16:30:23 #startmeeting fedora_coreos_meeting 16:30:23 Meeting started Wed Jan 29 16:30:23 2020 UTC. 16:30:23 This meeting is logged and archived in a public location. 16:30:23 The chair is dustymabe. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:30:23 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:30:23 The meeting name has been set to 'fedora_coreos_meeting' 16:30:35 #topic roll call 16:30:36 .hello2 16:30:37 bgilbert: bgilbert 'Benjamin Gilbert' 16:30:39 .hello2 16:30:39 .hello lucab 16:30:41 dustymabe: dustymabe 'Dusty Mabe' 16:30:42 .hello miabbott 16:30:44 kaeso[m]: lucab 'Luca Bruno' 16:30:46 miabbott: miabbott 'Micah Abbott' 16:30:55 .hello sinnykumari 16:30:56 ksinny: sinnykumari 'Sinny Kumari' 16:30:57 .hello2 16:30:58 o/ 16:31:00 jlebon: jlebon 'None' 16:31:17 .hello mnguyen 16:31:20 mnguyen_: mnguyen 'Michael Nguyen' 16:31:56 #chair bgilbert kaeso[m] miabbott ksinny jlebon mnguyen_ 16:31:56 Current chairs: bgilbert dustymabe jlebon kaeso[m] ksinny miabbott mnguyen_ 16:32:14 nice to be back for a meeting this week! sorry I missed last week 16:33:36 #topic Action items from last meeting 16:33:50 * bgilbert to work with x3mboy on FCOS infographic 16:33:50 * slowrie to help us get the azure image upload/boot tested 16:34:13 I will pick up working with marketing on an infographic 16:34:39 to let bgilbert focus on some of the tooling we want to get some polish on 16:34:46 #action dustymabe to work with x3mboy on FCOS infographic 16:34:51 dustymabe: +1, thanks! 16:34:55 np 16:35:16 looks like slowrie isn't around today 16:35:20 slowrie did some initial testing in Azure; ran into DNS problems 16:35:31 https://github.com/coreos/fedora-coreos-tracker/issues/356 16:35:38 miabbott: so are those issues with our image or issues with his account or issues with azure? 16:35:52 ahh thanks for the link 16:36:04 #info slowrie did some initial testing in Azure; ran into DNS problems https://github.com/coreos/fedora-coreos-tracker/issues/356 16:36:36 .hello red_beard 16:36:37 red_beard: Sorry, but you don't exist 16:36:40 .hello redbeard 16:36:41 red_beard: redbeard 'Brian 'redbeard' Harrington' 16:36:59 ok that seems it for action items 16:37:28 there are no meeting tickets I see so I'll run through a few topics and then open it up for open floor - if you want a main topic send me a DM and I'll bring it up 16:38:09 #topic devconf.cz fedora coreos lab 16:38:26 #link https://dustymabe.com/2020/01/23/devconf.cz-2020-fedora-coreos-lab/ 16:38:40 we had a great turnout at the hands on lab we did at devconf.cz this past weekend 16:38:54 there was not an empty seat in the room and some people were standing 16:39:35 So this is great news - as it at least shows from a small sample size that we have interest in FCOS 16:39:49 dustymabe++ 16:39:50 \o/ 16:39:53 that's awesome 16:39:57 nice! 16:40:12 If you are interested in running the lab yourself the link above can be used to run it at home so give it a spin if you want to learn more about FCOS 16:40:35 dustymabe++ 16:40:35 abai: Karma for dustymabe changed to 5 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 16:40:56 thanks darkmuggle ashcrow walters miabbott for coming and helping out with the lab 16:40:57 Also, I like the way lab content has been prepared, prerhaps people can follow that later on who didn't get chance to attend in person 16:41:02 dustymabe++ 16:41:26 ok that was a quick FYI 16:41:47 #topic FCOS "out of preview" retrospective 16:42:14 I'm throwing this topic in here to see if there's anything we learned from the out of preview release 16:42:29 obviously there is a lot we know we have to improve on (automation, etc) 16:42:40 but anything in particular that is new that we learned that we need to do? 16:43:22 I'm doing a double stable+testing release, and it turns out the process doesn't parallelize a lot 16:43:58 kaeso[m]: which parts? the jenkins pipeline runs themselves? 16:44:11 (I think this the first double we do) 16:44:19 kaeso[m]: yes, now that we're doing double releases each time, i think we need to prioritize https://github.com/coreos/fedora-coreos-pipeline/issues/68 16:44:53 dustymabe: yes, so far the jenkins builds 16:44:53 yep, I think that should help a lot 16:45:17 i can tackle it soon 16:45:42 jlebon: I added a label to sync over to internal JIRA tracking 16:45:45 thanks! 16:45:57 +1 16:46:11 so with an uptick in user interest, did we learn anything about our users and how they are trying to use FCOS that maybe we didn't expect? 16:47:33 we definitely get the users asking questions about how to install packages, but we've been doing a good job of trying to push them towards containers I think 16:47:46 Personally I wasn't expecting this one https://github.com/coreos/fedora-coreos-tracker/issues/352#issuecomment-577733058 16:48:27 kaeso[m]: yeah I half expected that 16:48:59 that's definitely an interesting one for us to try to solve a bit down the road I think 16:49:14 kaeso[m]: heh, RHEL AH cloud kickstarts actually do this too -- pull down system containers and install them 16:49:50 the whole offline install and customize pet node, I mean 16:51:17 ahh gotcha 16:51:32 anything else we want to say about the preview release before moving on to another topic 16:52:22 #topic open floor 16:52:32 anyone with anything for open floor? 16:53:18 #info we are in the process of doing a new release for the testing and stable streams (thanks kaeso[m]) 16:53:39 https://github.com/coreos/fedora-coreos-tracker/issues/355 16:53:58 this one showed up yesterday 16:54:07 #info we now have some basic upgrade testing: https://github.com/coreos/fedora-coreos-pipeline/pull/190 16:54:57 kaeso[m]: interesting 16:55:00 #info there is interest from Exoscale (EU cloud provider) to make FCOS available in their cloud 16:55:17 I didn't know Vultr, but as a platform looks quite sane. Does Fedora have accounts there? 16:55:53 kaeso[m]: I don't think so. vultr themselves reached out to us a few years ago but we didn't have the resources to look at them 16:56:06 i think they are probably willing to give us accounts if we have time to work on it 16:56:25 starting with just providing the platform-id stamped images should be straightforward 16:56:46 that would be nice, they seem to even support arbitrary PXE 16:57:38 miabbott: do you have a contact? 16:57:52 jlebon: yep, they're going to open an issue in our tracker too 16:57:57 jlebon: sanja just provided an introduction via email today 16:58:08 +1 16:58:08 and what dustymabe said :) 16:59:56 anything else for open floor? 17:00:36 Now that we are out of preview and also got this conference out of the way I'm looking forward to steady progress through bi-weekly releases 17:00:43 and settling into some cadence there 17:00:46 :) 17:01:04 dustymabe: based on questions i've seen raised (even internally) i think a push will need to be done walking folks through use cases 17:01:35 red_beard: i.e. educating people on the philosophy behind FCOS and also when not to use it? 17:01:40 last night the most recent question was: "how to monitor system performance on RHCOS? In RHEL there are tools like sar, vmstat, iostat, etc. Similarly, do we have any tools or ways to monitor system performance" 17:01:58 and if folks are asking about RHCOS, they need to know about FCOS. 17:03:04 red_beard: so the use case it "how to monitor system performance on FCOS?" and we'd provide some docs there? 17:03:12 docs/guidance 17:03:20 red_beard: that's an interesting one, are people running manual interactive tools to monitor clusters of RHEL nodes? 17:03:32 i think (for now) a fedora magazine article is sufficient 17:03:54 kaeso[m]: abso-lutely 17:04:30 red_beard: my understanding is that there is an higher level direction of getting those out of the node as metrics in a continuous way, and then do performance monitoring in an aggregated way 17:05:36 i'm merely pointing out that i'm directly seeing users who are confused and think that *COS is dark witchery and needs all new tools 17:06:01 kaeso[m]: you and i spent lots of time educating folks about that, now there is a whole new crew of folks who can share that message ;) 17:06:33 fair 17:06:58 anything else for open floor? 17:07:51 will close out meeting in a few minutes if nothing else comes up 17:09:23 #endmeeting