fesco
LOGS
18:00:01 <nirik> #startmeeting FESCO (2015-07-29)
18:00:01 <zodbot> Meeting started Wed Jul 29 18:00:01 2015 UTC.  The chair is nirik. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:01 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:00:01 <nirik> #meetingname fesco
18:00:01 <zodbot> The meeting name has been set to 'fesco'
18:00:02 <nirik> #chair ajax dgilmore hguemar jwb nirik paragan rishi thozza sgallagh
18:00:02 <nirik> #topic init process
18:00:02 <zodbot> Current chairs: ajax dgilmore hguemar jwb nirik paragan rishi sgallagh thozza
18:00:19 <jwb> hello
18:00:26 * rishi is here
18:00:29 <thozza> hi all
18:00:31 <sgallagh> /me is here-ish
18:00:48 <paragan> hi
18:01:01 <dgilmore> hey
18:01:30 <jreznik_pp> /me is here from phone and dinner with my parents, semi available
18:01:35 <jreznik_pp> And hi :)
18:02:14 <nirik> ok. lets go ahead and dive in then. ;)
18:02:22 <nirik> #topic ticket #1427	List of release blocking deliverables
18:02:22 <nirik> .fesco 1427
18:02:24 <zodbot> nirik: #1427 (List of release blocking deliverables) – FESCo - https://fedorahosted.org/fesco/ticket/1427
18:02:28 <nirik> jreznik_pp: where are we with this?
18:02:57 <nirik> https://fedoraproject.org/wiki/Fedora_Program_Management/ReleaseBlocking/Fedora23
18:03:15 <nirik> we need spins I guess and atomic/vagrant/whatever
18:03:24 <jreznik_pp> So there's wiki page, some initial idea of structure. I don't want to ask everyone to fill it in and then change
18:03:39 <jreznik_pp> So the first question what data do we want to gather?
18:04:25 <nirik> I think imagename/location/blocking or not is a good start
18:04:33 <jreznik_pp> It's a draft... But then we can put it into some json whatever for machine use or PDC
18:04:47 <nirik> we should also have liasons talk to each working group and actually confirm we have it right.
18:05:25 <jreznik_pp> Sure, that's the plan. As I said I was looking for a feedback not to do more than one round of reviews
18:05:55 <jreznik_pp> So I can finish it tomorrow/after tomorrow and ask for review/fixes
18:06:03 <nirik> sounds good to me.
18:06:23 <jreznik_pp> Sorry it took so long time...
18:07:18 <thozza> do we want to add some of the approved changes as well?
18:07:19 <nirik> anyone else have any thoughts on it?
18:07:21 <jreznik_pp> But with my transition, I didn't have much time for Fedora (and actually break for a few months was a nice thing;-)
18:07:58 <nirik> thozza: well, this is deliverables. Do any of them have specific deliverables?
18:08:18 <jreznik_pp> If yes, it will be an item on the list
18:08:30 <thozza> right, I think only the atomic stuff afaik
18:08:33 <jreznik_pp> But not change itself
18:08:48 <nirik> I suppose we could try and list size target too?
18:09:15 <jreznik_pp> nirik, maybe but I'm not that fan of size targets...
18:09:19 <rishi> nirik: That could be interesting for the Workstation live images.
18:09:20 <paragan> yes size information should be added
18:09:31 <jreznik_pp> Especially strictly enforced
18:09:53 <nirik> just a thought. we could leave it off for now, but it would help QA if there is a hard limit somewhere.
18:10:27 <jreznik_pp> nirik, we have the list so why not, just my personal opinion
18:10:31 <rishi> Hmm... on second thoughts, we did give up on the CD size limitation.
18:10:49 <jreznik_pp> That's what I'm talking about ;-))
18:10:49 <rishi> So, I guess, not a big deal anymore.
18:11:10 <nirik> ok, anything else on this? or shall we move on?
18:11:15 <jreznik_pp> I'll take a look
18:11:33 <rishi> nirik: I don't have anything else.
18:11:50 <nirik> jreznik_pp: lets leave the ticket open and you can update it when ready and we can review. :)
18:11:55 <nirik> thanks for working on it.
18:12:01 <jreznik_pp> Sure, np
18:12:08 <nirik> #topic ticket #1455	F23 System Wide Change: Standardized Passphrase Policy
18:12:08 <nirik> .fesco 1455
18:12:12 <zodbot> nirik: #1455 (F23 System Wide Change: Standardized Passphrase Policy) – FESCo - https://fedorahosted.org/fesco/ticket/1455
18:12:15 <nirik> ok, this bit o fun...
18:12:45 <nirik> I put a proposal in, but if folks prefer I can start a devel list discussion for feedback first...
18:12:56 <nirik> or if it looks ok we can try it and adjust as we go.
18:13:14 <paragan> +1 to https://fedorahosted.org/fesco/ticket/1455#comment:30
18:14:30 * nirik listens to the crickets chirp
18:14:56 * rishi is reading the ticket
18:15:00 <rishi> Still catching up after vacations.
18:15:18 <nirik> no worries. I understand if folks prefer more discussion...
18:16:15 <sgallagh> /me voted +1 in the ticket, but I'll do so again here
18:17:22 <thozza> nirik: +1 for your proposal
18:17:34 <jwb> hm
18:18:37 <nirik> thats +4 (I am also +1 to my own proposal)
18:18:39 <jwb> this really mostly depends on what libpasswdquality scores something as, right?
18:19:08 <nirik> well, not score, but if it returns an exception or not.
18:19:34 <sgallagh> jwb: Right, the major disconnect was that libpwquality doesn't intend for the score to be used in decision making
18:19:40 <nirik> score should be used only for indicatiors...
18:19:46 <sgallagh> It's a fuzzy value intended only for analog strength meters
18:19:56 <nirik> yeah
18:20:16 <jwb> hm
18:20:45 <jwb> i suppose +1.
18:20:53 <nirik> we can always revisit...
18:21:06 <rishi> nirik: Is https://fedorahosted.org/fesco/ticket/1455#comment:12 the actual draft of the policy?
18:21:29 <nirik> note that I will file anaconda bugs to match the policy, but likely they won't land until beta
18:21:37 <dgilmore> i am +1
18:22:05 <nirik> rishi: or https://fedorahosted.org/fesco/ticket/1455#comment:30
18:22:55 * nirik thinks workstation intends to override the policy for their product, but I guess we will see.
18:23:35 <sgallagh> I lost track of the thread on the WS mailing list
18:23:49 <sgallagh> stickster: You were going to summarize, I thought? :)
18:23:59 <nirik> well, they were talking about it before the policy even existed, which I thought was not too useful
18:24:05 <jwb> i think he's been rather busy and is still trying to get it done
18:24:38 <rishi> nirik: That looks sane to me.
18:24:38 <rishi> I was half-expecting something more verbose. Glad to see that it is just 4 bullet points.
18:24:58 <nirik> well, we can add to it, I just wanted something so we all started from the same place at least
18:25:09 <jwb> i'm worried that it hinges too much on pwdquality implementation, but not enough to not vote for it
18:25:30 <nirik> jwb: yeah, it's hard to otherwise quantify the set of passwords tho. ;(
18:25:39 <jwb> correct
18:25:44 <nirik> ok, thats +7
18:25:52 <rishi> Umm... isn't the libpwquality score dependent on minlen and dfiok too?
18:26:08 <nirik> no
18:26:19 <rishi> ok
18:26:30 <nirik> if you give it something too short or with too few different characters, it will just return an exception...
18:26:34 <nirik> saying why...
18:26:42 <nirik> there's no score in that case
18:27:26 <nirik> #agreed Change is accepted. (+7,0,0)
18:27:39 <nirik> #action nirik to file bugs on components to match policy
18:27:49 <nirik> #action nirik to write up wiki page with policy and announce it.
18:28:20 <nirik> #topic ticket #1463	upgrades for F23 and beyond
18:28:20 <nirik> .fesco 1463
18:28:21 <zodbot> nirik: #1463 (upgrades for F23 and beyond) – FESCo - https://fedorahosted.org/fesco/ticket/1463
18:28:43 <nirik> There's a change now here, but since our program manager is out we haven't announced/discussed it on list yet.
18:28:55 <nirik> do we want to vote on this change today? Or wait ? or
18:28:56 <sgallagh> So the conversation has begun, but it's not fully fleshed out yet
18:29:22 <nirik> sgallagh: you talked with packaging team on it?
18:29:35 <sgallagh> I have, and they're amenable to solving this.
18:29:40 <jreznik_pp> Jan didn't tell me to cover this for him... So I can't help here
18:30:00 <nirik> jreznik_pp: it happened after he was out I think. ;)
18:30:15 <sgallagh> They're not certain they want to implement this in dnf proper, but they're fine with it as a plugin that Fedora opts to ship by default
18:30:19 <jreznik_pp> Aha, so I missed it :(
18:30:35 <sgallagh> (It may move into the plugins-core or dnf itself after a probationary period)
18:30:59 <sgallagh> But the general sense here is that systemd's offline-updates approach makes sense and that it's ultimately a special-case of updates.
18:31:29 * nirik nods.
18:31:33 <sgallagh> We (FESCo) may want to make a ruling on whether we think distro-upgrades should do distro-sync or just the usual upgrade logic, but otherwise I think it's going to be pretty close to the PoC wwoods wrote.
18:31:42 <nirik> I am ok +1ing this change now, but we should still send it to the list to get wider attention...
18:32:30 <jwb> that seems backwards
18:33:10 * zbyszek wonders whether to submit dnf-plugins-core package for review, or to wait
18:33:12 <nirik> well, if we just approve this, other people who don't pay attention to fesco meetings may go "wha?"
18:33:32 <jwb> right.  which is why i said approving and then announcing seems backwards
18:33:33 <nirik> we need to make it widely known that this is happening so docs, qa, etc, etc can all see it.
18:33:42 <sgallagh> zbyszek: Hold that thought for Open Floor, please?
18:33:48 <rishi> Wondering about https://fedorahosted.org/fesco/ticket/1463#comment:4 ...
18:33:53 <nirik> well, we can hold off approving, but I don't see any alternative.
18:34:00 <nirik> I guess someone else could come up with one.
18:34:04 <rishi> How does this affect the plans to have system upgrade in PackageKit?
18:34:25 <nirik> rishi: for f22-f23 not at all.
18:34:25 <sgallagh> rishi: Richard Hughes has been part of the discussion I've been involved in
18:35:02 <rishi> sgallagh: Ok. That answers my question. :)
18:35:09 <jwb> sgallagh, it is wonderful that you are having these discussions.  where are they being discussed?
18:35:13 <sgallagh> He's suggested resurrecting an older PackageKit function that got removed called UpgradeSystem() and implementing this there in the same manner.
18:35:34 <sgallagh> Unclear if he'd do that by calling out to DNF (in some appropriate wrapper) or reimplement
18:35:43 <sgallagh> But it looks like he plans to use the same logic and approach
18:36:21 <rishi> sgallagh: Ok. I will trust hughsie and the dnf guys to sort it out between themselves and do the right thing.
18:36:36 <nirik> proposal: send change to list for comment, revisit next week.
18:36:36 <rishi> I was just curious if the left hand knew what the right is doing.
18:36:41 <rishi> nirik: +1
18:36:42 <thozza> nirik: +1
18:36:42 <sgallagh> jwb: They've been private, but will be public within the next day or so
18:37:04 <jwb> sgallagh, disappointing that we're going to vote based on information that is not public.
18:37:08 <jwb> so -1
18:37:29 <paragan> I am okay with nirik proposal +1
18:37:36 <thozza> jwb: we are voting about postponing to next week
18:37:50 <paragan> there is some discussion happening on rpm-ecosystem list
18:38:01 <jwb> thozza, that isn't clear to me
18:38:01 <paragan> http://lists.rpm.org/pipermail/rpm-ecosystem/Week-of-Mon-20150727/thread.html
18:38:13 <sgallagh> Oh ok. I knew it was headed there. DIdn't know it had startd.
18:38:16 <thozza> jwb: (20:36:35) nirik: proposal: send change to list for comment, revisit next week.
18:38:17 <jwb> oh, i see nirik's line now
18:38:21 <nirik> well, I put up that as a proposal, but whicheveer
18:38:31 <jwb> sure, i'll +1 deferring until discussion is actually public
18:38:39 <nirik> perhaps next week we will have good public info. ;)
18:38:54 <sgallagh> Ah, I thought I was subscribed there. I'm not. Whoops
18:39:21 <nirik> thats +5 for sending to list and revisitig next week
18:39:37 <nirik> any other votes?
18:39:43 <sgallagh> 0
18:39:56 <sgallagh> I don't see a lot of alternatives
18:39:59 <nirik> #agreed will send change proposal to list and revist next week (+5,0,0)
18:40:07 <dgilmore> +1 to nirik
18:40:10 <nirik> sgallagh: me either, but perhaps the collective is smarter than all of us. ;)
18:40:16 <nirik> #undo
18:40:16 <zodbot> Removing item from minutes: AGREED by nirik at 18:39:59 : will send change proposal to list and revist next week (+5,0,0)
18:40:20 <nirik> #agreed will send change proposal to list and revist next week (+6,0,0)
18:40:32 <nirik> jreznik_pp: can you send the change to devel-announce? or would you like me to?
18:40:57 <jreznik_pp> nirik, pls do it
18:41:05 <nirik> ok. can do
18:41:05 <jreznik_pp> I'm on phone now
18:41:15 <nirik> #action nirik to send proposed change to list to gather feedback
18:41:18 <nirik> #topic ticket #1466	non-responsive maintainer exception process for skottler
18:41:18 <nirik> .fesco 1466
18:41:19 <zodbot> nirik: #1466 (non-responsive maintainer exception process for skottler) – FESCo - https://fedorahosted.org/fesco/ticket/1466
18:42:05 <nirik> so, I don't think I have enough info here to say much... I would probibly support a mediator to talk to the concerned parties and recommend an action to us.
18:42:27 <dgilmore> the CVE's should be fixed
18:42:42 <nirik> that I can agree with.
18:42:44 <dgilmore> but I think there needs to be some mediation here
18:44:05 <swilkerson> may I add additional information here?
18:44:25 <nirik> swilkerson: sure if you like.
18:44:26 <rishi> I guess we need to reach out to skottler to resolve the "hostile takeover" issue.
18:44:32 <swilkerson> skottler has publicly stated "I’m not particuarly interested in working with Nagios Enterprises" which in my opinion is not in the best interest of the Fedora users.  He currently is the maintainer for both nagios-plugins and nrpe packages where Nagios Enterprises is the upstream.  I believe this is contrary to the responsibilities of a package mai
18:44:33 <swilkerson> ntainer, who should be close to upstream projects, working with the project to build a better user experience, expediting security fixes, etc.
18:44:38 <swilkerson> Additionally, skottler has also stated that once a different project he is working on is available he may orphan the nagios-plugins package, seemingly just because he doesn't like Nagios Enterprises, despite the fact that the nagios-plugins package is widely used in Fedora, CentOS and RHEL communities.
18:45:08 <jwb> great.  then we can move on and skip all this crap
18:45:14 <jwb> i'll talk to sam
18:45:29 <dgilmore> jwb: cheers
18:46:05 <nirik> fine with me. If he intends to orphan it, someone else could take it over and get things updated.
18:46:32 <nirik> proposal: jwb to talk to maintainer and work out some mutially agreeable solution
18:46:34 <rishi> swilkerson: Is skottler opposed to having you as a co-maintainer?
18:46:40 <swilkerson> yes
18:46:47 <rishi> Oh. That is sad.
18:47:00 * nirik is +1 to the proposal.
18:47:00 <swilkerson> because he wants to push the package to his project
18:47:05 <swilkerson> monitoring-plugins
18:47:35 <swilkerson> I actually don't care who maintains the package, but don't want to see that happen
18:47:46 <swilkerson> and want to see the CVE's fixed
18:48:14 <swilkerson> so I offered to take it over when David Cafaro contacted me
18:48:38 <rishi> nirik: Yeah, I am +1 to the proposal too.
18:49:14 <rishi> swilkerson: What does "push the package to his project monitoring-plugins" mean?
18:49:26 <paragan> nirik, +1
18:49:30 <thozza> nirik: +1
18:49:39 <rishi> Is skottler planning to ship something else under the nagios-plugins package name?
18:49:53 <jwb> no
18:49:53 <swilkerson> rishi: yes
18:50:25 * nirik doesn't think getting into this without skottler here to speak for himself is going to be that productive.
18:51:29 <nirik> so thats +4 I think, any other votes?
18:51:36 <rishi> nirik: True, but I could not help clarifying this statement because it sounded too crazy at first.
18:51:40 <swilkerson> It was proposed somewhere in this https://bugzilla.redhat.com/show_bug.cgi?id=1054340
18:51:55 <swilkerson> about a year ago
18:52:47 <dgilmore> +!
18:52:50 <dgilmore> +1
18:53:06 * nirik isn't going to comb thru 86 comments right now. ;)
18:53:47 <nirik> ok, thats 5
18:54:06 <nirik> #agreed jwb to talk to maintainer and work out some mutially agreeable solution (+5,0,0)
18:54:21 <nirik> jwb: keep the ticket posted?
18:54:25 <jwb> yeah
18:54:42 <nirik> #topic ticket #1467     F23 Changes - Progress at Change Checkpoint: Completion deadline (testable)
18:54:43 <nirik> .fesco 1467
18:54:44 <zodbot> nirik: #1467 (F23 Changes - Progress at Change Checkpoint: Completion deadline (testable)) – FESCo - https://fedorahosted.org/fesco/ticket/1467
18:54:50 <nirik> this came in after the agenda went out...
18:55:15 <nirik> btw, swilkerson, thanks for coming. :)
18:55:59 <swilkerson> np
18:56:04 <jreznik_pp> Yep, too late
18:56:28 <jreznik_pp> It's more reminder for some FESCo members to take a look on theirs changes
18:56:51 <nirik> do the change owners move the bugs to MODIFIED or ?
18:56:55 <jreznik_pp> And tomorrow I'll work with Jan to poke the rest of folks
18:57:14 <jreznik_pp> Yep, MODIFIED means it's testable and in correct state
18:57:42 <nirik> kushal: is cloud motd and cloud networkd ready to test? (I thought so, but they are on the list here)
18:58:17 <paragan> I will check "Glibc locale subpackaging" with my teammate
18:58:20 <kushal> nirik, I have an image with networkd (locally built one). motd is not ready to test yet (but will be in the coming weeks).
18:58:56 <nirik> looks like freeze crept up on lots of people.
18:59:12 <thozza> The default DNS resolver is testable to some extent, the changes agreed with GNOME and NM are still being worked on... We reached agreement kind of late :-/
18:59:22 <dgilmore> nirik: this whole release has snuck up on folks
18:59:36 <jreznik_pp> Or less poking from me as I usually spent quite a lot of time hunting folks to update change bugs
19:00:14 <nirik> so, what do we want to do here?
19:00:26 <ajax> hi
19:00:29 <jreznik_pp> For the next meeting we will report final results
19:01:29 <nirik> ok, but sounds like some of these things didn't land, we let them land late? or allow them to be FE's?
19:02:20 <jreznik_pp> Sort it out case by case
19:02:24 <nirik> I guess many of them we just don't know. ok
19:02:29 <sgallagh> How are composes going right now? Are we likely to slip for infra/rel-eng stuff anyway?
19:02:39 <nirik> we got a tc2. ;)
19:02:45 <dgilmore> sgallagh: they are fine
19:02:47 <sgallagh> OK
19:02:57 <dgilmore> sgallagh: though atomic is completely busted
19:03:07 <sgallagh> /me sighs
19:03:18 <nirik> dgilmore: is that the 'cannot find package atomic' thing?
19:03:36 <sgallagh> Well, the policy is there to keep us on track. I think we allow them to propose FEs, but the blocker process folks don't have to take them
19:03:40 <dgilmore> but I think that is the same issue that was causing us updates pushing problems and has been dealt with
19:03:41 <kushal> dgilmore, Is there any way to get messages for any failed createImage task from koii?
19:03:45 <nirik> if so, I think it got sorted since we hit it with f22-updates-testing hell.
19:03:45 <dgilmore> nirik: yes
19:03:49 <nirik> right
19:03:57 <dgilmore> kushal: that is off topic for here
19:04:31 <nirik> so, move on? or anything else on these?
19:04:47 <kushal> dgilmore, Oh, sorry. (will ask in the right place).
19:05:27 <nirik> #topic Next weeks chair
19:05:29 <nirik> who wants it?
19:06:24 <rishi> Not me, because I would be too busy preparing for GUADEC next week.
19:06:25 <thozza> I may not be available next week, since I'll be traveling to US for Flock (little bit earlier)
19:07:29 <nirik> anyone? anyone? buller. I guess I could run it again, but would rather not. ;)
19:07:49 <paragan> I can then chair next week :)
19:07:57 <nirik> paragan: thanks!
19:08:07 <nirik> #action paragan to chair next week.
19:08:12 <nirik> #topic Open Floor
19:08:18 <nirik> anyone have things for open floor?
19:08:26 <swilkerson> jwb: One last bit of information on .1466 - Their is a ticket to change the upstream source from nagios-plugins to monitoring-plugins - https://bugzilla.redhat.com/show_bug.cgi?id=1054340
19:09:46 <swilkerson> and Anitya for nagios-plugins already has been changed to look at Homepage:  http://www.monitoring-plugins.org/download.html
19:10:36 <nirik> alright
19:10:46 <nirik> (well, not alright, but thanks for the info)
19:11:08 <nirik> if nothing else will close out in a minute here.
19:11:16 * zbyszek wonders whether to submit dnf-plugins-core package for review, or to wait
19:11:34 <sgallagh> zbyszek: Isn't there already a package for that
19:11:36 <sgallagh> ?
19:11:43 <paragan> zbyszek, that is already in Fedora
19:11:54 <thozza> it sure is
19:11:56 <sgallagh> zbyszek: http://koji.fedoraproject.org/koji/packageinfo?packageID=17599
19:11:56 <nirik> you mean the upgrade plugin?
19:12:01 <zbyszek> I meant dnf-plugin-fedup of course, sorry.
19:12:11 <nirik> right
19:12:17 <paragan> zbyszek, submit a pull request to upstream
19:12:35 <sgallagh> zbyszek: Yeah, I'd much rather see this as part of the upstream plugin packages
19:12:36 <nirik> well, I guess I would say hold off until you know what dnf folks prefer?
19:12:45 <sgallagh> ideally dnf-plugins-core
19:12:58 <zbyszek> OK. I will wait then.
19:13:02 <nirik> if they are willing to take it,, great, but it sounded like they wanted it to be standalone until it proved itself?
19:13:07 <thozza> I think some people may want to use it regardless of what will be decided and implemented in the end.
19:13:17 <zbyszek> nirik: yeah
19:13:32 <thozza> but yes, whatever DNF folks prefer
19:13:43 <sgallagh> nirik: I think they're willing to carry it in plugins-extras at minimum
19:13:48 <nirik> sgallagh: ok.
19:13:56 <nirik> yeah, just work that out with them. :)
19:14:14 <sgallagh> And we (Fedora) can always choose to force that plugin to be part of the standard install
19:15:34 <nirik> sure.
19:15:41 <nirik> ok, thanks for coming everyone.
19:15:44 <nirik> #endmeeting