14:00:01 #startmeeting sahara 14:00:05 Meeting started Thu Apr 5 14:00:01 2018 UTC and is due to finish in 60 minutes. The chair is tellesnobrega. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:06 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:09 The meeting name has been set to 'sahara' 14:00:30 show off hands 14:00:33 o/ 14:00:43 o/ 14:01:12 tosky... 14:02:07 o/ 14:02:53 hi zchkun 14:03:21 I guess we can start 14:03:22 hi jeremyfreudberg 14:03:32 #topic News/Updates 14:04:25 I've been working on plugins split, it looks better now, had some interesting conversation and improved the implementation idea 14:04:50 working with mediator pattern to make the sahara and plugins as uncoupled as possible 14:05:03 sooo... I'm testing and restesting image generation (so found an issue in a review) 14:05:22 also, I started working on boot from volume but not much was done, just studying 14:05:28 the vanilla job finally started, but it failed with some timeouts, so probably until we have the CI or until -infra enables nested virtualization we are out of luck 14:06:00 nothing new to report from me, really (busy school week) 14:06:39 we have few reviews almost ready to be merged 14:07:10 jeremyfreudberg, school :( hopefully it is almost over 14:07:58 I'm also working on fixing stuff that tosky finds failing 14:08:28 next week should be an easier week, and the semester ends in a month 14:08:47 cool 14:09:20 any more news? 14:10:34 #topic Sahara CI 14:10:55 jeremyfreudberg, I saw that evgeny replied your email, what is the plan from now on? 14:12:17 was about to click send on the reply button 14:12:27 nice 14:13:01 for those not in the email thread, the question was about whether we need the separate vm for Jenkins/Nodepool/Zuul or if those services can share a node with devstack 14:13:38 i suppose it's easier to have things separate, but just want to see what my options are 14:13:48 in terms of plan, i will probably give evgeny access and see what he can do 14:14:12 I think it will be easier having it separated too, but lets see how it goes 14:15:34 ok, lets move on 14:15:57 #topic Rocky 1 14:16:42 just wanted to give a heads up that rocky 1 is around the corner and we should take care of the important features, at least specs asap 14:17:11 two weeks from today, more or less 14:17:25 yes 14:17:29 i'll see what time i have to write some specs 14:17:39 thanks jeremyfreudberg 14:17:56 also, we should start working on plugins upgrades 14:18:15 specially the complex ones (CDH and vanilla) 14:18:29 agreed 14:18:36 it seems that we will see the light with ambari with some new external (and hopefully stable :) contribution 14:18:59 yes 14:19:07 that is why I left ambari out of the list there 14:19:13 we already have people working on it 14:21:18 ok, nothing else on rocky 1 14:21:49 do any of you have specific topic to talk about? 14:22:34 not really whole topics, just two points to mention during open discussion 14:22:37 if not lets go to open discussion 14:22:43 tosky, any topics? 14:22:44 #topic Open Discussion 14:22:49 guess not 14:23:02 anyone planning to run for TC? 14:23:18 definitely not 14:23:56 i'd like to run some day, but not this time around 14:24:03 hmm, not right now, maybe some point in the future 14:24:08 right now I won't have the time 14:24:35 don't want to brace all in the same time and fail it 14:24:49 right 14:24:51 alright, other thing to mention is this puppet-sahara review https://review.openstack.org/#/c/558442/ 14:25:33 i don't think we've properly deprecated this on our end yet, so we should be mindful 14:25:49 iirc we don't use it 14:26:06 we do and we don't, i think 14:26:09 at least not in the code which imports keystone_authtoken when the new trustee section is defined 14:26:33 there are no many references to auth_uri in sahara code 14:26:42 few in configuration (and it should be changed) 14:27:02 one in the tools/get_auth_token.py, which is an tool that should be rewritten anyway for keystone v3, or removed 14:27:40 oh right, i was confused (we actually don't even use auth_uri when trustee is missing, we use keystone_authtoken/auth_urL) 14:27:46 so, it's no problem then 14:28:07 good 14:28:20 tosky, can you list the reviews that are "ready" to merge here 14:28:27 so everyone can take a look 14:29:36 that would be nice 14:30:15 so, https://review.openstack.org/#/c/557682/ (HDP 2.6, probably ready) 14:30:36 https://review.openstack.org/#/c/556844/ change in zuul image building, technical refactoring 14:30:47 https://review.openstack.org/#/c/547669/ experimental spark multinode working 14:30:53 it looks ready 14:30:54 https://review.openstack.org/#/c/555204/ document building 14:31:09 https://review.openstack.org/#/c/538728/ and this may be ready (just a small comment on my side) 14:31:52 and that's it iirc 14:32:13 I would like to bring this up https://review.openstack.org/#/c/508099/ 14:32:18 its been there for a while 14:32:33 if we all could take some time to review and get it going again would be awesome 14:32:43 as well as https://review.openstack.org/#/c/539856/ 14:32:46 we need to test it out 14:32:55 indeed 14:33:50 anything else? 14:33:55 not from me 14:34:06 tosky? 14:36:26 I guess not 14:36:43 I have a problem that it reports timeout error when I create vanilla 2.7.1 clusters 14:37:00 zchkun, do you have logs for that? 14:37:02 I don't know if you've met 14:37:35 yes, can you give any other details (for example, which provisioning step) 14:37:38 the log displays only timeout 14:38:01 nothing from me 14:38:07 what version? master? 14:39:17 queens 14:39:28 sorry guys, have to sign off. zchkun: if you have trouble finding details in log file, there is the dashboard too 14:40:24 zchkun, it is very hard to identify the problem, but if you can pin point the step where it fails, and paste logs it can help 14:40:48 I will try to deploy sahara queens locally and see if I can deploy vanilla cluster 14:40:56 and let you know if I have problems 14:41:40 ok, thank you 14:42:10 sorry , i will tidy my problem up 14:42:30 zchkun, no worries, we just want to be able to help better :) 14:42:49 ok :) 14:42:51 let me know if you have more details 14:43:20 I guess we can close now 14:44:55 thanks all 14:44:58 #endmeeting