal package set down it makes sense to break freeze for this package to be split.
18:43:21 <sgallagh> kalev: I just built the fix 15 minutes ago
18:43:23 <pwhalen> ack
18:43:31 <roshi> and that's all we got for today
18:43:47 <roshi> and we're about out of time
18:44:03 <roshi> are there any accepted blockers or FE's that people want to look at with the remaining time?
18:44:30 <roshi> #info this is the first blocker review meeting that used the whole timeslot for F21
18:44:46 <kalev> wohoo, achievement unlocked
18:45:27 <roshi> :)
18:46:05 <adamw> =)
18:46:10 <adamw> er, acceptedblockers
18:46:28 <roshi> ?
18:46:48 <adamw> hold on, just checking the list
18:46:53 <roshi> np
18:48:47 <adamw> https://bugzilla.redhat.com/show_bug.cgi?id=1139015 and https://bugzilla.redhat.com/show_bug.cgi?id=1134524 are the ones really holding us up
18:48:56 <adamw> and https://bugzilla.redhat.com/show_bug.cgi?id=1088933 is *still* lying around
18:50:09 <adamw> it worries me that both 1139015 and 1134524 are kinda releng-y, could be causing more slippage
18:50:17 <roshi> 1088933 needs pinged
18:50:17 <adamw> jreznik: anything we can do there?
18:50:24 <adamw> i'll ping it
18:51:30 <roshi> well, 1139015 is the one we just talked about today
18:51:50 <roshi> and AIUI the other filtered groups bug is being worked on
18:52:49 <jreznik> ops, sorry - I had to go out with my dog
18:53:14 <adamw> 1139015 has been known for some days, though, and we haven't really figured out the problem
18:53:42 <roshi> true
18:53:53 <adamw> the 'make netinst work' is indeed being worked on, i'd be happy if we could somehow throw more resources than just dgilmore at it but i think he's happier doing it himself
18:53:59 <jreznik> 933 - dgilmore promised to finish the patch, it should be corner cases he's working on
18:54:01 <roshi> what truncates composes?
18:54:52 <jreznik> so maybe the question is, if we really want all corner cases covered or some initial patch should be ok for alpha
18:56:09 <kalev> in today's meeting, fesco briefly discussed the possibility to drop some deliverables to just get something out and get past the alpha release
18:56:50 <kalev> nothing got decided though, just some discussion
18:56:54 <adamw> kalev: unless 'both net installers' were in that list, it doesn't really help :)
18:57:16 <adamw> i can see dropping WS netinst for alpha, dropping server netinst seems less plausible
18:57:30 <jreznik> yep
18:57:48 <adamw> and honestly, at least as I understand the plan for fixing netinst, i'd really like it to be in at alpha so we can test it
18:58:06 <adamw> it's not hugely complex but i'd feel happier having it there all the way through the cycle so we catch any gremlins
18:58:27 <roshi> true
18:58:32 <kalev> yep
18:59:03 <jreznik> still it would be nice to release this year:)
18:59:11 <adamw> sure
18:59:33 <sgallagh> adamw: Actually, we *did* discuss dropping both netinsts for Alpha (only)
18:59:33 <adamw> we definitely need to do more testing on TC6 and catch other outstanding blockers
18:59:34 <jreznik> I'll reping Dennis again for 1088933, maybe we really don't need perfect patch for Alpha
18:59:42 <adamw> we don't want to sit around and twiddle thumbs on the netinst bug
18:59:44 <sgallagh> But that if-and-only-if it's the last thing holding up release
18:59:52 <sgallagh> Which it is not, currently
19:00:07 <adamw> jreznik: i've pinged the bug
19:00:08 <jreznik> sgallagh: almost it is
19:00:19 <adamw> jreznik: i'm actually kinda working on the assumption that's not *really* blocking alpha atm, but imbw
19:01:40 <roshi> is there anything we can do from this meeting now?
19:01:47 <adamw> go forth and test
19:01:54 <roshi> works for me
19:01:59 <adamw> find the alpha validation tests we haven't run and run 'em, stat
19:02:03 <roshi> gonna make a sandwich first though
19:02:15 <roshi> thanks for coming folks!
19:02:19 <roshi> #endmeeting