05:00:12 #startmeeting ironic 05:00:13 Meeting started Tue Apr 28 05:00:12 2015 UTC and is due to finish in 60 minutes. The chair is devananda. Information about MeetBot at http://wiki.debian.org/MeetBot. 05:00:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 05:00:17 The meeting name has been set to 'ironic' 05:00:20 #chair NobodyCam 05:00:21 Current chairs: NobodyCam devananda 05:00:32 o/ 05:00:39 o/ 05:00:47 o/ 05:00:49 good morning/afternoon/evening/night ! 05:00:57 :) 05:01:12 our agenda for today (https://wiki.openstack.org/wiki/Meetings/Ironic) is surprisingly empty ... 05:01:14 o/ 05:01:24 light agenda today! 05:01:42 I have two topics to cover, though, which could take up the whole meeting :) 05:02:04 #topic Kilo RC status 05:02:05 o/ 05:02:25 #undo 05:02:25 o/ 05:02:26 Removing item from minutes: 05:02:31 actually - I'm going to give folks more time to show up 05:02:35 we're really light right now 05:03:39 * NobodyCam runs to grab soda in the lull 05:03:59 fwiw, my two topics: kilo rc statuss & release notes; summit planning 05:04:04 o/ 05:04:37 rameshg87, wanyen: glad you two could make it :) 05:04:48 :) 05:05:04 devananda: it's 10.30 am for me :) 05:05:12 * rameshg87 cannot have a reason for not attending :D 05:05:19 heh 05:05:20 :) 05:05:25 it's 10pm here 05:05:34 oh :( 05:05:59 which makes it 1am for Shrews and TheJulia ... (you two should be asleep! don't answer this :) ) 05:06:09 ok - let's get started 05:06:15 cinerama: ohhai! 05:06:21 devananda: hi! 05:06:38 #topic Kilo release status 05:06:51 first off, we've published an RC2 build last week 05:06:56 please please go test it 05:07:42 and if you find any huge bugs, please tag them "kilo-rc-potential" so they show up here 05:07:45 https://bugs.launchpad.net/ironic/+bugs?field.tag=kilo-rc-potential 05:07:48 oops 05:07:49 #link https://bugs.launchpad.net/ironic/+bugs?field.tag=kilo-rc-potential 05:08:07 we've had two get tagged. i dont think either one is wort ha backport though 05:08:15 https://bugs.launchpad.net/ironic/+bug/1384379 05:08:15 Launchpad bug 1384379 in OpenStack Compute (nova) "versions resource uses host_url which may be incorrect" [High,In progress] - Assigned to shihanzhang (shihanzhang) 05:08:18 https://bugs.launchpad.net/ironic/+bug/1448103 05:08:19 Launchpad bug 1448103 in Ironic "disabling erase_devices clean_step in agent_ilo driver doesn't work" [Undecided,Fix committed] - Assigned to Anusha (anusha-iiitm) 05:08:52 wanyen: I dont see anusha here, but can you speak to why that bug fix was proposed to stable/kilo? 05:08:52 devananda: and i think #1 doesn't have a fix in ironic yet 05:09:06 rameshg87: yea, I haven't seen any code up for it 05:09:20 devananda: anusha here :) 05:09:40 devananda: thought it is required for backporting 05:10:17 ramineni: ohhai! 05:10:22 devananda: because of #2 bug, disk erase cannot be disabled for agent_ilo driver if cleaning is enabled 05:10:50 is there a case you want cleaning enabled but NOT to run disk erase step? 05:10:51 devananda: And as teh clean step 'erase_devcies' takes lot of time, 05:11:10 devananda: so, it is important to be disabled , if user doesnt want it 05:11:22 ramineni: which "user" ? 05:11:27 you mean operator, right? 05:11:31 devananda: yes 05:11:34 devananda: may be a single tenant environment ? idk. may be people won't care about erasing disks 05:11:40 ?? 05:11:46 ok - so there is already a flag to disable cleaning 05:11:56 why would someone want to do cleaning but NOT do disk erase? 05:12:05 devananda: yes, there is 05:12:24 devanand, admin may want other cleaning steps but want to disable disk erase only. 05:13:02 devananda: Disk erase being very time-consuming, operator may not do it very frequently 05:14:27 hm. i'm not thrilled that this is a driver-specific option, but I see the need for the fix, and it's pretty simple 05:14:45 thanks for explaining. seems OK to me 05:14:54 devananda: thanks 05:15:21 any volunteers t owork on https://bugs.launchpad.net/ironic/+bug/1384379 ? 05:15:21 Launchpad bug 1384379 in OpenStack Compute (nova) "versions resource uses host_url which may be incorrect" [High,In progress] - Assigned to shihanzhang (shihanzhang) 05:15:41 shih ??? 05:15:51 if no one's working on a fix, we shouldn't target it to kilo ... 05:16:03 I thought disabling disk erase is generic not ilo driver specific. 05:16:30 wanyen: the CONF option is defined in ilo driver 05:16:48 "if CONF.ilo.clean_priority_erase_devices is not None:" 05:16:58 wanyen: there is an option for agent driver and one for ilo driver 05:17:03 I see. But I thought IPA allow disk erase to be disabled as well. 05:17:36 wanyen: yes , through agent.agent_erase_devices_priority= 05:17:58 ok, BadCub, can you track that host_url bug ^ and if we don thave someone working on it by EOD tomorrow, let's untarget it 05:18:00 wanyen: have a seprate option for that as well 05:18:12 devananda: i will try to see if something is possible https://bugs.launchpad.net/ironic/+bug/1384379 and will have an update when you come back tomorrow 05:18:12 Launchpad bug 1384379 in OpenStack Compute (nova) "versions resource uses host_url which may be incorrect" [High,In progress] - Assigned to shihanzhang (shihanzhang) 05:18:28 rameshg87: great. thanks! 05:18:29 devananda: got it 05:18:56 rameshg87: pls assign bug to yourself in LP 05:19:16 devananda: doesn't look like it is targeted at this point 05:19:22 any other questions re: Kilo release before I move on? 05:19:37 BadCub: no, but it has kilo-rc-potential tag 05:19:47 * rameshg87 did it 05:19:48 shouldn't https://review.openstack.org/#/c/177213 have a conf file update too 05:19:49 yep, see that 05:20:20 NobodyCam: why? 05:20:27 NobodyCam: nope. the option was added in previous patches. it just wasn't *used* propertly 05:20:47 #topic Liberty summit planning 05:20:50 #link https://etherpad.openstack.org/p/liberty-ironic-design-summit-ideas 05:21:00 we've gathered ideas up there ^ 05:21:14 and I've collated them into a draft of our schedule, at the top 05:21:18 please take a look 05:21:47 also, please discuss items inline below, and if you have other topics, you can still add them to the 'pad 05:25:43 * rameshg87 wonders if everyone is on etherpad 05:26:09 * devananda thinks so 05:26:18 yep 05:26:25 devananda: Have added one item :) 05:26:34 I am on etherpad 05:26:45 * mrda thinks we've got a lot to discuss at summit 05:26:50 stendulker: where? 05:27:14 devananda: in etherpad 05:27:15 oh, got it 05:27:24 devananda: ok :) 05:28:06 for the two fishbowl sessions -- these are the large ~200 person rooms 05:28:33 my goal there is to present the most important changes & objectives, and gather feedback from the wider audience 05:29:00 are the any special IRC meeting held where we discuss items ? 05:29:06 I'm currently thinking that its worth discussing the driver composition & support issue in that context 05:29:25 devananda, cinder integration and networking will not be discussed at Ironic sessions? 05:29:27 but open to ohter ideas, too, if someone thinks another topic is better suited to that slot 05:29:51 wanyen: that's my hope, because each of those projects will have a session fo rus 05:30:00 devananda, cinder integration and networking issues will not be discussed at the ironic sessions? 05:30:07 oh. that's great. 05:30:31 devananda: do we know if all 200 seats will be dev's, or will there be other folk there too ie sales, etc 05:30:54 devananda, I see. 05:31:00 I've already gotten the ok from mestery re: neutron session for us to discuss ml2 gateway plugin 05:31:11 and the track in cinder seems to be on their radar as well 05:31:23 ditto for a nova session on clustered hypervisors 05:31:26 oh! that's not on our 'pad 05:32:11 devananda: line 281 05:32:15 devananda, are there any session with operators to gather input? 05:33:53 wanyen: yes. the ops summit is hte place for that 05:34:07 NobodyCam: we don't really control who's in the room, but the goal is for it to be ATC's only 05:34:37 devananda: ahh ok Ty :) 05:34:39 the smaller rooms should be "just us devs" but the 2 fishbowls will probably attract a lot of others 05:35:05 * devananda adds "HA Nova compute nodes / Clustered Hypervisor Support [discuss in Nova track]" section to the 'pad 05:35:44 NobodyCam: oh, thta's definitely not a topic for the fishbowl 05:36:35 only a handfull of people understand nova's code well enough to reason about how to fix it. it'll be a working session between some nova-cores and some ironic-cores 05:38:18 pshige: not sure what you mean by "special IRC meeting" 05:38:31 2nd meeting of SPEC Core, or something 05:39:03 pshige: you mean related to summit schedule? 05:39:14 yes 05:39:24 nothing planned, aside from this 05:39:31 we could certainly have one 05:40:12 anyone want a separat emeeting to discuss the schedule before next monday? 05:41:01 we could 05:41:09 * devananda wonders if everyone's asleep, or just busy reading the 'pad 05:41:18 * mrda thinks it looks pretty good already 05:41:30 mrda: :) 05:41:32 * rameshg87 thinks one week summit is not enough 05:41:58 rameshg87: +1 05:42:03 rameshg87: indeed 05:42:05 rameshg87: +1 05:42:28 also, I'm dead after a week of this already ... 05:42:49 devananda: are we into next topic ? 05:43:01 devananda: i wanted to share some updates on proliant ci 05:43:04 I think we're done with this topic 05:43:22 rameshg87: oh! ok. I was skipping driver updates today, but if you'd like 05:43:32 devananda: okay. i will just go ahead with mine 05:43:43 #topic open discussion 05:43:44 * naohirot I'm interested in proliant ci 05:43:59 devananda: we have made some changes in devstack and tempest for this - there are about 5 in devstack and 2 in tempest 05:44:14 devananda: i have tested and run ilo drivers with these changes and works fine 05:44:26 devananda: i triggered a sandbox review and it ran - https://review.openstack.org/#/c/176253/ 05:44:41 looks for "HP Proliant CI check" 05:44:51 it ran the test (took about 30 mins) - 05:45:38 rameshg87: fyi, the link to artifacts is broken - http://127.0.0.1/53/176253/1/check/iscsi-ilo-bios-ci/6c65f33 05:45:46 that's clerly not the right IP :) 05:45:48 i am just working on that 05:45:56 but that's awesome 05:45:59 it doesn't have upload logs yet - because i don't have a place to upload logs :( 05:46:06 yes 05:46:08 yet 05:46:15 can you summarize the changes you had to make to devstack & tempest? 05:46:25 https://etherpad.openstack.org/p/IloDriverGateTesting 05:46:49 tempest patches are there in bottom 05:46:50 rameshg87: I hear that your company might run a public cloud ... :) 05:47:20 devananda: yeah just checking internally regarding that :) 05:47:37 cool. lemme know if I can help 05:47:58 sure, thanks 05:48:07 devananda: important thing - i need ironic folks review on https://review.openstack.org/#/c/174733/ 05:48:24 lib/ironic changes in devstack for enabling ilo drivers for devstack 05:49:23 #info iLO CI system passed a sandbox test, has several patches up to enable the work 05:49:29 lol gota love this error: http://logs.openstack.org/33/174733/6/check/check-grenade-dsvm-neutron/414fa90/logs/error.txt.gz 05:49:30 #link https://review.openstack.org/#/c/174733/ 05:49:59 NobodyCam: yeah needs a recheck ? :) 05:50:25 :) 05:50:45 * ten minutes * 05:50:50 devananda: i will continue working on it. if all internal formalities are complete, we should have it pretty soon. 05:51:16 rameshg87: fantastic! 05:51:54 anyone have anything else? 05:52:11 not me 05:52:15 nope 05:52:19 nope 05:52:30 devananda: nope 05:52:32 oki doki - I'm happy to give everyone back 7 minutes :) 05:52:40 thank you all! 05:52:42 g'night all! thanks! 05:52:46 bye 05:52:50 bye 05:52:51 \o 05:52:52 good night 05:52:54 goodnight / goodday 05:52:57 devananda: good night 05:52:57 good night 05:52:59 bye 05:53:01 and thank you devananda for all the work ont the summit planning 05:53:07 bye 05:53:10 #endmeeting