17:00:14 <kushal> #startmeeting Fedora Cloud WG
17:00:14 <zodbot> Meeting started Wed Aug  5 17:00:14 2015 UTC.  The chair is kushal. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:00:23 <kushal> #topic Roll Call
17:00:24 <roshi> .hello roshi
17:00:25 <zodbot> roshi: roshi 'Mike Ruckman' <mruckman@redhat.com>
17:00:25 <dustymabe> .hellomynameis dustymabe
17:00:27 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dusty@dustymabe.com>
17:00:29 <kushal> .hellomynameis kushal
17:00:29 <number80> .hello hguemar
17:00:30 <zodbot> kushal: kushal 'Kushal Das' <kushaldas@gmail.com>
17:00:33 <zodbot> number80: hguemar 'Haïkel Guémar' <karlthered@gmail.com>
17:00:40 <sgallagh> /me lurks
17:01:32 <sayan> .hellomynameis sayanchowdhury
17:01:33 <zodbot> sayan: sayanchowdhury 'Sayan Chowdhury' <sayan.chowdhury2012@gmail.com>
17:01:49 <kushal> #chair roshi dustymabe number80 sgallagh sayan gholms
17:01:49 <zodbot> Current chairs: dustymabe gholms kushal number80 roshi sayan sgallagh
17:01:54 <kushal> Anyone else?
17:02:04 <rtnpro> .hello rtnpro
17:02:05 <zodbot> rtnpro: rtnpro 'Ratnadeep Debnath' <rtnpro@gmail.com>
17:02:11 <kushal> #chair rtnpro
17:02:11 <zodbot> Current chairs: dustymabe gholms kushal number80 roshi rtnpro sayan sgallagh
17:02:20 <jbrooks> .hello jasonbrooks
17:02:21 <zodbot> jbrooks: jasonbrooks 'Jason Brooks' <JBROOKS@REDHAT.COM>
17:02:47 <dustymabe> jbrooks: welcome
17:02:52 <kushal> jbrooks, Hello
17:02:57 <jbrooks> Hey
17:03:16 <kushal> #chair jbrooks
17:03:16 <zodbot> Current chairs: dustymabe gholms jbrooks kushal number80 roshi rtnpro sayan sgallagh
17:04:01 <kushal> #topic Previous action items
17:04:23 <kushal> Anyone remembers them? I can not find any log on the site :(
17:04:41 <roshi> jas, lemme look
17:05:07 <roshi> http://meetbot-raw.fedoraproject.org/fedora-meeting-1/2015-07-29/cloud_wg.2015-07-29-17.09.html
17:05:30 <kushal> Oh, why I could not found them :(
17:05:42 <kushal> kushal to file releng ticket for updated cloud images
17:05:43 <kushal> kushal to update releng ticket #6219 with what packages were updated
17:05:45 <kushal> jbrooks to update ticket #105 with status
17:05:46 <dustymabe> kushal: great Indian firewall :)
17:05:47 <kushal> dustymabe jbrooks to work on #105 next monday (2015-08-03)
17:05:49 <kushal> roshi to update ticket #14 with status
17:05:51 <kushal> lalatenduM to update ticket #84 with his post to the mailing list
17:05:52 <roshi> I just go to meetbot-raw.fp.o and click the date - easier than the new UI I think
17:05:53 <kushal> dustymabe, rofl
17:06:00 <kushal> roshi, I tried that only.
17:06:22 <roshi> I haven't updated #14 yet - but will do it
17:06:41 <dustymabe> so I didn't work with jbrooks but I think I figured out his problem
17:06:56 <jbrooks> dustymabe, I saw your comment on my gist, and I updated the gist
17:07:00 <jbrooks> So it's working now
17:07:12 <kushal> I did both the items, and now we have https://fedorahosted.org/rel-eng/ticket/6219#comment:1
17:07:17 <jbrooks> I'm wondering if it'd make sense to create an rpm w/ this service file
17:07:43 <dustymabe> In case anyone was wondering what the problem was I wrote a short blog post about it:
17:07:45 <dustymabe> http://dustymabe.com/2015/08/03/installingstarting-systemd-services-using-cloud-init/
17:07:45 <jbrooks> It'd be much simpler, and the kickstart case would be simpler, too
17:08:02 <jbrooks> https://gist.github.com/jasonbrooks/20cd3ebba36f851b957b
17:08:45 <dustymabe> jbrooks: interesting.. I don't know
17:08:56 <dustymabe> I imagine we are going to hit a similar problem over and over again
17:09:07 <dustymabe> the problem with making an rpm out of it is that this is Atomic Host
17:09:19 <dustymabe> and you can't just install an rpm
17:09:31 <jbrooks> Yeah, we'd be putting it in the tree
17:09:41 <jbrooks> Otherwise it wouldn't be useful
17:09:46 <dustymabe> but you could include it in the tree. but that assumes that people "want" cockpit
17:09:59 <dustymabe> is that a safe assumption
17:10:17 <jbrooks> Disable by default -- we're already putting cockpit stuff in the tree -- one more service file is tiny
17:10:19 <dustymabe> if it were then I don't know why we aren't just including it in the tree as well
17:10:37 <dustymabe> oh yeah. what cockpit stuff?
17:10:37 <jbrooks> I know I've seen some discussion of this somewhere...
17:10:49 <jbrooks> the cockpit-bridge and cockpit-shell
17:10:59 <kushal> Okay, what are those?
17:11:08 <dustymabe> ok. so the people surrounding that discussion are probably ones to ask then
17:11:22 <jbrooks> Those are the minimal bits needed on a host to be manageable w/ cockpit
17:11:36 <jbrooks> And then the app itself is what's now off in a container
17:12:24 <kushal> jbrooks, Okay.
17:12:35 <kushal> I still wonder why don't just that too as in rpm?
17:12:39 <kushal> * just keep
17:12:40 <dustymabe> jbrooks: I guess the only thing for me is I would want the name of the service to be explicit about the fact it is starting a container
17:13:09 <dustymabe> so it doesn't conflict with the names of the services from the actual cockpit rpm
17:13:13 <jbrooks> dustymabe, The makes sense -- we could be providing an example of how to do this right
17:13:24 <jbrooks> Ah, hmm, well, something to consider
17:13:52 <dustymabe> yes. because theoretically people could install the cockpit rpms on non-atomic and just use it "not" in a container
17:14:12 <dustymabe> and then things start to get confusing if we don't name the services differently
17:14:24 <dustymabe> my 2 cents
17:14:26 <jbrooks> Yep, that'd make sense
17:14:41 <dustymabe> cool :)
17:15:00 <dustymabe> tickets?
17:15:07 <jbrooks> Maybe I'll update the ticket w/ some mention of this...
17:15:08 <dustymabe> kushal ^^
17:15:13 <dustymabe> jbrooks: ++
17:15:24 <kushal> dustymabe +1 to that installation point.
17:16:00 <kushal> I guess we can now move to the tickets.
17:16:42 <kushal> #topic  https://fedorahosted.org/cloud/ticket/105
17:16:56 <kushal> #topic  Missing Cockpit RPMs in Fedora Atomic 22 https://fedorahosted.org/cloud/ticket/105
17:17:23 <kushal> jbrooks, dustymabe may be you two can update that ticket now :)
17:17:35 <jbrooks> will do
17:18:09 * jsmith arrives late, sorry
17:18:16 * dustymabe wavews
17:18:21 * dustymabe waves
17:18:27 <gholms> o/
17:18:41 * roshi wavews too, sounds fun :p
17:18:53 <kushal> #chair gholms jsmith
17:18:53 <zodbot> Current chairs: dustymabe gholms jbrooks jsmith kushal number80 roshi rtnpro sayan sgallagh
17:20:16 <kushal> #topic Investigate systemd-networkd https://fedorahosted.org/cloud/ticket/14
17:20:32 <kushal> I have released one AMI (public) for testing the same.
17:20:41 <kushal> I am not sure if anyone from here tested that.
17:20:48 <dustymabe> kushal: I did
17:20:48 <kushal> I forgot to update the ticket.
17:20:54 <kushal> dustymabe, Ah, any comments?
17:20:57 <roshi> I didn't get the ticket updated either
17:21:10 <dustymabe> I sent you some feedback (I think I commented on your blog post or in the ticket)
17:21:12 <dustymabe> can't remember
17:21:19 <kushal> dustymabe, yup, on the qcow2 image
17:21:27 <kushal> I think about /etc/resolve.conf
17:21:36 <kushal> That is why in the second post, I mentioned that
17:21:47 <dustymabe> ooh. there was a new post?
17:21:54 <kushal> http://kushaldas.in/posts/testing-systemd-networkd-based-fedora-22-ami-s.html
17:22:11 <dustymabe> missed that one
17:22:15 <kushal> dustymabe, :)
17:22:33 <dustymabe> any chance you posted this series of blogs posts to the fedora cloud list?
17:22:35 <kushal> dustymabe, So the first post contained the qcow2 image, this one has the AMI
17:22:39 <dustymabe> might be worth it to promote
17:22:49 <kushal> dustymabe, will do so tonight.
17:23:02 * kushal should make a separate list of writings he has to do.
17:23:41 <dustymabe> ok yeah. that is one change that would need to be communicated though right?
17:23:48 <dustymabe> the resolv.conf thing?
17:24:13 <kushal> dustymabe, Yes, it is link in default, other wise one can right normal resolve.conf
17:24:16 <dustymabe> or would it make sense to disable that behavior
17:24:29 <kushal> dustymabe, we have to use the systemd resolver.
17:24:48 <dustymabe> ok.
17:24:52 * dustymabe needs to read up on that
17:25:14 <dustymabe> so in all reality this is going to be slated for F24 right ?
17:25:19 <dustymabe> I think it has missed F23
17:26:32 <kushal> dustymabe, that way, none of the features are in for F23
17:26:36 <kushal> in alpha
17:26:44 <kushal> dustymabe, I have showed this work, and asked for beta
17:27:02 <dustymabe> roshi: ^^
17:27:09 <dustymabe> I know you had an opinion on this
17:27:16 <kushal> dustymabe, yes he should  :)
17:27:20 <lalatenduM> RE: lalatenduM to update ticket #84 with his post to the mailing list -> I have done that
17:27:43 <roshi> kushal: if it's not in alpha, we can't change it mid release
17:27:56 <kushal> roshi, Okay.
17:27:58 <roshi> the earliest we can get networkd into the cloud image is F24 at this point
17:28:05 <kushal> roshi, understood.
17:28:19 <roshi> and we still need to have a long discussion as to if it's a good idea to switch, if the benefits outweigh the cost
17:28:28 <dustymabe> we could get it into rawhide now though couldn't we - if we wanted to?
17:28:38 <roshi> first of which is to write down what we expect from our network stack
17:28:43 <kushal> roshi, but that same will happen to any other feature? say Two week atomic?
17:28:57 <roshi> can't really do a comparison of two stacks w/o a standard to compare to
17:29:09 <roshi> atomic is a spin - they can change all they want
17:29:16 <kushal> roshi, Okay.
17:29:18 <dustymabe> roshi: the "what is required by network" discussion could happen at flock?
17:29:30 <roshi> since it's all on them to produce, test and distribute
17:29:35 <dustymabe> might be good to carve out some time for it
17:29:46 <roshi> we could for sure get the ball rolling on it
17:30:00 <kushal> dustymabe, we should include it on the cloud meeting then.
17:30:30 <dustymabe> is there a place where information about that lives that can be updated?
17:30:56 <roshi> about the meeting at flock?
17:31:10 <dustymabe> yeah.. like agenda items?
17:31:17 * gholms likes this plan
17:31:49 <roshi> we should start a page on the wiki
17:31:49 <dustymabe> yeah we can get you three in a room and let you fight it out :)
17:31:54 <kushal> dustymabe, I will create a wiki page.
17:31:54 <dustymabe> whoever lives wins
17:31:58 * roshi will create one and post to the cloud list
17:32:02 <roshi> or kushal can :)
17:32:04 <kushal> roshi, will doit
17:32:07 <dustymabe> or is that whomever ?
17:32:07 <kushal> heheh
17:32:09 <dustymabe> sigh
17:32:26 <dustymabe> #englishasafirstlanguage
17:32:27 <kushal> roshi, Okay, I will do it.
17:32:42 <roshi> haha
17:32:46 <roshi> sounds good kushal
17:32:51 <kushal> #action kushal will create cloud meeting agenda page for flock and post it to the list.
17:33:08 <dustymabe> I've got something for that agenda
17:33:20 <dustymabe> I'll add once the page is created
17:33:26 <kushal> Remember we will have that chat with CentOS team there too.
17:33:28 <roshi> could be an evening dinner discussion :)
17:33:30 <kushal> dustymabe, Thanks :)
17:33:45 <roshi> kushal: yeah, we should let them know about the wiki page so they can add stuff as well
17:33:51 <kushal> roshi, Okay
17:34:16 <kushal> Moving to next ticket.
17:34:28 <kushal> #topic Care and Feeding, Fedora Dockerfiles https://fedorahosted.org/cloud/ticket/84
17:35:11 <kushal> sgallagh, needs an update to an image.
17:35:20 <dustymabe> yeah he just opened a new ticket
17:35:27 <dustymabe> and I gave it keyword of meeting
17:35:36 <kushal> dustymabe, Thanks :)
17:35:44 <dustymabe> https://fedorahosted.org/cloud/ticket/114
17:35:53 <kushal> Update fedora/memcached in Docker Registry
17:36:21 <dustymabe> yeah. we need to get this whole process more figured out
17:37:04 <kushal> dustymabe, hopefully with layer image build process we can automate it.
17:37:30 <dustymabe> yeah. either way we need someone to pick up the torch on this exact item and carry it through to completion
17:37:46 <dustymabe> meaning knocking on doors.. cold calls, etc..
17:37:50 <dustymabe> anyone up for it?
17:38:15 <dustymabe> the whole process needs to be improved but this item is a blocker for something else so we need to just get it done for now
17:38:22 <roshi> I'm short on cycles and don't know much about docker
17:38:30 <kushal> dustymabe, Yeah, I will ping scoiler
17:38:41 <kushal> only problem is that I will be traveling after tomorrow night.
17:38:44 <kushal> For Flock
17:39:10 <dustymabe> kushal: ok. in that case i'll pick it up for now. If I need to hand off for some reason I will reach back out to you
17:39:18 <kushal> dustymabe, Okay
17:39:27 <dustymabe> i'll get in touch with scollier
17:39:47 <kushal> Moving to next ticket
17:39:50 <dustymabe> lalatenduM: did you see what scollier wrote ?
17:39:53 <dustymabe> kushal: one sec
17:39:58 <lalatenduM> dustymabe: yes
17:40:02 <dustymabe> he created new branches I think?
17:40:06 * kushal pushes the break hard.
17:40:10 <lalatenduM> dustymabe: yup
17:40:26 <dustymabe> does that help you do whatever work you were planning to do?
17:40:27 <lalatenduM> We need CI now
17:40:37 <dustymabe> ok just wondering
17:40:42 <dustymabe> kushal: go ahead
17:40:50 <lalatenduM> because even if I change the dockerfiles for dnf , I need a easy way to test them
17:40:55 <kushal> #topic Create text for Fedora cloud Flyer https://fedorahosted.org/cloud/ticket/108
17:40:59 <kushal> #topic Create text for Fedora cloud Flyer https://fedorahosted.org/cloud/ticket/108
17:41:05 <lalatenduM> dustymabe: yes, the branches will help
17:41:12 <dustymabe> kushal: number80 gave update in the ticket
17:41:18 <kushal> Yup.
17:41:32 <kushal> Moving to next bigger ticket then.
17:41:36 <dustymabe> seems like he has not had time. if you want to write down a paragraph to send them then I'm sure
17:41:43 <dustymabe> he wouldn't mind
17:42:19 <kushal> dustymabe, True, but I was hoping someone better than me (in writing) should do it.
17:42:36 <dustymabe> :) - if you write down some content we can clean it up
17:42:52 <dustymabe> I just need the content -> I'm bad at that but should get better now that I can focus on things more
17:43:21 <roshi> I can help clean it up
17:43:41 <dustymabe> yeah so if you put some content in the ticket we can clean it up and get it sent off
17:44:04 <kushal> Okay, will do that, but by then we all be in Flock :)
17:44:33 <kushal> #topic Producing Updated Cloud/Atomic Images https://fedorahosted.org/cloud/ticket/94
17:44:49 <kushal> I put a ticket to rel-eng
17:45:08 <kushal> we can see that we may want to do another spin with the given rpm and release that instead.
17:45:13 <kushal> Any thoughts?
17:45:30 <dustymabe> yeah let's do another spin to get the libuser stuff
17:45:36 <dustymabe> and then promote that for people to test
17:46:50 <kushal> Yeah.
17:46:58 <kushal> dustymabe, We are actually taking too much to test :(
17:47:19 <kushal> I hope that automatic testing from the two week atomic side will help us in future
17:47:27 <roshi> we just need some people dedicated to testing
17:47:34 <roshi> and documenting it
17:47:36 <dustymabe> roshi: agreed
17:47:52 <roshi> QA folks can pretty much only test on testcloud - unless they have their own cloud accounts they can get into
17:48:12 <dustymabe> we can try to recruit more from the community - do we have any paid openings too?
17:48:35 <roshi> not that I know of
17:48:45 <dustymabe> might be something we should bring up
17:48:49 <roshi> the infra team is hiring for oddshocks position
17:48:56 <dustymabe> if people don't know we need people then we won't get any
17:49:02 <roshi> but unsure how much testing time that'd be
17:49:14 <roshi> someone to care for fedimg and work with releng
17:49:52 <roshi> http://paul.frields.org/2015/07/21/fedora-engineering-team-opening/
17:49:57 <roshi> but that's all I know of
17:51:24 <dustymabe> ok
17:51:32 <kdas__> roshi, that would be
17:51:37 <dustymabe> either way if we want to release a new 22 image then we have to test it some
17:51:38 <kdas__> kushal here, network issue, can someone run and finish the meeting.
17:51:40 <kdas__> roshi, I will be handling it for now.
17:51:42 <kdas__> s/handling/maintaining
17:51:44 <kdas__> kushal here, network issue, can someone run and finish the meeting.
17:51:57 <roshi> sure
17:52:08 <kdas__> dustymabe, I did the testing for that updated image
17:52:50 <roshi> #topic 32 bit images
17:52:54 <roshi> #link https://fedorahosted.org/cloud/ticket/106
17:52:56 <dustymabe> kdas__: ok. well I still say we send out an email to the list before we release. if no one else tests then that is fine too
17:53:30 <kdas__> dustymabe, Yes, I am also saying that we need more people.
17:53:47 <dustymabe> kdas__: who should we talk to about needing more people?
17:54:21 <roshi> #topic Open Floor
17:54:24 <kdas__> dustymabe, More blog posts, and emails
17:54:33 <roshi> we're about out of time, so it seemed like a good time for open floor
17:54:38 <dustymabe> roshi: ok
17:54:45 <dustymabe> for the 32bit image thing
17:54:49 <roshi> dustymabe: talking about paid people or community members?
17:54:50 <kdas__> For openfloor, I need to ask about dustymabe's email on the list.
17:54:54 <dustymabe> who do we talk to about making that change?
17:55:01 <dustymabe> roshi: both :)
17:55:41 <roshi> dustymabe: for the 32bit image, it's a releng ticket and fesco sign off I think
17:55:45 <roshi> but that's a guess
17:55:58 <roshi> since it's still a primary arch
17:56:13 <dustymabe> ok. i'm not familiar with the whole process but next steps would be good. I'll make a comment in the ticket
17:56:23 <dustymabe> kdas__: ask away
17:56:28 <roshi> kk
17:56:39 <roshi> as for people - i don't know how that goes
17:56:40 <kdas__> dustymabe, I mean if anyone knows more about how that tool works :)
17:56:42 <roshi> for RH at least
17:56:54 <kdas__> rtnpro, ^^^^^^^^^^^^^^^^^^^^ may be you can look into it.
17:57:08 <kdas__> rtnpro, look for dustymabe's latest email on the cloud list.
17:58:10 <dustymabe> kdas__: ok. yeah I think some people will respond on the list to confirm that the behavior is not what we want
17:58:18 <dustymabe> and then we can file a bug and patch it
17:59:18 <rtnpro> yes
17:59:41 <kdas__> #action rtnpro to test docker-storage as mention in the cloud list
18:01:10 <dustymabe> ok. we done here?
18:02:00 <kdas__> I guess so
18:02:01 * roshi sets the fuse
18:02:03 <roshi> 3...
18:02:06 <roshi> thanks for coming folks
18:02:20 <dustymabe> roshi: kdas__ thanks for running
18:02:24 <roshi> 2...
18:02:33 <roshi> np np - kushal did most of the work :p
18:02:35 <roshi> 1...
18:02:39 <kdas__> dustymabe, heh, I need disconnect again :)
18:02:40 <roshi> #endmeeting