17:05:08 #startmeeting atomic-wg 17:05:08 Meeting started Wed Feb 22 17:05:08 2017 UTC. The chair is dustymabe. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:05:08 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:05:08 The meeting name has been set to 'atomic-wg' 17:05:12 #topic roll call 17:05:17 .hellomynameis dustymabe 17:05:17 dustymabe: dustymabe 'Dusty Mabe' 17:05:20 hi everyone 17:05:24 marc84: hi 17:05:50 .hello sayanchowdhury 17:05:51 sayan: sayanchowdhury 'Sayan Chowdhury' 17:06:13 hi jberkus 17:06:18 .hello bowlofeggs 17:06:19 bowlofeggs: bowlofeggs 'Randy Barlow' 17:06:22 .fas jasonbrooks 17:06:22 jbrooks: jasonbrooks 'Jason Brooks' 17:06:28 dustymabe: I can volunteer for the meeting if you want 17:06:30 .hello jberkus 17:06:31 yay we are graced with presences of eggs 17:06:31 jberkus: jberkus 'Josh Berkus' 17:06:38 so many eggs 17:06:43 sayan, thanks 17:06:56 sayan: go for it 17:06:57 sayan: i'll let you pick up when if I "disappear" or have to leave 17:07:09 dustymabe: sure 17:07:18 #chair bowlofeggs sayan jbrooks walters bowlofeggs marc84 17:07:18 Current chairs: bowlofeggs dustymabe jbrooks marc84 sayan walters 17:07:25 who am I missing? 17:08:26 ok I'll get started 17:08:33 #topic action items from last meeting 17:08:57 * maxamillion to close pagure issue 202 and complete the sync 17:08:58 * maxamillion to comment and close this ticket as it had already been 17:09:00 agreed on 17:09:02 * dustymabe to create a ticket with this discussion in it 17:09:04 * jberkus to open issues around each proposed dockerfile requirements 17:09:06 change 17:09:18 feh, didn't get that done, keep it on for next meeting 17:09:31 #action jberkus to open issues around each proposed dockerfile requirements change 17:09:38 I opened a ticket for the discussion 17:09:41 https://pagure.io/atomic-wg/issue/228 17:10:13 #202 is closed 17:10:20 so maxy did that 17:10:26 so we look to be good 17:10:33 ok items for today 17:10:54 #topic decide on version scheme and image naming scheme for f26 17:10:58 #link https://pagure.io/atomic-wg/issue/229 17:11:11 would be nice if we could get some more comments in the ticket on this one 17:11:21 and/or discussion here 17:11:30 i'll give you all a moment to read 17:12:29 commented 17:12:57 ok, will give it another minute, if we are good with comments in the ticket and no discussion here then we'll move on 17:14:10 ok cool, will carry on discussion in the ticket 17:14:12 thanks 17:14:38 #topic clarify policy on atomic host support for older Fedora "number" releases 17:14:43 #link https://pagure.io/atomic-wg/issue/228 17:14:51 sigh 17:14:55 wait a sec 17:15:06 #topic clarify policy on atomic host support for older Fedora "number" releases 17:15:18 #link https://pagure.io/atomic-wg/issue/228 17:15:38 i've got a couple of +1 a couple of -1 17:15:44 err. one -1 17:16:06 should we talk more about this here or leave discussion in the ticket? 17:16:20 if we talk here I'll try to summarize what we decide and put it in the ticket too 17:17:18 either way I think we need more people to weigh in on the ticket and give their opinion/vote 17:17:42 I'll comment on the ticket after the meeting 17:17:48 sayan: thanks 17:18:10 anyone else? have anything on this? orrr NEXT? 17:19:24 i feel like I'm talking to myself a lot today 17:19:31 move to next ticket 17:19:47 .fas rtnpro 17:19:48 rtnpro: rtnpro 'Ratnadeep Debnath' 17:19:49 .fas rtnpro 17:19:51 rtnpro: rtnpro 'Ratnadeep Debnath' 17:19:57 #topic move ostree ref to be "container-host" 17:20:08 #link https://pagure.io/atomic-wg/issue/198 17:20:22 so really this is just an FYI and a last opportunity for anyone to object 17:20:41 we are moving to something like fedora/rawhide/x86_64/atomic-host for f26 17:20:52 from fedora-atomic/rawhide/x86_64/docker-host 17:21:20 I'll be merging that change on thursday night (don't want to do it right before I become unavailable) 17:21:50 one minute for comments and then I'll move on to the next issue 17:22:24 added comment 17:23:20 #topic Future of Fedora Dockerfiles 17:23:23 https://pagure.io/atomic-wg/issue/180 17:23:31 i'm pretty sure this one can be closed 17:23:48 #action dusty to close Future of Fedora Dockerfiles #180 17:24:11 #topic Ship fedora-motd in F24 atomic image 17:24:24 #link https://pagure.io/atomic-wg/issue/160 17:24:27 yes, let me update 17:25:06 I have fixed the 2 blocker issues: 1) don't cache update info from pam module 2) move from cron to system.d timer 17:25:29 I'd reckon folk in this group to review the code 17:25:56 once I get up votes for it, I can re package it and push it to the repos 17:26:11 walters: any comments about fedora-motd and potential issues with adding it ? 17:26:21 haven't had a chance to look 17:26:24 i.e. any other stuff that needs to be done (other than proper integration and testing) 17:26:36 yeah I agree. I really need to look at it myself 17:26:40 big picture i want to go to automatic updates 17:26:50 which somewhat (but not entirely) obviates this 17:26:53 walters, what do you mean? 17:26:58 walters: agree, but if a user opts out this would be nice 17:27:37 rtnpro: he means the system automatically updates when there is an update available 17:27:43 dustymabe, aha! 17:27:52 now that we only release ostree content every two weeks 17:27:56 rather than every day 17:27:56 * roshi is here 17:27:59 this makes much more sense 17:28:02 sorry, was distracted :) 17:28:02 roshi: :) 17:28:08 #chair roshi 17:28:08 Current chairs: bowlofeggs dustymabe jbrooks marc84 roshi sayan walters 17:28:13 #chair rtnpro 17:28:14 Current chairs: bowlofeggs dustymabe jbrooks marc84 roshi rtnpro sayan walters 17:28:15 there's levels of sophistication we can apply, for example download but not apply by default, but *do* automatically apply for critical CVE if not applied after a week 17:28:23 dustymabe, walters, in the future, we can let motdgen enable/disable plugins on the fly' 17:28:31 how's that going to work given that a reboot is required? 17:28:50 jberkus: I think he is saying that it would implement a reboot as well 17:29:03 walters, that will be great, but it's out of scope of motdgen 17:29:10 users would chhose to be ok with this, or opt out 17:29:12 dustymabe: that sounds like a solution for a tiny minority of users 17:29:12 jberkus, +1 17:29:21 and we could never have it on by default 17:29:29 rtnpro: agreed, outside the scope of motdgen 17:29:36 even for IoT devices, automated unscheduled reboots would not be OK 17:29:40 so we need to take a look at motdgen and get back to you 17:29:52 dustymabe, cool 17:30:02 jberkus: let's move this to fedora-cloud or we may open a ticket on it 17:30:07 ok 17:30:09 and have discussion there 17:30:12 next topic 17:30:29 #topic Spec for IRC bot to notify about blockers 17:30:44 #link https://pagure.io/atomic-wg/issue/169 17:30:52 roshi: i'm going to remove the meeting label from this one 17:31:00 sounds good 17:31:02 you said you wanted to look at it, maybe in some downtime 17:31:08 will update it 17:31:10 thanks 17:31:18 #topic design, deploy and document Fedora OpenShift Playground (FOSP) 17:31:19 yeah, I thought I took the meeting label off of it :-/ 17:31:26 #link https://pagure.io/atomic-wg/issue/153 17:31:30 wait, I thought we closed that one? 17:31:43 I'm going to close the IRC one 17:31:52 jberkus: roshi wanted to "look at it" 17:32:07 so we left it open 17:32:09 IRC one is needed? 17:32:09 ok, but let's take the meeting tag off it 17:32:17 yes, we just agreed to do that 17:32:22 we can remove that or close that ticket? 17:32:24 sayan: that would be a "nice to have but extremely low priority" 17:32:44 dustymabe: I have no idea what's going on with the HW for the FOSP 17:32:49 I'll bug misc 17:32:53 we are still waiting for it 17:32:53 ok to current topic 17:33:05 misc: how long does it take to order hardware? 17:33:21 jberkus: too much... 17:33:22 no wonder this "cloud" thing has taken off 17:33:43 if we haven't spent the money, what about just having a slush fund for AWS/GCE? 17:33:48 seems like the better route 17:34:10 jberkus: this is probably the wrong place to have that discussion 17:34:13 jberkus: quaid can give you are more detailed reports on the status regarding the order 17:34:23 do you mind taking it up with misc individually? 17:35:21 ok moving to open floor 17:35:25 #topic open floor 17:35:31 anyone have any items? 17:35:42 actually I do - 17:35:49 we have a few VFADs we want to have 17:35:53 yes, I have one 17:36:03 1. discussion on fedora atomic and "rolling release" and what all that could look like 17:36:29 2. kushal is going to give us a deep dive on gotun and how it works and we'll write a new test for it 17:36:44 i need to create a page with future VFAD topics 17:36:55 #action dusty to create atomic-wg page for future VFAD topics 17:36:56 why are we getting a deep dive on gotun? 17:37:14 roshi: would be nice for more of us to understand it 17:37:27 tunir/autocloud/gotun will all be going away when we move to taskotron, would make more sense to have the cycles spent there 17:37:43 since we're actively trying to move to taskotron 17:37:49 roshi: if that's the case then why the hell did kushal just do a rewrite of tunir in Go? 17:38:07 I have no idea 17:38:09 roshi: when are we planning the move? 17:38:12 roshi: what's the status of taskotron, anyway? 17:38:31 I've got a task to run the upstream atomic-host-tests in taskotron already 17:38:42 working with tflink and company to get it into production 17:38:56 roshi: does taskotron handle testing of AMIs? 17:39:21 I've ported the tunirtests (except non-gating) to ansible, to run in the same manner 17:39:23 #action jberkus to follow-up with misc and quaid about location of HW 17:39:41 it can, we just have to give it the bits to talk to aws 17:39:42 sigh. in the very least we need more communication around this topic 17:39:54 so we can VFAD on what the hell we should be doing with regard to tests 17:40:00 fwiw, I've been saying this for ages now 17:40:04 years even 17:40:07 moving to next open floor topic 17:40:11 sayan: you're up 17:40:35 dustymabe: should I be talking about the fedimg issue here? 17:40:48 sayan: sure 17:41:18 I saw the BZ issue after returning from PyCon 17:41:26 no worries 17:41:55 libcloud does not support making the snapshot public :( 17:42:12 I have submitted a PR today: https://github.com/apache/libcloud/pull/990 17:42:33 dustymabe, roshi do we have an action item for the group around taskotron? 17:42:40 sayan: that's nice - 17:42:50 I don't know how much time will it take to merge 17:42:56 jberkus: what do you mean? 17:43:04 sayan: do we keep running into limitations of "libcloud" when it comes to the full API coverage that we need? 17:43:17 seems like we keep wanting to do things, but "libcloud doesn't support it" 17:43:23 dustymabe: yes, quite often 17:43:24 "we need more communicaiton around this topic" (taskotron/testing) 17:43:31 what does "more communication" translate into 17:43:40 I was thinking of re-writing using boto 17:43:44 jberkus: VFAD would be one 17:43:51 jberkus: VFAD would work 17:43:54 sayan: I think that would be smart 17:43:55 ok 17:44:16 maybe open a ticket for that 17:44:22 shit they just called me to the desk 17:44:22 dustymabe: in that case for the time being let me fix the issue using boto for now 17:44:25 i have to go 17:44:33 dustymabe: have a safe flight 17:44:35 can someone please send the meeting minutes out 17:44:37 thanks 17:44:40 bye all 17:44:41 dustymabe: I will do 17:44:45 later dustymabe 17:46:39 anything else for the open floor? 17:46:57 where's the list of potential vfads? 17:47:18 we should have one on the wiki somewhere ... 17:47:21 * roshi has nothing for open floor 17:47:34 I don't know that we have a good list of them 17:47:46 dustymabe was saying we needed to create a vfad topic list 17:47:51 jberkus: want to set up a spot on the wiki 17:47:55 ? 17:48:02 ok 17:48:12 new page, or just add it to /Cloud ? 17:48:19 eh, new page 17:48:21 will do 17:48:23 new page +1 17:48:34 #action jberkus to create vfad idea list page 17:49:09 #action sayan to create ticket on moving fedimg from libcloud to boto 17:50:06 can I go ahead and close the meeting? 17:50:27 yes 17:50:36 #endmeeting