| *** dimsum__ has quit IRC | 00:03 | |
| *** mordred has joined #openstack-ironic | 00:04 | |
| mordred | TheJulia, JayF, jamielennox, NobodyCam: does python-ironicclient support keystone sessions yet? | 00:05 |
|---|---|---|
| *** naohirot has joined #openstack-ironic | 00:05 | |
| mordred | ooh! it looks like it does | 00:08 |
| jamielennox | yea, I've seen that code but I've never tested it | 00:14 |
| *** davideagnello has quit IRC | 00:15 | |
| *** shoutm has quit IRC | 00:18 | |
| mordred | jamielennox: I think I just fiured out how to rip out a BUNCH of hacks in shade a) by having occ do handle making ksa auth objects like you suggsted and b) using the ironicclient session codepath | 00:21 |
| *** shoutm has joined #openstack-ironic | 00:21 | |
| mordred | jamielennox: there's one part of the patch I think you won't like - but if you have violent reactions to it, I can just write a plugin | 00:21 |
| jamielennox | mordred: i always like to see less of shade. send me a link to the review, when i have some time i want to get occ, osc, shade and see if we can come up with a single leading scheme to get rid of all these problems | 00:23 |
| *** shadower has joined #openstack-ironic | 00:23 | |
| mordred | jamielennox: I think they're all streamlined in shade/occ now (with this patches) - the main 'hack' now is that ironic has a no-keystone operational mode | 00:23 |
| jamielennox | I'm coming around to occ being dicts only but in which case it shouldn't validate auth params at all | 00:23 |
| mordred | jamielennox: as in - it should just defer auth dict params to ksa's execptions, yeah? | 00:24 |
| *** puranamr has quit IRC | 00:24 | |
| jamielennox | mordred: so ksa plugins specifically don't _require_ keystone as a backed, there just a lot of reuse we can do if they do | 00:24 |
| jamielennox | right, do the dict merge, validation should be the job of whoever actually creates the plugin | 00:25 |
| mordred | jamielennox: I think I'm of the same mind - hacking my way towards somethinglike that right now | 00:25 |
| *** puranamr has joined #openstack-ironic | 00:26 | |
| jamielennox | and using ksa right? not ksc? | 00:26 |
| mordred | oh yeah | 00:26 |
| mordred | jamielennox: for the iroinc case, I'm jsust oing "if auth_type == 'None): auth_type = 'admin_token' (which is the thing I could imagine you not liking) | 00:27 |
| mordred | other option is ot make a "None" plugin that is pretty much the exact same as admin_token | 00:27 |
| mordred | but which is mildly clearly in terms of intent for usage | 00:27 |
| jamielennox | cool, i think that the clients shouldn't notice the difference | 00:27 |
| mordred | yeah | 00:27 |
| mordred | o - sahde is using ksa now - that's landed :) | 00:28 |
| jamielennox | there is a default value for both argparse and conf | 00:28 |
| mordred | so, consider yourself in production | 00:28 |
| jamielennox | yay?! | 00:28 |
| mordred | :) | 00:28 |
| mordred | actually, I don't know that we've released shade and restarted nodepool since I landed that patch | 00:28 |
| jamielennox | but yea, you should be able to set a default plugin that gets loaded if nothing set | 00:29 |
| mordred | oh - well, this is different | 00:29 |
| mordred | this is explicitly configyuring to use no-auth | 00:29 |
| jamielennox | I'm not sure setting admin_token as default is good though | 00:29 |
| mordred | rather than not setting an auth type | 00:30 |
| jamielennox | ok | 00:30 |
| mordred | one sec - patches incoming | 00:30 |
| jamielennox | umm, i had a patch for this in ksc that i abandoned for ksa so you could register your own plugins without setup tools | 00:31 |
| jamielennox | but otherwise yea, either check before loading out catch the plugin not found makes sense | 00:31 |
| jamielennox | s/out/or | 00:32 |
| mordred | jamielennox: https://review.openstack.org/#/q/status:open+topic:occ-ksa-auth,n,z | 00:32 |
| mordred | jamielennox: here is untested first stab | 00:33 |
| jamielennox | mordred: do load_from_options rather than plugin_class, there is some validation there | 00:36 |
| mordred | jamielennox: but I think load_from_options wants options objects? | 00:37 |
| jamielennox | oh? | 00:38 |
| mordred | does it not? it'll just work with kwargs? | 00:38 |
| jamielennox | yep, it gets options from self.get_options | 00:39 |
| mordred | ok. I'll try tat | 00:39 |
| jamielennox | https://github.com/openstack/keystoneauth/blob/master/keystoneauth1/loading/base.py | 00:39 |
| jamielennox | bottom | 00:39 |
| jamielennox | On phone on a train so not quick with links | 00:39 |
| mordred | yeah. I just was unclear since it say "from_options" and I'm not using options in any way | 00:39 |
| mordred | jamielennox: on plane - same thing :) | 00:40 |
| jamielennox | OK, so maybe we need to make that clearer at some point | 00:40 |
| jamielennox | so why the admi _token default? | 00:41 |
| mordred | jamielennox: it also looked like it expects the argument list to be normalized to o.dest key names before it's run ... so I may need to work on that a little bit | 00:41 |
| jamielennox | what is ironic passing through there that they won't set --os-auth-type? | 00:42 |
| mordred | jamielennox: the use caes for admin_token is that in keystone-ess ironic setup, the operator has an ironic endpoint and that's all they need because it's a no-auth | 00:42 |
| mordred | they will set os-auth-type to None | 00:42 |
| jamielennox | yep, standard admin_token use case, but why not set os-auth-type to admin_token? | 00:43 |
| mordred | and then config['auth']['endpoint'] = $the_ironicendpoint | 00:43 |
| mordred | right now, backwards compat | 00:43 |
| mordred | all of the code consuming this is listing auth_type as "None" | 00:43 |
| jamielennox | ergh | 00:43 |
| mordred | and also - they actually don't have a token | 00:43 |
| mordred | so it might be a confusing end user experience to use a thing called "*_token" and then set the token to nothing - which is why I was thinking of writing a no-auth auth plugin | 00:44 |
| mordred | (the second thing is the thing I'm more concerned with - we can migrate people to a new name) | 00:44 |
| jamielennox | OK, well no auth would be OK. the other thing is just not create a plugin at all, set endpoint_override to url and authenticate=false on reqs | 00:45 |
| jamielennox | in theory this works, i would need to test it to make sure | 00:46 |
| mordred | jamielennox: can I do that at session creation time? | 00:46 |
| jamielennox | endpoint override is adapter time | 00:46 |
| mordred | oh - what's an adapter? | 00:47 |
| jamielennox | you can create session with no auth and our will default to unauthenticated | 00:47 |
| jamielennox | then it depend how the client calls session methods | 00:47 |
| jamielennox | if the client never asks for the service catalog it'd be OK | 00:48 |
| *** dimsum__ has joined #openstack-ironic | 00:48 | |
| *** puranamr has quit IRC | 00:49 | |
| mordred | hrm. I'm not sure where to provide endpoint_override in that model in such a way that the client doesn't have to know to pass it all the time - but maybe that'll just work out - lemme look | 00:49 |
| jamielennox | i think it won't matter, you must be passing a full url to request() anyway because there's no catalog | 00:51 |
| mordred | so - righ tnow we're constructing the ironic Client differently based on auth or no-auth | 00:51 |
| jamielennox | override means use this instead of catalog when you request based on path | 00:51 |
| *** puranamr has joined #openstack-ironic | 00:51 | |
| mordred | and in one of those, we're passing in an endpoint override | 00:51 |
| mordred | I'd like to stop doing that, because it means I have to know at Client creation time what the auth plugin is so I can check if it's None | 00:52 |
| mordred | which means I can't just delegate creation of all of that | 00:52 |
| jamielennox | ironic client must be doing something weird internally there to handle that without a session then | 00:53 |
| mordred | the nice thing about a no-auth auth p,ugin is that the code flow for getting a working session that does the right thing is the same regardless. the downside is that it means creating an auth plugin that does a short circuit in an area where there is already a facility for it | 00:53 |
| jamielennox | sure, admin_token is essentially a noauth already without passing token | 00:54 |
| mordred | yah. so I think this will 'work' as a first step - but I think there is definitely room to make it more betterer | 00:55 |
| jamielennox | yep, so my only concern there is where you have it is the default for everything consuming occ, not just ironic | 00:55 |
| mordred | well, it's only the default if you explicitly pass in "None" as your auth_type - but I'm guessing there is another valid usecase for doing that? | 00:56 |
| jamielennox | none as string i don't think so, but from cli none value would be common | 00:56 |
| mordred | ah. nod | 00:57 |
| mordred | k. I'll work with TheJulia to get things to move from None to admin_token explicitly | 00:57 |
| jamielennox | yep, that's best longer term, when i have a decent screen I'll have a look at how shade loads ironic client | 00:58 |
| mordred | jamielennox: ++ | 00:58 |
| mordred | jamielennox: also - with occ, there will never be None as an auth_type value unless it's added in a config file | 00:58 |
| mordred | jamielennox: so even from a command line, you'll never get a None passed through to anything without requesting it explicitly. that may also be something we want to visit | 00:59 |
| jamielennox | yea, I'm keen to see if we can standardise occ cli loading with ksa. i was just waiting for it to be released and then to have some time | 01:05 |
| mordred | ++ | 01:05 |
| openstackgerrit | Wang Wei proposed openstack/ironic: Add description how to restart ironic services in Fedora/RHEL7/CentOS7 https://review.openstack.org/220879 | 01:07 |
| openstackgerrit | Gopi Krishna S proposed openstack/ironic-specs: Add out-of-band RAID configuration support for Cisco Hardware https://review.openstack.org/192142 | 01:19 |
| *** Marga_ has joined #openstack-ironic | 01:19 | |
| *** Marga_ has quit IRC | 01:20 | |
| *** Marga_ has joined #openstack-ironic | 01:21 | |
| *** shoutm has quit IRC | 01:23 | |
| openstackgerrit | Wang Wei proposed openstack/ironic: Add supported environment 'VMware' to comments https://review.openstack.org/220880 | 01:27 |
| *** shoutm has joined #openstack-ironic | 01:28 | |
| *** stendulker has joined #openstack-ironic | 01:32 | |
| *** yuanying has quit IRC | 01:32 | |
| *** yuanying has joined #openstack-ironic | 01:35 | |
| *** Marga_ has quit IRC | 01:35 | |
| *** zhenguo has joined #openstack-ironic | 01:35 | |
| *** dimsum__ has quit IRC | 01:40 | |
| *** shoutm has quit IRC | 01:48 | |
| *** shoutm has joined #openstack-ironic | 01:50 | |
| *** david-ly_ has joined #openstack-ironic | 01:51 | |
| *** krtaylor has quit IRC | 01:52 | |
| *** yuanying has quit IRC | 01:52 | |
| *** yuanying has joined #openstack-ironic | 01:53 | |
| *** baoli has joined #openstack-ironic | 01:54 | |
| *** david-lyle has quit IRC | 01:54 | |
| *** stendulker has quit IRC | 01:56 | |
| *** yuanying has quit IRC | 02:01 | |
| *** yuanying has joined #openstack-ironic | 02:01 | |
| *** yuanying has quit IRC | 02:02 | |
| *** krtaylor has joined #openstack-ironic | 02:05 | |
| *** spandhe has joined #openstack-ironic | 02:13 | |
| *** yuanying has joined #openstack-ironic | 02:13 | |
| *** davideagnello has joined #openstack-ironic | 02:15 | |
| *** yuanying has quit IRC | 02:15 | |
| *** baoli has quit IRC | 02:16 | |
| *** davideagnello has quit IRC | 02:21 | |
| *** puranamr has quit IRC | 02:25 | |
| *** puranamr has joined #openstack-ironic | 02:27 | |
| *** puranamr has quit IRC | 02:30 | |
| *** puranamr has joined #openstack-ironic | 02:31 | |
| *** yuanying has joined #openstack-ironic | 02:38 | |
| *** praneshp_ has joined #openstack-ironic | 02:39 | |
| *** dimsum__ has joined #openstack-ironic | 02:40 | |
| *** praneshp has quit IRC | 02:41 | |
| *** praneshp_ is now known as praneshp | 02:41 | |
| *** zz_natorious is now known as natorious | 02:41 | |
| *** dimsum__ has quit IRC | 02:46 | |
| *** natorious is now known as zz_natorious | 02:48 | |
| *** amotoki has joined #openstack-ironic | 02:48 | |
| *** yuanying has quit IRC | 02:50 | |
| *** shoutm_ has joined #openstack-ironic | 02:53 | |
| *** yuanying has joined #openstack-ironic | 02:54 | |
| *** shoutm has quit IRC | 02:54 | |
| *** baoli has joined #openstack-ironic | 02:55 | |
| *** shoutm_ has quit IRC | 03:03 | |
| *** shoutm has joined #openstack-ironic | 03:03 | |
| *** puranamr has quit IRC | 03:04 | |
| *** chenglch has joined #openstack-ironic | 03:04 | |
| *** puranamr has joined #openstack-ironic | 03:05 | |
| openstackgerrit | chenglch proposed openstack/ironic: Add iscsi_opc and agent_opc to manage the OpenPOWER servers https://review.openstack.org/209824 | 03:05 |
| *** shoutm_ has joined #openstack-ironic | 03:22 | |
| *** shoutm has quit IRC | 03:24 | |
| *** puranamr has quit IRC | 03:26 | |
| *** ramineni_ has joined #openstack-ironic | 03:29 | |
| *** persia has quit IRC | 03:30 | |
| *** persia has joined #openstack-ironic | 03:32 | |
| *** zz_natorious is now known as natorious | 03:41 | |
| *** yuanying_ has joined #openstack-ironic | 03:42 | |
| *** natorious is now known as zz_natorious | 03:44 | |
| *** deray has joined #openstack-ironic | 03:44 | |
| *** yuanying has quit IRC | 03:44 | |
| *** shoutm_ has quit IRC | 03:46 | |
| *** puranamr has joined #openstack-ironic | 03:54 | |
| *** rameshg871 has joined #openstack-ironic | 03:54 | |
| *** vvb has joined #openstack-ironic | 03:55 | |
| *** baoli has quit IRC | 03:55 | |
| *** vvb has left #openstack-ironic | 03:56 | |
| *** jamielennox is now known as jamielennox|away | 03:56 | |
| *** puranamr has quit IRC | 03:57 | |
| *** puranamr has joined #openstack-ironic | 03:58 | |
| *** shoutm has joined #openstack-ironic | 04:03 | |
| *** lazy_prince has joined #openstack-ironic | 04:05 | |
| *** puranamr has quit IRC | 04:18 | |
| *** puranamr has joined #openstack-ironic | 04:19 | |
| *** saripurigopi has joined #openstack-ironic | 04:39 | |
| *** deray has quit IRC | 04:43 | |
| *** deray has joined #openstack-ironic | 04:53 | |
| *** bizarrochristy has joined #openstack-ironic | 05:04 | |
| *** davideagnello has joined #openstack-ironic | 05:16 | |
| *** shoutm_ has joined #openstack-ironic | 05:31 | |
| *** shoutm has quit IRC | 05:34 | |
| *** spandhe has quit IRC | 05:36 | |
| *** davideagnello has quit IRC | 05:38 | |
| *** ukalifon has joined #openstack-ironic | 05:44 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Validate the input of properties https://review.openstack.org/215505 | 05:56 |
| *** alexpilotti_ has joined #openstack-ironic | 06:01 | |
| *** VikasC has joined #openstack-ironic | 06:09 | |
| openstackgerrit | Merged openstack/ironic: Updated from global requirements https://review.openstack.org/220667 | 06:19 |
| openstackgerrit | Tan Lin proposed openstack/ironic: Support trusted boot using iPXE https://review.openstack.org/207278 | 06:20 |
| *** baoli has joined #openstack-ironic | 06:26 | |
| *** baoli has quit IRC | 06:31 | |
| *** Haomeng has quit IRC | 06:34 | |
| *** Haomeng|2 has joined #openstack-ironic | 06:34 | |
| *** puranamr has quit IRC | 06:35 | |
| *** mgoddard has joined #openstack-ironic | 06:40 | |
| *** lsmola has joined #openstack-ironic | 06:47 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Improve the ability to resolve capability value https://review.openstack.org/190900 | 06:48 |
| *** mgoddard has quit IRC | 06:50 | |
| zigo | Where to report bugs for python-seamicroclient? | 07:01 |
| zigo | FYI: https://paste.debian.net/310789/ | 07:01 |
| zigo | (when running the Debian package in Experimental) | 07:01 |
| *** ifarkas has joined #openstack-ironic | 07:09 | |
| *** chenglch|2 has joined #openstack-ironic | 07:12 | |
| *** chenglch has quit IRC | 07:12 | |
| *** ijw_ has joined #openstack-ironic | 07:18 | |
| openstackgerrit | Pranesh Pandurangan proposed openstack/ironic-specs: Add a new command to count ironic nodes https://review.openstack.org/221173 | 07:27 |
| *** Guest32267 is now known as d0ugal | 07:32 | |
| *** d0ugal has joined #openstack-ironic | 07:32 | |
| *** davideagnello has joined #openstack-ironic | 07:38 | |
| *** davideagnello has quit IRC | 07:43 | |
| *** achanda has joined #openstack-ironic | 07:43 | |
| *** mgoddard has joined #openstack-ironic | 07:45 | |
| *** ktolstoy has joined #openstack-ironic | 07:56 | |
| *** chlong has quit IRC | 07:56 | |
| *** ijw_ has quit IRC | 08:02 | |
| *** mbound has joined #openstack-ironic | 08:06 | |
| *** lucasagomes has joined #openstack-ironic | 08:06 | |
| *** mbound has quit IRC | 08:07 | |
| *** jcoufal has joined #openstack-ironic | 08:09 | |
| *** dtantsur|afk is now known as dtantsur | 08:13 | |
| dtantsur | Morning Ironic | 08:13 |
| *** romainh has joined #openstack-ironic | 08:14 | |
| openstackgerrit | Anton Arefiev proposed openstack/python-ironicclient: Introduce tempest-lib to functional tests https://review.openstack.org/216701 | 08:18 |
| *** athomas has joined #openstack-ironic | 08:20 | |
| lucasagomes | dtantsur, morning! | 08:25 |
| *** achanda has quit IRC | 08:25 | |
| dtantsur | lucasagomes, hey man, how're you doing? | 08:25 |
| lucasagomes | dtantsur, all is well! How are you? | 08:25 |
| dtantsur | lucasagomes, pretty fine. I finally got some good sleep :) only about worried about all that political things around Europe (refugees, "refugees", etc) | 08:26 |
| lucasagomes | dtantsur, oh nice. Yesterday I was playing hearthstone until kinda late, maybe I need some more sleep now heh | 08:27 |
| dtantsur | hehe | 08:28 |
| *** achanda has joined #openstack-ironic | 08:28 | |
| lucasagomes | dtantsur, yeah, it's complicated right now. Ireland will take some of the refugee AFAIK, dunno how many | 08:28 |
| dtantsur | I haven't had good sleep for 1.5 months due to damned weather | 08:28 |
| lucasagomes | dtantsur, too hot over there? | 08:28 |
| dtantsur | lucasagomes, yeah, it was like +32-35 for hell lot of time; impossible to sleep in our flat | 08:29 |
| dtantsur | also, Czech people are pretty noisy in mornings :D | 08:29 |
| lucasagomes | dtantsur, holy monkeys yeah that's pretty hot | 08:29 |
| lucasagomes | lol | 08:29 |
| dtantsur | lucasagomes, how was summer in Ireland? Was it a nice day? :D | 08:29 |
| lucasagomes | dtantsur, lol ah, I like cold so for me was fine | 08:30 |
| lucasagomes | but we had I think 2 weeks only which temperatures got up to 20+ | 08:30 |
| *** jistr has joined #openstack-ironic | 08:30 | |
| lucasagomes | the thermometers here usually maxes out at 26C | 08:30 |
| lucasagomes | saves mercury you know? | 08:30 |
| dtantsur | LOOOL | 08:31 |
| lucasagomes | heh | 08:31 |
| dtantsur | ours is -30 to +50 | 08:31 |
| lucasagomes | damn hah yeah dude it's so extreme | 08:31 |
| *** davechen has joined #openstack-ironic | 08:31 | |
| dtantsur | lucasagomes, so lemme tell you a thing: last summer we left a thermometer exposed to the sun, and it broke down after reaching +50 Oo | 08:32 |
| *** davechen has left #openstack-ironic | 08:33 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow abort for CLEANWAIT states https://review.openstack.org/201552 | 08:33 |
| *** deray has joined #openstack-ironic | 08:33 | |
| lucasagomes | dtantsur, lol wth! mannn | 08:33 |
| lucasagomes | dtantsur, end of the year I'm going to brazil and it's summer there | 08:34 |
| lucasagomes | I will def complain about the weather here on IRC | 08:34 |
| *** romcheg has joined #openstack-ironic | 08:34 | |
| * lucasagomes is expecting 35C+ | 08:35 | |
| lucasagomes | :-( | 08:35 |
| dtantsur | heh | 08:35 |
| *** ktolstoy has quit IRC | 08:35 | |
| dtantsur | good luck man! at least do some swimming, we didn't have chance this year | 08:35 |
| lucasagomes | ++ yeah will do for sure, work in the swimming pool | 08:37 |
| betherly | morning Ironic! | 08:37 |
| lucasagomes | betherly, morning! | 08:37 |
| betherly | do i read complaints re hot weather? | 08:37 |
| betherly | :o :o :o | 08:37 |
| lucasagomes | betherly, hah you probably will never understand, but they are real! | 08:38 |
| betherly | doeseth mine ears deceiveth me? | 08:38 |
| betherly | haha | 08:38 |
| dtantsur | betherly, morning | 08:41 |
| dtantsur | lucasagomes, I'd like to start telling people to call their drivers iscsi_smth, not pxe_smth. WDYT? | 08:41 |
| dtantsur | first victim is sambetts with his CIMC driver :) | 08:41 |
| rameshg871 | dtantsur: +1 | 08:43 |
| lucasagomes | dtantsur, I think it's fine | 08:43 |
| rameshg871 | dtantsur: now that we have 255 characters as well | 08:43 |
| * rameshg871 thinks of much fancier names | 08:43 | |
| lucasagomes | I think we should have a template at least for drivers? <boot>_<deploy>_<power> or something like that | 08:43 |
| lucasagomes | drivers name* | 08:44 |
| dtantsur | pxe_iscsi_cimc? well, maybe... | 08:44 |
| lucasagomes | dtantsur, yeah, well at least it tells exactly what the driver does | 08:44 |
| *** pelix has joined #openstack-ironic | 08:44 | |
| *** derekh has joined #openstack-ironic | 08:46 | |
| dtantsur | rameshg871, lucasagomes, we should probably start creating aliases for existing drivers, wdyt? | 08:48 |
| rameshg871 | dtantsur: I was about to say that. I think we should do that | 08:48 |
| rameshg871 | dtantsur: but what about your work on dynmically using different interfaces ? | 08:48 |
| rameshg871 | dtantsur: might need to consider that as well. | 08:49 |
| lucasagomes | dtantsur, ++ | 08:49 |
| dtantsur | fair enough, I plan on landing it in M | 08:49 |
| dtantsur | I hope deva will leave us a time slot for it | 08:49 |
| dtantsur | * on the summit | 08:49 |
| *** stendulker has joined #openstack-ironic | 08:51 | |
| *** achanda has quit IRC | 08:52 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic-specs: New driver interface for BIOS configuration https://review.openstack.org/209612 | 08:53 |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic-specs: Intel Node Manager policies support https://review.openstack.org/214657 | 08:54 |
| *** VikasC has quit IRC | 08:57 | |
| sambetts | Morning lucasagomes, dtantsur thanks for name suggestion :) btw wasn't there a plan at the summit to change all the names? | 08:57 |
| dtantsur | sambetts, morning | 08:58 |
| dtantsur | sambetts, yeah, we're working on the spec still: https://review.openstack.org/#/c/188370/ | 08:58 |
| sambetts | dtantsur: ah cool | 08:58 |
| *** VikasC has joined #openstack-ironic | 09:04 | |
| sambetts | dtantsur: Is there somewhere I can link out to from my doc which explains agent deploy in more depth/ | 09:07 |
| sambetts | ? | 09:07 |
| dtantsur | I'm not 100% sure we have it | 09:08 |
| dtantsur | if not, you can just leave a note, that iSCSI deploy writes image from a conductor, while Agent deploy writes image from the ramdisk | 09:09 |
| *** wuhg has quit IRC | 09:11 | |
| sambetts | dtantsur: even that seems vague :( hmmm ... | 09:11 |
| dtantsur | pretty vague, still at least gives some clues | 09:13 |
| dtantsur | in the long run, we need to properly document it | 09:13 |
| sambetts | Unlike with | 09:16 |
| sambetts | iSCSI deploy in Agent deploy, the ramdisk is responsible for writing the image to the | 09:16 |
| sambetts | disk, instead of the conductor. | 09:16 |
| *** e0ne has joined #openstack-ironic | 09:16 | |
| sambetts | dtantsur: ^ thats what I've added to the bottom of the agent section | 09:16 |
| sambetts | WDYT? | 09:16 |
| dtantsur | LGTM | 09:17 |
| sambetts | :) | 09:21 |
| *** yog_ has joined #openstack-ironic | 09:29 | |
| lucasagomes | sambetts, morning! | 09:36 |
| *** davideagnello has joined #openstack-ironic | 09:39 | |
| *** Marga_ has joined #openstack-ironic | 09:41 | |
| *** e0ne has quit IRC | 09:41 | |
| sambetts | Morning lucasagomes o/ | 09:43 |
| *** davideagnello has quit IRC | 09:44 | |
| *** dimsum__ has joined #openstack-ironic | 09:46 | |
| *** jlvillal has quit IRC | 09:50 | |
| *** jlvillal has joined #openstack-ironic | 09:50 | |
| *** dims_ has joined #openstack-ironic | 09:50 | |
| openstackgerrit | Merged openstack/ironic: Add vendor interface to ipminative driver https://review.openstack.org/215185 | 09:51 |
| zhenguo | dtantsur: thanks for review the nodes-tagging patch, the inconsistent operations is because node-tags-list/node-tags-update are for multiple tags, and node-tag-add/node-tag-remove for individual one. | 09:54 |
| *** dimsum__ has quit IRC | 09:55 | |
| dtantsur | zhenguo, it's still better to have one command prefix "node-tag" or "node-tags" IMO | 09:55 |
| dtantsur | it's confusing otherwise | 09:55 |
| dtantsur | zhenguo, treat as that we have a collection "node-tags". And we have operations on it: list, update, add, remove | 09:55 |
| zhenguo | dtantusr: ok, will change all tags to tag | 09:56 |
| *** naohirot has quit IRC | 09:57 | |
| *** shoutm_ has quit IRC | 09:57 | |
| *** e0ne has joined #openstack-ironic | 09:57 | |
| zhenguo | dtantusr: will address that in the coming patch set, and thanks very much for reviewing this again :) it's almost die. | 09:58 |
| dtantsur | sorry for taking long - too many things to do before the upcoming release(s) | 09:59 |
| *** chenglch|2 has quit IRC | 10:01 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-specs: xUpdate nodes-tagging spec https://review.openstack.org/192935 | 10:04 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic-specs: Update nodes-tagging spec https://review.openstack.org/192935 | 10:05 |
| *** priteau has joined #openstack-ironic | 10:14 | |
| *** praneshp has quit IRC | 10:18 | |
| *** jlvillal has quit IRC | 10:22 | |
| openstackgerrit | Wang Wei proposed openstack/ironic: Add description how to restart ironic services in Fedora/RHEL7/CentOS7 https://review.openstack.org/220879 | 10:27 |
| *** romainh has quit IRC | 10:29 | |
| *** shoutm has joined #openstack-ironic | 10:39 | |
| *** baoli has joined #openstack-ironic | 10:42 | |
| *** bizarrochristy has quit IRC | 10:44 | |
| *** vvb has joined #openstack-ironic | 10:45 | |
| *** vvb has quit IRC | 10:45 | |
| *** Marga_ has quit IRC | 10:46 | |
| *** baoli has quit IRC | 10:47 | |
| *** VikasC has quit IRC | 10:57 | |
| dtantsur | holy cow, qemu-img pull the whole Gtk libraries stack on the agent ramdisk | 10:57 |
| dtantsur | no surprise it's so big >_< | 10:58 |
| *** Marga_ has joined #openstack-ironic | 10:58 | |
| *** jlvillal has joined #openstack-ironic | 11:03 | |
| lucasagomes | dtantsur, heh yeah man | 11:10 |
| lucasagomes | it alsos pulls libX11 stuff | 11:11 |
| dtantsur | I'm asking qemu folks right now, if it's really required | 11:11 |
| sambetts | :( I hate that with things that use the gnome librarys | 11:11 |
| dtantsur | otherwise we should consider alternatives | 11:11 |
| sambetts | its like you want to use this tiny app, here have 2 gig of extra stuff | 11:11 |
| openstackgerrit | Merged openstack/ironic: Use oslo.messaging serializers https://review.openstack.org/188701 | 11:11 |
| openstackgerrit | Merged openstack/python-ironicclient: Replace ConfigParser with six.moves.configparser. https://review.openstack.org/212350 | 11:12 |
| lucasagomes | dtantsur, not sure if we have much alternatives to qemu-img | 11:13 |
| * lucasagomes doesn't know any | 11:13 | |
| * rameshg871 goes home | 11:15 | |
| *** rameshg871 has quit IRC | 11:15 | |
| *** romainh has joined #openstack-ironic | 11:16 | |
| dtantsur | so | 11:16 |
| dtantsur | [13:15] <davidgiluk> dtantsur: Yes, but a qemu-img I built from source has many less | 11:16 |
| dtantsur | [13:16] <davidgiluk> dtantsur: Do you want a file a fedora bug against that? | 11:16 |
| dtantsur | lucasagomes, ^^ | 11:17 |
| lucasagomes | oh | 11:17 |
| lucasagomes | dtantsur, worth checking dependencies from other distros | 11:17 |
| *** deray has quit IRC | 11:17 | |
| dtantsur | sambetts, have Ubuntu or Debian at hand? | 11:17 |
| lucasagomes | cause yeah it can be a packaging bug in fedora indeed | 11:17 |
| sambetts | dtantsur: debian | 11:18 |
| dtantsur | sambetts, could you check: 1. qemu-img package deps, 2. qemu-img binary ldd output? | 11:18 |
| *** lucasagomes is now known as lucas-hungry | 11:19 | |
| * lucas-hungry bbl will grab some food | 11:19 | |
| *** stendulker has quit IRC | 11:19 | |
| betherly | lucas-hungry: food sounds like a good idea | 11:20 |
| *** Marga_ has quit IRC | 11:20 | |
| sambetts | dtantsur: http://paste.openstack.org/show/450111/ | 11:25 |
| dtantsur | so, not gtk, not X | 11:27 |
| openstackgerrit | Merged openstack/python-ironicclient: Fixes file cache TypeError https://review.openstack.org/220475 | 11:28 |
| sambetts | dtantsur: Not in that list as far as I can tell | 11:28 |
| dtantsur | sambetts, what version of qemu-img do you have? | 11:28 |
| sambetts | qemu-img version 2.1.2, Copyright (c) 2004-2008 Fabrice Bellard | 11:29 |
| sambetts | dtantsur: ^ | 11:29 |
| sambetts | dtantsur: Full package info http://paste.openstack.org/show/450119/ | 11:29 |
| *** ramineni_ has quit IRC | 11:31 | |
| dtantsur | reported: https://bugzilla.redhat.com/show_bug.cgi?id=1260996 | 11:35 |
| openstack | bugzilla.redhat.com bug 1260996 in qemu "qemu-img requires a whole stack of X and Gtk libraries" [Unspecified,New] - Assigned to virt-maint | 11:35 |
| *** thrash|wknd is now known as thrash | 11:36 | |
| *** deray has joined #openstack-ironic | 11:36 | |
| *** davideagnello has joined #openstack-ironic | 11:40 | |
| *** early has quit IRC | 11:41 | |
| *** saripurigopi has quit IRC | 11:41 | |
| * TheJulia reads log and determines more coffee required | 11:44 | |
| *** davideagnello has quit IRC | 11:45 | |
| *** early has joined #openstack-ironic | 11:49 | |
| sambetts | Morning TheJulia o/ | 11:50 |
| dtantsur | TheJulia, o/ | 11:51 |
| dtantsur | lucas-hungry, so, we've found the problem, it's in qemu ./configure script. once it is fixed, IPA on Fedora will be substantially smaller | 11:52 |
| TheJulia | good morning | 11:53 |
| TheJulia | dtantsur: out of curiosity, how much smaller? | 11:55 |
| dtantsur | TheJulia, not sure now, didn't try. But it pulls in ~100 more libs, and gtk libs pull in some artwork | 11:58 |
| dtantsur | TheJulia, I also have a patch removing .pyc and .pyo files: https://review.openstack.org/#/c/221243/ | 11:59 |
| dtantsur | (not too much of saving, only 2MB compressed though) | 12:00 |
| *** karimb has joined #openstack-ironic | 12:01 | |
| *** dtantsur is now known as dtantsur|brb | 12:01 | |
| *** e0ne has quit IRC | 12:01 | |
| *** alexpilotti_ has quit IRC | 12:03 | |
| *** e0ne has joined #openstack-ironic | 12:04 | |
| *** mbound has joined #openstack-ironic | 12:05 | |
| TheJulia | mordred: jamielennox|away: you guys make me worry. ;) Afaik, the client doesn't do anything funky, the server just doesn't care if it gets the auth data, and if it does it discards it. My only concern admin_token is that the client library upon load in the standard token path attempts to validate the token and kaboom, but if admin_token bypasses that, then that might just work. | 12:06 |
| *** mbound has quit IRC | 12:07 | |
| *** mbound has joined #openstack-ironic | 12:07 | |
| TheJulia | dtantsur|brb: gtk getting pulled in seems like deja vu to me, I seem to remember discussing the same issue with lucas-hungry a few weeks ago | 12:08 |
| *** baoli has joined #openstack-ironic | 12:11 | |
| *** ifarkas has quit IRC | 12:18 | |
| *** Marga_ has joined #openstack-ironic | 12:21 | |
| *** chlong has joined #openstack-ironic | 12:22 | |
| *** ifarkas has joined #openstack-ironic | 12:22 | |
| *** daemontool_ has quit IRC | 12:23 | |
| *** Marga_ has quit IRC | 12:26 | |
| *** e0ne has quit IRC | 12:26 | |
| openstackgerrit | Sam Betts proposed openstack/ironic: Add Cisco IMC PXE Driver https://review.openstack.org/219253 | 12:30 |
| *** alexpilo_ has joined #openstack-ironic | 12:30 | |
| *** alexpilotti has quit IRC | 12:31 | |
| *** alexpilo_ has quit IRC | 12:32 | |
| *** alexpilotti has joined #openstack-ironic | 12:33 | |
| *** lucas-hungry is now known as lucasagomes | 12:33 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Check image size before provisioning for agent driver https://review.openstack.org/215254 | 12:38 |
| *** shoutm has quit IRC | 12:39 | |
| *** smoriya has quit IRC | 12:42 | |
| *** daemontool_ has joined #openstack-ironic | 12:42 | |
| openstackgerrit | Anton Arefiev proposed openstack/python-ironicclient: Introduce tempest-lib to functional tests https://review.openstack.org/216701 | 12:46 |
| mordred | TheJulia: yeah - admin_token is a keystoneauth auth plugin that then allows you to get a Session object you can pass in to ironcclient.Client | 12:46 |
| mordred | TheJulia: which is great, because it means we can get rid of the "if auth_type==None" blocks in shade | 12:47 |
| mordred | since the logic will be the same no matter what the config is | 12:47 |
| mordred | in theory | 12:47 |
| mordred | still untested | 12:47 |
| TheJulia | in theory :) | 12:47 |
| TheJulia | I tested the last idea before admin_token and it blew up kind of nicely :) Sounds like it will work and be good in the long run | 12:48 |
| mordred | yeah - the new keystoneauth library is pretty awesome | 12:48 |
| mordred | best mistype of the morning: "git retch origin" | 12:51 |
| *** trown|outttypeww is now known as trown | 12:53 | |
| *** amotoki has quit IRC | 12:54 | |
| *** dtantsur|brb is now known as dtantsur | 12:57 | |
| *** jpeeler has joined #openstack-ironic | 12:59 | |
| *** slagle_ is now known as slagle | 13:03 | |
| *** alexpilotti has quit IRC | 13:04 | |
| *** alexpilotti has joined #openstack-ironic | 13:05 | |
| *** mbound has quit IRC | 13:06 | |
| *** e0ne has joined #openstack-ironic | 13:08 | |
| *** alexpilotti has quit IRC | 13:09 | |
| *** marzif has joined #openstack-ironic | 13:09 | |
| *** zz_natorious is now known as natorious | 13:09 | |
| *** marzif has quit IRC | 13:10 | |
| *** marzif has joined #openstack-ironic | 13:11 | |
| *** mbound has joined #openstack-ironic | 13:13 | |
| *** coreycb` is now known as coreycb | 13:14 | |
| *** rloo has joined #openstack-ironic | 13:14 | |
| *** romainh has quit IRC | 13:15 | |
| *** mbound has quit IRC | 13:15 | |
| openstackgerrit | Merged stackforge/pyghmi: Implement get/set of DCMI asset tag and MCI https://review.openstack.org/220636 | 13:18 |
| *** Marga_ has joined #openstack-ironic | 13:20 | |
| *** alexpilotti has joined #openstack-ironic | 13:20 | |
| openstackgerrit | Juliana Motira proposed stackforge/pyghmi: Add custom OEM get iKVM Key Status for Lenovo ThinkServers https://review.openstack.org/220608 | 13:20 |
| jroll | morning all :) | 13:22 |
| jroll | zigo: I assume the best place is https://github.com/seamicro/python-seamicroclient/issues | 13:22 |
| jroll | zigo: we don't maintain that library, so I'm not really sure, sorry :( | 13:22 |
| *** daemontool_ has quit IRC | 13:22 | |
| zigo | jroll: Thanks. | 13:22 |
| jroll | np | 13:22 |
| *** mbound has joined #openstack-ironic | 13:23 | |
| *** marzif has quit IRC | 13:23 | |
| *** marzif_ has joined #openstack-ironic | 13:23 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: Enable colorized logging in devstack plugin https://review.openstack.org/221298 | 13:24 |
| *** Marga_ has quit IRC | 13:24 | |
| dtantsur | jroll, morning | 13:25 |
| jroll | ohai dtantsur | 13:25 |
| *** alexpilotti has quit IRC | 13:26 | |
| *** marzif_ has quit IRC | 13:26 | |
| *** marzif_ has joined #openstack-ironic | 13:26 | |
| *** alexpilotti has joined #openstack-ironic | 13:27 | |
| *** alexpilotti has quit IRC | 13:27 | |
| jroll | soooo anything blocking us from shipping a release now? | 13:27 |
| *** alexpilotti has joined #openstack-ironic | 13:27 | |
| jroll | probably https://bugs.launchpad.net/ironic/+bug/1493233 hmm | 13:28 |
| openstack | Launchpad bug 1493233 in Ironic "Unit test failures when building 4.0.0 debian package" [Undecided,New] | 13:28 |
| sambetts | morning jroll | 13:28 |
| *** deray has quit IRC | 13:29 | |
| jroll | butbutbut why | 13:29 |
| *** marzif has joined #openstack-ironic | 13:29 | |
| *** marzif_ has quit IRC | 13:29 | |
| rloo | morning jroll, hi dtantsur, sambetts | 13:30 |
| dtantsur | rloo, o/ | 13:30 |
| dtantsur | jroll, metaclass conflict was already fixed | 13:30 |
| jroll | morning rloo :) | 13:30 |
| dtantsur | jroll, the seamicro issue is more weird | 13:30 |
| dtantsur | hell, people still use exceptions.args... | 13:31 |
| *** romainh has joined #openstack-ironic | 13:31 | |
| jroll | it's weird to me that it only fails when packaging? | 13:32 |
| lucasagomes | jroll, zigo morning! | 13:32 |
| jroll | hey lucasagomes :) | 13:32 |
| lucasagomes | jroll, nop, unless you wanna take a look at that cleanwait stuff to be part of the release | 13:32 |
| lucasagomes | if not, I'm good as-is! | 13:33 |
| lucasagomes | and that debian package thing | 13:33 |
| zigo | hi | 13:33 |
| *** alexpilotti has quit IRC | 13:34 | |
| jroll | lucasagomes: yeah, I'll take a look, would like it to be there but 4.2 is also fine for that I think | 13:34 |
| jroll | zigo: looking at doing a release today, could you see if these are still problems on master? (I'm also testing locally) https://bugs.launchpad.net/ironic/+bug/1493233 | 13:36 |
| openstack | Launchpad bug 1493233 in Ironic "Unit test failures when building 4.0.0 debian package" [Undecided,New] | 13:36 |
| *** lazy_prince has quit IRC | 13:36 | |
| *** Marga_ has joined #openstack-ironic | 13:37 | |
| jroll | py27 tests are passing for me :/ | 13:37 |
| *** Marga_ has quit IRC | 13:37 | |
| zigo | https://github.com/seamicro/python-seamicroclient/issues/1 <--- Issue opened. | 13:37 |
| zigo | jroll: Ok, I can try to build now. | 13:37 |
| *** Marga_ has joined #openstack-ironic | 13:37 | |
| jroll | thanks | 13:37 |
| zigo | jroll: Give me a few minutes. | 13:37 |
| jroll | dtantsur: I'm replacing these ex.args things btw | 13:37 |
| jroll | zigo: yeah, no rush :) | 13:38 |
| dtantsur | ack | 13:38 |
| zigo | jroll: Rush for me, I go pick-up my son at school in 30 minutes ! | 13:38 |
| jroll | :) | 13:38 |
| lucasagomes | jroll, fair enuff! | 13:39 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Replace remaining usage of exc.args[0] https://review.openstack.org/221302 | 13:41 |
| jroll | dtantsur: ^ | 13:42 |
| jroll | zigo: ^ I'm hoping that patch will fix the remaining issue, if you still have them, on master | 13:42 |
| jroll | (but tests pass in the gate so I'm not sure why that's causing problems for you to begin with) | 13:42 |
| jroll | (maybe because seamicroclient is already installed rather than being mocked, testing that now) | 13:44 |
| jroll | sho nuff | 13:46 |
| jroll | zigo: you'll need that patch, sorry for the red herring :( | 13:46 |
| zigo | jroll: Which patch? The 221302 ? | 13:46 |
| jroll | zigo: yep, just adding closes-bug now | 13:47 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Replace remaining usage of exc.args[0] https://review.openstack.org/221302 | 13:47 |
| jroll | dtantsur: which patch fixed the metaclass thing? | 13:47 |
| dtantsur | jroll, don't remember, but it was ~1st patch after 4.0.0 | 13:48 |
| jroll | heh | 13:48 |
| *** ifarkas_ has joined #openstack-ironic | 13:49 | |
| *** ifarkas has quit IRC | 13:49 | |
| zigo | I'm currently upgrading ironicclient to 0.8.0 | 13:51 |
| jroll | dtantsur: having trouble finding it :( | 13:52 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-inspector: [WIP] Deprecate returning patches from plugins https://review.openstack.org/220987 | 13:52 |
| *** liliars has joined #openstack-ironic | 13:52 | |
| jroll | dtantsur: https://bugs.launchpad.net/ironic/+bug/1488270 maybe? | 13:54 |
| openstack | Launchpad bug 1488270 in Ironic "oslo.versionedobjects 0.8.0 release causing py27/py34 unittests to fail in ironic" [Critical,Fix committed] - Assigned to Tan Lin (tan-lin-good) | 13:54 |
| jroll | and by proxy https://github.com/openstack/ironic/commit/6186a9de | 13:54 |
| zigo | jroll: Package is building. | 13:54 |
| zigo | :) | 13:55 |
| dtantsur | jroll, yeah, I think so | 13:55 |
| jroll | zigo: cool, though I realized you'll also want https://github.com/openstack/ironic/commit/6186a9de :/ | 13:55 |
| sambetts | dtantsur: just going through my inspector api discovery patch, how do you propose being able to discover the /v1/introspection/<uuid>? | 13:55 |
| jroll | zigo: also, 4.1.0 coming today with both of these patches | 13:55 |
| dtantsur | sambetts, it should be visible in /v1/introspection (which we don't have right now) | 13:56 |
| *** r-daneel has joined #openstack-ironic | 13:57 | |
| zigo | jroll: This last one doesn't apply to top of master. | 13:57 |
| jroll | zigo: oh, it's in master, sorry. I meant if you wanted to package from 4.0.0 | 13:58 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Replace remaining usage of exc.args[0] https://review.openstack.org/221302 | 13:58 |
| jroll | ^ with nits fixed | 13:58 |
| zigo | jroll: From master, I still have 2 unit test failures. | 13:59 |
| zigo | jroll: Let me build again, as I lost the build logs... | 13:59 |
| zigo | (when trying that other patch) | 13:59 |
| jroll | :( | 13:59 |
| *** jcoufal has quit IRC | 14:00 | |
| *** amotoki has joined #openstack-ironic | 14:00 | |
| zigo | jroll: https://paste.debian.net/310883/ | 14:00 |
| zigo | There you are. | 14:00 |
| zigo | Both issues are in test_ipminative.py | 14:00 |
| jroll | zigo: thanks, will check it out | 14:00 |
| zigo | I'm running with python-mock 1.3.0 | 14:01 |
| *** jistr is now known as jistr|call | 14:01 | |
| jroll | what version of pyghmi? | 14:01 |
| zigo | That's latest release ... | 14:01 |
| zigo | jroll: 0.6.17 | 14:02 |
| jroll | thanks | 14:02 |
| zigo | Shall I upgrade it too ? | 14:02 |
| *** alexpilotti has joined #openstack-ironic | 14:02 | |
| jroll | zigo: I think the issue is that in the gate we don't actually install pyghmi, we mock it | 14:02 |
| jroll | (same for seamicro) | 14:02 |
| jroll | apparently our mock isn't up to date with the real world | 14:02 |
| *** Marga_ has quit IRC | 14:03 | |
| zigo | Maybe ... | 14:03 |
| *** marzif_ has joined #openstack-ironic | 14:03 | |
| *** marzif has quit IRC | 14:03 | |
| zigo | I got 10 more minutes,. I can try to upgrade pyghmi | 14:03 |
| jroll | I'm not sure that will help, I should be able to reproduce this locally and get you sorted out | 14:03 |
| jroll | running tests with latest pyghmi now | 14:04 |
| zigo | Upgraded to latest pyghmi, rebuilding ironic now. :) | 14:05 |
| jroll | cool | 14:05 |
| zigo | jroll: It worked ! :) | 14:06 |
| jroll | zigo: ooo, they passed | 14:06 |
| jroll | \o/ | 14:06 |
| zigo | Yup, here as well... | 14:06 |
| zigo | So, I'll do python-pyghmi (>= 0.8.0) in my Ironic package. | 14:07 |
| jroll | ok | 14:07 |
| zigo | I know there's been a controvercy about global-requirements.txt not accepting pyghmi. | 14:07 |
| zigo | Though I wouldn't mind having a file within Ironic itself telling about this kind of things. | 14:07 |
| *** lazy_prince has joined #openstack-ironic | 14:07 | |
| jroll | zigo: so, 4.1.0 coming today with all of these fixed... if you indeed want to package 4.0.0 just need https://github.com/openstack/ironic/commit/6186a9de and https://review.openstack.org/221302 pulled in | 14:08 |
| *** killer_prince has joined #openstack-ironic | 14:08 | |
| jroll | right, so... | 14:08 |
| zigo | Like: package-maintainer-version-hints.txt ! :) | 14:08 |
| jroll | driver-requirements.txt | 14:08 |
| zigo | Oh ! :) | 14:08 |
| zigo | Super nice. | 14:08 |
| jroll | we could probably do better with versions in there, but hey :) | 14:08 |
| zigo | Well, set >= 0.8.0 for pyghmi there then. | 14:08 |
| jroll | yep | 14:09 |
| zigo | I'll make sure to have all of them up-to-date in Debian before I package the last stable then. | 14:09 |
| zigo | Now, time for me to go. | 14:09 |
| zigo | Thanks for all. | 14:09 |
| jroll | thank you sir! | 14:09 |
| zigo | Oh, one more thing: I did ironic-discoverd. | 14:09 |
| zigo | I haven't uploaded, did I? :/ | 14:09 |
| dtantsur | zigo, is it kilo or liberty? for liberty it's better to have ironic-inspector (the same thing renamed) | 14:09 |
| zigo | Oh, thanks for letting me know. | 14:10 |
| zigo | I'll do that one instead then. | 14:10 |
| * zigo loggs off ... | 14:10 | |
| openstackgerrit | Merged openstack/ironic: Remove policy 'admin' rule support https://review.openstack.org/221066 | 14:11 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add version info for pyghmi in driver-requirements.txt https://review.openstack.org/221317 | 14:11 |
| jroll | pins pyghmi | 14:11 |
| *** marzif_ has quit IRC | 14:11 | |
| *** marzif has joined #openstack-ironic | 14:11 | |
| *** Marga_ has joined #openstack-ironic | 14:13 | |
| rloo | hey, is gate failing? that's what it shows on our etherpad | 14:16 |
| *** killer_prince has quit IRC | 14:16 | |
| *** lazy_prince has quit IRC | 14:16 | |
| dtantsur | rloo, it's outdated | 14:17 |
| *** lazy_prince has joined #openstack-ironic | 14:17 | |
| *** killer_prince has joined #openstack-ironic | 14:17 | |
| jroll | lucasagomes: question on https://review.openstack.org/#/c/201552/ | 14:17 |
| * lucasagomes looks | 14:17 | |
| rloo | dtantsur: ? golden now? should I delete the stuff about agent gates failing? | 14:17 |
| dtantsur | rloo, I think so | 14:17 |
| *** Marga_ has quit IRC | 14:17 | |
| rloo | dtantsur: dsvm jobs are still non-voting in think | 14:17 |
| rloo | dtantsur: ok | 14:17 |
| jroll | btw, this makes those voting rloo https://review.openstack.org/#/c/219712/ | 14:18 |
| dtantsur | rloo, yeah, jobs are not voting | 14:18 |
| rloo | jroll: yeah, noted in the etherpad, but infra is ignoring i guess. | 14:18 |
| rloo | we need to check those non-voting before approving. | 14:19 |
| jroll | yep | 14:19 |
| jroll | bugged infra about it | 14:19 |
| rloo | thx jroll | 14:20 |
| *** jistr|call is now known as jistr | 14:20 | |
| lucasagomes | jroll, answered. Lemme know if it makes sense, I will remove that comment too | 14:21 |
| lucasagomes | C&P problems | 14:21 |
| jroll | lucasagomes: comment makes sense to me | 14:22 |
| lucasagomes | jroll, cool. Only the nit then? | 14:23 |
| * lucasagomes fixes it | 14:23 | |
| jroll | yep! | 14:23 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow abort for CLEANWAIT states https://review.openstack.org/201552 | 14:24 |
| lucasagomes | boom! done ^ | 14:24 |
| jroll | thanks! | 14:25 |
| rloo | i've got a question for folks, mostly cores. Are people focussing this week on reviewing patches for features? | 14:25 |
| jroll | you all ok with only running pxe_ssh job in check (voting) but not in gate so we don't break co-gates? | 14:26 |
| *** dims_ has quit IRC | 14:26 | |
| rloo | jroll: yup. if check fails, it won't gate, right? | 14:26 |
| jroll | correct | 14:27 |
| jroll | though check isn't run at +A time | 14:27 |
| *** dimsum__ has joined #openstack-ironic | 14:27 | |
| jroll | just when patches are pushed | 14:27 |
| lucasagomes | jroll, yup, we still running pxe_ipa and agent_ssh in gate right? | 14:27 |
| rloo | jroll: hmm. ok, i'm fine. just means i shouldn't +A now, before jenkins finishes the check. | 14:27 |
| jroll | rloo: AIUI it won't start gate jobs if check fails | 14:28 |
| lucasagomes | yeah it won't | 14:28 |
| lucasagomes | it does check then gate | 14:28 |
| lucasagomes | so... | 14:28 |
| rloo | jroll, lucasagomes: oh, then we're good | 14:28 |
| jroll | the risk is: check passes, +A, rebase for gate, rebase breaks pxe_ssh | 14:28 |
| *** baoli has quit IRC | 14:29 | |
| rloo | jroll: but we won't know if rebase breaks pxe_ssh, or will it be non-voting in gate? | 14:29 |
| jroll | it just won't run at all | 14:29 |
| lucasagomes | so just to be clear the idea is to not run pxe_ssh on other's project gate right? | 14:29 |
| lucasagomes | not ironic itself | 14:29 |
| jroll | lucasagomes: relatedly, we should make a pxe_ipa or agent_ssh postgres job | 14:29 |
| rloo | jroll: why not run & non-vote? | 14:29 |
| lucasagomes | jroll, yeah, most projects dropped postgres, but I still see value on it | 14:30 |
| jroll | lucasagomes: don't run in our gate, because our gate resets devstack/tempest/devstack-gate gates if it fails | 14:30 |
| lucasagomes | oh | 14:30 |
| jroll | rloo: 1) waste of resources, 2) do you ever really go back and see if gate passed? :) | 14:30 |
| rloo | oh. | 14:30 |
| *** baoli_ has joined #openstack-ironic | 14:30 | |
| *** killer_prince has quit IRC | 14:30 | |
| lucasagomes | jroll, what makes me feel safer about it is the fact that we deprecated the endpoints for the bash ramdisk (which pxe_ssh uses) | 14:31 |
| lucasagomes | so I'm ok with the idea | 14:31 |
| jroll | yeah, we'll be removing this job soon anyway | 14:31 |
| lucasagomes | yeah | 14:31 |
| lucasagomes | hopefully | 14:31 |
| lucasagomes | we need the tripleo patches landed | 14:31 |
| jroll | x.x | 14:32 |
| rloo | jroll: yeah, i'm ok. whatever you do, please send out email about it | 14:32 |
| *** vishwanathj has joined #openstack-ironic | 14:32 | |
| jroll | boo email :P | 14:32 |
| jroll | lucasagomes: now that I'm thinking about it... do we want to set last_error for aborted clean? | 14:33 |
| NobodyCam | good morning Ironicers :) | 14:35 |
| jroll | hey hey NobodyCam :) | 14:35 |
| NobodyCam | :) | 14:35 |
| *** jcoufal has joined #openstack-ironic | 14:37 | |
| lucasagomes | jroll, oh it wouldn't be a bad idea | 14:37 |
| * lucasagomes checks | 14:37 | |
| jroll | lucasagomes: yeah, forgot to put my ops hat on when I reviewed that :P | 14:37 |
| *** teju has joined #openstack-ironic | 14:38 | |
| *** Marga_ has joined #openstack-ironic | 14:38 | |
| *** Marga_ has quit IRC | 14:38 | |
| lucasagomes | jroll, yeah, I haven't thought about it. Cause since it's a operator input I thought he would know that he aborted it | 14:38 |
| *** Marga_ has joined #openstack-ironic | 14:38 | |
| *** Marga_ has quit IRC | 14:39 | |
| jroll | lucasagomes: yeah, but think about many operators and if I abort one and leave it | 14:39 |
| lucasagomes | but yeah overall I think it makes sense to always have last_error to indicate what was done | 14:39 |
| lucasagomes | yeah | 14:39 |
| jroll | +1 | 14:39 |
| * jroll changes vote | 14:39 | |
| jroll | also if you could look at https://review.openstack.org/#/c/221302/ when you have a sec | 14:40 |
| lucasagomes | ack | 14:40 |
| jroll | thanks | 14:40 |
| *** Marga_ has joined #openstack-ironic | 14:42 | |
| dtantsur | morning NobodyCam | 14:43 |
| * jroll bbiab | 14:45 | |
| lucasagomes | jroll, I;ve seem a similar patch https://review.openstack.org/#/c/217434/ | 14:46 |
| dtantsur | ifarkas_, sambetts, nobody minds if I release python-ironic-inspector-client? | 14:49 |
| ifarkas_ | dtantsur, not at all, go ahead | 14:49 |
| NobodyCam | morning dtantsur :) | 14:50 |
| NobodyCam | mornign lucasagomes :) | 14:50 |
| *** david-ly_ is now known as david-lyle | 14:53 | |
| *** tsekiyama has joined #openstack-ironic | 14:55 | |
| lucasagomes | NobodyCam, hi there! | 14:56 |
| lucasagomes | morning | 14:56 |
| *** ifarkas_ is now known as ifarkas | 15:00 | |
| *** dlpartain has joined #openstack-ironic | 15:04 | |
| *** lazy_prince has quit IRC | 15:05 | |
| *** teju has quit IRC | 15:08 | |
| *** ijw_ has joined #openstack-ironic | 15:14 | |
| *** ijw_ has quit IRC | 15:15 | |
| *** rameshg87 has joined #openstack-ironic | 15:16 | |
| rameshg87 | rloo: hi | 15:16 |
| rloo | hi rameshg87 | 15:17 |
| rameshg87 | rloo: regarding https://review.openstack.org/198238, I have replied to your comment on how it works in detail. may be I will wait for you to get back to that and then discuss with you when required. | 15:18 |
| rloo | rameshg87: thx. i'm going to look at that patch in the next hour I think. | 15:18 |
| rameshg87 | rloo: if I am around (should be) and you have time, we can discuss. thanks. | 15:18 |
| rloo | rameshg87: ok. this is the last patch for RAID, right? | 15:19 |
| rameshg87 | rloo: yes. | 15:19 |
| rloo | rameshg87: will be good if we can get it in this week :-) | 15:19 |
| rameshg87 | rloo: yeah :) | 15:20 |
| rloo | rameshg87: btw, did you touch base with JoshNang about zapping? not sure i saw any patches up. | 15:20 |
| rameshg87 | rloo: I didn't touch base with JoshNang. I have two days before I go for a small vacation till end of this week. I am quite to happy to help with something until then | 15:21 |
| rameshg87 | rloo: may be we can sort it out today with JoshNang | 15:22 |
| rloo | rameshg87: so you are saying you are around until Thurs this week? | 15:22 |
| rloo | rameshg87: then back next Monday? | 15:22 |
| rameshg87 | rloo: yes | 15:22 |
| *** mgoddard has quit IRC | 15:22 | |
| *** dlpartain has left #openstack-ironic | 15:23 | |
| JoshNang | rameshg87: if you can get the client change (should be small), i should be able to get the api bit up in the next day or two. | 15:23 |
| rloo | rameshg87: we should try to get your patches done by tomorrow then. | 15:23 |
| rloo | JoshNang: we don't need to client changes this week. | 15:23 |
| *** mgoddard has joined #openstack-ironic | 15:23 | |
| rameshg87 | oh yeah | 15:23 |
| JoshNang | ah true | 15:23 |
| rameshg87 | JoshNang: is there anything else I can pick up ? | 15:24 |
| rloo | rameshg87: how many patches of yours are in the features we want? | 15:24 |
| dtantsur | yeah, would be super-cool to have RAID and zapping | 15:24 |
| rloo | rameshg87: if you could review JoshNang's first patch for zapping/cleaning. the one that caches. | 15:24 |
| rameshg87 | rloo: only raid and boot/deploy interface split are the work which I have contributed for this release | 15:25 |
| rloo | dtantsur: if you have time, I think rameshg87's first patch is ready/almost ready. https://review.openstack.org/#/c/196007/24 | 15:25 |
| *** dims_ has joined #openstack-ironic | 15:25 | |
| * dtantsur reviews | 15:25 | |
| rloo | rameshg87: and your code is done wrt boot/deploy, right? | 15:25 |
| rameshg87 | rloo: done for pxe drivers | 15:25 |
| rameshg87 | rloo: stendulker has picked up the virtual media one for ilo. and I have gone through it once. | 15:26 |
| rameshg87 | rloo: I will go through it again today. | 15:26 |
| rloo | rameshg87: ok, let me focus on 198238 then so we can turn it around faster. | 15:26 |
| rameshg87 | rloo: thanks | 15:26 |
| JoshNang | rameshg87: yeah review would be great on https://review.openstack.org/#/c/218532/. and i need to write the api bit to accept zapping steps and the 'zap' action | 15:26 |
| rameshg87 | JoshNang: sure. I will start going through it now. | 15:27 |
| dtantsur | I'm fine with the RAID patch (one tiny thing to follow-up). Anyone else wants to have a look before I press the Big Green Button? lucasagomes? NobodyCam? | 15:28 |
| dtantsur | (patch https://review.openstack.org/#/c/196007/) | 15:28 |
| rloo | dtantsur: you are fast. I'm sure it took me more than a few minutes to review that one :-) | 15:28 |
| *** bradjones has joined #openstack-ironic | 15:28 | |
| *** bradjones has quit IRC | 15:28 | |
| *** bradjones has joined #openstack-ironic | 15:28 | |
| lucasagomes | rloo, re https://review.openstack.org/#/c/208924/ yum is present on older versions of RHEL and fedora | 15:28 |
| lucasagomes | like very old versions | 15:29 |
| dtantsur | rloo, I already started this morning, just forgot to finish :) | 15:29 |
| *** dimsum__ has quit IRC | 15:29 | |
| rloo | lucasagomes: yeah, but I don't think those older versions work with ironic. at least, we don't know. | 15:29 |
| rloo | lucasagomes: or do we? | 15:29 |
| rloo | dtantsur: that explains it! | 15:29 |
| dtantsur | rloo, I only know about RHEL7 (7.1 even) | 15:29 |
| lucasagomes | rloo, yeah maybe we don't have thing packaged for it | 15:29 |
| rloo | dtantsur, lucasagomes: I'd feel safer just leaving the lowest version, what we originally had. | 15:30 |
| lucasagomes | rloo, RHEL7/Fedora 21: ? | 15:30 |
| lucasagomes | ok | 15:30 |
| * lucasagomes will edit soon | 15:30 | |
| rloo | lucasagomes: cuz putting 'or lower' means yum works, but we don't know if those packages work in those OS, even if installed. | 15:31 |
| rloo | lucasagomes: thx | 15:31 |
| lucasagomes | rloo, fair enuff! | 15:32 |
| lucasagomes | thanks | 15:32 |
| rloo | lucasagomes: :) | 15:32 |
| jroll | lucasagomes: oh, you're right, I should abandon my patch (.args[0] stuff) | 15:32 |
| lucasagomes | jroll, yeah, there's a bit of dicussion going on the other | 15:32 |
| lucasagomes | perhaps we can change the commit message to tag the bug there | 15:33 |
| jroll | rloo: idk if I answered your question earlier. yes I'll be focusing on reviewing features esp network stuff | 15:33 |
| lucasagomes | jroll, I'm just unsure about that __unicode__ | 15:33 |
| jroll | lucasagomes: yeah, I'll take a look there | 15:33 |
| lucasagomes | cool | 15:33 |
| *** baoli_ has quit IRC | 15:33 | |
| rloo | jroll: good. if i get time, i'll look at network. looks like I should focus on RAID cuz ramesh is only around til Thurs. | 15:34 |
| *** baoli has joined #openstack-ironic | 15:34 | |
| rloo | jroll: also, wanted to ask/mention. you put priorities in the etherpad, i have been looking at google doc. would be useful to have one place to look at priorities/status. | 15:34 |
| jroll | rloo: oh, maybe should look at raid then heh | 15:34 |
| jroll | rloo: yeah, put them there just to group them together... basically took anything high prio on the google doc | 15:35 |
| rloo | jroll: the API part of the spec is good to go I think. if you want to take a look quickly: https://review.openstack.org/#/c/196007/. | 15:35 |
| rloo | jroll: but we need people to look at network too, there are a lot of patches there. | 15:35 |
| jroll | rloo: yep, trying to look at both today | 15:36 |
| rloo | jroll: ok, i'll look at google doc then. i suppose it would be good to confirm that folks agree with those priorities but i'm fine with them :-) | 15:36 |
| jroll | rloo: agree, those were primarily set by me/deva/chris/other people at midcycle | 15:37 |
| jroll | feel free to raise concerns :) | 15:37 |
| rloo | jroll: I don't have concerns (otherwise I would have raised them already). | 15:37 |
| jroll | rloo: I figured :) | 15:37 |
| rloo | jroll: Just that there wasn't any 'official' communication about what happened in the mid-cycle, outside of the mid-cycle, if you know what i mean. | 15:38 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Refactor IRMCVirtualMediaIscsiDeploy by applying new BootInterface https://review.openstack.org/221371 | 15:38 |
| jroll | rloo: :( I do know what you mean | 15:38 |
| rameshg87 | JoshNang: hi | 15:39 |
| rameshg87 | JoshNang: I was just going through the first patch https://review.openstack.org/#/c/218532/5/ironic/drivers/modules/agent_base_vendor.py | 15:40 |
| JoshNang | o/ | 15:40 |
| rameshg87 | JoshNang: may be you should also take a look at my inband raid configuration patch - https://review.openstack.org/#/c/198238/. we could be actually touching a lot of common code | 15:40 |
| JoshNang | rameshg87: ahh yeah, i remember reviewing this, looks like we are | 15:42 |
| *** devlaps has joined #openstack-ironic | 15:43 | |
| JoshNang | the only major bit seems to be get_clean_steps, which is going to have to change a lot for caching anyway | 15:43 |
| rameshg87 | JoshNang: wondering if we need a similar kind of logic to cache the raid steps as well (I mean steps returned by agent saying it is part of 'raid' interface) | 15:44 |
| dtantsur | can't we land everything simple first, and then get_clean_steps? | 15:44 |
| dtantsur | I pretty much believe that zapping will be very useful even without this API (if we don't manage to land it in L) | 15:45 |
| rameshg87 | dtantsur: an example for simple please :) | 15:45 |
| dtantsur | rameshg87, everything that does not require a solution on caching :) | 15:46 |
| dtantsur | s/simple/straightforward/ is what I meant | 15:46 |
| JoshNang | dtantsur: yeah, definitely | 15:46 |
| *** mgoddard1 has joined #openstack-ironic | 15:46 | |
| dtantsur | having said that, I'm leaving for a day :) | 15:47 |
| JoshNang | o/ | 15:47 |
| dtantsur | I'll review anything RAID/zapping related that will remain unmerged in the morning | 15:47 |
| jroll | night dtantsur :) | 15:47 |
| *** dtantsur is now known as dtantsur|afk | 15:47 | |
| JoshNang | rameshg87: if you have any 'get_raid_levels' logic, yeah caching it would be good | 15:47 |
| JoshNang | (logic that gets pulled from the agent, that is) | 15:48 |
| lucasagomes | jroll, the api to set the last_error on process_event sucks a bit... I will have to spawn a new thread to set it (as callback) | 15:48 |
| lucasagomes | oh or maybe not | 15:48 |
| lucasagomes | wait... I may be able to set last error before calling process event | 15:48 |
| *** mgoddard has quit IRC | 15:48 | |
| jroll | lucasagomes: I mean, callback is ok right? | 15:49 |
| lucasagomes | jroll, yeah but I think it doesn't need it | 15:50 |
| * lucasagomes tests | 15:50 | |
| jroll | lucasagomes: I also think https://review.openstack.org/#/c/217434/ is fine | 15:50 |
| jroll | yeah | 15:50 |
| lucasagomes | I can set last_error and call process_event(), if it succeed it will save() | 15:50 |
| lucasagomes | jroll, right, yeah I was just unsure about that __unicode()__ | 15:51 |
| lucasagomes | jroll, perhaps edit the commit message to tag that bug? | 15:51 |
| jroll | yep, was just going to do that | 15:51 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Try to standardize retrieval of an Exception's description https://review.openstack.org/217434 | 15:52 |
| jroll | whee | 15:52 |
| *** karimb has quit IRC | 15:52 | |
| jroll | once that lands we should be good for 4.1, if this abort patch is close I may just wait for that | 15:53 |
| * zigo uploade python-ironicclient 0.8.0 to Experiental, and will do a -2 release with Python 3 support (going through the FTP master NEW queue...) | 15:53 | |
| zigo | Is there a man page for ironicclient ? | 15:54 |
| *** romainh has left #openstack-ironic | 15:55 | |
| zigo | jroll: ^ | 15:55 |
| jroll | zigo: hmmmm, we have the readme and I think some official docs? | 15:55 |
| zigo | Yeah, but no man page, right? | 15:55 |
| zigo | That's fine, just asking... | 15:55 |
| jroll | no man page afaik | 15:55 |
| zigo | Thanks. | 15:55 |
| jroll | this is probably the closest http://docs.openstack.org/developer/python-ironicclient/cli.html | 15:55 |
| *** VikasC has joined #openstack-ironic | 15:56 | |
| *** r-daneel has quit IRC | 15:58 | |
| *** Marga_ has quit IRC | 16:00 | |
| *** tsekiyama has quit IRC | 16:00 | |
| *** shoutm has joined #openstack-ironic | 16:03 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow abort for CLEANWAIT states https://review.openstack.org/201552 | 16:04 |
| lucasagomes | boom ^ needed the worker thread anyway :-/ | 16:04 |
| lucasagomes | | last_error | Clean operation aborted | | 16:05 |
| jroll | nice thanks | 16:05 |
| jroll | I wonder why just setting it doesn't work | 16:05 |
| *** zhenguo has quit IRC | 16:08 | |
| *** ijw_ has joined #openstack-ironic | 16:10 | |
| *** yog_ has quit IRC | 16:11 | |
| *** jistr has quit IRC | 16:11 | |
| *** shoutm has quit IRC | 16:13 | |
| jroll | lucasagomes: just waiting for jenkins on that exception patch? | 16:13 |
| lucasagomes | jroll, yup | 16:14 |
| jroll | cool, thanks | 16:14 |
| jroll | I can keep an eye on it | 16:14 |
| lucasagomes | tho I'm ok to push the button if jenkins doesn't return until I call it a day | 16:14 |
| lucasagomes | (soon) | 16:14 |
| jroll | yeah, either way, I'll be around | 16:15 |
| *** ijw_ has quit IRC | 16:15 | |
| *** ifarkas has quit IRC | 16:18 | |
| lucasagomes | cool | 16:18 |
| *** Marga_ has joined #openstack-ironic | 16:23 | |
| rloo | rameshg87: if you're still around, I commented on https://review.openstack.org/#/c/198238/ | 16:26 |
| rameshg87 | rloo: just noticed. also just noticed your comment on https://review.openstack.org/#/c/216946/3/ironic/drivers/base.py | 16:27 |
| rameshg87 | rloo: which one goes first ? :) | 16:27 |
| rloo | rameshg87: the one in drivers/base.py is the big question. | 16:29 |
| rloo | rameshg87: i think the rest are minor. | 16:29 |
| rloo | rameshg87: or /big/bigger/, not that big I don't think. | 16:30 |
| rameshg87 | rloo: the line "if not node.target_raid_config, then return", as I explained was added just to make create_configuration as part of cleaning (because I heard people talk about that there is a use-case for it) | 16:31 |
| *** dims_ has quit IRC | 16:31 | |
| rloo | rameshg87: when does the RAID's validate() get called? | 16:31 |
| *** dimsum__ has joined #openstack-ironic | 16:32 | |
| *** achanda has joined #openstack-ironic | 16:32 | |
| rameshg87 | rloo: it has to be called before cleaning | 16:32 |
| rloo | rameshg87: when someone does that validate() API call. I don't know if RAID validate should return true of false there | 16:33 |
| rloo | rameshg87: 'it has to be called before cleaning'. do you know if it is, and where in the code it is being called? | 16:33 |
| *** hurgleburgler1 has joined #openstack-ironic | 16:33 | |
| rloo | rameshg87: it is also possible that we forgot to code it. dunno. | 16:34 |
| rameshg87 | rloo: just checking, a sec. I thought before we start cleaning, we call validate on all the interfaces participating in cleaning. | 16:34 |
| rameshg87 | rloo: looks like I was mistaken. we don't seem to do it. | 16:35 |
| rameshg87 | omg :( | 16:35 |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Add constraint target to tox.ini https://review.openstack.org/220627 | 16:35 |
| rameshg87 | we call only power.validate() | 16:35 |
| rameshg87 | not management.validate() or some-other-interface-that-participate-in-validate.validate() | 16:36 |
| rloo | rameshg87: so that is a bug perhaps. maybe check with JoshNang on that. | 16:36 |
| rameshg87 | JoshNang: ^^^ if you are around ? | 16:36 |
| JoshNang | i don't think calling validate is reasonable: https://github.com/openstack/ironic/blob/master/ironic/drivers/base.py#L224 | 16:37 |
| rloo | rameshg87: it is possible that it is left to the clean step to validate. | 16:37 |
| JoshNang | at least for deploy interface | 16:37 |
| rameshg87 | JoshNang: https://github.com/openstack/ironic/blob/master/ironic/conductor/manager.py#L895-L904 | 16:37 |
| JoshNang | sure, power validate is good, that should be the same whether you're deploying or cleaning or X. management might be ok too | 16:38 |
| rameshg87 | JoshNang: rloo: so, I guess it might make sense to validate every other interface other than deploy | 16:38 |
| rameshg87 | right ? | 16:38 |
| JoshNang | rameshg87: it makes sense, but i'd check all the docstrings, or we may be breaking expectations in our driver interface by calling it during cleaning | 16:39 |
| *** mbound has quit IRC | 16:39 | |
| *** romcheg has quit IRC | 16:39 | |
| *** mdbooth has quit IRC | 16:39 | |
| *** romcheg has joined #openstack-ironic | 16:40 | |
| *** Guest60469 is now known as mgagne | 16:40 | |
| *** mgagne has joined #openstack-ironic | 16:40 | |
| rameshg87 | or let me put it this way | 16:40 |
| *** romcheg has quit IRC | 16:40 | |
| *** alexpilotti has quit IRC | 16:40 | |
| JoshNang | either way, calling raid.validate is fine :) | 16:40 |
| rameshg87 | if there is an interface except deploy participating in cleaning/zapping, it might make sense to do validate() on that interface | 16:40 |
| rameshg87 | because if you are going to do some operations on that interface, you want to be sure that you have everythign required to operate using that interface | 16:41 |
| JoshNang | sounds good to me | 16:41 |
| rameshg87 | rloo: ^^ | 16:41 |
| rloo | rameshg87: yes, I think that makes sense. Now, what if someone wants to add a clean step to delete a RAID config, but they plan on manually create a RAID config somehow. does RAID.validate() return true or false? | 16:42 |
| *** derekh has quit IRC | 16:42 | |
| rloo | rameshg87: I mean, should a target_raid_config be needed in raid.validate()? | 16:43 |
| rloo | rameshg87: also, when do you think the validate() should be done. Right before invoking that interface's clean step, or calling all .validate's before doing any clean step? | 16:43 |
| *** davideagnello has joined #openstack-ironic | 16:44 | |
| JoshNang | rloo: i'd do it right by that power.validate() block before cleaning starts...if those fail, the api can return a failure to the set-provision-state request, rather than saying 'ok starting cleaning' and then failing the node later (possibly right away) | 16:45 |
| rameshg87 | rloo: I think validate() should be called before doing any clean step, just like we do https://github.com/openstack/ironic/blob/master/ironic/conductor/manager.py#L895-L904 | 16:45 |
| *** amotoki has quit IRC | 16:46 | |
| JoshNang | heh ++ | 16:46 |
| rloo | JoshNang: I don't think we can do it when we do power.validate(). At that point, we may not know what the clean steps are. | 16:46 |
| rloo | JoshNang: but yeah, we can do it before we start any clean step. | 16:46 |
| rameshg87 | rloo: that is our strategy most of the times to catch errors at the earliest | 16:46 |
| *** alexpilotti has joined #openstack-ironic | 16:47 | |
| rloo | rameshg87, JoshNang: so before line 932: https://github.com/openstack/ironic/blob/master/ironic/conductor/manager.py#L932 | 16:47 |
| rloo | rameshg87, JoshNang: I'm assuming the call to set_node_cleaning_steps() is when we know what the steps are. | 16:47 |
| JoshNang | correct....if you don't have to boot the ramdisk, otherwise it quits here: https://github.com/openstack/ironic/blob/master/ironic/conductor/manager.py#L929 | 16:48 |
| rameshg87 | JoshNang: I guess it's other way round. right ? | 16:48 |
| rameshg87 | if you have to boot the ramdisk, it quits there | 16:49 |
| rameshg87 | ?? | 16:49 |
| JoshNang | right, that's what i meant | 16:49 |
| rameshg87 | okay :) | 16:49 |
| *** tsekiyama has joined #openstack-ironic | 16:50 | |
| rameshg87 | rloo: okay, coming back to your first question. given this, we would end up calling raid.validate() even if delete was what operator wanted. | 16:51 |
| *** alexpilotti has quit IRC | 16:51 | |
| rameshg87 | rloo: and we will end up throwing an error if we tried to validate node.target_raid_config. right ? | 16:51 |
| rloo | rameshg87: I need to think about that. I wonder/don't know if we want to support the case of ONLY deleting, not creating. | 16:52 |
| rloo | rameshg87: the RAID interface is for creating configs, so that seems like a corner case. | 16:52 |
| rloo | rameshg87: I am fine now, leaving the code in validate as-is. But I'm not sure it is right. but if it isn't, it is a bug. And we need to focus on getting features in. | 16:53 |
| rloo | rameshg87: don't know if it is a good analogy or not, but when we have a deploy interface, we assume we're deploying and cleaning up after that. we don't assume that the user wants to somehow manually deploy, but have ironic clean up. | 16:54 |
| rameshg87 | rloo: agreed, it seems same. | 16:55 |
| rloo | rameshg87: i was being a bit of a devil's advocate before. although who knows, maybe there is a usecase for it. dunno. | 16:55 |
| rameshg87 | rloo: but I agree with your point. we need to help people by throwing some real errors if they forgot to put node.target_raid_config | 16:55 |
| rloo | rameshg87: yes. so if eg someone calls the validate API, I think it makes sense to return False there, for RAID. | 16:56 |
| *** achanda has quit IRC | 16:56 | |
| *** achanda_ has joined #openstack-ironic | 16:56 | |
| *** e0ne has quit IRC | 16:56 | |
| rloo | rameshg87: but anyway -- sorry, maybe it wasn't that big an issue. do you have time to look at my comments from https://review.openstack.org/#/c/198238/ | 16:57 |
| rameshg87 | rloo: just looking. | 16:57 |
| rameshg87 | rloo: will try to reply now only, and may be wait if you are free. | 16:58 |
| rloo | rameshg87: i'm 'free' | 16:58 |
| rameshg87 | okay :) | 16:58 |
| rloo | rameshg87: but did want to get some lunch at some point. | 16:58 |
| *** tsekiyama has quit IRC | 17:00 | |
| *** alexpilotti has joined #openstack-ironic | 17:01 | |
| *** mgoddard1 has quit IRC | 17:02 | |
| *** harlowja has joined #openstack-ironic | 17:03 | |
| *** mgoddard has joined #openstack-ironic | 17:03 | |
| rloo | rameshg87: most of those comments are minor. I think the only thing confusing me now is when/when not to use those clean_step decorators. | 17:03 |
| rameshg87 | rloo: okay, may be I will try to tell my point about that first | 17:03 |
| rameshg87 | rloo: agent ramdisk may or may not have a hardware manager that can do raid configuration | 17:04 |
| rameshg87 | rloo: if we would have decorated that method, we would be running the steps everytime if we have a non-zero priority | 17:05 |
| rameshg87 | rloo: but instead, we wanted to run those methods only if agent ramdisk reported that it is capable of doing so. | 17:05 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Update docs for Fedora 22 https://review.openstack.org/208924 | 17:05 |
| lucasagomes | aigh folks, I'm going to call it a day | 17:05 |
| rameshg87 | rloo: that place is the get_clean_steps() method | 17:05 |
| rameshg87 | lucasagomes: g'night | 17:05 |
| lucasagomes | have a good evening everyone! | 17:05 |
| *** trown is now known as trown|lunch | 17:05 | |
| betherly | have a good evening lucasagomes | 17:05 |
| rloo | rameshg87 & JoshNang: do you both have the same understanding wrt the clean_step decorator? ^^ | 17:06 |
| lucasagomes | you too (-: see y'all | 17:06 |
| rloo | night lucasagomes | 17:06 |
| JoshNang | yup! | 17:06 |
| *** lucasagomes is now known as lucas-dinner | 17:06 | |
| *** ukalifon has quit IRC | 17:06 | |
| *** Marga_ has quit IRC | 17:07 | |
| rloo | JoshNang, rameshg87: Ok. I'll revisit that, the docstrings, etc. for the decorator, because it seems misleading. Ie, the create_configuration and delete_configuration are clean steps, so I would have thought they'd need the decorator. | 17:07 |
| *** Marga_ has joined #openstack-ironic | 17:07 | |
| rloo | JoshNang, rameshg87: cuz the decorator is called 'clean_step' :) | 17:07 |
| rameshg87 | rloo: okay | 17:08 |
| *** achanda_ has quit IRC | 17:08 | |
| rameshg87 | rloo: but that's why I thought off a logging | 17:08 |
| rameshg87 | rloo: https://review.openstack.org/#/c/198238/18/ironic/drivers/modules/agent.py - L518 | 17:08 |
| *** Marga_ has quit IRC | 17:08 | |
| rameshg87 | rloo: does that seem an odd place ? | 17:08 |
| *** Marga_ has joined #openstack-ironic | 17:09 | |
| *** athomas has quit IRC | 17:09 | |
| rloo | rameshg87: but that's an odd place. the method is called get_clean_steps(). | 17:09 |
| rameshg87 | rloo: because if someone set node.target_raid_config and then we figured out that agent ramdisk cannot support it, we need to inform the operator somehow | 17:09 |
| rloo | rameshg87: i don't really see why we need to inform that operator about that. | 17:10 |
| *** ijw_ has joined #openstack-ironic | 17:10 | |
| rameshg87 | rloo: hmm..but it could be someone thought their hardware supports raid configuration, only to find out agent ramdisk couldn't support it | 17:10 |
| rameshg87 | but then I am assuming operator doesn't know their hardware :D | 17:11 |
| rloo | rameshg87: then they use the API-that-doesn't-exist-yet to see what the clean steps are | 17:11 |
| rameshg87 | oh, it could also be because they don't have the right hardware manager | 17:11 |
| *** Marga_ has quit IRC | 17:11 | |
| rameshg87 | oh yeah, they can .. | 17:11 |
| rameshg87 | may be we need to document it | 17:11 |
| rloo | rameshg87: yup. maybe after we have that API... | 17:12 |
| rameshg87 | for inband raid configuration, check if your bare metal supports raid configuration, by looking at clean steps | 17:12 |
| rameshg87 | I will remove the log then | 17:12 |
| rloo | thx rameshg87 | 17:12 |
| rloo | rameshg87: i think the rest of my comments, you can comment on. i'm going to disappear for a bit. starving... | 17:13 |
| rameshg87 | rloo: sure, will see if I can address comments straight away | 17:13 |
| rameshg87 | rloo: thanks a lot for quick reviews | 17:13 |
| rloo | rameshg87: wow! don't stay up just cuz of this. but thanks! | 17:13 |
| *** tsekiyama has joined #openstack-ironic | 17:14 | |
| *** Marga_ has joined #openstack-ironic | 17:15 | |
| rameshg87 | rloo: just a quick thing | 17:15 |
| rameshg87 | rloo: https://review.openstack.org/#/c/198238/18/ironic/drivers/modules/deploy_utils.py L1272 | 17:15 |
| rameshg87 | rloo: do you have suggestion for another name ? | 17:15 |
| rameshg87 | if that doesn't convey the meaning | 17:15 |
| jlvillal | rameshg87: 'reset'? | 17:18 |
| jlvillal | set? | 17:19 |
| rameshg87 | jlvillal: or does 'change' look simpler | 17:19 |
| rameshg87 | jlvillal: actually I noticed that she suggested something in that comment itself. I missed it. | 17:19 |
| rameshg87 | jlvillal: does 'change' work ? | 17:19 |
| jlvillal | rameshg87: change would work for me. Or set. | 17:19 |
| jlvillal | rameshg87: change has an implication that it already exists. set does not imply that, but it could exist. | 17:20 |
| rameshg87 | jlvillal: okay, so set seems bettter. because it already exists when it is returned from agent ramdisk. | 17:20 |
| rameshg87 | I mean change seems better | 17:20 |
| * rameshg87 scratches head | 17:21 | |
| *** achanda has joined #openstack-ironic | 17:21 | |
| rameshg87 | right ? | 17:21 |
| jlvillal | rameshg87: You say tomato I say tomato ;) | 17:21 |
| jlvillal | rameshg87: change works for me! :) | 17:21 |
| jlvillal | Either works. | 17:21 |
| rameshg87 | :D | 17:21 |
| rameshg87 | thanks | 17:21 |
| jlvillal | rameshg87: But I are an engineer ;) | 17:22 |
| rameshg87 | jlvillal: I think I am :) | 17:22 |
| *** tsekiyama has quit IRC | 17:24 | |
| openstackgerrit | Sam Betts proposed openstack/ironic-inspector: Add API Discovery to Ironic Inspector https://review.openstack.org/215067 | 17:26 |
| *** ukalifon1 has joined #openstack-ironic | 17:28 | |
| *** lsmola has quit IRC | 17:29 | |
| *** pelix has quit IRC | 17:33 | |
| *** pelix has joined #openstack-ironic | 17:34 | |
| *** praneshp has joined #openstack-ironic | 17:35 | |
| *** saripurigopi has joined #openstack-ironic | 17:37 | |
| jlvillal | Anyone here using tftpd-hpa? As our documentation says to use. | 17:46 |
| jlvillal | I'm looking at: http://docs.openstack.org/developer/ironic/deploy/install-guide.html | 17:47 |
| openstackgerrit | Merged openstack/ironic: Add supported environment 'VMware' to comments https://review.openstack.org/220880 | 17:47 |
| *** e0ne has joined #openstack-ironic | 17:47 | |
| jroll | we use xinet whatever | 17:48 |
| jlvillal | jroll: Okay. I was looking at the part where it says: Enable tftp map file, modify /etc/xinetd.d/tftp as below and restart xinetd service | 17:48 |
| jlvillal | But it seems like tftpd-hpa doesn't use xinetd, but has a /etc/init.d/tftpd file | 17:49 |
| *** priteau has quit IRC | 17:49 | |
| jroll | jlvillal: that said, we manage tftp outside of ironic so I'm not sure I'll be much help :P | 17:49 |
| jlvillal | jroll: Thanks, trying to set up a real bare metal setup inside my cube :) | 17:49 |
| jlvillal | I've only done VM method until now. | 17:49 |
| jroll | heh | 17:49 |
| jroll | yeah docs could be wrong on this | 17:49 |
| jlvillal | jroll: Thanks. | 17:49 |
| jroll | np, idk if that helped :) | 17:50 |
| *** aarefiev22 has joined #openstack-ironic | 17:50 | |
| TheJulia | jlvillal: I think it can be launched either way, just depends on distribution default most likely | 17:51 |
| jlvillal | TheJulia: Thanks. I'm playing on Ubuntu 14.04 | 17:51 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add support for inband raid configuration agent ramdisk https://review.openstack.org/198238 | 17:53 |
| TheJulia | jlvillal: same here, I'm using xinetd, if you want I can show you what my xinetd.d/tftp file looks like | 17:53 |
| *** alexpilotti has quit IRC | 17:53 | |
| jlvillal | TheJulia: Thanks! Did you use tftpd or tftpd-hpa? | 17:54 |
| jlvillal | TheJulia: Yes I would like to see :) | 17:54 |
| *** trown|lunch is now known as trown | 17:54 | |
| jlvillal | I am thinking about modifying /etc/defaults/tftpd-hpa file to add options. | 17:54 |
| TheJulia | I'm fairly sure its hpa, one moment | 17:55 |
| jlvillal | And using the /etc/init.d/tftpd-hpa file | 17:55 |
| jlvillal | Instead of doing xinetd. But my opinion can change :) | 17:55 |
| * rameshg87 goes to sleep | 17:56 | |
| TheJulia | jlvillal: tftp-hpa with xinetd | 17:57 |
| rameshg87 | good night all | 17:57 |
| *** rameshg87 has quit IRC | 17:57 | |
| jlvillal | TheJulia: Thanks | 17:58 |
| TheJulia | jlvillal: heh, looks like I've also got the script, but xinetd is getting the requests and passing it over on my dev machine | 17:58 |
| jlvillal | TheJulia: But would like to see your xinetd file | 17:58 |
| *** baoli has quit IRC | 18:00 | |
| TheJulia | jlvillal: sure, one moment | 18:00 |
| jlvillal | TheJulia: Thanks! | 18:00 |
| *** baoli has joined #openstack-ironic | 18:01 | |
| *** r-daneel has joined #openstack-ironic | 18:02 | |
| TheJulia | jlvillal: http://paste.openstack.org/show/3oprAImjo0CpPOIvgx5K/ I have this feeling like user is totally wrong, and I changed it and it failed to work with a quick test, but now I'm going to keep digging at it since it just seems very wrong to me. | 18:05 |
| jlvillal | TheJulia: Thank you. | 18:05 |
| *** aarefiev22 has quit IRC | 18:06 | |
| jlvillal | TheJulia: This is what I tried: http://paste.openstack.org/show/450612/ | 18:07 |
| jlvillal | So my data is going into /var/lib/tftpboot/ | 18:07 |
| jlvillal | I ended up making my map-file a NOOP for now. And I could do a 'tftp' client and get my pxelinux.0 file | 18:07 |
| jlvillal | I did a: service tftpd-hpa start | 18:08 |
| *** saripurigopi has quit IRC | 18:08 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Remove explicit ELEMENTS_PATH definition https://review.openstack.org/221436 | 18:08 |
| TheJulia | jlvillal: awesome :) | 18:12 |
| jlvillal | TheJulia: Thanks, but a lot more to go. Now to see what I else I need to do before I setup my NUC :) | 18:12 |
| *** alexpilotti has joined #openstack-ironic | 18:14 | |
| *** alexpilotti has quit IRC | 18:18 | |
| *** e0ne has quit IRC | 18:29 | |
| openstackgerrit | Jarrod Johnson proposed stackforge/pyghmi: Implement NTP support for Lenovo Thinkserver https://review.openstack.org/221448 | 18:29 |
| *** openstackgerrit has quit IRC | 18:31 | |
| *** openstackgerrit has joined #openstack-ironic | 18:32 | |
| *** Nisha has joined #openstack-ironic | 18:39 | |
| *** achanda has quit IRC | 18:40 | |
| *** achanda has joined #openstack-ironic | 18:41 | |
| betherly | heading off for the day. have a good day all! | 18:42 |
| *** baoli has quit IRC | 18:43 | |
| *** romcheg has joined #openstack-ironic | 18:44 | |
| *** alexpilotti has joined #openstack-ironic | 18:45 | |
| *** garthb has joined #openstack-ironic | 18:49 | |
| openstackgerrit | Jarrod Johnson proposed stackforge/pyghmi: Implement NTP support for Lenovo Thinkserver https://review.openstack.org/221448 | 18:50 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add support for inband raid configuration agent ramdisk https://review.openstack.org/198238 | 18:50 |
| *** alexpilo_ has joined #openstack-ironic | 18:52 | |
| *** alexpilotti has quit IRC | 18:52 | |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Move Redhat-specific libvirt tasks into file to be included https://review.openstack.org/221453 | 18:54 |
| *** pelix has quit IRC | 18:56 | |
| *** achanda has quit IRC | 18:57 | |
| *** derekh has joined #openstack-ironic | 19:00 | |
| *** ukalifon1 has quit IRC | 19:03 | |
| *** alexpilo_ has quit IRC | 19:03 | |
| *** alexpilotti has joined #openstack-ironic | 19:03 | |
| *** Marga_ has quit IRC | 19:06 | |
| *** baoli has joined #openstack-ironic | 19:09 | |
| *** baoli has quit IRC | 19:15 | |
| *** baoli has joined #openstack-ironic | 19:16 | |
| *** adam_g` is now known as adam_g | 19:27 | |
| *** adam_g has quit IRC | 19:28 | |
| *** adam_g has joined #openstack-ironic | 19:28 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Ensure all methods in util.py have docstrings. https://review.openstack.org/220313 | 19:32 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Revise TFTP settings and note configuration https://review.openstack.org/221476 | 19:38 |
| *** achanda has joined #openstack-ironic | 19:38 | |
| *** alexpilotti has quit IRC | 19:39 | |
| *** alexpilotti has joined #openstack-ironic | 19:39 | |
| *** achanda has quit IRC | 19:45 | |
| *** derekh is now known as derekh_afk | 19:58 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Correct log file storage location https://review.openstack.org/218841 | 20:01 |
| *** dimsum__ is now known as dims | 20:03 | |
| openstackgerrit | Merged openstack/ironic: Try to standardize retrieval of an Exception's description https://review.openstack.org/217434 | 20:03 |
| *** dims is now known as Guest80071 | 20:04 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Remove explicit ELEMENTS_PATH definition https://review.openstack.org/221436 | 20:08 |
| *** Guest80071 is now known as dims__ | 20:08 | |
| openstackgerrit | Merged openstack/bifrost: Fix doc typo for configdrive role https://review.openstack.org/220856 | 20:11 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Make the erase_devices clean step abortable https://review.openstack.org/202137 | 20:15 |
| JayF | ^ fixed nits in the commit message and landed it, it already had a +2 from dmitry | 20:15 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Remove outdated TODO file https://review.openstack.org/221490 | 20:23 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Improve logging for VMs to identify IPA issues https://review.openstack.org/220112 | 20:24 |
| openstackgerrit | Jarrod Johnson proposed stackforge/pyghmi: Implement NTP support for Lenovo Thinkserver https://review.openstack.org/221448 | 20:27 |
| openstackgerrit | Merged openstack/bifrost: Correct log file storage location https://review.openstack.org/218841 | 20:38 |
| *** ekarlso has quit IRC | 20:38 | |
| *** achanda has joined #openstack-ironic | 20:41 | |
| *** alexpilo_ has joined #openstack-ironic | 20:44 | |
| *** alexpilotti has quit IRC | 20:44 | |
| *** alexpilotti has joined #openstack-ironic | 20:45 | |
| *** alexpilo_ has quit IRC | 20:45 | |
| *** Marga_ has joined #openstack-ironic | 20:46 | |
| *** derekh_afk has quit IRC | 20:48 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-python-agent: Tuples returned by clean steps cause API error https://review.openstack.org/221498 | 20:51 |
| *** achanda has quit IRC | 20:51 | |
| *** achanda has joined #openstack-ironic | 20:54 | |
| *** ekarlso has joined #openstack-ironic | 20:57 | |
| *** ekarlso has quit IRC | 20:57 | |
| *** ekarlso has joined #openstack-ironic | 20:57 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Fix HACKING.rst and canary test for merges https://review.openstack.org/220113 | 20:58 |
| *** gmmaha has joined #openstack-ironic | 20:59 | |
| *** gmmaha has quit IRC | 20:59 | |
| *** Nisha has quit IRC | 20:59 | |
| *** gmmaha has joined #openstack-ironic | 21:00 | |
| *** harshs has joined #openstack-ironic | 21:04 | |
| *** harshs has left #openstack-ironic | 21:04 | |
| jroll | everybody cool with releasing ironic 4.1.0 now that this landed? https://github.com/openstack/ironic/commit/fe2c6207f3ca385e9904d591d064232beb3514f1 | 21:06 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Improve logging for VMs to identify IPA issues https://review.openstack.org/220112 | 21:06 |
| *** romcheg has quit IRC | 21:07 | |
| *** karimb has joined #openstack-ironic | 21:07 | |
| *** harshs has joined #openstack-ironic | 21:07 | |
| *** gmmaha has quit IRC | 21:07 | |
| rloo | jroll: nooooooo. Yeah, go for it! | 21:07 |
| jroll | :) | 21:08 |
| *** gmmaha has joined #openstack-ironic | 21:08 | |
| NobodyCam | :) | 21:08 |
| jroll | rloo: https://review.openstack.org/#/c/221503/ | 21:08 |
| *** marzif has quit IRC | 21:09 | |
| *** marzif has joined #openstack-ironic | 21:09 | |
| *** romcheg has joined #openstack-ironic | 21:09 | |
| rloo | jroll: do we need to update the release notes? | 21:09 |
| *** trown is now known as trown|outttypeww | 21:09 | |
| jroll | rloo: good point | 21:09 |
| jroll | I can do that after cross project mtg | 21:09 |
| rloo | jroll: ok. | 21:10 |
| jroll | that's in doc/ so I expect it doesn't need to be in the actual release | 21:10 |
| * jroll starts drafting it | 21:11 | |
| rloo | jroll: oh, i just -1'd your patch cuz of that. i think we do want it in tree cuz it is packaged with the release. or at least, deva wanted the release notes in before cutting 4.0.0 i think. | 21:11 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Fix HACKING.rst and canary test for merges https://review.openstack.org/220113 | 21:12 |
| jroll | I seem to remember them landing after 4.0.0 | 21:12 |
| jroll | but not sure | 21:12 |
| *** baoli has quit IRC | 21:12 | |
| jroll | 6 days ago, definitely was after 4.0 | 21:12 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Fix HACKING.rst and canary test for merges https://review.openstack.org/220113 | 21:12 |
| JayF | I mean, why shouldn't we write release notes and land them real quick? | 21:12 |
| JayF | If we're releasing shouldn't we know what we're releasing? | 21:13 |
| rloo | jroll: yeah, which is why deva didn't really want to announce 4.0.0. | 21:13 |
| jroll | JayF: define "land them real quick" :) | 21:13 |
| jroll | oh. well. | 21:13 |
| JayF | jroll: Heh. I'm just saying, shouldn't the tools exist to answer the question "what's the change?" | 21:13 |
| jroll | I'm fine with either way, probably notes before release is better | 21:13 |
| JayF | jroll: and putting that into release notes seems trivial | 21:13 |
| rloo | JayF: ha ha. I suspect it isn't 'real quick' cuz it means you have to 1. know what changed; 2. update the docn; 3. submit patch; 4. get it merged | 21:13 |
| jroll | JayF: yeah, but then making them english :) | 21:13 |
| jroll | 3.a. have rloo correct all the english | 21:14 |
| rloo | jroll: i'm fine if someone else does 3.a :D | 21:14 |
| rloo | jroll: and happier if i don't even have to look at the patch! :D | 21:14 |
| jroll | heh | 21:15 |
| rloo | jroll: did you want this in for 4.1.0 too? (I know, it is documentation) https://review.openstack.org/#/c/221317/ | 21:18 |
| jroll | rloo: I'm apathetic | 21:18 |
| jroll | how is that merge conflict | 21:18 |
| jroll | oh because abandoned patch, ugh | 21:19 |
| rloo | jroll: the downside of making a patch dependent on another patch | 21:19 |
| jroll | yeah | 21:19 |
| jroll | I'm not worried about that in 4.1 | 21:19 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Allow Test VM RAM setting to be tunable https://review.openstack.org/221507 | 21:20 |
| *** thrash is now known as thrash|g0ne | 21:33 | |
| jroll | ok, I think this is the list of commits since 4.0 https://gist.github.com/jimrollenhagen/13b25f0963448d4418b5 | 21:34 |
| jroll | going to go off of that for release notes | 21:34 |
| *** achanda has quit IRC | 21:35 | |
| *** achanda has joined #openstack-ironic | 21:35 | |
| *** romcheg has quit IRC | 21:35 | |
| rloo | jroll: looks good to me. I'd like to see 'Add release notes for 4.1.0' on top :) | 21:36 |
| jroll | rloo: heh yeah | 21:37 |
| jroll | I don't think "adds release notes" belongs in the release notes :P | 21:37 |
| rloo | jroll: i just approved a couple of patches i think. but they weren't important/release-news-worthy I don't think. | 21:37 |
| rloo | jroll: Agree, just in the list of changes though :) | 21:38 |
| jroll | heh, yeah | 21:38 |
| *** devlaps has quit IRC | 21:38 | |
| mrda | Morning Ironic | 21:38 |
| NobodyCam | morning mrda | 21:40 |
| mrda | o/ | 21:40 |
| *** achanda_ has joined #openstack-ironic | 21:44 | |
| *** achanda has quit IRC | 21:44 | |
| jroll | hiya mrda | 21:44 |
| mrda | hey jroll | 21:45 |
| jlvillal | mrda: Good morning | 21:47 |
| mrda | hey jlvillal | 21:47 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Fix HACKING.rst and canary test for merges https://review.openstack.org/220113 | 21:49 |
| *** priteau has joined #openstack-ironic | 21:49 | |
| *** jamielennox|away is now known as jamielennox | 21:51 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add 4.1.0 release notes https://review.openstack.org/221517 | 21:51 |
| * jlvillal imagines circular dependency of 'add release notes about adding release note about adding release note...' :) | 21:51 | |
| jroll | rloo and friends ^ | 21:51 |
| rloo | whee, /me is a friend of jroll. | 21:52 |
| openstackgerrit | Merged openstack/bifrost: Move Redhat-specific libvirt tasks into file to be included https://review.openstack.org/221453 | 21:52 |
| jroll | rloo: idk, I listed you separately from 'friends' ;) | 21:52 |
| * jlvillal is unsure of his status ;) | 21:52 | |
| openstackgerrit | Merged openstack/bifrost: Revise TFTP settings and note configuration https://review.openstack.org/221476 | 21:52 |
| rloo | jroll: yeah, I wondered about that :) | 21:52 |
| jroll | we can still be friends, I already did my full-channel ping for the day :P | 21:53 |
| *** priteau has quit IRC | 21:54 | |
| mrda | :) | 21:54 |
| rloo | I think jroll's friends are anyone that reviews his patch :D | 21:55 |
| jroll | true story | 21:56 |
| mrda | or buys him breakfast burritos or beer | 21:56 |
| *** lucas-dinner has quit IRC | 21:56 | |
| rloo | jroll: i have to take off. i'll check in later, maybe 2 hours or so, if you need more +2s or whatever. | 21:56 |
| jroll | rloo: cool, ty | 21:56 |
| *** rloo is now known as rloo_afk | 21:57 | |
| *** achanda_ has quit IRC | 21:58 | |
| *** achanda has joined #openstack-ironic | 22:02 | |
| *** derekh has joined #openstack-ironic | 22:03 | |
| *** Marga_ has quit IRC | 22:07 | |
| *** Marga_ has joined #openstack-ironic | 22:07 | |
| *** alexpilotti has quit IRC | 22:08 | |
| *** alexpilotti has joined #openstack-ironic | 22:08 | |
| *** achanda has quit IRC | 22:09 | |
| *** achanda has joined #openstack-ironic | 22:09 | |
| *** achanda_ has joined #openstack-ironic | 22:13 | |
| *** achanda has quit IRC | 22:13 | |
| openstackgerrit | Merged openstack/bifrost: Improve logging for VMs to identify IPA issues https://review.openstack.org/220112 | 22:15 |
| openstackgerrit | Merged openstack/ironic: Handle missing is_whole_disk_image in pxe._build_pxe_config_options https://review.openstack.org/221003 | 22:21 |
| openstackgerrit | Merged openstack/ironic: Improve the ability to resolve capability value https://review.openstack.org/190900 | 22:21 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Add 4.1.0 release notes https://review.openstack.org/221517 | 22:22 |
| jroll | mrda: thanks for the nit | 22:22 |
| jroll | rloo_afk: ^ updated | 22:22 |
| jroll | NobodyCam: if you could also look at that, would be fantastic | 22:22 |
| NobodyCam | looks | 22:22 |
| *** derekh has quit IRC | 22:24 | |
| *** r-daneel has quit IRC | 22:26 | |
| *** alex_xu has quit IRC | 22:28 | |
| *** alex_xu has joined #openstack-ironic | 22:30 | |
| NobodyCam | ahh Deprecated the 'parallel' option to periodic task decorator got me .. we've only added a deprecation warning for parallel=False | 22:32 |
| NobodyCam | :p | 22:32 |
| *** Sukhdev_ has joined #openstack-ironic | 22:36 | |
| jroll | NobodyCam: because True is the default, no? | 22:46 |
| jroll | the point is to not make that optional :P | 22:47 |
| *** Marga__ has joined #openstack-ironic | 22:49 | |
| *** krtaylor has quit IRC | 22:49 | |
| *** Marga_ has quit IRC | 22:51 | |
| *** karimb has quit IRC | 22:53 | |
| *** achanda has joined #openstack-ironic | 22:55 | |
| *** achanda_ has quit IRC | 22:55 | |
| *** achanda_ has joined #openstack-ironic | 23:01 | |
| *** achanda has quit IRC | 23:01 | |
| *** Sukhdev_ has quit IRC | 23:17 | |
| *** Sukhdev has quit IRC | 23:19 | |
| *** Sukhdev has joined #openstack-ironic | 23:19 | |
| *** alexpilotti has quit IRC | 23:21 | |
| *** alexpilotti has joined #openstack-ironic | 23:22 | |
| *** achanda has joined #openstack-ironic | 23:23 | |
| *** achanda_ has quit IRC | 23:23 | |
| *** chlong has quit IRC | 23:27 | |
| *** chlong has joined #openstack-ironic | 23:28 | |
| jlvillal | bnemec: ping? | 23:29 |
| jlvillal | bnemec: unping :) | 23:32 |
| *** VikasC has quit IRC | 23:35 | |
| *** Sukhdev has quit IRC | 23:38 | |
| *** ijw_ has quit IRC | 23:40 | |
| *** Sukhdev has joined #openstack-ironic | 23:40 | |
| mrda | hey jroll? | 23:47 |
| mrda | Or lucasagomes? | 23:48 |
| jroll | mrda: kinda here, what's up | 23:48 |
| mrda | https://review.openstack.org/#/c/206614/ and https://review.openstack.org/#/c/209457 are two reviews for bug https://bugs.launchpad.net/nova/+bug/1477490 | 23:48 |
| openstack | Launchpad bug 1477490 in OpenStack Compute (nova) "Ironic: Deleting while spawning can leave orphan ACTIVE nodes in Ironic" [Undecided,In progress] - Assigned to Lucas Alvares Gomes (lucasagomes) | 23:48 |
| mrda | Is the first one superceeded? | 23:48 |
| *** shoutm has joined #openstack-ironic | 23:49 | |
| mrda | (jroll: sorry for bugging you late in your day) | 23:49 |
| jroll | mrda: I don't think so, I think they're both valid | 23:49 |
| jroll | but not totally sure | 23:49 |
| jroll | oh you're fine, I made the choice to answer :P | 23:49 |
| mrda | ok, it just seems #1 is now blocked for a spec, and #2 isn't, so I'm confused | 23:50 |
| mrda | but that's not your concern :) | 23:50 |
| mrda | I'm often confused | 23:50 |
| jroll | heh | 23:50 |
| jroll | they're both needed to close the bug, afai | 23:50 |
| jroll | k | 23:50 |
| mrda | ta, thanks | 23:51 |
| jroll | np | 23:51 |
| *** naohirot has joined #openstack-ironic | 23:53 | |
| *** smoriya has joined #openstack-ironic | 23:54 | |
| *** alexpilotti has quit IRC | 23:54 | |
| *** alexpilotti has joined #openstack-ironic | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!