19:00:02 #startmeeting Ironic 19:00:02 #chair devananda 19:00:03 Meeting started Mon Dec 16 19:00:02 2013 UTC and is due to finish in 60 minutes. The chair is NobodyCam. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:00:04 Welcome everyone to the Ironic meeting. 19:00:05 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:00:07 The meeting name has been set to 'ironic' 19:00:08 Current chairs: NobodyCam devananda 19:00:18 Of course the agenda can be found at: 19:00:18 #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting 19:00:24 #topic Greetings, roll-call and announcements 19:00:25 Who's here for the Ironic Meeting? 19:00:28 o/ 19:00:29 o/ 19:00:36 o/ 19:00:37 me 19:00:40 welcome all :) 19:00:51 devananda: is trapped in traffic today 19:00:53 Me! (Pretty new to Ironic, hi all!) 19:01:03 hi matty_dubs 19:01:04 o/ 19:01:19 me 19:01:27 :) 19:01:36 + 19:01:45 Haomeng:) 19:02:16 great to see everyone 19:02:34 one announcement 19:02:59 we want to start a new topic in the meeting 19:03:26 o/ 19:03:32 to go over the BP, and bug reports targeted at the next release.. just make sure we are on track 19:03:49 (we'll get to that shortly) 19:04:15 ahh also.. no meeting next week 19:04:23 holiday and all 19:04:33 NobodyCam: is there a meeting the week after that? 19:04:46 humm 19:04:50 NobodyCam: Dec 30 19:04:57 that would be the 1st 19:04:59 oh 19:05:18 I'm not sure I'll be able to make that one 19:05:41 but we'll play it by ear and see how many folks are going to be around 19:05:44 NobodyCam: could you just make an action item to let us know when the next meeting is :-) 19:06:18 #action (nobodycam) find out when next meeting will be and update agenda with correct time 19:06:32 I will be on holidays after this week so I won't be able to make the next two meetings 19:06:40 :) 19:06:40 same 19:06:44 :) 19:07:07 rloo_: so first glance looks like no meeting 19:07:11 :-p 19:07:20 NobodyCam: so can we decide now that it will be Jan whatever? 19:07:26 Jan 6 19:07:35 I am here for Ironic meeting, sorry to be late 19:07:42 I will be on vac. back on Jan 2nd. 19:07:55 ya 6th looks like it 19:08:11 but please check the agenda I'll make sure its updated 19:08:21 will do! 19:08:35 any announcements from any one else? 19:08:59 moving on. 19:09:00 #topic Outstanding, in-progress or Action Item updates 19:09:29 devananda: is working hard on https://blueprints.launchpad.net/ironic/+spec/instance-mapping-by-consistent-hash 19:10:03 we found a slight issue with the nova driver on friday 19:10:13 I'm still working on tempest and infra 19:10:26 romcheg1: please check the mailing lists 19:10:27 NobodyCam, that's the problem with the http_client ? 19:10:45 devananda: said there was some good info from the last nova meeting 19:11:15 lucasagomes: ya 19:11:42 o/ 19:11:48 NobodyCam: will do 19:11:52 :) 19:12:02 NobodyCam, right, I started reviewing the driver, didn't finish tho (it was almost end the day here) so I will post my comments tomorrow 19:12:15 lucasagomes: TY 19:12:59 ok lets jump in to the new section: Regular Topics 19:13:09 #topic Regular Topics 19:13:30 #link https://blueprints.launchpad.net/ironic 19:13:56 lets take a few minutes and go over the Bp's and bugs 19:15:02 only looking at Essential and high priority 19:15:56 so: romcheg1 anything on https://blueprints.launchpad.net/ironimigration-from -nova 19:16:23 NobodyCam: I started making a research on that 19:16:40 :) still on track :) 19:16:41 There are no chassis in nova 19:17:16 But in fact every node has it's chassis 19:17:27 NobodyCam: that link doesn't work for me 19:17:38 rloo_: https://blueprints.launchpad.net/ironic/+spec/migration-from-nova 19:17:45 thx jdob 19:17:56 :) 19:17:58 Currently I'm not sure how to handle that 19:18:51 j/win 3 19:18:54 romcheg1: lets hook up with devananda on that asap.. prob tomorrow morning by the time he's on line 19:18:55 romcheg1, yea, I think there's not much we can do in this case 19:19:01 nova doesn't have the chassis abstraction 19:19:07 so I would just put everything in one chassis 19:19:33 is the migration supposed to be automated? 19:19:49 rloo_: as much as posiable 19:19:51 and/or can info be collected first? 19:19:54 possiable 19:20:01 (like what chassis they'd like the nodes associated with) 19:20:25 otherwise, i'd go with lucasagomes' suggestion. 19:20:29 next one's are on devananda's plate so we'll a update when hes back on line 19:20:47 do we have Sun Jing here today? 19:21:36 about bug https://bugs.launchpad.net/ironic/+bug/1260099 I think we need "maintenance" provision state and save current state in target_provision _state 19:21:39 Launchpad bug 1260099 in ironic "Can not take a node offline for maintenance" [High,Confirmed] 19:21:39 ok need to get a update on "use ipmitool to provide serial console access" 19:21:52 She puts some comments in the BP 19:22:06 She is working on it as far as I know. 19:22:19 see my comment ^ 19:22:28 I had some discussion last week. 19:22:32 with her 19:22:47 I can reach Jing tmorrow morning 19:23:06 She said she'll use ipmitool + shellinabox first. 19:23:12 #link https://bugs.launchpad.net/ironic 19:23:30 haomeng2: great 19:23:30 then use jbjohn's pyghmi. 19:23:47 There should be two console implementaions 19:24:02 haomeng2: can you also check that she is still on track with that one 19:24:13 one use ipmitools for use who choose ipmitool as the power driver 19:24:23 sure:) 19:24:39 linggao: we only need one to meet graduation requirments 19:24:54 and the other pyghmi for user who choose to use ipminative for the ipminative. 19:24:58 got it 19:25:15 haomeng2: she can just up date the status on the bp as I know timming is tuff 19:25:15 Yes, the other one can be done later after the icehouse release. 19:25:35 IMO pyghmi is preferred, but not completed (no sensors, etc.) 19:25:41 ok 19:25:52 lots of high level bugs 19:26:05 not sure we have the time to go thru each one 19:26:08 :) 19:26:23 thou I beleieve we have every one here 19:26:31 yuriyz, yes. We'll plan to add sensor support when jbjonso comes back in Jan. 19:26:40 is there anyone blocked on anything releated to the bugs? 19:26:58 yuriyz, +1 19:28:41 please if you are working on a critical or high level bug please update the status as best you can. we have several moving parts at this point are are workig towards release 19:29:04 ok 19:29:11 If you help unblocking anything please reach out for help 19:29:16 gah 19:29:22 ok 19:29:24 *if you NEED help..... 19:29:46 any question/ comments? 19:30:18 ok moving on. 19:30:20 #topic Integration and testing 19:31:02 romcheg1: you gave a quick update above .. anythig to add? 19:31:13 NobodyCam: Nope 19:31:15 :) 19:31:59 just quick update on the dib element other then the issue we found friday (and this morning) things are moving along well 19:33:11 dkehn: can you give a up date on the neutrn intergration work? 19:33:59 https://review.openstack.org/#/c/60904 wait for final, in the middle of some neutron bugs, next on the list to get to neutron interface 19:34:14 #link https://review.openstack.org/#/c/60904 19:34:45 Last week we were working on PoC functional testing scheme. We sent pull request https://review.openstack.org/#/c/62410/2 please have a look. We are open for critisism. By the moment we've made a couple of fake tests and a set of shell scripts which allow to run testing on ubuntu machine. Some comments are available in the beginning of the ironic/irci/README.md file. 19:34:56 dkehn: great.. just off the top of your head do you have a time frame for that(we wont hold you to it) 19:35:23 NobodyCam: hmm, let me think about that 19:35:29 vkozhukalov: ahh I missed the pull request... we'll take a look in a bit 19:35:49 vkozhukalov, added myself to the review list, I will take a look at it (probably tomorrow) 19:35:56 #link https://review.openstack.org/#/c/62410 19:36:31 lucasagomes, thanks. will wait for feedback. 19:36:58 awesome... good to move on? any Questios / coments? 19:37:21 moving on. 19:37:25 #topic Python-IronicClient 19:37:44 We need more tests around the client 19:38:14 NobodyCam, +1, specially on the shell part 19:38:45 romcheg1, https://bugs.launchpad.net/python-ironicclient/+bug/1235431 19:38:47 Launchpad bug 1235431 in python-ironicclient "command-line shell lacks unit tests" [High,Triaged] 19:38:59 quick update: seems this patch unknowingly broke some of the client. https://github.com/openstack/python-ironicclient/commit/1fd3fcf59548ca53ea742a140457daabd579fe72 19:39:30 this was dissuced in channel this morning 19:39:55 NobodyCam, actually it didn't 19:40:11 :-p 19:40:13 the ironic driver on nova was using the libraries wrong 19:40:19 ahh :) 19:41:15 any other comments on client? 19:41:44 ok moving on 19:41:53 #topic Nova-driver 19:42:09 just quickly 19:43:00 we will prob start to break up the nova driver so that we are not tring to land a multi k line patch all at once 19:43:35 any Q / c on the nova-driver 19:43:40 May I ask a really dumb question here? What does 'Nova-driver' refer to? The Ironic driver for Nova? 19:43:49 matty_dubs: yep 19:44:02 #link https://review.openstack.org/#/c/51328/ 19:44:38 lucasagomes: quicker then I 19:44:45 :) I had it open 19:44:48 inclouding scheduler? 19:44:48 hehehe 19:44:55 NobodyCam: Thanks :) 19:45:16 haomeng2: we should not need to change scheduler 19:45:35 bm one. got 19:45:41 :) 19:46:04 good to move on? 19:46:22 #topic API discussion 19:46:25 lucasagomes: go 19:46:27 lol 19:46:29 :) 19:46:50 heh nothing much to say actually 19:46:56 there was two patches approved yesterday 19:47:03 but I had to rebase (and fix it) 19:47:12 so need to be reviewed again 19:47:18 :) 19:47:22 #link https://review.openstack.org/#/c/57495/ 19:47:27 #link https://review.openstack.org/#/c/60299/ 19:47:34 lucasagomes: I'll take a look after meeting 19:48:00 I also started drafting the blueprint to abort the deployment of the node 19:48:08 I put some infos how the API would look like 19:48:29 #link https://blueprints.launchpad.net/ironic/+spec/abort-deployment 19:48:30 awesome 19:48:38 so I will add more stuff on that tomorrow as well 19:48:44 :) 19:48:59 any other questions / comments on the api 19:49:36 ok then for the final tan minutes: 19:49:38 #topic Food for Thought / Open Discussion 19:49:46 open floor! 19:50:00 any comments / questions from anybody 19:50:43 So, I'm looking at https://bugs.launchpad.net/ironic/+bug/1260337 and wonder if there's a good way to generate test data. 19:50:44 Launchpad bug 1260337 in ironic "API objects to generate sample data for the auto-documentation" [Wishlist,In progress] 19:50:54 Maybe this just belongs in the main channel vs. this meeting though 19:51:12 Being new, I just worry I'd generate test data that didn't represent a real deployment. 19:51:13 * NobodyCam looks 19:51:24 can we have bp to discover hardware server node? 19:51:57 matty_dubs: I would start by looking at the test fixtures 19:52:01 Right now I'm just creating nodes and the like manually 19:52:04 NobodyCam: Oh, good thinking. 19:52:10 haomeng2 https://blueprints.launchpad.net/ironic/+spec/discovery-ramdisk 19:52:14 haomeng2: do you have ideas on how to do that 19:52:25 :) 19:52:32 #link https://blueprints.launchpad.net/ironic/+spec/discovery-ramdisk 19:52:59 ping ipmi port and try to get mac 19:53:12 will try 19:53:21 NobodyCam: I'm kui, Is there some tips to speed up the setup of dev test for ironic ? e.g. http_proxy, and make use of pip-cache / deploy-kexec elements in dib 19:53:28 haomeng2: add comments to the bp yuriyz linked in 19:53:44 that is in-band 19:53:48 kushi_: yes proxy helps a lot 19:54:09 haomeng2, you can talk to Sun Jing on it. 19:54:18 kushi_: see footnotes at the bottom of: 19:54:20 we need outofband maybe 19:54:20 #link http://docs.openstack.org/developer/tripleo-incubator/devtest.html 19:54:27 ok 19:55:11 linggao:thank 19:55:54 any thing else on anyone mind 19:56:19 NobodyCam: yes, I have set up one. 19:56:20 haomeng2, np. we had discovered two kinds of hw discovery. One was discover through ramdisk. Otherer was autodiscovery and registration of beremetal nodes. 19:57:09 lianggao:ok.will diss with jing 19:57:15 discovered/discussed 19:57:46 we may want/need to use both methods 19:58:02 NobodyCam +1 19:58:04 NobodyCam, +1 19:58:31 :) last two minutes 19:58:41 yes +1 19:59:24 if I dont see you before the upcomming holiday's please have a great time and be safe!!! 19:59:46 great meeting everyone 19:59:49 Have great holidays, everyone! 20:00:03 getting back to my dream:) 20:00:07 anything we missed here we can go over in channel 20:00:13 thank you all 20:00:15 haomeng2 +1 :) 20:00:16 haomeng2, thanks for coming :) 20:00:21 #endmeeting