| *** naohirot has joined #openstack-ironic | 00:02 | |
| *** andreykurilin_ has quit IRC | 00:04 | |
| jroll | devananda: this works fine without the corresponding ironic patch, just shows blank, should I still -2 until that lands? https://review.openstack.org/#/c/150724/ | 00:05 |
|---|---|---|
| *** andreykurilin_ has joined #openstack-ironic | 00:05 | |
| jroll | also shameless plug to make agent jobs vote on ironic changes: https://review.openstack.org/#/c/152340/ | 00:08 |
| *** EmilienM is now known as EmilienM|afk | 00:11 | |
| naohirot | good morning ironic, jroll | 00:12 |
| jroll | heya naohirot :) | 00:12 |
| naohirot | jroll: good evening :) | 00:12 |
| devananda | jroll: yes | 00:14 |
| jroll | devananda: lame | 00:15 |
| * jroll changes vote again | 00:15 | |
| devananda | jroll: well | 00:15 |
| devananda | jroll: i mean, it's odd to land a feature in the client that doesn't exist in the server | 00:16 |
| devananda | jroll: we dont have to -2 it -- but we shouldn't land it | 00:16 |
| jroll | right | 00:16 |
| jroll | ok, I'm going to leave my +2 | 00:16 |
| jroll | meh, I'll change it, just noticed ruby -2'd a client patch for the same reason | 00:17 |
| rloo | jroll: what did I do? Oh, yeah, I don't think a client patch should go in before the feature. What if, eg the feature isn't approved or the API part gets changed during the review process or ... | 00:20 |
| jroll | yeah, agree | 00:20 |
| rloo | jroll: having said that, I have seen others approve changes in client before the feature goes in | 00:20 |
| jroll | yeah | 00:20 |
| rloo | jroll: probably good if we were consistent though, so people don't think that I'm singling them out ;) | 00:20 |
| jroll | right :P | 00:21 |
| jroll | I'm fine with it either way, not a huge deal | 00:21 |
| rloo | jroll: and the only reason for -2 is so that I don't forget/trip on it again, the -1 can disappear. | 00:22 |
| jroll | yep | 00:23 |
| openstackgerrit | Dan Prince proposed openstack/ironic: Partial revert of 4606716 until we debug further https://review.openstack.org/152328 | 00:24 |
| *** derekh has quit IRC | 00:26 | |
| *** penick has quit IRC | 00:32 | |
| *** penick has joined #openstack-ironic | 00:50 | |
| jroll | why oh why | 00:52 |
| jroll | in devstack: ironic --debug node-list | 00:52 |
| jroll | take the curl command that --debug outputs, and run it, 401 | 00:52 |
| jroll | wtf | 00:52 |
| jroll | does devstack use one-time tokens now? | 00:54 |
| *** xxj has quit IRC | 01:00 | |
| *** Haomeng|2 has quit IRC | 01:00 | |
| *** xxj has joined #openstack-ironic | 01:01 | |
| *** Haomeng|2 has joined #openstack-ironic | 01:03 | |
| *** victor_lowther has quit IRC | 01:07 | |
| *** victor_lowther has joined #openstack-ironic | 01:13 | |
| *** chenleji has quit IRC | 01:13 | |
| *** chenleji has joined #openstack-ironic | 01:14 | |
| *** zhenzanz has joined #openstack-ironic | 01:17 | |
| mrda | thanks for the review jroll | 01:17 |
| jroll | mrda: np :) | 01:17 |
| jroll | it's so close! | 01:18 |
| *** chenleji has quit IRC | 01:18 | |
| mrda | getting there :) | 01:18 |
| *** chenleji has joined #openstack-ironic | 01:18 | |
| jroll | the get() being still just uuid really tripped me up | 01:18 |
| jroll | I thought something else was validating input data | 01:18 |
| jroll | or I was tricked | 01:18 |
| mrda | ya | 01:19 |
| jroll | this has 2x+2 if anyone else wants to get eyes on it https://review.openstack.org/#/c/148128/ | 01:19 |
| jroll | I'm off for the night, see y'all tomorrow | 01:20 |
| *** davideagnello has quit IRC | 01:24 | |
| *** achanda has quit IRC | 01:31 | |
| *** andreykurilin_ has quit IRC | 01:33 | |
| mrda | jroll: Are you still here? | 01:33 |
| mrda | jroll: or have you _really_ gone? :P | 01:34 |
| mrda | If not jroll, any other comments from cores welcomed :) | 01:34 |
| mrda | In https://review.openstack.org/#/c/141737/10/ironic/api/controllers/v1/node.py jroll asked whether validate should use node_ident instead of node_uuid and node_name | 01:35 |
| *** jerryz has joined #openstack-ironic | 01:36 | |
| mrda | The reason for this is because the current interface does a HTTP GET on /nodes/validate?node_uuid=... | 01:36 |
| mrda | and so changing this to node_ident will change the API | 01:36 |
| mrda | and I don't want to do this, lest I get slapped | 01:36 |
| mrda | So do people agree that node_uuid?=... and ?node_name=... (both optinal) is the way to go? | 01:37 |
| openstackgerrit | Zhenzan Zhou proposed openstack/ironic: Add policy mask_password to mask passwords in driver_info https://review.openstack.org/150688 | 01:37 |
| mrda | devananda: ^^ if you don't mind taking a look (or like I said any other core) | 01:37 |
| devananda | ooh | 01:39 |
| devananda | mrda: IFF we want to rename that parameter, we should: version bump (because that's cool) and return a deprecation warning in the HTTP header (or something like that) if the old parameter name is supplied instead | 01:41 |
| devananda | we can't just remove validate?node_uuid=... from the API | 01:41 |
| mrda | right, the other option is to support node_uuid and node_name | 01:41 |
| devananda | right | 01:41 |
| mrda | which way would you prefer? | 01:41 |
| devananda | consistency | 01:42 |
| mrda | (I like adding another optional param personally, but happy to switch...) | 01:42 |
| mrda | so change it | 01:42 |
| mrda | ? | 01:42 |
| devananda | AIUI, you're changing all places in the API where a node uuid was passed in previously | 01:42 |
| devananda | and allowing either uuid or name in those places | 01:42 |
| mrda | that is true, but... | 01:43 |
| devananda | so an operator who starts using ironic after this patch may reasonably never know or care about uuids | 01:43 |
| devananda | except when they have to make a call to the /validate?node_name=nnnn | 01:43 |
| mrda | ... there's no API change, as the URL just supports either a UUID or name (positionally) | 01:43 |
| devananda | then they need to know it's a name not a uuid | 01:43 |
| mrda | this, since it's a HTTP GET, is an API change | 01:43 |
| devananda | right | 01:43 |
| mrda | if we had /validate/node_ident as an end-point, this wouldn't be an issue :) | 01:44 |
| mrda | but since it's a param we do | 01:44 |
| devananda | mrda: I hate to ask this, but it just crossed my very tired and jetlagged brain | 01:45 |
| mrda | sure | 01:45 |
| mrda | (hope you're enjoying Grenoble by the way) | 01:45 |
| devananda | mrda: if at some point we were to add multi-tenancy (or rather, assigning nodes to tenants, and giving them some control thereof) | 01:45 |
| devananda | mrda: the uniqueness of a name, because it's part of the URL, is global -- not tenant local | 01:45 |
| mrda | that would be an API bump :) | 01:45 |
| devananda | it would conflict, wouldn't it? | 01:45 |
| mrda | I think we would make (tenant X name) unique | 01:46 |
| devananda | admin being the default tenant, then | 01:46 |
| mrda | ...whcih would be fine, since we would migrate to the new structure, adding in information | 01:46 |
| devananda | (because uniqueness on ( NULL X name ) is not enforceable | 01:46 |
| mrda | if we wanted to remove tenants then it would be an issue | 01:47 |
| devananda | k k, cool | 01:47 |
| devananda | glad i asked. that cleared up nicely :) | 01:47 |
| mrda | Cool | 01:47 |
| mrda | So, change the API and go to node_ident? | 01:47 |
| mrda | Or support both node_uuid and node_name? | 01:47 |
| devananda | or just /validate?node=xxxx | 01:47 |
| mrda | (sure, that's a variant on option 1) | 01:48 |
| devananda | and leave node-uuid for backwards compat, with a deprecation header | 01:48 |
| mrda | if that's what best, sure | 01:48 |
| devananda | or even without a deprecation header, i ncase someone later on needs to be really explicit | 01:48 |
| mrda | (need to see how to do a deprecation header then) | 01:48 |
| devananda | (I dunno either offhand) | 01:48 |
| mrda | ok, I'll push up a new patch and we can discuss in the review | 01:49 |
| devananda | but the microversion stuff did add result headers | 01:49 |
| devananda | anyway ... cheers. i'm delighted by how close it is | 01:49 |
| mrda | thanks for your time - and enjoy Grenoble! :) | 01:49 |
| devananda | thanks ... I'd enjoy some sleep ... :) | 01:49 |
| *** chuckC_ has quit IRC | 01:53 | |
| *** ChuckC_ has joined #openstack-ironic | 01:55 | |
| *** davideagnello has joined #openstack-ironic | 01:56 | |
| *** davideagnello has quit IRC | 01:56 | |
| *** ChuckC has quit IRC | 01:59 | |
| *** chenglch has joined #openstack-ironic | 02:11 | |
| *** mrmoje has quit IRC | 02:16 | |
| *** penick has quit IRC | 02:20 | |
| *** ChuckC has joined #openstack-ironic | 02:30 | |
| *** ChuckC_ has quit IRC | 02:34 | |
| *** Marga_ has quit IRC | 02:38 | |
| *** Marga_ has joined #openstack-ironic | 02:39 | |
| *** spandhe has quit IRC | 02:43 | |
| *** Marga_ has quit IRC | 02:43 | |
| *** ramineni has joined #openstack-ironic | 02:45 | |
| *** davideagnello has joined #openstack-ironic | 02:51 | |
| *** BadCub_ has quit IRC | 02:51 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC Management module for iRMC Driver https://review.openstack.org/146803 | 02:59 |
| jroll | 01:36:07 mrda | The reason for this is because the current interface does a HTTP GET on /nodes/validate?node_uuid=... <- really? ew. I think /v1/nodes/uuid/validate also works, though, maybe, idk | 02:59 |
| mrda | eeew | 03:00 |
| * mrda sees black magic all around | 03:01 | |
| jroll | basically yeah | 03:01 |
| mrda | MOAR TESTING | 03:02 |
| mrda | for me :) | 03:02 |
| mrda | Might as well push up the newest version for more review :) | 03:03 |
| *** rloo has quit IRC | 03:03 | |
| openstackgerrit | Michael Davies proposed openstack/ironic: Add logical name support to Ironic https://review.openstack.org/141737 | 03:03 |
| openstackgerrit | Michael Davies proposed openstack/ironic: Add logical name support to Ironic https://review.openstack.org/141737 | 03:04 |
| jroll | mrda: if you wanted even more black magic, you could accept ?node_uuid=blah for backwards compat, but map it to uuid or name :P | 03:05 |
| jroll | that's more like evil magic though | 03:05 |
| mrda | I could do that. But I like my sanity where it is right now. | 03:06 |
| mrda | ...which is teetering :) | 03:06 |
| *** eghobo has quit IRC | 03:15 | |
| *** zhenzanz_ has joined #openstack-ironic | 03:21 | |
| *** zhenzanz has quit IRC | 03:21 | |
| *** zhenzanz_ is now known as zhenzanz | 03:21 | |
| *** chuckC_ has joined #openstack-ironic | 03:21 | |
| *** annegentle has quit IRC | 03:28 | |
| *** pcrews has quit IRC | 03:34 | |
| *** Guest81568 has joined #openstack-ironic | 03:41 | |
| NobodyCam | adam_g: will you have a testing update on the whiteboard before teh meeting? | 03:49 |
| jroll | NobodyCam: he's in france, I think it's safe to assume no :P | 03:53 |
| *** pcrews has joined #openstack-ironic | 03:53 | |
| jroll | would love to hear folks' thoughts on https://review.openstack.org/#/c/150688/6/etc/ironic/policy.json | 03:54 |
| *** stendulker has joined #openstack-ironic | 03:54 | |
| openstackgerrit | Merged openstack/ironic: Move oslo.db references to oslo_db https://review.openstack.org/151955 | 04:00 |
| *** davideagnello has quit IRC | 04:02 | |
| *** davideagnello has joined #openstack-ironic | 04:03 | |
| *** harlowja has quit IRC | 04:08 | |
| *** achanda has joined #openstack-ironic | 04:12 | |
| *** penick has joined #openstack-ironic | 04:12 | |
| *** Haomeng has joined #openstack-ironic | 04:18 | |
| *** Haomeng|2 has quit IRC | 04:18 | |
| *** pcrews has quit IRC | 04:26 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Update etc/ironic/ironic.conf.sample https://review.openstack.org/152385 | 04:33 |
| jroll | naohirot: you can just roll that in with your patch that actually updates the sample config, it's fine | 04:34 |
| naohirot | jroll: this is the reflection of the comment https://review.openstack.org/#/c/151958/2/etc/ironic/ironic.conf.sample | 04:36 |
| jroll | naohirot: it's fine, that's no reason to block a patch | 04:37 |
| naohirot | jroll: thanks | 04:37 |
| jroll | left a note on the review | 04:38 |
| naohirot | jroll: I made the same comment in https://review.openstack.org/#/c/152385/ | 04:38 |
| naohirot | jroll: thanks | 04:39 |
| jroll | no problem | 04:39 |
| naohirot | jroll: I added Vladyslav to the reviews | 04:40 |
| *** rameshg87 has joined #openstack-ironic | 04:41 | |
| *** coolsvap|afk is now known as coolsvap | 04:42 | |
| *** penick_ has joined #openstack-ironic | 04:48 | |
| *** chuckC_ has quit IRC | 04:48 | |
| *** penick has quit IRC | 04:49 | |
| *** penick_ is now known as penick | 04:49 | |
| * NobodyCam is almost awake for the meeting | 04:55 | |
| NobodyCam | lol | 04:55 |
| *** achanda has quit IRC | 04:55 | |
| jroll | heh | 04:55 |
| jroll | same here | 04:55 |
| * jroll could use a coffee but doesn't want to make a whole pot | 04:56 | |
| jroll | should be a short one, though, I hope | 04:57 |
| * jroll reads this sensor spec thing | 04:58 | |
| *** lazy_prince is now known as killer_prince | 04:59 | |
| *** chuckC_ has joined #openstack-ironic | 05:01 | |
| openstackgerrit | Michael Davies proposed openstack/ironic: Add logical name support to Ironic https://review.openstack.org/141737 | 05:08 |
| *** Haomeng has quit IRC | 05:11 | |
| *** Haomeng|2 has joined #openstack-ironic | 05:12 | |
| *** achanda has joined #openstack-ironic | 05:16 | |
| *** eghobo has joined #openstack-ironic | 05:21 | |
| *** eghobo has quit IRC | 05:35 | |
| *** ijw has quit IRC | 05:45 | |
| *** killer_prince is now known as lazy_prince | 05:45 | |
| *** rwsu is now known as rwsu-afk | 05:51 | |
| *** davideagnello has joined #openstack-ironic | 05:51 | |
| *** davideagnello has quit IRC | 05:56 | |
| jroll | phew | 05:58 |
| NobodyCam | night all | 05:58 |
| NobodyCam | and thank you | 05:58 |
| jroll | g'night | 05:58 |
| Haomeng|2 | jroll: good night | 05:58 |
| Haomeng|2 | NobodyCam: night! | 05:58 |
| naohirot | NobodyCam: jroll: good night | 05:59 |
| mrda | night :) | 05:59 |
| openstackgerrit | Zhenzan Zhou proposed openstack/ironic: Add policy mask_password to mask passwords in driver_info https://review.openstack.org/150688 | 06:01 |
| *** stendulker_ has joined #openstack-ironic | 06:08 | |
| *** stendulker has quit IRC | 06:10 | |
| *** datajerk has quit IRC | 06:44 | |
| *** eghobo has joined #openstack-ironic | 06:55 | |
| *** penick has quit IRC | 06:58 | |
| *** Marga_ has joined #openstack-ironic | 06:59 | |
| *** Marga_ has quit IRC | 07:00 | |
| *** Marga_ has joined #openstack-ironic | 07:00 | |
| *** eghobo has quit IRC | 07:02 | |
| *** achanda has quit IRC | 07:04 | |
| *** stendulker has joined #openstack-ironic | 07:09 | |
| *** stendulker_ has quit IRC | 07:12 | |
| *** mrda is now known as mrda-away | 07:14 | |
| *** achanda has joined #openstack-ironic | 07:29 | |
| *** yog_ has joined #openstack-ironic | 07:43 | |
| *** achanda has quit IRC | 07:54 | |
| *** chlong has quit IRC | 07:56 | |
| *** achanda has joined #openstack-ironic | 08:01 | |
| *** pradipta_away is now known as pradipta | 08:13 | |
| *** dtantsur|afk is now known as dtantsur | 08:20 | |
| dtantsur | Morning Ironic | 08:20 |
| Haomeng|2 | dtantsur: good morning:) | 08:28 |
| dtantsur | Haomeng|2, o/ | 08:29 |
| Haomeng|2 | dtantsur: :) | 08:29 |
| *** foexle has joined #openstack-ironic | 08:31 | |
| *** lucasagomes has joined #openstack-ironic | 08:31 | |
| *** achanda has quit IRC | 08:32 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add support for local boot https://review.openstack.org/146189 | 08:35 |
| *** jcoufal has joined #openstack-ironic | 08:35 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add Node.driver_internal_info https://review.openstack.org/150784 | 08:41 |
| rameshg87 | dtantsur, hi | 08:42 |
| dtantsur | rameshg87, o/ | 08:42 |
| rameshg87 | dtantsur, for adding proliantutils to stackforge | 08:42 |
| rameshg87 | dtantsur, i was following this: http://docs.openstack.org/infra/manual/creators.html | 08:42 |
| rameshg87 | dtantsur, You may need to create the bug management team in Launchpad. If you do so, set the owner of the team to ‘openstack-administrators’. | 08:42 |
| rameshg87 | dtantsur, i created a bug management team | 08:43 |
| openstackgerrit | Tan Lin proposed openstack/ironic: Clean driver_internal_info when changes nodes' driver https://review.openstack.org/151990 | 08:43 |
| dtantsur | I didn't do it :) | 08:43 |
| dtantsur | it's probably for official projects... | 08:43 |
| rameshg87 | dtantsur, ah okay ..thought so :) | 08:43 |
| rameshg87 | dtantsur, so hope without that also it is fine | 08:43 |
| dtantsur | yeah, nevermind for now... | 08:44 |
| dtantsur | I'm still the only bug driver in discoverd, I guess :D | 08:44 |
| * dtantsur needs to change it eventually... | 08:44 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Last-minute release notes update for 1.0.0 https://review.openstack.org/152428 | 08:44 |
| *** ifarkas has joined #openstack-ironic | 08:44 | |
| *** ndipanov has joined #openstack-ironic | 08:50 | |
| *** spandhe has joined #openstack-ironic | 08:50 | |
| *** kan_ has joined #openstack-ironic | 08:51 | |
| dtantsur | lucasagomes, hey hey! how is midcycle going? | 08:52 |
| lucasagomes | dtantsur, hey man, it's all good | 08:53 |
| lucasagomes | just started | 08:53 |
| *** andreykurilin_ has joined #openstack-ironic | 08:54 | |
| openstackgerrit | Merged stackforge/ironic-discoverd: Last-minute release notes update for 1.0.0 https://review.openstack.org/152428 | 08:56 |
| *** jistr has joined #openstack-ironic | 08:58 | |
| *** spandhe has quit IRC | 09:03 | |
| *** spandhe has joined #openstack-ironic | 09:03 | |
| *** lazy_prince has quit IRC | 09:03 | |
| *** spandhe has quit IRC | 09:09 | |
| *** killer_prince has joined #openstack-ironic | 09:17 | |
| *** killer_prince is now known as lazy_prince | 09:17 | |
| *** MattMan has quit IRC | 09:19 | |
| *** athomas has joined #openstack-ironic | 09:21 | |
| *** derekh has joined #openstack-ironic | 09:21 | |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Finalize name + small readme fixes https://review.openstack.org/152435 | 09:25 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC Management module for iRMC Driver https://review.openstack.org/146803 | 09:33 |
| *** Marga_ has quit IRC | 09:37 | |
| *** mrmoje has joined #openstack-ironic | 09:37 | |
| openstackgerrit | Merged stackforge/ironic-discoverd: Finalize name + small readme fixes https://review.openstack.org/152435 | 09:38 |
| *** Marga_ has joined #openstack-ironic | 09:38 | |
| *** pelix has joined #openstack-ironic | 09:40 | |
| *** Marga_ has quit IRC | 09:42 | |
| lucasagomes | https://github.com/rackerlabs/ironic-neutron-plugin | 09:43 |
| *** stelfer has joined #openstack-ironic | 09:44 | |
| lucasagomes | stelfer, https://github.com/rackerlabs/ironic-neutron-plugin | 09:44 |
| *** MattMan has joined #openstack-ironic | 09:45 | |
| *** mgoddard has joined #openstack-ironic | 09:48 | |
| rameshg87 | ifarkas, hi | 09:52 |
| ifarkas | rameshg87, hi | 09:52 |
| rameshg87 | ifarkas, regarding the raid configuration, are you also proposing a vendor passthru for it ? | 09:52 |
| ifarkas | rameshg87, yes but separately | 09:53 |
| rameshg87 | ifarkas, okay, so for drac there will support for doing it in zapping and also by a vendor-passthru ? | 09:53 |
| ifarkas | rameshg87, yes | 09:54 |
| rameshg87 | ifarkas, okay | 09:55 |
| *** jiangfei has quit IRC | 09:55 | |
| *** jiangfei has joined #openstack-ironic | 09:56 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: New driver interface for RAID configuration https://review.openstack.org/135899 | 10:04 |
| *** davideagnello has joined #openstack-ironic | 10:07 | |
| openstackgerrit | Zhenzan Zhou proposed openstack/ironic: Add policy mask_password to mask passwords in driver_info https://review.openstack.org/150688 | 10:12 |
| *** zhenzanz has quit IRC | 10:18 | |
| dtantsur | discoverd 1.0.0 gone live \o/ | 10:19 |
| *** Haomeng has joined #openstack-ironic | 10:28 | |
| *** Haomeng|2 has quit IRC | 10:28 | |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Add MANAGEABLE state and associated transitions https://review.openstack.org/150073 | 10:30 |
| *** chenglch has quit IRC | 10:49 | |
| openstackgerrit | Frederic Lepied proposed stackforge/ironic-discoverd: Add eDeploy plugin https://review.openstack.org/146599 | 10:49 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Start 1.1 development https://review.openstack.org/152465 | 10:55 |
| yog_ | Hi who do i login into to ubuntu or fedora image built by diskiimage builder after deploying using ironic .. | 10:58 |
| openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Add eDeploy plugin https://review.openstack.org/146599 | 11:02 |
| *** ramineni has quit IRC | 11:02 | |
| openstackgerrit | Merged openstack/ironic: region missing in endpoint selection https://review.openstack.org/148128 | 11:03 |
| openstackgerrit | Merged stackforge/ironic-discoverd: Start 1.1 development https://review.openstack.org/152465 | 11:05 |
| *** Haomeng has quit IRC | 11:11 | |
| *** Haomeng has joined #openstack-ironic | 11:12 | |
| *** andreykurilin_ has quit IRC | 11:19 | |
| dtantsur | yog_, how did you build it? also did you use --key-name argument to nova boot? | 11:21 |
| *** jistr has quit IRC | 11:21 | |
| *** stelfer has quit IRC | 11:21 | |
| *** chenleji has quit IRC | 11:22 | |
| *** chenleji has joined #openstack-ironic | 11:22 | |
| yog_ | dtantsur, used diskimage builder ..disk-image-create ubuntu baremetal | 11:25 |
| dtantsur | yog_, then --key-name argument should be used to provide a key pair to nova IIRC | 11:26 |
| yog_ | dtantsur, yes i did used nova boot --flavor baremetal --image $image --key-name default testing | 11:26 |
| dtantsur | yog_, then (unless something went terribly wrong) you should be using this key | 11:26 |
| *** mgoddard has quit IRC | 11:26 | |
| *** jistr has joined #openstack-ironic | 11:27 | |
| yog_ | dtantsur, thanks .. will double check if i had messed up somewhr or not | 11:29 |
| *** jcoufal_ has joined #openstack-ironic | 11:29 | |
| *** jcoufal has quit IRC | 11:32 | |
| *** rameshg87 has quit IRC | 11:32 | |
| *** jcoufal_ has quit IRC | 11:34 | |
| *** lazy_prince has quit IRC | 11:56 | |
| *** jcoufal has joined #openstack-ironic | 11:59 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic: Add MANAGEABLE state and associated transitions https://review.openstack.org/150073 | 12:04 |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic: Add states required for node-inspection https://review.openstack.org/147857 | 12:04 |
| *** chlong has joined #openstack-ironic | 12:06 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC Management module for iRMC Driver https://review.openstack.org/146803 | 12:06 |
| dtantsur | Haomeng, wrt https://blueprints.launchpad.net/ironic-discoverd/+spec/in-band-ipmi-auto-configuration have you seen https://blueprints.launchpad.net/ironic-discoverd/+spec/setup-ipmi-credentials ? | 12:07 |
| *** zhenzanz has joined #openstack-ironic | 12:07 | |
| *** EmilienM|afk is now known as EmilienM | 12:08 | |
| *** Marga_ has joined #openstack-ironic | 12:09 | |
| *** derekh has quit IRC | 12:12 | |
| *** zhenzanz has quit IRC | 12:13 | |
| *** derekh has joined #openstack-ironic | 12:13 | |
| *** killer_prince has joined #openstack-ironic | 12:16 | |
| *** killer_prince is now known as lazy_prince | 12:16 | |
| *** derekh has quit IRC | 12:16 | |
| *** derekh has joined #openstack-ironic | 12:17 | |
| *** dtantsur is now known as dtantsur|bbl | 12:20 | |
| *** jcoufal has quit IRC | 12:23 | |
| openstackgerrit | Nisha Agarwal proposed openstack/ironic: Add states required for node-inspection https://review.openstack.org/147857 | 12:27 |
| *** jcoufal has joined #openstack-ironic | 12:28 | |
| *** erwan_taf has quit IRC | 12:28 | |
| *** dlpartain has joined #openstack-ironic | 12:30 | |
| openstackgerrit | Merged openstack/ironic: Add Node.driver_internal_info https://review.openstack.org/150784 | 12:32 |
| *** dlpartain has left #openstack-ironic | 12:32 | |
| *** davideagnello has quit IRC | 12:36 | |
| *** Haomeng|2 has joined #openstack-ironic | 12:40 | |
| openstackgerrit | Merged openstack/ironic: Clean driver_internal_info when changes nodes' driver https://review.openstack.org/151990 | 12:40 |
| *** Haomeng has quit IRC | 12:41 | |
| *** Haomeng has joined #openstack-ironic | 12:45 | |
| *** Haomeng|2 has quit IRC | 12:46 | |
| *** jmankov has joined #openstack-ironic | 12:48 | |
| *** stelfer has joined #openstack-ironic | 12:51 | |
| *** alex_xu_ has joined #openstack-ironic | 12:53 | |
| *** athomas_ has joined #openstack-ironic | 12:56 | |
| yog_ | yog_, Hi, to deploy a image on a specific baremetal do i need to create unique flavor for each machine? so should i use system name as key par.. example nova flavor-key my-baremetal-flavor set sys_name=<system name> | 12:57 |
| *** alex_xu has quit IRC | 12:57 | |
| *** subscope has quit IRC | 12:57 | |
| *** athomas has quit IRC | 12:57 | |
| *** kbyrne has quit IRC | 12:57 | |
| *** jmanko has quit IRC | 12:57 | |
| *** kbyrne has joined #openstack-ironic | 12:58 | |
| devananda | yog_: why do you want to specify the particular bare metal node to deploy to? | 12:58 |
| *** subscope has joined #openstack-ironic | 12:58 | |
| yog_ | devananda, at preset I just want to install image on a particular system using ironic.. | 12:59 |
| yog_ | kind of like replacement to install using cobler | 13:00 |
| yog_ | devananda, rather then have a provisioning environment base on cobbler .. want to use ironic to do that | 13:01 |
| *** dprince has joined #openstack-ironic | 13:04 | |
| openstackgerrit | Dan Prince proposed openstack/ironic: Partial revert of 4606716 until we debug further https://review.openstack.org/152328 | 13:05 |
| *** stendulker has quit IRC | 13:05 | |
| openstackgerrit | Dan Prince proposed openstack/ironic: Partial revert of 4606716 until we debug further https://review.openstack.org/152328 | 13:06 |
| yog_ | devananda, did try using flaovr with sys_name=<system name> and ironic node with properties sys_name=<system name>, but its failed with not able to find suitable hosts, is there other way to deploy an image to specific ironic node | 13:12 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Add MANAGEABLE state and associated transitions https://review.openstack.org/150073 | 13:15 |
| devananda | yog_: sure. but Nova's job is to find a suitable node based on the flavor, available hardware, quotas, etc. | 13:18 |
| devananda | (among nova's jobs is ...) | 13:18 |
| devananda | yog_: you didn't say why the exact placement is important to you, and without knowing why, I can't help you address the need / usecase | 13:19 |
| devananda | yog_: it is technically possible to deploy to bare metal without nova at all, but I don't recommend it for most use cases | 13:20 |
| yog_ | devananda, at present I cobbler to install OS on hardware, want to replay with ironic, machines are owned by different people, so when a person likes to have certain os to be installed his system, i be able to do that | 13:22 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Refactor ilo.deploy.get_single_nic_with_vif_port_id() https://review.openstack.org/152506 | 13:24 |
| devananda | so Ironic does not currently have a means to constrain specific tenants to specific machines | 13:24 |
| devananda | or specific machines to specific tenants | 13:24 |
| devananda | and even if it did -- the tenant wouldn't need to select exactly which node they wanted to deploy | 13:24 |
| devananda | you could approximate this with host aggregates in Nova today, I think | 13:25 |
| devananda | and let the scheduler do the matching | 13:25 |
| openstackgerrit | Stig Telfer proposed openstack/ironic-specs: Fix check for missing subheadings https://review.openstack.org/152507 | 13:25 |
| devananda | eg, by writing a scheduler filter so that tenant X would only get nodes that have node.properties {'extra_specs': 'tenant-X'} | 13:26 |
| *** jjohnson2 has joined #openstack-ironic | 13:31 | |
| yog_ | devananda, mm.. thanks for the info | 13:31 |
| jroll | morning all | 13:34 |
| jroll | how's the midcycle? I see links to our neutron code :D | 13:35 |
| *** mlsoule has joined #openstack-ironic | 13:39 | |
| *** alexpilotti has joined #openstack-ironic | 13:40 | |
| *** rloo has joined #openstack-ironic | 13:44 | |
| *** yog_ has quit IRC | 13:47 | |
| *** Marga_ has quit IRC | 13:50 | |
| *** mjturek has joined #openstack-ironic | 14:01 | |
| Shrews | morning | 14:01 |
| jroll | \o Shrews | 14:02 |
| Shrews | devananda: thx for the response regarding the pre-emptive lock check | 14:05 |
| Shrews | devananda: i guess my concern is that it bypasses the retry logic we built in in an attempt to reduce these types of exceptions | 14:06 |
| devananda | Shrews: oh - fair point. but is that bad? | 14:07 |
| Shrews | devananda: dunno? that's why i'm asking :) | 14:07 |
| devananda | Shrews: I mean, is that going to prevent the retries from working? | 14:07 |
| devananda | also, lucasagomes and I wwere just discussing how we need to be a little bit better with the microversion support | 14:07 |
| jroll | oh? | 14:08 |
| jroll | I was just going to hack on that a bit | 14:08 |
| devananda | namely, there were several patches that landed since Juno closed, which changed the API | 14:08 |
| Shrews | will it introduce more NodeLocked exceptions? possibly. is that bad? well, we did take steps to reduce them | 14:08 |
| devananda | and, well, aren't discoverable | 14:08 |
| devananda | Shrews: oh. yah. you're totally correct | 14:08 |
| lucasagomes | morning Shrews jroll | 14:08 |
| jroll | heya lucasagomes :) | 14:09 |
| *** dtantsur|bbl is now known as dtantsur | 14:09 | |
| devananda | Shrews: want to comment and fix while I hack on the API a bit more? | 14:09 |
| Shrews | devananda: will do | 14:09 |
| devananda | cheers | 14:09 |
| jroll | devananda: so about things we've already changed this cycle, that aren't discoverable? what's the game plan? | 14:10 |
| jroll | (is there one?) | 14:11 |
| rloo | hello ironickers | 14:11 |
| dtantsur | devananda, Shrews, jroll, rloo _o/ | 14:12 |
| jroll | hey dtantsur :) | 14:13 |
| devananda | jroll: yep. working on it now | 14:13 |
| trown | dtantsur: congrats on discoverd release | 14:15 |
| *** BadCub_ has joined #openstack-ironic | 14:15 | |
| jroll | ^ | 14:15 |
| dtantsur | thanks :) it took some time... | 14:15 |
| jroll | ohai trown rloo BadCub_ :) | 14:15 |
| trown | o/ hi everybody | 14:16 |
| BadCub_ | Morning Ironic | 14:19 |
| Shrews | dtantsur: rloo: trown: BadCub_: <insert nick here>: morning | 14:19 |
| rloo | morning Shrews, jroll, BadCub_, dtantsur, trown ;) | 14:21 |
| *** Nisha has joined #openstack-ironic | 14:26 | |
| jroll | so thinking about the ceilometer conversation last night... http://docs.openstack.org/developer/ceilometer/measurements.html | 14:27 |
| jroll | ceilometer doesn't just take arbitrary key/value data | 14:27 |
| jroll | ? | 14:27 |
| jroll | (for measurements) | 14:27 |
| * jroll asks ceilometer people | 14:28 | |
| Shrews | probably a good idea ;) | 14:29 |
| jroll | http://docs.openstack.org/developer/ceilometer/measurements.html#naming-convention | 14:29 |
| jroll | tells me I'm probably right | 14:30 |
| * Shrews thinks that's a thing that tells him how far his ceiling has travelled | 14:30 | |
| jroll | :D | 14:30 |
| Shrews | mine is zero, but NobodyCam's is off the charts | 14:30 |
| jroll | lol | 14:30 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Raise minimum API version to 1.6 https://review.openstack.org/152527 | 14:31 |
| devananda | jroll: ^ | 14:31 |
| jroll | huh, you can have a variable | 14:31 |
| jroll | let's make our meters hardware:key | 14:31 |
| jroll | where key is variable | 14:31 |
| jroll | :D | 14:31 |
| jroll | thanks devananda | 14:31 |
| * jroll looks | 14:31 | |
| dtantsur | devananda, don't see much sense in using 6 instead of 1... except that we already landed 1.1 (or even 2) | 14:32 |
| devananda | dtantsur: read the commit message please | 14:33 |
| jroll | devananda: can we move (MIN|MAX)_VER to base, and make them the full version? | 14:33 |
| dtantsur | devananda, I did. | 14:34 |
| devananda | dtantsur: ok. can you explain? | 14:34 |
| openstackgerrit | David Shrewsbury proposed openstack/ironic: Add MANAGEABLE state and associated transitions https://review.openstack.org/150073 | 14:35 |
| NobodyCam | good morning Ironicers | 14:35 |
| dtantsur | devananda, I don't see point in pretending that we had these versions. people installing from master can already rely on these changes without microversion | 14:35 |
| openstackgerrit | Adam Gandelman proposed openstack/ironic: Do not pass PXE net config from bootloader to ramdisk https://review.openstack.org/152551 | 14:35 |
| dtantsur | the only point of this change is to break that people | 14:35 |
| devananda | dtantsur: a user can not discover the difference between Juno's API and what they get by default from the current API | 14:36 |
| Shrews | devananda: pushed a new patch to 150073 to remove that pre-emptive check | 14:36 |
| devananda | if they don't specify the X-OpenStack-Ironic-API-Version: 1.3, then they will get 1.0 by default | 14:36 |
| devananda | Shrews: thanks much | 14:36 |
| *** davideagnello has joined #openstack-ironic | 14:36 | |
| devananda | dtantsur: but in getting 1.0, they actually will *not* get the same behavior that Juno had | 14:36 |
| dtantsur | devananda, then we need to break the compatibility between commits and hide all that changes under one version. but why have 6 fake versions? | 14:36 |
| devananda | dtantsur: and I think it's reasonable to assume that a 1.0 version is the same as the thing you were using before there was a version tag | 14:36 |
| dtantsur | I'm not arguing against one more version... | 14:37 |
| devananda | dtantsur: so that someone could, later on, go back and wrap each feature under a different version tag | 14:37 |
| devananda | dtantsur: ahh ok. so I would be fine also wrapping them all into one version | 14:37 |
| jroll | devananda: 1.8 and 1.9 haven't landed yet, why would that patch bump the version for them? | 14:37 |
| dtantsur | yep, that's what I mean :) | 14:37 |
| dtantsur | devananda, jroll, we might force all people to rebase on top of this patch instead | 14:38 |
| jroll | dtantsur: sure, so why bump the version before that stuff lands | 14:38 |
| devananda | saying that 1.0 == juno and is unsupported, 1.1 == where we add microversioning, 1.2 == the NOSTATE->AVAIALBLE change | 14:38 |
| devananda | jroll: sure. just the order I was working in locally | 14:38 |
| devananda | I can rebase quickly if that's what folks want | 14:38 |
| devananda | lucasagomes: do you agree? tldr; instead of raising MIN from 1.0 to 1.6, raise it from 1.0 to 1.1 | 14:39 |
| dtantsur | devananda, and 1.1 also means features that we didn't version properly, right? | 14:39 |
| devananda | dtantsur: it would, yes | 14:39 |
| devananda | 1.0 == Juno (not supported) and 1.1 == first version we versioned | 14:39 |
| devananda | I am fine with that | 14:39 |
| *** erwan_taf has joined #openstack-ironic | 14:39 | |
| dtantsur | lgtm (though we will definitely break inter-commit backward compatibility) | 14:39 |
| lucasagomes | yeah I'm fine with that, we can leave it open to someone to update the API and include the missing changes as part of the Kilo 1.0 | 14:39 |
| devananda | yea, we will :( but this is a small enough window I think it's OK | 14:40 |
| devananda | for a new feature that most clients are not using | 14:40 |
| devananda | and in fact, our client lib doesn't support yet ;) | 14:40 |
| devananda | so I think the impact of breaking this /now/ is near 0 | 14:40 |
| devananda | and it lets us be much clearer later on | 14:40 |
| jroll | oh, driver internal info landed, cool | 14:40 |
| devananda | yah | 14:41 |
| jroll | wait, why would bumping minimum to 1.1 break compatibility? | 14:41 |
| * jroll continues reading and agrees | 14:41 | |
| jroll | I don't think anyone is reasonably depending on this yet | 14:41 |
| devananda | jroll: IFF someone is currently specifying any version ... yea, that's it | 14:41 |
| jroll | yeah | 14:41 |
| jroll | I'm not | 14:42 |
| jroll | I doubt tripleo is | 14:42 |
| devananda | cool. /me rebases | 14:42 |
| devananda | I'm sure they're not | 14:42 |
| devananda | because they would only be using our python client | 14:42 |
| devananda | which doesn't set headers yet | 14:42 |
| jroll | right, ok | 14:42 |
| *** jgrimm is now known as zz_jgrimm | 14:42 | |
| devananda | so I'll rebase this on master and cahnge the versions, then rebase teh MANAGEABLE patch on that, then move the MIN/mAX stuff to base ? | 14:42 |
| jroll | let's move min/max before the MANAGEABLE patcH? | 14:43 |
| jroll | I guess it doesn't matter | 14:43 |
| jroll | and then I can rebase my things on top of that | 14:43 |
| lucasagomes | jroll, quick, probably dumb. If I want to inject extra metadata to the configdrive how I do it? | 14:45 |
| jroll | lucasagomes: --user-data iirc | 14:45 |
| jroll | not sure what gets passed to that, though | 14:46 |
| devananda | jroll: yea, moving those around shouldn't affect any API bits | 14:48 |
| lucasagomes | jroll, right, hmm... I'm still looking, but, not sure if we need to be able to add something in Ironic that nova can look at and add to the config drive, some extra metadata. | 14:48 |
| lucasagomes | For e.g, one use case is that we want to name some nics that we found in Ironic to something specific. Like we don't want it to be eth0 eth1 etc because that's arbitrary and doesn't guarantee order | 14:49 |
| lucasagomes | we would like to say, look this port here will be called nic1 | 14:49 |
| lucasagomes | looking what would be the best way to do that | 14:49 |
| lucasagomes | ideais are welcome ^ | 14:50 |
| lucasagomes | ideas* | 14:50 |
| jroll | lucasagomes: hrm, I don't think that's possible | 14:50 |
| jroll | but there's a thing | 14:50 |
| jroll | https://review.openstack.org/#/c/102649/ | 14:50 |
| jroll | that's what we did to work around the same thing | 14:50 |
| lucasagomes | jroll, I was thinking of having some extra_configdrive key in the extra field of the node, which the driver would look at and inject it as part of the configdrive | 14:50 |
| lucasagomes | so there we could add arbritary things | 14:50 |
| * lucasagomes looks | 14:50 | |
| jroll | lucasagomes: right, should work, I just don't see the use case honestly | 14:51 |
| lucasagomes | jroll, well, you can do it for all ur instances so ur sure that for e.g the interface accessing the network X is called Y across all ur instances | 14:52 |
| jroll | hrm | 14:53 |
| lucasagomes | jroll, oh yeah, I know that neutron patch. that's a good one | 14:53 |
| *** absubram has joined #openstack-ironic | 14:53 | |
| lucasagomes | but the extra_configdrive is arbritary. the nic renaming is just one use case | 14:53 |
| jroll | lucasagomes: can cloud-init even set the if name? | 14:53 |
| jroll | s/if/interface/ | 14:53 |
| jroll | right | 14:53 |
| jroll | idk, maybe it makes sense | 14:53 |
| * lucasagomes not sure, but that could be added | 14:54 | |
| lucasagomes | I think know os-network-config does | 14:54 |
| lucasagomes | (I know...) | 14:54 |
| lucasagomes | but just saying, that's a doable thing to do | 14:54 |
| jroll | right | 14:54 |
| jroll | seems weird to rely on interface names, though | 14:54 |
| lucasagomes | anyway, I will put a patch for that on top of the one in Nova | 14:54 |
| jroll | especially because the person with the instance doesn't have ironic access, theoretically | 14:55 |
| lucasagomes | jroll, but that's why we want to name it accordingly | 14:56 |
| lucasagomes | like MAC X name Y | 14:56 |
| lucasagomes | right, that's not the user setting it | 14:56 |
| lucasagomes | it's the deployer actually | 14:56 |
| lucasagomes | some extra configuration that applies to baremetal | 14:56 |
| jroll | hrm, yeah | 14:57 |
| jroll | ohhh. | 14:57 |
| jroll | yeah, that could be good. | 14:57 |
| jroll | we have some things in our baremetal images that are only in those images | 14:57 |
| lucasagomes | yeah | 14:57 |
| jroll | if we could specify them via ironic, might be nice | 14:57 |
| jroll | ok, +1 | 14:57 |
| lucasagomes | jroll, awesome! thanks | 14:57 |
| lucasagomes | right, I will put a patch on top of the one we have for the configdrive | 14:57 |
| lucasagomes | so we don't mix things up | 14:58 |
| jroll | yeah | 14:58 |
| jroll | agree | 14:58 |
| * naohirot good night all | 14:58 | |
| jroll | night naohirot | 14:58 |
| lucasagomes | g'night naohirot | 14:58 |
| naohirot | jroll: lucasagomes: see you tomorrow:) | 14:59 |
| *** yog_ has joined #openstack-ironic | 14:59 | |
| *** naohirot has quit IRC | 14:59 | |
| NobodyCam | anyone have a second for a quick review on AMT Spec patch. has 2 +2's if not I'll land it! :) hehehehe | 14:59 |
| NobodyCam | https://review.openstack.org/#/c/141269/ | 14:59 |
| jroll | looping back around, ceilometer indeed does not take arbitrary metrics, it appears: https://github.com/openstack/ceilometer/blob/master/ceilometer/ipmi/notifications/ironic.py#L162-175 | 14:59 |
| * jroll cries | 15:00 | |
| jroll | NobodyCam: I will not review that | 15:00 |
| NobodyCam | jroll: you did already lol | 15:00 |
| *** Nisha has quit IRC | 15:00 | |
| *** Nisha_away has joined #openstack-ironic | 15:00 | |
| jroll | exactly :P | 15:00 |
| NobodyCam | my question is why is ceilometer not adding a Ironic plugin | 15:01 |
| *** coolsvap is now known as coolsvap|afk | 15:01 | |
| NobodyCam | vs using ipmi to talk to us | 15:02 |
| jroll | NobodyCam: so there's two things | 15:03 |
| jroll | they can either poll this ipmi agent thing running on BM | 15:03 |
| jroll | or ironic can emit notifications on rabbit | 15:03 |
| jroll | with measurements | 15:03 |
| NobodyCam | humm | 15:05 |
| NobodyCam | they couldn't hit our api to get_sensor_data? | 15:05 |
| NobodyCam | brb | 15:05 |
| jroll | NobodyCam: we'd have to talk to BMCs syncronously, pls no | 15:06 |
| NobodyCam | oh ther is that | 15:07 |
| * NobodyCam takes large slug of coffee | 15:08 | |
| devananda | ironic emitting notifications to ceilometer ++ | 15:10 |
| jroll | NobodyCam: http://farm7.staticflickr.com/6158/6181080354_0fe5278a9f.jpg | 15:11 |
| jroll | devananda: did you uh, see the meeting logs from last night? | 15:11 |
| devananda | no | 15:11 |
| devananda | i should read those, huh? | 15:12 |
| jroll | so we want to emit something more generic | 15:12 |
| jroll | like hardware.temperature, not hardware.ipmi.temperature | 15:12 |
| devananda | righ | 15:12 |
| devananda | right | 15:12 |
| jroll | and uh | 15:12 |
| jroll | ceilometer defines a list of acceptable metric names | 15:12 |
| devananda | what's wrong with taht? | 15:12 |
| NobodyCam | jroll: lol | 15:12 |
| jroll | you can't just give it arbitrary keys | 15:12 |
| devananda | so Jim should (hae already) change(d) ceilometer | 15:12 |
| jroll | oh? | 15:13 |
| jroll | hrm | 15:13 |
| NobodyCam | oh really? | 15:13 |
| jroll | jmankov: ^ is that done? | 15:13 |
| NobodyCam | I didn't get that impression lastnight | 15:13 |
| jroll | right | 15:14 |
| jroll | chris dent was very opposed to this spec | 15:14 |
| devananda | oh | 15:14 |
| jroll | https://review.openstack.org/#/c/130359/ | 15:14 |
| NobodyCam | that what I got, chris's comments are on rev 6 | 15:14 |
| jroll | I guess he's planning on doing it | 15:15 |
| jroll | it just makes me rage at this point | 15:15 |
| lucasagomes | folks I'm going to land https://review.openstack.org/#/c/152328 | 15:17 |
| lucasagomes | this broke fedora, I know it's a temporary fix right now and we have a bug open for that | 15:17 |
| lucasagomes | to keep track so I think it's good to go | 15:17 |
| NobodyCam | lucasagomes: sure, do you why it broke? | 15:18 |
| devananda | jroll: clearly i should go read the meeting logs | 15:18 |
| jroll | lucasagomes: right on, can you also re-open the bug that this fixed? | 15:18 |
| NobodyCam | devananda: please do! | 15:18 |
| jroll | devananda: I mean, we agreed to make it generic | 15:18 |
| lucasagomes | NobodyCam, I haven't looked into yet | 15:18 |
| jroll | I still need to actually review the spec | 15:18 |
| lucasagomes | NobodyCam, but the commit message says that this is a temp fix and the real fix will come later | 15:18 |
| lucasagomes | this is just to unblock fedora | 15:18 |
| lucasagomes | I think it's fair | 15:19 |
| lucasagomes | jroll, sure, that's a partial-fix | 15:19 |
| lucasagomes | so it's all good | 15:19 |
| lucasagomes | the bug won't close | 15:19 |
| jroll | ah ok | 15:19 |
| NobodyCam | ok, lucasagomes happy to help where I can | 15:19 |
| lucasagomes | NobodyCam, thanks, yeah dprince will take a look at it when he finds some tie | 15:20 |
| lucasagomes | time | 15:20 |
| NobodyCam | :) | 15:20 |
| dprince | NobodyCam: hi, yes. I hit this at a bad time. I would love to look into it with a bit more detail. | 15:20 |
| NobodyCam | I looked thru the log he added to the bug. but didn't see some of the logging I expected to see | 15:21 |
| dprince | NobodyCam: I'm not a fan of the sleep... but the new code doesn't seem to work sufficiently to eliminate it. At least not on Fedora. | 15:21 |
| NobodyCam | hey dprince :) | 15:21 |
| NobodyCam | yea | 15:21 |
| tteggel | morning ironic | 15:21 |
| dprince | NobodyCam: One thing I would ask is has anybody else (besides me) tried this on real hardware? | 15:21 |
| NobodyCam | I'm ok with revert but also want to know whats under the hood | 15:21 |
| lucasagomes | dansmith just +2'd the confidrive patch \o/! | 15:22 |
| dprince | NobodyCam: it seems to work in the virtualized CI w/ Fedora... just not for me in my dev environment on real machines. So it is definately a race of some sort I think. | 15:22 |
| jroll | NobodyCam: with chris dent +1'ing the ceilometer spec, I don't understand his objections to the ironic spec at all... | 15:22 |
| *** stendulker has joined #openstack-ironic | 15:22 | |
| *** novice_xyz has joined #openstack-ironic | 15:23 | |
| NobodyCam | humm I have not looked at the ceilometer spec :( | 15:23 |
| NobodyCam | I need to | 15:23 |
| jroll | it basically just says s/hardware.ipmi/hardware/ best I can tell | 15:25 |
| *** jlvillal has quit IRC | 15:27 | |
| *** jlvillal_ has joined #openstack-ironic | 15:28 | |
| *** rameshg87 has joined #openstack-ironic | 15:29 | |
| *** novice_xyz has quit IRC | 15:31 | |
| *** achanda has joined #openstack-ironic | 15:32 | |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic: Raise minimum API version to 1.1 https://review.openstack.org/152527 | 15:38 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Add support for driver-specific periodic tasks https://review.openstack.org/147523 | 15:40 |
| dtantsur | rloo, ^^^ | 15:40 |
| *** achanda has quit IRC | 15:40 | |
| rloo | dtantsur: ahh, ok, in middle of the logical name one; will look at yours next. | 15:40 |
| dtantsur | cool | 15:40 |
| *** zz_jgrimm is now known as jgrimm | 15:41 | |
| gilliard | Helloo Ironic! | 15:46 |
| *** coolsvap|afk is now known as coolsvap | 15:46 | |
| NobodyCam | good morning gilliard | 15:46 |
| jroll | hiya gilliard :) | 15:47 |
| *** coolsvap is now known as coolsvap_ | 15:50 | |
| *** kozhukalov has joined #openstack-ironic | 15:50 | |
| kozhukalov | jroll: around? | 15:51 |
| jroll | kozhukalov: what's up | 15:51 |
| kozhukalov | jroll, great i have a q about IPA | 15:51 |
| kozhukalov | in agent ironic driver there is a method which is called lookup | 15:52 |
| kozhukalov | and there is a comment there about why we need this method | 15:52 |
| devananda | Shrews: looks like somehow your patch broke a unit test | 15:52 |
| *** coolsvap_ is now known as coolsvap|afk | 15:52 | |
| devananda | which doesn' tmake sense to me yet | 15:52 |
| Shrews | devananda: yeah. looking at it | 15:52 |
| kozhukalov | as far as i understand it is called when a node is booted without uuid kernel parameter | 15:52 |
| jroll | kozhukalov: right... | 15:53 |
| kozhukalov | is it supposed that sometimes a node will be booted with uuid kernel param? | 15:53 |
| dtantsur | devananda, what was the final idea about having transient states like INSPECTED? (that's about https://review.openstack.org/#/c/147857/ ) | 15:53 |
| *** coolsvap|afk is now known as swapnilkulkarni | 15:54 | |
| jroll | kozhukalov: yes, if using ironic without using neutron for dhcp (static dhcp configs instead) | 15:54 |
| *** swapnilkulkarni is now known as swapnil | 15:54 | |
| *** swapnil is now known as swapnilk | 15:54 | |
| kozhukalov | jroll: kind of default tftp file? like /tftproot/pxelinux.cfg/default? right? | 15:55 |
| jroll | kozhukalov: static pxe configs | 15:55 |
| jroll | kozhukalov: for example in onmetal we have a single ipxe.txt that goes to all ramdisks, no node uuid in it (because our dhcp server has no concept of ironic existing) | 15:56 |
| kozhukalov | jroll: great, that clarified the picture, thanks a lot, now i understand | 15:57 |
| jroll | you're welcome | 15:57 |
| * NobodyCam jumps on conf call | 15:57 | |
| kozhukalov | jroll: we do the same for fuel, we have this default tftp file and all nodes boot with this ramdisk w/o any node specific kernel parameters | 15:58 |
| jroll | kozhukalov: right on :) | 15:58 |
| stendulker | jroll: Hi | 16:00 |
| jroll | stendulker: hello | 16:00 |
| stendulker | jroll: good morning | 16:00 |
| *** ijw has joined #openstack-ironic | 16:02 | |
| *** ijw has quit IRC | 16:02 | |
| stendulker | jroll: as discussed other day I implemented only _reboot_to_instance() method for ILOAgent driver VendorPassthru | 16:02 |
| *** ijw has joined #openstack-ironic | 16:02 | |
| stendulker | jroll: But this method do not get called, but method from agent.py gets called | 16:03 |
| jroll | oh, right | 16:03 |
| stendulker | jroll: But if I were to define heartbeat() as well, overridden method gets called | 16:03 |
| jroll | right. | 16:03 |
| stendulker | jroll: only with _reboot_to_instance() method pdb show something like this... | 16:04 |
| jroll | stendulker: yeah, I guess you need to override heartbeat(). that kinda sucks | 16:04 |
| jroll | that's just how python works, I wasn't thinking when I recommended it, I guess | 16:04 |
| stendulker | jroll: http://paste.openstack.org/show/166262/ | 16:04 |
| *** rameshg87 has quit IRC | 16:05 | |
| jroll | right, that seems normal | 16:05 |
| jroll | sorry, it was a bad recommendation | 16:05 |
| jroll | stendulker: so, remind me what the difference is in your _reboot_to_instance method? | 16:05 |
| stendulker | jroll: ok, I think, module points to agent.AgentVendorInterface instead of ILO one | 16:05 |
| jroll | right | 16:06 |
| stendulker | jroll: so I think I can avoid redefining heartbeat() | 16:06 |
| stendulker | jroll: I mean I cannot avoid redefining heartbeat() :) | 16:06 |
| jroll | right | 16:06 |
| jroll | so what's the difference with _reboot_to_instance here? | 16:07 |
| jroll | perhaps we can just improve that method, rather than re-implementing it | 16:07 |
| stendulker | jroll: Actuall I want to add calls to turn on the secure boot if that is the flavor requested, once deploy is complete | 16:07 |
| jroll | any reason we can't do that in the base agent driver? | 16:08 |
| stendulker | jroll: So wanted to avoid heartbeat() and do it in _rebbot_to_instance() | 16:08 |
| jroll | just call task.driver.management.set_boot_mode() or whatever? | 16:08 |
| stendulker | jroll: It would be a vendor specific call | 16:08 |
| stendulker | jroll: And not all drivers would have that call | 16:09 |
| jroll | right, but it's in the management interface, right? | 16:09 |
| jroll | it could just be ignored by other ManagementInterface implementations | 16:09 |
| jroll | or raise an exception | 16:09 |
| stendulker | jroll: It is. ok, i can handle the NotImplemented exception | 16:09 |
| jroll | yeah, let's just do that I think | 16:10 |
| jroll | I'd have to see the code | 16:10 |
| jroll | but I think it should be fine | 16:10 |
| stendulker | jroll: I think this should be possible | 16:10 |
| jroll | stendulker: awesome | 16:10 |
| stendulker | jroll: Was trying to constrain code within ilo driver :) | 16:10 |
| *** dkingshott has joined #openstack-ironic | 16:11 | |
| stendulker | jroll: thank you for your help. | 16:11 |
| jroll | stendulker: you're welcome | 16:11 |
| stendulker | jroll: One more thing, more of an academic query | 16:12 |
| jroll | yes? | 16:12 |
| stendulker | jroll: Why does the inheritance does not help in getting right _reboot_ti_instance() method during passthru call | 16:12 |
| stendulker | jroll: tried undersatnding base.py, but just got lost... | 16:13 |
| jroll | stendulker: because the AgentVendorInterface class doesn't know about its subclasses | 16:13 |
| jroll | _reboot_to_instance() is called by something in AgentVendorInterface, not by the base | 16:13 |
| stendulker | jroll: But when ILOAgent driver declared its class as new class, shouldn't it have taken that as 'self' | 16:14 |
| stendulker | jroll: And in the absence of heartbeat() call base class method | 16:14 |
| jroll | stendulker: anything in the ILOAgent class will take ILOAgent instance as 'self', but the _reboot_to_instance call isn't in the ILOAgent class | 16:15 |
| rloo | dtantsur: wrt 147523, how do I use the commit value mentioned in the commit msg, to get to the code? | 16:15 |
| stendulker | jroll: When reboot_to_instance() is re-defined in ILOAgent | 16:15 |
| jroll | rloo: 'git show $commit' or alternatively click it in the gerrit view | 16:16 |
| stendulker | jroll: scenario is ILOAgent has reboot_to_instance() but no heartbeat() | 16:16 |
| dtantsur | rloo, cd oslo-incubator && git checkout <hash> | 16:16 |
| jroll | stendulker: right, _reboot_to_instance() is called from AgentVendorInterface, it will start at that class and go up, it will not look down in to subclasses | 16:16 |
| rloo | jroll: I tried clicking it in the gerrit view and it didn't work. will try the other ways... | 16:16 |
| jroll | rloo: right, didn't realize it was in a different project | 16:17 |
| rloo | jroll: ahh, that's why. I keep forgetting oslo-incubator isn't 'openstack'... | 16:17 |
| *** pcrews has joined #openstack-ironic | 16:17 | |
| *** lazy_prince is now known as killer_prince | 16:17 | |
| jroll | rloo: it is... but it's not ironic | 16:17 |
| openstackgerrit | Merged openstack/ironic: Partial revert of 4606716 until we debug further https://review.openstack.org/152328 | 16:18 |
| stendulker | jroll: So it origically heartbeat() call came to ILOAgent which got redirected to Agent as heartbeat is not defined. And then it would always call Agent interfaces. | 16:18 |
| rloo | dtantsur: your suggestion didn't work. Are you certain you have the correct commit value? | 16:19 |
| jroll | stendulker: correct | 16:19 |
| stendulker | jroll: thank you | 16:19 |
| dtantsur | rloo, hmm.. works for me actually (both checkout and show). do you have up-to-date oslo-incubator? | 16:20 |
| jroll | dtantsur: rloo: that sha is a merge commit, I think you want fbd77a7b9b569210f0df18f33c90ef69c9166a2f | 16:20 |
| jroll | (is the actual commit) | 16:20 |
| jroll | and yeah, need to pull oslo-incubator | 16:20 |
| rloo | dtantsur: argh, i had an older version of oslo-incubator. | 16:21 |
| rloo | jroll: ++, that worked (in gerrit view) | 16:21 |
| Shrews | devananda: ugh. i may have just stumbled on a reason why your pre-emptive check needs to remain | 16:21 |
| Shrews | oh, hrm... may be a way around this | 16:24 |
| ChuckC | jroll: hi | 16:26 |
| *** datajerk has joined #openstack-ironic | 16:27 | |
| jroll | ChuckC: hello | 16:28 |
| ChuckC | jroll: question about the SF midcycle | 16:28 |
| jroll | uh huh | 16:28 |
| ChuckC | jroll: do you expect any empty seats? | 16:28 |
| ChuckC | wondering if I might be able to have folks try out a neutron patch for mac address updates? | 16:29 |
| jroll | ChuckC: this says 7 tickets left, you're welcome to grab one https://www.eventbrite.com/e/openstack-ironic-kilo-midcycle-sprint-in-san-francisco-tickets-15184923515 | 16:29 |
| jroll | you probably could, yes, it's a developer sprint, not a planning meeting | 16:29 |
| ChuckC | jroll: ya, I didn't want to take an ironic dev's spot, so thought I would ask first.... | 16:29 |
| jroll | ChuckC: yeah, spots are open, go ahead | 16:30 |
| ChuckC | jroll: thanks | 16:31 |
| *** Marga_ has joined #openstack-ironic | 16:31 | |
| jroll | no problem :) | 16:31 |
| *** Marga_ has quit IRC | 16:31 | |
| *** Marga_ has joined #openstack-ironic | 16:32 | |
| Shrews | devananda: So, w/o the pre-emptive check, if a node is DEPLOYING, we will return 400 (invalid state) whereas currently/previously we return 409 (conflict). So without it we are changing API behavior, which I'm guessing we do not want. | 16:32 |
| Shrews | The old check on rpc_node.provision_state and rpc_node.target_provision_state was essentially doing the same thing. | 16:34 |
| Shrews | given this new info, i think the reservation check is, unfortunately, needed for backward compatibility | 16:36 |
| NobodyCam | devananda: if you get a free minute can you take a look at https://review.openstack.org/#/c/150096 you have a -1 on it. but I think that was because the patch needed some time to bake in to master | 16:38 |
| openstackgerrit | Vladimir Kozhukalov proposed openstack/ironic-specs: Add a new driver for Fuel Agent https://review.openstack.org/138115 | 16:45 |
| *** jerryz has quit IRC | 16:46 | |
| *** dprince has quit IRC | 16:48 | |
| *** eghobo has joined #openstack-ironic | 16:50 | |
| *** vdrok has quit IRC | 16:51 | |
| *** athomas_ has quit IRC | 16:52 | |
| openstackgerrit | David Shrewsbury proposed openstack/ironic: Add MANAGEABLE state and associated transitions https://review.openstack.org/150073 | 16:54 |
| *** ChuckC__ has joined #openstack-ironic | 16:56 | |
| *** chuckC_ has quit IRC | 16:57 | |
| *** ChuckC__ is now known as ChuckC_ | 16:57 | |
| *** ChuckC has quit IRC | 17:00 | |
| *** Marga_ has quit IRC | 17:01 | |
| jlvillal_ | Is 'spacing' a defined unit of measurement? I'm looking at https://review.openstack.org/#/c/147523/3..9/doc/source/dev/architecture.rst | 17:02 |
| * jlvillal_ is new to Ironic, so trying to learn... | 17:02 | |
| *** athomas has joined #openstack-ironic | 17:02 | |
| jroll | jlvillal_: I believe it's seconds, but we should clarify that | 17:03 |
| jlvillal_ | If seconds I would prefer the argument to be 'seconds' :) I'm silly that way... | 17:04 |
| rloo | jroll, jlvillal_, yeah, it is secs, I mentioned it ;) | 17:04 |
| jroll | jlvillal_: this is a parameter to a thing in oslo, we just pass it along | 17:04 |
| * devananda wanders away, tired and hungry and grumpy | 17:05 | |
| devananda | see ya'll tomorrow :) | 17:05 |
| NobodyCam | have a good night devananda | 17:05 |
| jroll | later devananda :) | 17:05 |
| rloo | jlvillal_: The 'spacing' name was used in oslo's periodic_task decorator, this new decorator invokes the other one | 17:05 |
| jroll | don't be grumpy! | 17:05 |
| jlvillal_ | jroll: rloo: Thanks for the info | 17:05 |
| rloo | bye devananda | 17:05 |
| *** Nisha_away has quit IRC | 17:05 | |
| *** Nisha_away has joined #openstack-ironic | 17:05 | |
| *** jlvillal_ is now known as jlvillal | 17:05 | |
| rloo | jlvillal: but good questions. we're always open to improving our documentation, although sometimes it gets a bit difficult. I did suggest that we remove 'spacing' so people don't ask ;) | 17:06 |
| jlvillal | I would +1 that. | 17:08 |
| * rloo wonders why devananda was grumpy | 17:09 | |
| jlvillal | I've just started working on Ironic. So trying to learn by doing code-reviews and watching YouTube videos... | 17:09 |
| rloo | jlvillal: feel free to suggest it, the documentation is meant to help not confuse. | 17:09 |
| rloo | jlvillal: welcome. and doing code-reviews is a great way to ramp up. and helps us all at the same time. | 17:10 |
| jlvillal | rloo: Any idea who I would contact with problems signing the ICLA? OpenStack Individual Contributor License Agreement. They are working on getting me added to our corporate agreement. | 17:12 |
| jlvillal | But I keep getting an error when I try to sign the individual :( | 17:12 |
| *** ChuckC has joined #openstack-ironic | 17:13 | |
| *** rwsu-afk is now known as rwsu | 17:15 | |
| *** ChuckC_ has quit IRC | 17:16 | |
| jroll | jlvillal: #openstack-infra may be able to help | 17:22 |
| dtantsur | anyone else to review/approve driver-specific periodic tasks https://review.openstack.org/#/c/147523/ ? IIRC it's targeted to K2 | 17:23 |
| *** stelfer has quit IRC | 17:24 | |
| *** stelfer has joined #openstack-ironic | 17:25 | |
| jroll | dtantsur: after all these revisions that I've already +2'd, I think I want to test it locally | 17:26 |
| rloo | jroll: why test it now? ;) | 17:26 |
| dtantsur | jroll, lol sure | 17:26 |
| *** dprince has joined #openstack-ironic | 17:27 | |
| jlvillal | dtantsur: I had a question on: https://review.openstack.org/#/c/147523/9/ironic/tests/drivers/test_base.py | 17:27 |
| dtantsur | sure | 17:27 |
| *** derekh has quit IRC | 17:28 | |
| jlvillal | What is benefit of test on line 100, since same test on line 97. I'm trying to learn. | 17:28 |
| dtantsur | jlvillal, see that number has not changed? it's due to parallel=False on the second task | 17:29 |
| dtantsur | if it didn't work, number would be 2 | 17:29 |
| jlvillal | dtantsur: Or is it redundant? | 17:29 |
| jlvillal | dtantsur: Ah, thanks! | 17:29 |
| rloo | dtantsur: I wondered if people would understand that w/o a comment. | 17:29 |
| *** stelfer has quit IRC | 17:29 | |
| dtantsur | heh, our tests are not always clear :) | 17:30 |
| jroll | I was actually looking for a parallel=False test, and it took me a second to grok | 17:30 |
| rloo | dtantsur: yeah, that's why I didn't ask for a comment ;) | 17:30 |
| *** ifarkas has quit IRC | 17:31 | |
| *** lucasagomes has quit IRC | 17:33 | |
| *** jistr has quit IRC | 17:38 | |
| openstackgerrit | Sirushti Murugesan proposed openstack/ironic: Adds support for deploying whole disk images https://review.openstack.org/150142 | 17:42 |
| *** jcoufal has quit IRC | 17:44 | |
| *** krtaylor has quit IRC | 17:48 | |
| *** ijw has quit IRC | 17:48 | |
| *** Marga_ has joined #openstack-ironic | 17:49 | |
| *** ijw has joined #openstack-ironic | 17:49 | |
| *** kylestev has left #openstack-ironic | 17:51 | |
| openstackgerrit | Sirushti Murugesan proposed openstack/ironic: Adds support for deploying whole disk images https://review.openstack.org/150142 | 17:52 |
| *** krtaylor has joined #openstack-ironic | 17:54 | |
| *** kylestev has joined #openstack-ironic | 17:54 | |
| *** ChuckC_ has joined #openstack-ironic | 17:55 | |
| *** ChuckC has quit IRC | 17:59 | |
| *** MattMan has quit IRC | 18:02 | |
| *** spandhe has joined #openstack-ironic | 18:03 | |
| *** Nisha_brb has joined #openstack-ironic | 18:04 | |
| *** Nisha_away has quit IRC | 18:04 | |
| *** Marga_ has quit IRC | 18:08 | |
| *** ChuckC_ is now known as ChuckC | 18:13 | |
| *** athomas has quit IRC | 18:15 | |
| *** harlowja has joined #openstack-ironic | 18:15 | |
| *** Nisha_brb has quit IRC | 18:16 | |
| *** Nisha_brb has joined #openstack-ironic | 18:16 | |
| spandhe | anteaya: hey Anita, sorry I missed the meeting.. Actually I dont understand how I missed it.. I was planning on attending it.. I stayed up till 1:15 AM pacific time.. didnt see any activity on openstack-meetings.. and then I left | 18:16 |
| *** achanda has joined #openstack-ironic | 18:18 | |
| spandhe | anteaya: maybe something wrong with my IRC client.. I went through the meeting notes.. I actually wanted to get feedback on the script yesterday but unfortunately I missed the meeting | 18:18 |
| *** Marga_ has joined #openstack-ironic | 18:21 | |
| *** pradipta is now known as pradipta_away | 18:40 | |
| *** pelix has quit IRC | 18:50 | |
| NobodyCam | spandhe: our meeting are in openstack-meeting-3 | 18:51 |
| *** zer0c00l has quit IRC | 18:51 | |
| *** zer0c00l has joined #openstack-ironic | 18:51 | |
| NobodyCam | :( | 18:51 |
| *** mrmoje has quit IRC | 18:55 | |
| Nisha_brb | NobodyCam, wanted an opinion. In the https://review.openstack.org/#/c/103007/ (already merged), i wanted to replace "supported_boot_modes" with "BootMode" in code and its possible values with "uefi and bios" only. No secureboot in the list of possible values for this capability. Do i need to open the spec for this minor change or i can still get it through in code review | 18:56 |
| *** mrmoje has joined #openstack-ironic | 18:56 | |
| *** Nisha_brb is now known as Nisha | 18:58 | |
| NobodyCam | Nisha: we can always patch the spec if we need to... please point that out in the commit messages so reviewer know it going in | 19:00 |
| Nisha | commit message of the code review? | 19:01 |
| NobodyCam | yep | 19:01 |
| Nisha | ok thanks :) | 19:02 |
| *** stendulker has quit IRC | 19:14 | |
| *** swapnilk is now known as coolsvap_ | 19:15 | |
| *** coolsvap_ is now known as coolsvap | 19:19 | |
| *** dtantsur is now known as dtantsur|afk | 19:22 | |
| *** Guest81568 is now known as annegentle | 19:29 | |
| *** kozhukalov has quit IRC | 19:33 | |
| *** coolsvap is now known as coolsvap_ | 19:41 | |
| *** jrist has quit IRC | 19:48 | |
| *** shakamunyi has joined #openstack-ironic | 20:02 | |
| *** chlong has quit IRC | 20:05 | |
| *** Marga_ has quit IRC | 20:06 | |
| *** bradjones has quit IRC | 20:10 | |
| *** andreykurilin_ has joined #openstack-ironic | 20:12 | |
| *** bradjones has joined #openstack-ironic | 20:14 | |
| *** anderbubble has joined #openstack-ironic | 20:16 | |
| *** shakamunyi has quit IRC | 20:23 | |
| *** Haomeng has quit IRC | 20:32 | |
| mrda-away | Morning Ironic! | 20:32 |
| *** mrda-away is now known as mrda | 20:32 | |
| *** Haomeng has joined #openstack-ironic | 20:33 | |
| *** anteaya has quit IRC | 20:34 | |
| *** Marga_ has joined #openstack-ironic | 20:37 | |
| *** Hefeweizen has quit IRC | 20:38 | |
| *** Marga_ has quit IRC | 20:44 | |
| rloo | morning mrda. let me know if you have any questions about my comments to your logical names. | 20:48 |
| *** openstackgerrit has quit IRC | 20:50 | |
| mrda | thanks rloo - it's all good. | 20:51 |
| *** openstackgerrit has joined #openstack-ironic | 20:51 | |
| rloo | mrda: :) | 20:51 |
| mrda | I appreciate the reviews I got yersterday - hopefully I can address the comments and questions and we can get this merged RSN | 20:52 |
| *** anteaya has joined #openstack-ironic | 20:53 | |
| rloo | mrda: ++. (I'm around for another 2 hours or so. Otherwise, I can review tomorrow if it hasn't been approved yet. | 20:56 |
| mrda | thanks, I'll try and get a new revision up as quickly as I can | 20:57 |
| *** mrmoje has quit IRC | 20:59 | |
| rloo | mrda: no rush. there's still 2ish more days. | 21:00 |
| *** achanda has quit IRC | 21:03 | |
| mrda | rloo: I travel in 2 days to SFO, so it's a nice natural deadline | 21:06 |
| rloo | mrda: and the partying continues... ;) | 21:07 |
| JayF | I'm not sure having to fly across the world counts as a part | 21:07 |
| JayF | *party | 21:07 |
| JayF | lol | 21:07 |
| rloo | flight == rest, recover; rackspace-office == partying continues | 21:09 |
| mrda | lol | 21:10 |
| mrda | Aftyer 14.5 hours in a plane I'm not exactly ready for partying :) | 21:11 |
| *** Hefeweizen has joined #openstack-ironic | 21:21 | |
| *** achanda has joined #openstack-ironic | 21:21 | |
| NobodyCam | dprince: you around at all? | 21:23 |
| *** werhugenXYZ has joined #openstack-ironic | 21:37 | |
| *** yuriyz has quit IRC | 21:38 | |
| *** Marga_ has joined #openstack-ironic | 21:40 | |
| *** anderbubble has quit IRC | 21:41 | |
| *** Nisha has quit IRC | 21:41 | |
| *** anderbubble has joined #openstack-ironic | 21:41 | |
| *** yuriyz has joined #openstack-ironic | 21:42 | |
| anderbubble | I think I have Ironic pretty much working the way I expect now… with the caveat that a competing dhcp server on my network sometimes wins the race and ruins my day. :p | 21:42 |
| anderbubble | Thanks everyone for helping. | 21:42 |
| NobodyCam | anderbubble: woo hoo :) | 21:42 |
| anderbubble | I don’t think I ever said here, but mixing the dhcp-all-interfaces element into my image was a big win on my sans-eth0 nodes. | 21:43 |
| NobodyCam | oh nice :) | 21:44 |
| dprince | NobodyCam: hi, whats up? | 21:48 |
| NobodyCam | hey hey dprince | 21:48 |
| NobodyCam | had a question on that revert is 3 seconds enough? I was poking around and found this: https://review.openstack.org/#/c/121155/1 | 21:49 |
| dprince | NobodyCam: ah, yes. So it may be we should increase it to 10 | 21:51 |
| dprince | NobodyCam: honestly, I'm not seeing a personal issue w/ 3 though so perhaps we wait and keep it in mind. | 21:51 |
| NobodyCam | well I was thinking remove the new methods I added and just make the sleep configurable | 21:51 |
| dprince | NobodyCam: can you add this to the ticket? | 21:51 |
| dprince | NobodyCam: that we actually increased the sleep to 10 in Nova. | 21:52 |
| dprince | NobodyCam: code paths of slightly different all this bit of code was very similar | 21:52 |
| NobodyCam | dprince: almost exact | 21:52 |
| dprince | NobodyCam: code paths *ard* | 21:53 |
| * dprince can't type today | 21:53 | |
| NobodyCam | ya | 21:53 |
| NobodyCam | I can never type | 21:53 |
| NobodyCam | :-p | 21:53 |
| dprince | NobodyCam: yes. they are the same! | 21:53 |
| dprince | whew,, got it | 21:53 |
| NobodyCam | lol | 21:53 |
| NobodyCam | I believe the revert landed but I'll comment on the bug | 21:54 |
| dprince | NobodyCam: yes. And I would +1 the configurable sleep I think | 21:54 |
| NobodyCam | ack .. I may toss that up after lunch :0 | 21:54 |
| dprince | NobodyCam: this is one of those issues that appears to work in virt based CI or dev testing, but once you get on faster (real | 21:55 |
| dprince | NobodyCam: machines you are like, why did my installation just hang? | 21:55 |
| dprince | NobodyCam: but, nobody likes a hard coded sleep in the codebase either. | 21:55 |
| NobodyCam | ya and actually really hard to track down | 21:55 |
| *** foexle has quit IRC | 22:00 | |
| *** jlvillal has left #openstack-ironic | 22:19 | |
| *** jjohnson2 has quit IRC | 22:21 | |
| *** jlvillal has joined #openstack-ironic | 22:21 | |
| *** harlowja is now known as harlowja_away | 22:28 | |
| *** dprince has quit IRC | 22:34 | |
| *** davideagnello has quit IRC | 22:36 | |
| *** mjturek has quit IRC | 22:36 | |
| openstackgerrit | Erhan Ekici proposed openstack/ironic: Fixing Bug/1416298: f.write(password) -> f.write(str(password)) https://review.openstack.org/152702 | 22:46 |
| anderbubble | It looks like my instances aren’t getting iptables rules. I can ping them, but then I can’t ssh in (and port 22 is closed). How do the iptables rules get distributed to the boxes? What’s the chance that ssh just straight-up isn’t running on the centos7 image generated by disk-image-create? | 22:51 |
| openstackgerrit | Erhan Ekici proposed openstack/ironic: Fixing Bug/1416298: f.write(password) -> f.write(str(password)) https://review.openstack.org/152702 | 22:52 |
| *** ChuckC_ has joined #openstack-ironic | 22:55 | |
| *** ChuckC has quit IRC | 22:57 | |
| *** EmilienM is now known as EmilienM|afk | 23:03 | |
| *** jrist has joined #openstack-ironic | 23:03 | |
| *** ChuckC has joined #openstack-ironic | 23:03 | |
| *** ChuckC_ has quit IRC | 23:07 | |
| *** victor_lowther has quit IRC | 23:08 | |
| *** andreykurilin_ has quit IRC | 23:15 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements https://review.openstack.org/152713 | 23:16 |
| *** werhugenXYZ has quit IRC | 23:16 | |
| *** victor_lowther has joined #openstack-ironic | 23:17 | |
| *** victor_lowther has quit IRC | 23:20 | |
| *** victor_lowther has joined #openstack-ironic | 23:20 | |
| *** lucasagomes has joined #openstack-ironic | 23:30 | |
| *** harlowja_away is now known as harlowja | 23:32 | |
| *** andreykurilin_ has joined #openstack-ironic | 23:33 | |
| *** digimonde has joined #openstack-ironic | 23:34 | |
| *** digimonde has quit IRC | 23:40 | |
| *** ijw has quit IRC | 23:41 | |
| *** lucasagomes has quit IRC | 23:43 | |
| *** ChuckC_ has joined #openstack-ironic | 23:45 | |
| *** ChuckC has quit IRC | 23:46 | |
| *** jgrimm is now known as zz_jgrimm | 23:47 | |
| *** anderbubble has quit IRC | 23:51 | |
| openstackgerrit | Chris Krelle proposed openstack/ironic: remove new non-working iSCSI verify and refresh methods https://review.openstack.org/152734 | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
