08:02:22 #startmeeting 08:02:22 Meeting started Thu May 1 14:02:21 2014 UTC. The chair is jzb. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:02:22 Useful Commands: #action #agreed #halp #info #idea #link #topic. 08:02:43 chair mattdm red_trela rbergeron number80 samkottler geppetto 08:02:47 er 08:02:54 #chair mattdm red_trela rbergeron number80 samkottler geppetto 08:02:54 Current chairs: geppetto jzb mattdm number80 rbergeron red_trela samkottler 08:03:05 so early, such meeting, wow 08:03:11 roll call? 08:03:14 .fasinfo jzb 08:04:19 *red_trela 08:05:04 well, the others sure have a short attention span 08:05:20 .fasinfo jsmith 08:05:47 *jzb wonders why that's not working 08:05:48 *mattdm is back 08:05:50 sorry 08:06:08 *danofsatx-work has quit (Remote host closed the connection) 08:06:10 Looks like zodbot is asleep on the job :-p 08:06:21 well, let's get started... 08:06:21 .hellomynameis mattdm 08:06:50 *danofsatx-work (~danofsatx@fedora/danofsatx) has joined 08:07:23 #topic Crucial basic Docket Host Image decisions 08:07:36 #info https://fedorahosted.org/cloud/ticket/47 08:07:38 this is ticket 47, filed by red_trela 08:08:05 red_trela: you want to start the discussion? 08:08:09 pasting from ticket: 08:08:13 Whether to use (rpm-)ostree (aka Fedora Atomic Initiative) 08:08:16 Whether to remove yum/dnf (if going with ostree) 08:08:18 Whether to replace cloud-init by coreos' cloud-init or min-metadata-service 08:08:20 Whether to kick out python (if we decided against yum/dnf and cloud-init) 08:08:32 *zodbot_ has changed the topic to: Crucial basic Docket Host Image decisions 08:08:56 hi 08:08:58 oh hai zodbot_ 08:09:08 sure 08:09:42 i thought we were talking about dockah 08:09:45 so there's those 4 rather crucial decisions that we should make regarding the Docker Host Image (which is not the Atomic Image or whatever). 08:09:52 *mattdm nods 08:10:10 *rbergeron migrates to awakeishness 08:10:10 red_trela: do we want a non-Atomic docker host image? 08:10:43 jzb I think that we want to make sure that the generic image can be used that way if desired 08:10:44 uhm, do we? I thought we made the docker host image the atomic image last week - or did we just add an image? 08:11:08 i'm confused 08:11:12 *danofsatx-work has quit (Remote host closed the connection) 08:11:14 and as red_trela says, _the" docker host image should be Fedora Atomic. 08:11:16 red_trela: yeh, we all voted for 46 I thought 08:11:36 so I think the first point is confirmed. the rest are followups from there really 08:11:44 mattdm: exactly 08:11:49 *danofsatx-work (~danofsatx@fedora/danofsatx) has joined 08:11:52 ok 08:12:10 I think #2 and #4 pretty much follow on … there's little point in keeping yum/dnf if the host is managed by ostree 08:12:14 unless someone thinks we should vote on whether or not doing both (atomic and non-atomic) docker host images - current status is atomic only 08:12:35 +1 about removing python and its dependencies 08:12:56 #3 I'm not as sure about … someone want to bring up the pros. cons. of cloud-init with ostree? 08:13:07 but i don't think it's feasible before f22 08:13:09 geppetto: eventually rpm-ostree is supposed to allow installing packages 08:13:13 *m00g (~m00g@190.79.108.233) has joined 08:13:16 let's do these in order? 08:13:27 not sure if that will be in the F21 timeframe or not, but we should check in with Colin about that. 08:13:36 #1 I think we're all agreed on. So moving on to #2... 08:13:40 jzb: yeh, but it's doing so via. ostree which links to libhawkey 08:13:53 jzb: 3 package managers in F22 ftw :) 08:13:56 And I agree that yum/dnf don't make sense in ostree. in fact they are likely to break stuff. 08:14:20 unless geppetto says otherwise :) 08:14:26 no, I agree 08:14:44 anyone _not_ agree? 08:15:11 the sound of crickets is assent! 08:15:36 so, do we vote on that or is no opposition good enough to go with it? 08:15:47 #agreed fedora atomic docker-specific image will not include yum/dnf 08:15:57 *drieden (~drieden@host-69-95-41-10.spr.choiceone.net) has joined 08:16:05 red_trela I'm good with no opposition -- that's basically our governance plan 08:16:12 *danofsatx-work has quit (Remote host closed the connection) 08:16:48 so cloud init is more controversial. 08:16:49 *danofsatx-work (~danofsatx@fedora/danofsatx) has joined 08:17:03 so the next point would once again be cloud-init. I figure we already had this topic a short while ago and I think someone wanted to bring it to the list? 08:17:09 (also once again) 08:17:17 I know samkottler was just very recently saying he was interested in trying to make it work 08:17:38 make min-metadata-service work, that is, right? 08:17:38 From the previous discussion, sdake isn't concerned about heat requirements for the docker-specific image 08:17:53 red_trela yes 08:18:20 proposal: try it for alpha, fall back to cloud-init for beta if it doesn't work out 08:18:33 +1 08:18:37 and keep cloud-init in the generic image 08:19:01 +1 from me 08:19:06 the downside is... two things. counterbalanced by one really eing a subset of the other 08:19:36 i think i alreadt requested an account on fedora cloud for developping min-metadata-service but having it in f21 is very optimistic 08:19:57 I'm still not very convinced min-metadata-service is the best alternative, but I'm +1 to try any alternative ;) 08:20:16 -1 08:20:19 number80 what are the concerns? 08:20:42 making it work with aws, openstavk, requires a lot of testing 08:21:13 *danofsatx-work has quit (Remote host closed the connection) 08:21:13 another argument for it in my mind is that many of the things one could do with cloud-init are things one _shouldn't_ do with atomic 08:21:14 *itamar_ (~itamar@189-015-174-009.xd-dynamic.ctbcnetsuper.com.br) has joined 08:21:31 (like, install and configure packages) 08:21:31 *roshi walks in late 08:21:50 *danofsatx-work (~danofsatx@fedora/danofsatx) has joined 08:22:15 number80 we've got a little while before we _really_ have to decide for alpha ... do you want to defer the decision to closer to then to see how it is looking? 08:22:42 sure, we have many things running 08:22:54 okay, so: 08:22:57 (sorry for th slow typing) 08:22:59 *danofsatx-work has quit (Read error: Connection reset by peer) 08:23:25 proposal: defer min-metadata-service til closer to alpha, encourage people interested in it to continue development 08:23:37 number80: are those concerns for just the docker host image? or is that "if we tried to make it work for the generic image" 08:24:22 the thing really is, the last point (removing python) really depends on this - and doing that last-minute might not work well 08:24:58 putting python/cloud-init back in last-minute on the other hand, should be much less troublesome 08:25:19 good point red_trela 08:25:23 mattdm: defer it .. but not for generic image, that's comfirmed to stay w/cloud-init as we kust discussed. yes? 08:25:37 (just hoping I'm not summoning trouble here) 08:25:48 rbergeron yes. 08:26:22 rbergeron: both 08:26:26 and _especially_ since we are (resigned to?) keeping cloud-init in the generic image, we'll know that switching back to that should work. 08:26:28 okay. just making sure - we just listed out a long list of options ;) 08:26:47 i'd like to see more advancements before taking a decision 08:27:19 i'll change my vote +0 ;) 08:27:54 (also, i need sdake to bring me lunch today and cannot jeopardizr that. no, no, just kidding) 08:28:26 so, let's go with mattdm's proposal? 08:28:45 red_trela the defer one or the first one? 08:28:58 the try it for alpha one 08:29:09 I'm still +1 :) 08:29:29 *satellit (~Thomas@c-174-61-128-216.hsd1.wa.comcast.net) has joined 08:29:31 *zodbot_ is now known as zodbot 08:29:36 everyone is +1 or +0, so next :) 08:29:44 *zodbot has quit (Disconnected by services) 08:29:56 #agreed try min-metadata-service for fedora atomic image for alpha, fall back to cloud-init for beta if it doesn't work out 08:30:09 anyone opposed to kicking out python, if we really go without cloud-init? 08:30:13 I'll be +1 to then, so I can be everyone :) 08:30:27 #info cloud-init to stay for general-purpose cloud base image 08:30:28 +1 for kicking snakes 08:30:36 die python die 08:30:52 I love python, but let's get the scripting langages out of the minimal image 08:31:10 (die, javascript policykit requirement, die) 08:31:11 *pwhalen has quit (Ping timeout: 240 seconds) 08:31:25 *Renault (~Renault@AToulon-651-1-181-77.w83-113.abo.wanadoo.fr) has joined 08:31:30 js no language, js plague 08:31:30 currently, python is required by yum/dnf, cloud-init and nfs-utils only I think. so we already voted to (try) kick out most of those and the part of nfs-utils that need python can be put in a subpackage 08:32:04 red_trela yeah. 08:32:06 ...I opened a ticket in that regard weeks ago and haven't heard back, but it should be a easy fix 08:33:28 okay, so, any disagreement with this general route, assuming the prereqs line up? 08:34:32 none here 08:34:36 #agreed aim for getting python out of the fedora atomic image too, if all of the previous things line up. 08:35:04 *zodbot_ (supybot@fedora/bot/zodbot) has joined 08:35:18 great, that's all for this ticket :) 08:37:02 *jeid64 (~jeid64@youngmoney.rc.rit.edu) has joined 08:37:17 *aeperezt (~aeperezt@fedora/aeperezt) has joined 08:37:22 moar tickets? 08:37:47 #topic Spins Process for Cloud Images 08:37:53 #info https://fedorahosted.org/cloud/ticket/49 08:37:54 *egyDev has quit (Remote host closed the connection) 08:37:59 *jzb wonders what happened to the bot 08:38:17 *egyDev (~araby@fedora/ahmed-araby) has joined 08:38:30 bots have been slow lately, I've noticed 08:39:13 nirik mentioned that he was restarting something. but apparently the wiki is also broken 08:39:53 so anyway, I'm totally okay with the cloud image variants being spins, as long as the policies can be made flexible 08:39:53 ok, anyway 08:39:58 this ticket is just to make sure everyone is okay if we try to follow the spins process for the cloud images as closely as possible/reasonable 08:40:06 the current policies are very focused on livecds 08:40:22 it basically makes sense to not reinvent the wheel 08:40:46 mattdm: so how many "spins" are we going to need to create? 08:41:09 seems fine to me (knowing nothing about the processes :) 08:41:15 jzb as of now: Fedora Cloud Base, Fedora Atomic, Fedora Big Data Cloud Image 08:41:45 with the possibility of the big data one breaking into two for some different cases 08:42:00 K 08:42:30 so we're good on this ticket? 08:42:39 I am 08:42:48 we should probably come up with tickets for each spin we need, yes? 08:43:58 ok 08:44:15 *suresht has quit (Quit: Leaving) 08:44:16 jzb: yes 08:44:20 hokay 08:44:27 some administrative stuff to do for each image 08:44:28 #action jzb to create tickets for each Cloud Spin required. 08:44:35 admin stuff I can do :-) 08:44:39 *pwhalen (~pwhalen@unaffiliated/pwhalen) has joined 08:44:40 Sounds good. 08:44:41 *somvannda (~somvannda@120.136.30.97) has joined 08:44:51 jzb see also https://fedorahosted.org/cloud/ticket/50 08:44:59 #topic Deliverables and release engineering changes for Fedora.Next 08:45:02 right. :) 08:45:08 mattdm: indeed ;-) 08:45:11 this is really the same thing we just talked about, in retrospect :) 08:45:21 mattdm: you want to update the ticket? 08:45:49 *hanthana (~danishka@2404:e800:ea26:88:2216:d8ff:fe06:d27d) has joined 08:46:00 yeah. what I'll do is wait for you to make the new tickets just mentioned, and then close this one, and link to your new tickets in the rel-eng ticket 08:46:04 ticket ticket ticket ticket. 08:46:15 that's the ticket 08:46:30 *fabian_a (~fab@147.87.243.63) has joined 08:46:30 #action mattdm to update rel-eng trac with cloud spin links 08:46:37 #topic start communication/collaboration on cloud image updates 08:46:42 this is ticket 51 08:46:50 #info https://fedorahosted.org/cloud/ticket/51 08:47:02 *mattdm looks away 08:48:04 well, don't *everybody* look away! 08:48:26 *samkottler is here, sorry I'm late 08:48:31 i can look, not touch ;) 08:48:43 mattdm: I can put together a policy proposal on off-cycle updates 08:48:47 hi samkottler! how's the electricity? 08:48:55 mattdm: I assume "off-cycle" means "between releases"? 08:49:06 jzb that would be awesome. and, yes, that's what I meant. 08:49:11 it may not be the best term 08:49:14 I touched on this in the QA docs draft 08:49:23 *RGeri77 has quit (Quit: Még visszatérek........) 08:49:46 roshi thanks! there's obviously a lot of qa-related stuff here 08:49:50 #action come up with policy for updates 08:50:12 https://fedoraproject.org/wiki/User:Roshi/QA/Cloud_Docs under Test Overview 08:50:28 mattdm: good enough ;-) 08:50:51 *pwhalen has quit (Ping timeout: 252 seconds) 08:51:27 who's going to take on "QA contribution (commitments to help with the automation and possible manual testing)" 08:51:33 and "release engineering help" 08:51:34 ? 08:52:55 I can commit to help some, but I can't commit to being the lead on it -- need to have at least one other person really excited about this 08:53:12 I'll help with the testing for sure as I continue to get spun up on all the cloudy things 08:53:41 mattdm: well, I'm excited, but probably not the best person to be doing those things 08:53:47 I'd love to, but my plate is already very full so I really can't take on any more 08:53:49 roshi -- do you want to put yourself down as the lead for this? alternately or additionally, I can take an action item to search out more interest 08:54:29 *adimania has quit (Ping timeout: 246 seconds) 08:55:06 that could work 08:55:19 *roshi wonders what he just got himself into 08:55:27 :) 08:55:29 roshi it'll be fine. mwahhahahahaha. :) 08:55:53 proposal: put even more on roshi's plate, he's an easy victim :) 08:56:02 hopefully once we get the framework right it'll be easy for new volunteers to plug into and helpp 08:56:13 the plate is a lie 08:56:49 if it's a lie, we know for sure it can't be full 08:57:10 so you don't already have a "very full" plate then? :p 08:57:55 but yeah, that should work on my end as I want to learn more cloud-related things 08:58:12 *satellit has quit (Read error: Connection timed out) 08:58:15 thought, at this point I don't know what I can do to help with the "release engineering help." 08:58:22 s/thought/though 08:58:49 anyway, are there any seriously pressing matters? Otherwise I'd ask to adjourn the meeting as the hour is over in a moment and I need my sleep ;) 08:59:24 I think we're good unless anyone has new business 08:59:26 the only question I have is regarding feedback on the QA docs - is what I wrote out what we had in mind? 08:59:49 I sent the email out last week per the action item, but didn't really hear anything back 09:00:21 roshi: terribly sorry, I couldn't find the time to look at them yet but I'm having a long weekend so I'll try really hard to have a good look :) 09:00:27 roshi It's... um, on my plate to look at it. 09:00:41 yeah, what red_trela said :) 09:00:49 that's fine - just checking 09:00:51 roshi: I'll also review this weekend 09:01:03 roshi: for the first time in a few weeks, I actually will be home for a full week. 09:01:05 I assumed no-news was good news, but wanted to confirm it :) 09:01:23 OK 09:01:26 on another note, google people are still asking about our interest in gce 09:01:38 mattdm: what do we need to provide there? 09:01:59 *fol_tempus has quit (Remote host closed the connection) 09:01:59 jzb not exactly sure. I *think* we have cloud-init working? 09:02:09 that's one of the RFEs for min-metadata-service too 09:02:25 and given that, I think we just need someone to work with them and with release engineering to get it into place 09:02:38 at _least_ as an F21 deliverable 09:02:45 what do you need to know as a baseline to help out with that? 09:02:51 it's just hard to do without a gce test account 09:02:52 and maybe something semi-official (as with docker) before that 09:03:08 gce would be a good target to have images/whatever for - since devs will almost certainly be playing with it 09:03:12 red_trela I think we can hook those up. 09:03:27 *fol_tempus (~tempus@gateway/tor-sasl/foltempus) has joined 09:03:40 mattdm: maybe if we dangle the test account on list we can get someone to step up? 09:03:46 s/the/a/ 09:03:54 jzb maybe, yeah. samkottler you still around? 09:04:04 *pwhalen (~pwhalen@unaffiliated/pwhalen) has joined 09:04:15 yep 09:04:18 mattdm: I can understand why folks might be reluctant to volunteer if they feel they have to pay out of pocket to do the testing. 09:04:37 jzb yeah. samkottler do you know the state of the gce stuff better than i do? 09:04:39 jzb: GCE has offered us at least one testing account so if someone leads the charge they will get comped 09:04:54 so maybe we just need to say that on the list? 09:04:57 how does gce not have to go through legal as hp cloud does? 09:04:59 the main blocker to GCE is getting the tools packaged up 09:05:00 I know zooz was doing some work 09:05:10 red_trela: it has already gone through legal 09:05:15 and they cleared it 09:05:25 oh 09:05:29 the response was basically 'people can choose to sign the NDA if they please' 09:05:35 that was fast 09:05:53 red_trela: it happened months ago 09:06:18 well, missed that then, sorry 09:06:31 hp cloud is still stuck, though? 09:06:47 red_trela yes. and I pinged again recently. 09:07:26 *satellit_ (~Thomas@c-174-61-128-216.hsd1.wa.comcast.net) has joined 09:07:35 HP do at least build images already though, right? 09:07:50 just not *our* images 09:08:25 I think so 09:08:43 yes. in fact, they basically _do_ use our images with little modification 09:08:45 this is awesome. 09:08:53 but hp cloud would be the best spot to do generic latest-openstack testing :) 09:08:56 they're a very custom openstack cloud and I think they add their agent to our images 09:09:14 (they might do that, with the agent.) 09:10:25 OK, so are we good with GCE? 09:10:38 do we have an #action? 09:11:19 *satellit_e (~satellit@2601:8:8800:62a:255b:717c:55ca:d531) has joined 09:13:07 ? 09:13:08 mattdm: ^^ 09:13:26 jzb i guess..... 09:13:38 *itamar_ has quit (Ping timeout: 240 seconds) 09:13:50 #action mattdm to post about gce on cloud list, get right people connected 09:13:57 *vpavlin (~vpavlin@194.228.13.64) has joined 09:14:24 OK, I think we're ready to adjourn 09:14:27 any -1's to that? 09:14:35 not from me 09:15:21 none 09:15:31 OK 09:15:33 thanks all 09:15:36 #endmeeting