| *** [1]cdearborn has joined #openstack-ironic | 00:06 | |
| *** rama_y has quit IRC | 00:09 | |
| *** piet has quit IRC | 00:09 | |
| openstackgerrit | Michael Davies proposed openstack/python-ironicclient: Add env var for version cache timeout https://review.openstack.org/334387 | 00:12 |
|---|---|---|
| *** sdake has quit IRC | 00:15 | |
| *** jaybeale has quit IRC | 00:33 | |
| *** jaybeale has joined #openstack-ironic | 00:35 | |
| *** aNupoisc has quit IRC | 00:42 | |
| *** klindgren_ has quit IRC | 00:42 | |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [ilo] https://review.openstack.org/301947 | 00:43 |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [inspector] https://review.openstack.org/301955 | 00:44 |
| openstackgerrit | Merged openstack/ironic: Remove unnecessary calls to dict.keys() https://review.openstack.org/334476 | 00:45 |
| *** jaybeale has quit IRC | 00:45 | |
| *** jaybeale has joined #openstack-ironic | 00:45 | |
| *** Sukhdev has joined #openstack-ironic | 00:50 | |
| *** jaybeale has quit IRC | 00:50 | |
| *** jaybeale has joined #openstack-ironic | 00:56 | |
| openstackgerrit | Merged openstack/ironic: Fix parameter create-node.sh https://review.openstack.org/325732 | 00:57 |
| openstackgerrit | Merged openstack/ironic: Fail early if ramdisk type is dib, and not building https://review.openstack.org/328246 | 00:57 |
| openstackgerrit | Merged openstack/ironic: Fix releasenotes formatting error https://review.openstack.org/334712 | 00:58 |
| *** joprovost has joined #openstack-ironic | 01:00 | |
| *** baoli has joined #openstack-ironic | 01:01 | |
| *** baoli has quit IRC | 01:01 | |
| *** baoli has joined #openstack-ironic | 01:02 | |
| openstackgerrit | Michael Davies proposed openstack/python-ironicclient: Add env var for version cache timeout https://review.openstack.org/334387 | 01:04 |
| *** joprovost1 has joined #openstack-ironic | 01:05 | |
| *** joprovost has quit IRC | 01:06 | |
| *** joprovost1 is now known as joprovost | 01:06 | |
| *** piet has joined #openstack-ironic | 01:08 | |
| *** yibo has joined #openstack-ironic | 01:10 | |
| *** rcernin has quit IRC | 01:11 | |
| *** baoli has quit IRC | 01:15 | |
| openstackgerrit | Michael Davies proposed openstack/python-ironicclient: Add env var for version cache timeout https://review.openstack.org/334387 | 01:15 |
| *** stendulker has joined #openstack-ironic | 01:17 | |
| *** baoli has joined #openstack-ironic | 01:19 | |
| *** baoli_ has joined #openstack-ironic | 01:21 | |
| *** yuanying has quit IRC | 01:22 | |
| *** piet has quit IRC | 01:22 | |
| *** baoli has quit IRC | 01:24 | |
| *** yuanying has joined #openstack-ironic | 01:25 | |
| *** sabeen has quit IRC | 01:33 | |
| *** jaybeale has quit IRC | 01:36 | |
| *** PollyZ has joined #openstack-ironic | 01:40 | |
| *** amotoki has joined #openstack-ironic | 01:42 | |
| *** amotoki has quit IRC | 01:46 | |
| openstackgerrit | Yibo Cai proposed openstack/ironic: Support multi arch deployment https://review.openstack.org/322632 | 01:49 |
| *** Sukhdev has quit IRC | 01:52 | |
| *** piet has joined #openstack-ironic | 01:56 | |
| *** sacharya has joined #openstack-ironic | 01:59 | |
| openstackgerrit | Yuiko Takada Mori proposed openstack/ironic: Follow-up patch of 9a1aeb76da2ed53e042a94ead8640af9374a10bf https://review.openstack.org/334731 | 02:16 |
| *** fragatina has joined #openstack-ironic | 02:18 | |
| *** fragatina has quit IRC | 02:18 | |
| *** fragatina has joined #openstack-ironic | 02:19 | |
| *** sacharya_ has joined #openstack-ironic | 02:24 | |
| *** rbudden has quit IRC | 02:27 | |
| openstackgerrit | Ruby Loo proposed openstack/python-ironicclient: Simplify use of config_drive arg in osc https://review.openstack.org/334735 | 02:27 |
| *** sacharya has quit IRC | 02:27 | |
| *** stendulker has quit IRC | 02:28 | |
| *** jaybeale has joined #openstack-ironic | 02:30 | |
| openstackgerrit | Michael Davies proposed openstack/python-ironicclient: Add env var for version cache timeout https://review.openstack.org/334387 | 02:31 |
| *** mtreinish has quit IRC | 02:36 | |
| *** mtreinish has joined #openstack-ironic | 02:36 | |
| *** jaybeale has quit IRC | 02:36 | |
| *** chlong has quit IRC | 02:37 | |
| *** PollyZ has quit IRC | 02:37 | |
| *** jaybeale has joined #openstack-ironic | 02:38 | |
| *** PollyZ has joined #openstack-ironic | 02:39 | |
| *** rloo has quit IRC | 02:39 | |
| *** sdake has joined #openstack-ironic | 02:40 | |
| *** PollyZ_ has joined #openstack-ironic | 02:44 | |
| *** PollyZ has quit IRC | 02:44 | |
| *** hoangcx has joined #openstack-ironic | 02:44 | |
| *** jaybeale has quit IRC | 02:48 | |
| *** jaybeale has joined #openstack-ironic | 02:49 | |
| *** chlong has joined #openstack-ironic | 02:50 | |
| *** yuanying has quit IRC | 02:50 | |
| *** sdake has quit IRC | 02:56 | |
| *** joprovost has quit IRC | 02:58 | |
| *** jaybeale has quit IRC | 03:05 | |
| *** sdake has joined #openstack-ironic | 03:08 | |
| *** [1]cdearborn has quit IRC | 03:11 | |
| *** sabeen has joined #openstack-ironic | 03:19 | |
| *** PollyZ_ has quit IRC | 03:27 | |
| *** sdake has quit IRC | 03:27 | |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [ipmi] https://review.openstack.org/301967 | 03:31 |
| openstackgerrit | Merged openstack/ironic: Centralize config options - [irmc] https://review.openstack.org/302476 | 03:31 |
| *** PollyZ has joined #openstack-ironic | 03:32 | |
| *** amotoki has joined #openstack-ironic | 03:32 | |
| *** MattMan has quit IRC | 03:40 | |
| *** MattMan has joined #openstack-ironic | 03:41 | |
| *** PollyZ has quit IRC | 03:44 | |
| openstackgerrit | Dao Cong Tien proposed openstack/ironic: Nova-compatible serial console: socat console_utils https://review.openstack.org/328168 | 03:46 |
| *** praneshp has joined #openstack-ironic | 03:47 | |
| *** yuanying has joined #openstack-ironic | 03:48 | |
| *** jjohnson2 has quit IRC | 03:48 | |
| *** PollyZ has joined #openstack-ironic | 03:48 | |
| *** persia has quit IRC | 03:49 | |
| *** sabeen has joined #openstack-ironic | 03:49 | |
| *** links has joined #openstack-ironic | 03:50 | |
| *** persia has joined #openstack-ironic | 03:56 | |
| *** yuanying has quit IRC | 04:00 | |
| *** yuanying has joined #openstack-ironic | 04:01 | |
| *** PollyZ has quit IRC | 04:02 | |
| *** appu103_ has joined #openstack-ironic | 04:07 | |
| *** appu103_ is now known as aparnav | 04:07 | |
| *** baoli_ has quit IRC | 04:10 | |
| *** PollyZ has joined #openstack-ironic | 04:13 | |
| *** rama_y has joined #openstack-ironic | 04:15 | |
| *** sabeen has quit IRC | 04:20 | |
| *** sabeen has joined #openstack-ironic | 04:21 | |
| *** jtomasek_ has joined #openstack-ironic | 04:32 | |
| *** PollyZ has quit IRC | 04:36 | |
| *** amotoki has quit IRC | 04:37 | |
| *** fragatina has quit IRC | 04:39 | |
| *** PollyZ has joined #openstack-ironic | 04:41 | |
| *** sabeen has quit IRC | 04:44 | |
| *** sabeen has joined #openstack-ironic | 04:45 | |
| *** rama_y has quit IRC | 04:46 | |
| *** rcernin has joined #openstack-ironic | 04:47 | |
| *** yuanying has quit IRC | 04:49 | |
| *** rama_y has joined #openstack-ironic | 04:49 | |
| *** PollyZ has quit IRC | 04:52 | |
| *** amotoki has joined #openstack-ironic | 04:54 | |
| *** PollyZ has joined #openstack-ironic | 05:05 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic: Add validation of 'ilo_deploy_iso' in deploy.validate() https://review.openstack.org/334774 | 05:12 |
| *** jtomasek_ has quit IRC | 05:12 | |
| *** itamarl has joined #openstack-ironic | 05:13 | |
| *** M00nr41n has quit IRC | 05:13 | |
| *** ayoung has quit IRC | 05:14 | |
| *** fragatina has joined #openstack-ironic | 05:20 | |
| *** yibo has quit IRC | 05:27 | |
| *** ChubYann has quit IRC | 05:27 | |
| *** yogi has joined #openstack-ironic | 05:29 | |
| *** yibo has joined #openstack-ironic | 05:29 | |
| *** amotoki has quit IRC | 05:31 | |
| *** yogi_ has joined #openstack-ironic | 05:34 | |
| *** yogi_ has quit IRC | 05:34 | |
| *** piet has quit IRC | 05:35 | |
| *** yogi_ has joined #openstack-ironic | 05:36 | |
| *** rcernin has quit IRC | 05:37 | |
| *** rama_y has quit IRC | 05:37 | |
| *** moshele has joined #openstack-ironic | 05:39 | |
| *** amotoki has joined #openstack-ironic | 05:42 | |
| *** skramaja has joined #openstack-ironic | 05:42 | |
| *** rajinir has quit IRC | 05:44 | |
| *** praneshp has quit IRC | 05:47 | |
| *** jtomasek has quit IRC | 05:55 | |
| *** amotoki has quit IRC | 05:58 | |
| vdrok | sambetts|afk: http://logs.openstack.org/93/303093/9/check/gate-grenade-dsvm-ironic/1d65948/logs/grenade.sh.txt.gz not sure if it's some network glitch or not | 06:01 |
| * vdrok is afk today | 06:02 | |
| *** Sukhdev has joined #openstack-ironic | 06:02 | |
| *** praneshp has joined #openstack-ironic | 06:04 | |
| *** yuanying has joined #openstack-ironic | 06:05 | |
| *** amotoki has joined #openstack-ironic | 06:05 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Throwing an exception when creating a node with tags https://review.openstack.org/323680 | 06:05 |
| *** M00nr41n has joined #openstack-ironic | 06:07 | |
| *** rcernin has joined #openstack-ironic | 06:09 | |
| *** yuanying has quit IRC | 06:11 | |
| *** jaybeale has joined #openstack-ironic | 06:11 | |
| *** yuanying has joined #openstack-ironic | 06:12 | |
| *** jaybeale has quit IRC | 06:12 | |
| *** PollyZ has quit IRC | 06:13 | |
| *** jaybeale has joined #openstack-ironic | 06:14 | |
| *** mjura has joined #openstack-ironic | 06:15 | |
| *** _vsaienko has joined #openstack-ironic | 06:16 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Improve the readability of configuration drive doc part https://review.openstack.org/323675 | 06:17 |
| *** Sukhdev has quit IRC | 06:19 | |
| *** jaybeale has quit IRC | 06:19 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Improve the readability of configuration drive doc part https://review.openstack.org/323675 | 06:19 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic-specs: Enhance Driver Interface for Soft Power Off and Inject NMI https://review.openstack.org/186700 | 06:22 |
| *** jaybeale has joined #openstack-ironic | 06:23 | |
| *** jaybeale has quit IRC | 06:24 | |
| *** davidlenwell has quit IRC | 06:30 | |
| *** hshiina has joined #openstack-ironic | 06:30 | |
| *** _vsaienko has quit IRC | 06:32 | |
| *** derekh has joined #openstack-ironic | 06:35 | |
| *** davidlenwell has joined #openstack-ironic | 06:40 | |
| *** derekh has quit IRC | 06:43 | |
| openstackgerrit | Gerard Braad proposed openstack/ironic-webclient: Add install instructions to README https://review.openstack.org/333754 | 06:43 |
| *** pcaruana has joined #openstack-ironic | 06:46 | |
| *** rbartal has joined #openstack-ironic | 06:49 | |
| *** livelace has joined #openstack-ironic | 06:56 | |
| *** tesseract- has joined #openstack-ironic | 06:59 | |
| *** daemontool has joined #openstack-ironic | 07:04 | |
| *** praneshp has quit IRC | 07:12 | |
| openstackgerrit | Tang Chen proposed openstack/python-ironicclient: Use osc_lib instead of cliff https://review.openstack.org/334803 | 07:15 |
| *** yogi_ has quit IRC | 07:19 | |
| *** yogi has quit IRC | 07:20 | |
| *** athomas has joined #openstack-ironic | 07:30 | |
| *** ifarkas has joined #openstack-ironic | 07:31 | |
| *** yuanying has quit IRC | 07:34 | |
| *** yuanying has joined #openstack-ironic | 07:34 | |
| *** yuanying has quit IRC | 07:35 | |
| *** yuanying has joined #openstack-ironic | 07:36 | |
| openstackgerrit | Moshe Levi proposed openstack/ironic-python-agent: WIP Infiniband Support https://review.openstack.org/334816 | 07:47 |
| openstackgerrit | Tang Chen proposed openstack/python-ironicclient: Use osc_lib instead of cliff https://review.openstack.org/334803 | 07:50 |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:01 | |
| *** sacharya_ has quit IRC | 08:01 | |
| *** dmk0202 has joined #openstack-ironic | 08:06 | |
| alineb | morning all! | 08:07 |
| milan | morning alineb | 08:10 |
| milan | #pixiesay morning Ironic -m flexing | 08:10 |
| PixieBoots | ᕙʕ⇀ᴥ⇀ʔᕗ: morning Ironic | 08:10 |
| ifarkas | morning milan alineb and all | 08:13 |
| alineb | o/ milan ifarkas | 08:13 |
| *** _milan_ has joined #openstack-ironic | 08:14 | |
| *** dtantsur|afk is now known as dtantsur | 08:15 | |
| dtantsur | Morning Ironic, morning alineb, milan, ifarkas | 08:15 |
| *** milan has quit IRC | 08:17 | |
| _milan_ | morning dtantsur :) | 08:27 |
| _milan_ | dtantsur, I've just checked your comment | 08:27 |
| moshele | dtantsur: morning, I updated the IPA patch for vendor and product idshttps://review.openstack.org/#/c/279940/ also can you review the infiniband spec https://review.openstack.org/#/c/265635/ | 08:27 |
| _milan_ | dtantsur, the bad thing is smoke tests run before upgrade and then no more after | 08:28 |
| _milan_ | dtantsur, I'm not sure how to label our tempest test to be executed and where to put it | 08:28 |
| dtantsur | moshele, will check today | 08:30 |
| mgould | morning Ironic! | 08:30 |
| dtantsur | _milan_, mmm, good question, probably jlvillal or vdrok know that... | 08:30 |
| mgould | morning alineb _milan_ dtantsur ifarkas | 08:30 |
| mgould | morning moshele | 08:30 |
| dtantsur | morning mgould | 08:30 |
| moshele | morning mgould | 08:31 |
| dtantsur | _milan_, the grenade job is merged \o/ | 08:31 |
| _milan_ | dtantsur, yeah :D | 08:31 |
| _milan_ | \o/ merged experimental grenade for inspector | 08:31 |
| mgould | \o/ | 08:31 |
| alineb | morning mgould dtantsur | 08:32 |
| _milan_ | dtantsur, ack, I'll lift the WIP embargo from the grenade plugin and add the devstack and grenade patches it depends on | 08:32 |
| _milan_ | i'll update the qe ether pad for discussion on Wednesday QE meeting so people can elaborate on it | 08:33 |
| _milan_ | * discuss how to make the best use of inspector grenade & inspector tempest | 08:33 |
| *** jtomasek_ has joined #openstack-ironic | 08:34 | |
| _milan_ | dtantsur, another possibility is merge the plug-in and patch it as necessary if/once we decide about the test case selection | 08:34 |
| *** yogi has joined #openstack-ironic | 08:36 | |
| *** daemontool has quit IRC | 08:40 | |
| lucasagomes | morning all | 08:40 |
| dtantsur | morning lucasagomes | 08:40 |
| _milan_ | morning lucasagomes :) | 08:40 |
| _milan_ | oh, and morning mgould :) | 08:41 |
| dtantsur | _milan_, well, I'd prefer we at least try inspection after the upgrade :) but I'm fine with merging it and iterating further | 08:41 |
| _milan_ | dtantsur, yeah | 08:41 |
| *** mbound has joined #openstack-ironic | 08:42 | |
| dtantsur | _milan_, hmm https://github.com/openstack/ironic/commit/ac7f9cea9c312e7a2dfb9e1f23bfee0658bf1926 | 08:42 |
| _milan_ | dtantsur, interesting | 08:43 |
| dtantsur | vdrok, lol, awesome user pic https://github.com/openvdro :D | 08:43 |
| * dtantsur gets the joke, but people not speaking Russian probably won't | 08:44 | |
| dtantsur | _milan_, maybe that's a difference between a local environment and the gate? | 08:45 |
| *** vmud213 has joined #openstack-ironic | 08:46 | |
| _milan_ | dtantsur, could be, some things get set in the config of the plugin some in the job template, it's a bit difficult for me to figure out what to put where | 08:46 |
| dtantsur | _milan_, this should take care of it: https://github.com/openstack-infra/devstack-gate/blob/3ef48cec70c771242d743467768f22d1ecb55ecb/devstack-vm-gate.sh#L748-L791 | 08:47 |
| dtantsur | lets try in the gate, the new job should be available soon | 08:47 |
| _milan_ | yeah, I'm a bit more inclined to spec this in the job template cause it's not up to the plugin to say what tests a gate job should run imho | 08:48 |
| _milan_ | I mean if we need some tempest regex or other VARs set | 08:49 |
| dtantsur | yep | 08:49 |
| _milan_ | dtantsur, cool, so lemme lift the WIP and add the depends-on things in the plug-in | 08:50 |
| _milan_ | check experimental should then run through the grenade | 08:50 |
| dtantsur | yep | 08:52 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Introducing Inspector Grenade testing https://review.openstack.org/327667 | 08:53 |
| dtantsur | _milan_, don't we need to also enable ironic devstack plugin? | 08:57 |
| _milan_ | dtantsur, I've discussed that with vdrok and jlvillal on Friday and we concluded it should be part of the job template (it is now) | 08:57 |
| _milan_ | otherwise inspector would have to micro-manage ironic stuff :-/ | 08:58 |
| _milan_ | (me ran into that issue on Friday) | 08:58 |
| dtantsur | okie | 08:58 |
| * _milan_ checks experimental && fingers-cross | 08:58 | |
| dtantsur | that's awesome, thanks for working on it | 09:00 |
| _milan_ | dtantsur, thanks :) | 09:03 |
| *** jistr is now known as jistr|mtg | 09:04 | |
| *** yogi has quit IRC | 09:06 | |
| *** mjura_ has joined #openstack-ironic | 09:09 | |
| dtantsur | lucasagomes, re soft power off: we should probably start thinking how these parameters are to be used | 09:10 |
| lucasagomes | dtantsur, right | 09:11 |
| dtantsur | lucasagomes, I don't see point in setting timeout per boot request. it's something bound to an instance (or rather image) | 09:11 |
| dtantsur | fallback or not may be set per request on the other hand | 09:11 |
| lucasagomes | dtantsur, exactly | 09:11 |
| lucasagomes | dtantsur, that's why I think it's hard to have it in the driver_info | 09:11 |
| dtantsur | instance_info then? :) | 09:11 |
| lucasagomes | hoenstly, it may be related to the user usage of the image | 09:11 |
| lucasagomes | because it may take a longer time depending on how many process it has running, fs mounted etc | 09:12 |
| dtantsur | yeah, so I think the timeout is an instance property | 09:12 |
| *** mjura has quit IRC | 09:12 | |
| lucasagomes | dtantsur, still, I think the user using the CLI don't want to update neither instance_info or driver_info before invoking soft power off | 09:13 |
| dtantsur | I'm not that decided about the "fallback" thing | 09:13 |
| lucasagomes | yeah, that on a per-node basis seems pretty incosistent to me (the fallback) | 09:13 |
| dtantsur | lucasagomes, yes, but as I said, this is not something that should be set per request, it's an instance property for real | 09:13 |
| dtantsur | like, our deploy API does not set instance_info/disk_gb | 09:13 |
| gbraad | dtantsur: had some time again to look into my ironic/bifrost issue from yesterday: seems there is no agent-ipmitool | 09:14 |
| lucasagomes | dtantsur, but how that gets passed to the instance_info? | 09:14 |
| dtantsur | lucasagomes, node-update, like everything else there | 09:14 |
| gbraad | thanks again for the pointers. but it seems the markup is as used by openstack-infra/shade | 09:14 |
| dtantsur | I see | 09:14 |
| lucasagomes | dtantsur, well, the tenant should not update it right? it's more like an operator thing | 09:14 |
| lucasagomes | I can't assume someone has operator credentials to actually be able to issue a soft power off | 09:15 |
| dtantsur | lucasagomes, all our API is admin-only currently | 09:15 |
| dtantsur | so yes, you can | 09:15 |
| lucasagomes | dtantsur, we agreed on policies | 09:15 |
| lucasagomes | lemme see also if nova allows soft pwoer off | 09:15 |
| dtantsur | then we'll have to reconsider updating as well, because a lot of things depend on values in our magic JSONs | 09:15 |
| dtantsur | and I agree that's bad | 09:15 |
| dtantsur | lucasagomes, my second thought on the previous patchset was: if we allow timeout, it should be for all power actions, not only soft power off | 09:16 |
| dtantsur | so e.g. it should work for hard power off too | 09:16 |
| lucasagomes | dtantsur, I tend to agree with that | 09:17 |
| lucasagomes | --timeout seems geenric enough that could be coded for all operations | 09:17 |
| dtantsur | I'm fine with this option too. I just don't feel comfortable adding parameters only working for soft power off | 09:17 |
| lucasagomes | but it's outside the scope of the patch per-si | 09:17 |
| lucasagomes | perhaps, what if | 09:17 |
| lucasagomes | in that spec, let's have it all as ironic.conf options | 09:17 |
| lucasagomes | after that we can work on timeout being overriden for both hard and soft options | 09:18 |
| dtantsur | I'm fine with that, I'm not sure who requested user-settable options | 09:18 |
| lucasagomes | dtantsur, I think devananda 2 patch-sets before | 09:18 |
| dtantsur | hmm | 09:19 |
| openstackgerrit | Tan Lin proposed openstack/ironic-specs: Support rolling upgrades https://review.openstack.org/299245 | 09:19 |
| lucasagomes | dtantsur, https://review.openstack.org/#/c/186700/29/specs/approved/enhance-power-interface-for-soft-reboot-and-nmi.rst | 09:21 |
| lucasagomes | L167 | 09:21 |
| dtantsur | lucasagomes, so I think we can 1. have a common "timeout" option, 2. Do not fall back to hard power off for now. WDYT? | 09:22 |
| lucasagomes | dtantsur, seems fine... | 09:23 |
| lucasagomes | I'm not against fall back but I think it should be an ironic.conf thing | 09:23 |
| lucasagomes | so it's consistent across all nodes | 09:23 |
| dtantsur | to be honest, I'm not sure fall back is a good thing.. users can always request a hard power off if they want | 09:23 |
| dtantsur | on the other hand, hard fall back can destroy data | 09:24 |
| lucasagomes | yeah | 09:24 |
| lucasagomes | it's not a big deal to issue a hard power off if needed | 09:24 |
| dtantsur | ++ | 09:25 |
| dtantsur | lucasagomes, wanna summarize this discussion on the spec? | 09:25 |
| *** yuanying has quit IRC | 09:26 | |
| *** Nisha_away has joined #openstack-ironic | 09:26 | |
| lucasagomes | dtantsur, could be, do you mind doing it ? (Maybe just c&p the conversation!? | 09:27 |
| lucasagomes | ) | 09:27 |
| Nisha_away | dtantsur, need some help on inspector.... | 09:27 |
| *** amotoki has quit IRC | 09:27 | |
| dtantsur | lucasagomes, I can, but I'm on meeting right now | 09:29 |
| dtantsur | Nisha_away, ditto, I can try helping, but I have a lot of meeting ahead | 09:30 |
| Nisha_away | dtantsur, basically the server is not able to recv dhcp requests | 09:30 |
| Nisha_away | dtantsur, i have br-inspector, the inspector dhcp service and ironic dhcp server in same range | 09:31 |
| *** jistr|mtg is now known as jistr | 09:31 | |
| Nisha_away | dtantsur, pxe deploy works in the same env perfectly fine | 09:32 |
| *** derekh has joined #openstack-ironic | 09:33 | |
| Nisha_away | inspector is not able to boot up the ramdisk as dhcp is not reaching server | 09:33 |
| dtantsur | Nisha_away, probably DHCP misconfiguration, check http://docs.openstack.org/developer/ironic-inspector/install.html#configuration we've updated it recently | 09:35 |
| Nisha_away | dtantsur, thanks let me check | 09:39 |
| lucasagomes | yuriyz, around? I think I don't get the suggestion here https://review.openstack.org/#/c/327642/3/ironic/conductor/utils.py ? | 09:39 |
| lucasagomes | yuriyz, setting target_power_state to NOSTATE at the db api level for cleaning? But wouldn't that be outside of the patch scope? | 09:41 |
| lucasagomes | can be done, but I don't think we should change it in that patch. What you think? | 09:41 |
| *** david-lyle has quit IRC | 09:44 | |
| *** david-lyle has joined #openstack-ironic | 09:46 | |
| *** yogi has joined #openstack-ironic | 09:53 | |
| *** _vsaienko has joined #openstack-ironic | 09:54 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Implement new heartbeat for AgentDeploy https://review.openstack.org/333964 | 10:00 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Promote agent vendor passthru to core API https://review.openstack.org/330018 | 10:00 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: [WIP] Use new agent API if available https://review.openstack.org/334523 | 10:01 |
| Nisha_away | dtantsur, br-inspector ip has some settings for inspector? | 10:01 |
| dtantsur | Nisha_away, sorry, I don't get the question | 10:02 |
| Nisha_away | dtantsur, earlier (long back) when i had tried inspector i had manually done following: 1. create neutron port for br-inspector's mac with in the same dhcp ramge as of ironic 2. and then assigned the ip address to the br-inspector. | 10:03 |
| Nisha_away | dtantsur, now same is not working, | 10:04 |
| dtantsur | no neutron ports please, it's not needed | 10:04 |
| dtantsur | please follow the documentation or see our devstack plugin | 10:04 |
| dtantsur | https://github.com/openstack/ironic-inspector/blob/master/devstack/plugin.sh | 10:04 |
| Nisha_away | dtantsur, ok. thanks | 10:04 |
| Nisha_away | dtantsur, one more ques, we dont require neutron port to define the dhcp range at all? | 10:05 |
| Nisha_away | means like we do for ironic? | 10:05 |
| *** hoangcx has quit IRC | 10:06 | |
| dtantsur | Nisha_away, no, inspector does not interact with or require neutron | 10:06 |
| Nisha_away | ohk. thanks | 10:06 |
| Nisha_away | i will then refer the plugin | 10:06 |
| Nisha_away | that shud help me for configuration | 10:06 |
| *** baoli has joined #openstack-ironic | 10:09 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Add power_transition_started_at to the nodes table https://review.openstack.org/327642 | 10:10 |
| *** baoli_ has joined #openstack-ironic | 10:11 | |
| sambetts|afk | Morning all | 10:14 |
| *** sambetts|afk is now known as sambetts | 10:14 | |
| *** baoli has quit IRC | 10:14 | |
| * sambetts is zombie mode today and mostly focused downstream because of deadlines :'( | 10:14 | |
| dtantsur | morning sambetts | 10:16 |
| dtantsur | #pixiesay deadlines.. -m sad | 10:16 |
| PixieBoots | ʕ•͡ᴥ•ʔ: deadlines.. | 10:16 |
| dtantsur | oh, we don't have sad emotion, do we? | 10:16 |
| dtantsur | #pixiesay deadlines.. -m dead | 10:17 |
| dtantsur | #pixiesay deadlines.. -m dead | 10:17 |
| PixieBoots | ʕxᴥxʔ: deadlines.. | 10:17 |
| *** amotoki has joined #openstack-ironic | 10:17 | |
| sambetts | heh | 10:17 |
| * dtantsur is not the smartest person ever today | 10:17 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Implement new heartbeat for AgentDeploy https://review.openstack.org/333964 | 10:20 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Promote agent vendor passthru to core API https://review.openstack.org/330018 | 10:20 |
| * _milan_ relocates | 10:21 | |
| *** _milan_ has quit IRC | 10:22 | |
| openstackgerrit | Aline Bousquet proposed openstack/ironic-lib: Add support for metrics https://review.openstack.org/301526 | 10:25 |
| lucasagomes | hi folks, if you have some time... mind taking a look at https://review.openstack.org/#/c/248832/ ? | 10:28 |
| lucasagomes | cheers | 10:28 |
| *** amotoki has quit IRC | 10:30 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Implement new heartbeat for AgentDeploy https://review.openstack.org/333964 | 10:30 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Promote agent vendor passthru to core API https://review.openstack.org/330018 | 10:30 |
| *** amotoki has joined #openstack-ironic | 10:35 | |
| *** yogi__ has joined #openstack-ironic | 10:36 | |
| *** yogi__ has quit IRC | 10:40 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: [WIP] Implement new heartbeat for AgentDeploy https://review.openstack.org/333964 | 10:41 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Promote agent vendor passthru to core API https://review.openstack.org/330018 | 10:41 |
| *** Nisha_away has quit IRC | 10:41 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-python-agent: [WIP] Use new agent API if available https://review.openstack.org/334523 | 10:44 |
| dtantsur | this is getting closer ^^^ | 10:44 |
| *** yuanying has joined #openstack-ironic | 10:47 | |
| *** dtantsur is now known as dtantsur|lunch | 10:48 | |
| *** daemontool has joined #openstack-ironic | 10:48 | |
| *** amotoki has quit IRC | 10:50 | |
| *** jcoufal has joined #openstack-ironic | 10:54 | |
| *** vmud213 has quit IRC | 11:03 | |
| *** fragatin_ has joined #openstack-ironic | 11:22 | |
| *** moshele has quit IRC | 11:22 | |
| *** fragatina has quit IRC | 11:22 | |
| *** thrash|g0ne is now known as thrash | 11:25 | |
| *** vmud213 has joined #openstack-ironic | 11:27 | |
| *** fragatin_ has quit IRC | 11:30 | |
| *** milan has joined #openstack-ironic | 11:35 | |
| *** lucasagomes is now known as lucas-hungry | 11:42 | |
| *** krtaylor has joined #openstack-ironic | 11:42 | |
| *** appu103_ has joined #openstack-ironic | 11:48 | |
| *** aparnav has quit IRC | 11:50 | |
| *** amotoki has joined #openstack-ironic | 11:52 | |
| *** appu103_ has quit IRC | 11:52 | |
| *** raildo-afk is now known as raildo | 11:53 | |
| *** fragatina has joined #openstack-ironic | 11:54 | |
| *** sdake has joined #openstack-ironic | 11:58 | |
| *** sdake_ has joined #openstack-ironic | 12:00 | |
| *** piet has joined #openstack-ironic | 12:02 | |
| openstackgerrit | David Edery proposed openstack/ironic: Fix non-deployed nodes due to short nodes.instance_info https://review.openstack.org/334967 | 12:02 |
| *** sdake has quit IRC | 12:03 | |
| *** moshele has joined #openstack-ironic | 12:05 | |
| *** krtaylor has quit IRC | 12:05 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Follow up patch of I2293e03c05e13c716f78533680d128ba45ccda02 https://review.openstack.org/334969 | 12:06 |
| *** dtantsur|lunch is now known as dtantsur | 12:12 | |
| *** vmud213 has quit IRC | 12:13 | |
| *** yogi has quit IRC | 12:15 | |
| *** trown|outtypewww is now known as trown | 12:15 | |
| *** athomas has quit IRC | 12:18 | |
| TheJulia | Good morning | 12:19 |
| *** [1]cdearborn has joined #openstack-ironic | 12:20 | |
| milan | morning TheJulia | 12:21 |
| *** moshele has quit IRC | 12:22 | |
| *** athomas has joined #openstack-ironic | 12:23 | |
| dtantsur | morning TheJulia | 12:24 |
| *** e0ne has joined #openstack-ironic | 12:26 | |
| *** daemontool has quit IRC | 12:28 | |
| mat128 | morning TheJulia, milan and dtantsur | 12:29 |
| milan | mat128, morning | 12:29 |
| *** yogi has joined #openstack-ironic | 12:30 | |
| dtantsur | morning mat128 | 12:31 |
| *** baoli_ has quit IRC | 12:31 | |
| *** baoli has joined #openstack-ironic | 12:32 | |
| openstackgerrit | Caio Oliveira proposed openstack/python-oneviewclient: [WIP] Creating certificate manager for OneView https://review.openstack.org/334992 | 12:34 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Follow up patch of d52077f4fe8c668b258702e8298a4beaa19476d8 https://review.openstack.org/334969 | 12:35 |
| *** yuikotakadamori has joined #openstack-ironic | 12:36 | |
| *** daemontool has joined #openstack-ironic | 12:36 | |
| *** yuikotakadamori has quit IRC | 12:37 | |
| *** baoli has quit IRC | 12:37 | |
| openstackgerrit | Aline Bousquet proposed openstack/ironic-lib: Add support for metrics https://review.openstack.org/301526 | 12:38 |
| *** moshele has joined #openstack-ironic | 12:38 | |
| *** cdearborn has joined #openstack-ironic | 12:38 | |
| *** links has quit IRC | 12:40 | |
| *** yuikotakadamori has joined #openstack-ironic | 12:40 | |
| *** Nisha_away has joined #openstack-ironic | 12:43 | |
| *** wajdi has quit IRC | 12:51 | |
| *** links has joined #openstack-ironic | 12:53 | |
| TheJulia | good morning mat128, milan, daemontool | 12:53 |
| *** [1]cdearborn has quit IRC | 12:53 | |
| thiagop | Good morning, Ironicers | 12:54 |
| TheJulia | good morning thiagop | 12:56 |
| thiagop | TheJulia: o/ | 12:56 |
| *** moshele has quit IRC | 12:56 | |
| *** daemontool has quit IRC | 12:57 | |
| mat128 | o/ thiagop | 12:57 |
| *** _vsaienko has quit IRC | 12:57 | |
| thiagop | hey mat128 | 12:57 |
| *** rloo has joined #openstack-ironic | 12:57 | |
| gbraad | TheJulia, thanks for the email. I'll try some ofthe suggestions later and move to the conversation to the mailinglist | 12:58 |
| gbraad | at the moment I have to look after my kid. ;-) | 12:58 |
| gbraad | and yes, I added the spacing to the output (makes it slightly more readable) | 12:59 |
| TheJulia | gbraad: not exactly a fan of the mailing list, IRC is more preferable for troubleshooting issues | 13:00 |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add missing translation marker to clear_node_target_power_state https://review.openstack.org/334969 | 13:01 |
| TheJulia | Good to know about the spacing, that had me a little worried | 13:01 |
| *** jjohnson2 has joined #openstack-ironic | 13:02 | |
| gbraad | enterless strings can be hard to read. | 13:02 |
| gbraad | sure, but I am in a bad timezone for you I guess. China | 13:03 |
| TheJulia | ahh | 13:03 |
| TheJulia | That is quite a timezone difference | 13:03 |
| TheJulia | what time are you typically online in the morning? | 13:04 |
| gbraad | 8:30 | 13:05 |
| *** M00nr41n has quit IRC | 13:05 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: another test for ironic networking https://review.openstack.org/330210 | 13:06 |
| TheJulia | UTC? | 13:06 |
| *** Nisha_away has quit IRC | 13:06 | |
| gbraad | Beijing time... but I'll get back to you about this. | 13:07 |
| TheJulia | ok | 13:07 |
| gbraad | Anyways I might try this later tonight and let you know | 13:07 |
| TheJulia | Okay, otherwise have a good evening | 13:07 |
| gbraad | Thank you | 13:08 |
| *** lucas-hungry is now known as lucasagomes | 13:09 | |
| *** yuikotakadamori has quit IRC | 13:12 | |
| *** daemontool has joined #openstack-ironic | 13:13 | |
| *** baoli has joined #openstack-ironic | 13:14 | |
| *** rbudden has joined #openstack-ironic | 13:16 | |
| dtantsur | lucasagomes, mind reviewing one more agent API promotion patch https://review.openstack.org/#/c/332890/ please? | 13:17 |
| lucasagomes | dtantsur, will do in a sec | 13:17 |
| dtantsur | thnx | 13:17 |
| *** sdake_ has quit IRC | 13:23 | |
| *** sdake has joined #openstack-ironic | 13:24 | |
| *** daemontool has quit IRC | 13:25 | |
| *** rama_y has joined #openstack-ironic | 13:28 | |
| *** yuikotakadamori has joined #openstack-ironic | 13:28 | |
| *** snecklifter has quit IRC | 13:28 | |
| sean-k-mooney | hi everyone QQ re bifrost. do ye technically support Ubuntu 14.04 currently? | 13:28 |
| sean-k-mooney | i say some default overrides for 15.04+ but none for 14.04 | 13:29 |
| *** wajdi has joined #openstack-ironic | 13:31 | |
| rook | JayF would it be trivial to add a timing metric to this function? https://github.com/openstack/ironic-python-agent/blob/stable/mitaka/ironic_python_agent/extensions/standby.py#L66 | 13:32 |
| rook | JayF If I remember correctly, you are adding some metric gathering to ironic, right? | 13:32 |
| rook | rloo ^' | 13:39 |
| dtantsur | alineb, ^^^ | 13:40 |
| rloo | rook: I believe JayF is away for the next few days or more (I forgot how long). He said he would pop into irc but not frequently. | 13:40 |
| rloo | rook, yeah, alineb should be able to help you | 13:40 |
| rook | rloo ok... reviewing 301923 i don't get the sense this patch applies to what I am looking for. | 13:40 |
| *** daemontool has joined #openstack-ironic | 13:41 | |
| rook | however, maybe there are pieces that could be reused? | 13:41 |
| rook | rloo I am not too in-tune with how ipa writes the image -- I am guessing the function i posted above would be it. | 13:41 |
| rook | rloo - however, I think it would be a good metric for ipa to return to ironic when the disk writing has completed. | 13:41 |
| rloo | rook: possibly. i didn't look to see how they were adding metrics to ipa | 13:41 |
| rook | https://review.openstack.org/#/c/119981/ | 13:42 |
| rloo | rook: hopefully aline could help you. I didn't look and am in the middle of something, so don't want to look. | 13:43 |
| rloo | rook: you could also add a comment to that patch, with your question. if no one gets back to you. | 13:44 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Fix invalid chassis/nodes/chassis_uuid API url https://review.openstack.org/316149 | 13:44 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Fix all possible node subresources routing collisions https://review.openstack.org/315766 | 13:44 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Prevent URL collisions with sub-controllers: nodes/ports https://review.openstack.org/314514 | 13:44 |
| *** krtaylor has joined #openstack-ironic | 13:45 | |
| *** hkominos has quit IRC | 13:46 | |
| TheJulia | sean-k-mooney: 14.04 is supported and should work out of the box | 13:46 |
| *** hkominos has joined #openstack-ironic | 13:47 | |
| alineb | rook: yes, i think it should be possible, for instance with a context like it's done in https://review.openstack.org/#/c/119981/28/ironic_python_agent/extensions/standby.py | 13:48 |
| hkominos | Hello ironicers (and hopefully fuel-ironicers). I have a quick question. After a node is enrolled in the conductor does that mean that the conductor will manage its power settings? Cause i am having a weird behaviour of manually powering on the node and then the node is powered of automatically | 13:48 |
| *** piet has quit IRC | 13:49 | |
| dtantsur | hkominos, that's right, ironic will enforce the power state | 13:49 |
| *** xhku_ has joined #openstack-ironic | 13:50 | |
| rook | alineb: yeah, i am going through that commit now... looking at why jenkins is still throwing up | 13:50 |
| *** ayoung has joined #openstack-ironic | 13:52 | |
| alineb | rook: the patch depends on a another one in ironic-lib, so i don't think the tests will pass until the one in ironic-lib is merged | 13:53 |
| rook | hm, i thought I looked at that patch and it was... | 13:53 |
| *** xhku has quit IRC | 13:53 | |
| rook | ahh, no it wasn't | 13:54 |
| rook | alright alineb makes sense. | 13:54 |
| *** krtaylor has quit IRC | 13:54 | |
| jlvillal | milan, dtantsur: Regarding smoke tests and Grenade. In Ironic devstack-gate will run the tempest tests after the upgrade. Basically if you say grenade then devstack-gate should always run the tempest tests after the upgrade. Those results will be in the console log and not the grenade log | 13:55 |
| *** rama_y has quit IRC | 13:55 | |
| dtantsur | jlvillal, could you please check our job definitions (I think milan has a link) to see if we do run them? | 13:55 |
| jlvillal | dtantsur, If it tells devstack-gate that it is a grenade job then I think it must. | 13:56 |
| *** moshele has joined #openstack-ironic | 13:56 | |
| dtantsur | I see, I'll check it | 13:56 |
| *** ametts has joined #openstack-ironic | 13:57 | |
| alineb | rook: also, the patch in IPA will need some changes, but it will still tell you how we plan to call the context manages | 13:57 |
| *** moshele has quit IRC | 13:58 | |
| *** moshele has joined #openstack-ironic | 13:58 | |
| dtantsur | milan, it seems like we fail to shutdown inspector: http://logs.openstack.org/67/327667/15/experimental/gate-grenade-dsvm-ironic-inspector-nv/93eef2b/logs/grenade.sh.txt.gz#_2016-06-28_09_51_03_901 | 14:00 |
| milan | dtantsur, yeah, it's the setsid issue | 14:00 |
| milan | the pid file for inspector contains pid instead of pgid | 14:00 |
| milan | dtantsur, see the depends on | 14:01 |
| dtantsur | aha, I see | 14:01 |
| *** PollyZ has joined #openstack-ironic | 14:01 | |
| dtantsur | milan, we don't the other projects have the same problem? | 14:02 |
| dtantsur | how are we special? | 14:02 |
| *** baoli has quit IRC | 14:03 | |
| milan | dtantsur, somehow we use stack_it that boils down to _old_run_process in devstack/functions-common (if I'm not mistaken) | 14:03 |
| milan | dtantsur, in stack_it in inspector/devstack/plugin.sh that is | 14:04 |
| dtantsur | milan, is it possible we just use the same as ironic? | 14:04 |
| openstackgerrit | Saravanan KR proposed openstack/ironic-python-agent: Set the kernel args to the nodes at bootloader instal https://review.openstack.org/335052 | 14:04 |
| milan | we could update the plugin.sh, yeah, depends on how attractive my patches are for devstack guys | 14:05 |
| * milan didn't get much feedback on that though | 14:06 | |
| *** e0ne has quit IRC | 14:06 | |
| lucasagomes | dtantsur, re db lock for the heartbeat() | 14:07 |
| lucasagomes | dtantsur, it's writing to the node at driver_internal_info | 14:07 |
| lucasagomes | I wonder if we may introduce a race there, heartbeat vs some other node writting | 14:08 |
| *** PollyZ has quit IRC | 14:08 | |
| lucasagomes | very unlikely, maybe I'm over concerned, but I wanted to check first | 14:08 |
| dtantsur | lucasagomes, maybe; we can work around it by having a special db call to change last heartbeat. Also I can easily imagine a driver not doing anything in its heartbeat when no actions are expected | 14:09 |
| milan | jlvillal, dtantsur the job template: https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/devstack-gate.yaml#L2152-L2204 | 14:09 |
| milan | I've just spotted some export DEVSTACK_LOCAL_CONFIG+=$'\n'"TEMPEST_PLUGINS+=' /opt/stack/new/ironic-inspector'" it might be missing | 14:09 |
| dtantsur | milan, I think we should fix the plugin to match how ironic does it. landing patches to devstack can take serious time | 14:09 |
| milan | hmmm yeah | 14:10 |
| *** PollyZ has joined #openstack-ironic | 14:11 | |
| dtantsur | milan, yeah, I think we should switch to run_process | 14:11 |
| milan | dtantsur, https://github.com/openstack/ironic/blob/master/devstack/lib/ironic#L663 Ironic uses run process | 14:12 |
| milan | dtantsur, ack | 14:12 |
| dtantsur | thnx! would be awesome to get it passing soon :) | 14:13 |
| milan | dtantsur, the grenade patch for checking is_service_running can be worked around saying dnsmasq instead of ironic-inspector-dhcp | 14:13 |
| dtantsur | milan, or we can put "enable_service ironic-inspector-dhcp" to our setting | 14:13 |
| dtantsur | actually we do, hmm: https://github.com/openstack/ironic-inspector/blob/master/devstack/settings | 14:14 |
| milan | dtantsur, it is there already | 14:14 |
| milan | yeah | 14:14 |
| dtantsur | then what's the problem? | 14:14 |
| milan | right | 14:14 |
| milan | inspector/devstack/upgrade.sh does check that the services are up | 14:14 |
| milan | in the end | 14:14 |
| milan | after the upgrade | 14:14 |
| milan | as a sanity assertion | 14:14 |
| milan | it works for inspector but doesn't for -inspector-dhcp | 14:15 |
| NobodyCam | Good morning Ironicers | 14:15 |
| milan | we can check dnsmasq through is_service_running in there instead | 14:15 |
| *** mjura_ has quit IRC | 14:15 | |
| *** piet has joined #openstack-ironic | 14:15 | |
| *** PollyZ has quit IRC | 14:15 | |
| * milan just sad that patches that might fix it for other people will be worked around | 14:16 | |
| milan | but yeah | 14:16 |
| milan | #pixiesay nothing's ever ideal -m meh | 14:16 |
| PixieBoots | ¯\_(ツ)_/¯: nothing's ever ideal | 14:16 |
| milan | morning NobodyCam :) | 14:16 |
| NobodyCam | morning milan | 14:16 |
| NobodyCam | :) | 14:16 |
| dtantsur | milan, why doesn't is_service_running work for us? | 14:17 |
| dtantsur | morning NobodyCam | 14:17 |
| NobodyCam | morning dtantsur :) | 14:17 |
| milan | dtantsur, is_service_running uses ps | grep <service name> | 14:17 |
| dtantsur | milan, then how does it work for ir-api? the process is called ironic-api | 14:17 |
| milan | dtantsur, we don't match ironic-inspector-dhcp because it is dnsmasq | 14:17 |
| dtantsur | and ir-api is ironic-api. do you know how it works? | 14:18 |
| *** rook_ has quit IRC | 14:18 | |
| milan | it does lot of grep magic O:-) | 14:18 |
| * milan looks | 14:18 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Disable flat network driver https://review.openstack.org/335060 | 14:18 |
| milan | dtantsur, it's workaround in ironic: https://github.com/openstack/ironic/blob/master/devstack/upgrade/upgrade.sh#L123 | 14:20 |
| *** rook_ has joined #openstack-ironic | 14:20 | |
| *** rook_ has quit IRC | 14:21 | |
| dtantsur | milan, can we avoid using ensure_service_started for dnsmasq and use our own grep? | 14:21 |
| milan | dtantsur, we can do pretty much anything :D | 14:21 |
| dtantsur | so lets do it :) | 14:21 |
| dtantsur | if your patches get merged we can always remove our hacks | 14:22 |
| *** PollyZ has joined #openstack-ironic | 14:22 | |
| milan | yeah, a bit pragmatic | 14:22 |
| dtantsur | but I don't want to wait a couple of weeks to hear "don't do that" | 14:22 |
| milan | dtantsur, right, that could happen indeed. | 14:22 |
| milan | dtantsur, maybe if people gave those 2 tiny patches a review ;) | 14:23 |
| *** joprovost has joined #openstack-ironic | 14:23 | |
| *** baoli has joined #openstack-ironic | 14:24 | |
| dtantsur | welll... nobody here is core on grenade, I guess | 14:24 |
| *** baoli has quit IRC | 14:25 | |
| milan | dtantsur, right, +1 better than none still ;) | 14:25 |
| * milan hacks | 14:25 | |
| *** baoli has joined #openstack-ironic | 14:25 | |
| *** rajinir has joined #openstack-ironic | 14:25 | |
| dtantsur | will do ofc, but it does not help too much usually | 14:25 |
| *** jvgrant has joined #openstack-ironic | 14:26 | |
| *** itamarl has quit IRC | 14:26 | |
| *** M00nr41n has joined #openstack-ironic | 14:27 | |
| *** rbartal has quit IRC | 14:28 | |
| NobodyCam | jlvillal: happen to be around yet? | 14:28 |
| jlvillal | NobodyCam, I am but about to leave for 25 minutes | 14:29 |
| dtantsur | lucasagomes, fyi not requiring exclusive lock was requested by jroll on https://review.openstack.org/#/c/306418/11/specs/approved/agent-api.rst@194 and that's why it's on the spec | 14:29 |
| lucasagomes | dtantsur, a-ha, fair enough | 14:30 |
| *** rook_ has joined #openstack-ironic | 14:30 | |
| *** jvgrant has quit IRC | 14:30 | |
| milan | jlvillal, dtantsur wdyt about export DEVSTACK_LOCAL_CONFIG+=$'\n'"TEMPEST_PLUGINS+=' /opt/stack/new/ironic-inspector'" in the experimental inspector job template? | 14:31 |
| milan | it's missing in there now | 14:31 |
| lucasagomes | dtantsur, +2'd | 14:31 |
| dtantsur | thnx! | 14:31 |
| dtantsur | milan, I don't see it in ironic job definition though.. | 14:31 |
| lucasagomes | since it has been discussed in the spec already, we should keep going with it | 14:31 |
| dtantsur | lucasagomes, no necessary, if you strongly object, we can amend the spec | 14:31 |
| lucasagomes | dtantsur, sorry for the confusion, I thought it would worth asking anyway | 14:31 |
| lucasagomes | dtantsur, sure, I don't strongly object to it | 14:31 |
| lucasagomes | that's why I've voted mostly to ask about it | 14:32 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: Canery test for ironic networking changes https://review.openstack.org/330138 | 14:32 |
| dtantsur | cool, thanks lucasagomes | 14:32 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Introducing Inspector Grenade testing https://review.openstack.org/327667 | 14:32 |
| milan | dtantsur, ^ | 14:33 |
| dtantsur | milan, please update the commit message with your changes | 14:33 |
| milan | dtantsur, wdym? | 14:33 |
| dtantsur | milan, mention that you're fixing the plugin.sh (no rush, only if you create another patchset) | 14:33 |
| NobodyCam | hey hey Morning jlvillal | 14:34 |
| milan | dtantsur, aah | 14:34 |
| milan | OK | 14:34 |
| NobodyCam | just a quick question about https://review.openstack.org/#/c/272658 what are you seeing with the release note? | 14:35 |
| *** vgadiraj has joined #openstack-ironic | 14:38 | |
| *** links has quit IRC | 14:38 | |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Introducing Inspector Grenade testing https://review.openstack.org/327667 | 14:38 |
| dtantsur | thnx | 14:39 |
| *** rook_ has quit IRC | 14:39 | |
| *** rbudden has quit IRC | 14:41 | |
| *** rook_ has joined #openstack-ironic | 14:41 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Stop passing root device hints via kernel cmdline https://review.openstack.org/335070 | 14:41 |
| *** e0ne has joined #openstack-ironic | 14:42 | |
| *** yuikotakadamori has quit IRC | 14:46 | |
| dtantsur | lucasagomes, will it break mitaka ^^? | 14:47 |
| lucasagomes | dtantsur, yeah, wondering if we should put it on hold for now | 14:48 |
| *** krotscheck_dcm is now known as krotscheck | 14:48 | |
| lucasagomes | dtantsur, another way to do it would be to, pass root device hints _only_ if there's no operators on it :-/ | 14:48 |
| dtantsur | lucasagomes, I think we can only do it in Ocata :( grenade is using mitaka IPA | 14:48 |
| dtantsur | lucasagomes, I think it's a possible compromise, yes | 14:49 |
| dtantsur | declare that operators only work with newton IPA | 14:49 |
| *** mgould is now known as mgould|brb | 14:49 | |
| lucasagomes | yeah, that would be the case | 14:49 |
| *** e0ne has quit IRC | 14:50 | |
| *** xhku_ has quit IRC | 14:51 | |
| *** moshele has quit IRC | 14:52 | |
| jlvillal | NobodyCam, I thought I said nevermind on the release notes :) | 14:54 |
| jlvillal | NobodyCam, It wasn't your patch. I fixed it here: https://review.openstack.org/334712 | 14:54 |
| JayF | rook: rloo: Sure, we can metric things. If you want to get more metrics stuff in, the #1 thing to do to help is review alineb's ironic-lib patch to add metrics. Once that is in choosting to metric individual items will be trivial. | 14:57 |
| NobodyCam | jlvillal: :) | 14:57 |
| NobodyCam | ack just wanted to check | 14:57 |
| NobodyCam | morning lucasagomes JayF | 14:57 |
| lucasagomes | NobodyCam, JayF morning | 14:58 |
| NobodyCam | :) | 14:58 |
| JayF | gm | 14:59 |
| NobodyCam | :) | 14:59 |
| JayF | last day I'll be online working for probably a couple of weeks (reality is I'll probably work one day next week, but don't count on it), so if you need something from me speak now :) | 15:00 |
| NobodyCam | omg :) moving? | 15:00 |
| *** vgadiraj has quit IRC | 15:01 | |
| *** xhku has joined #openstack-ironic | 15:02 | |
| JayF | yes | 15:02 |
| NobodyCam | :) be safe | 15:02 |
| *** baoli has quit IRC | 15:03 | |
| JayF | That's what I'll be trying for, lol | 15:03 |
| *** baoli has joined #openstack-ironic | 15:03 | |
| *** [1]cdearborn has joined #openstack-ironic | 15:04 | |
| NobodyCam | lol | 15:04 |
| *** baoli has quit IRC | 15:04 | |
| *** rloo has quit IRC | 15:04 | |
| *** sabeen has joined #openstack-ironic | 15:04 | |
| *** rloo has joined #openstack-ironic | 15:04 | |
| *** baoli has joined #openstack-ironic | 15:05 | |
| *** amotoki has quit IRC | 15:07 | |
| *** rbudden has joined #openstack-ironic | 15:08 | |
| *** M00nr41n has quit IRC | 15:08 | |
| *** praneshp has joined #openstack-ironic | 15:09 | |
| *** sacharya has joined #openstack-ironic | 15:09 | |
| *** baoli has quit IRC | 15:09 | |
| *** M00nr41n has joined #openstack-ironic | 15:10 | |
| *** sacharya_ has joined #openstack-ironic | 15:12 | |
| *** jistr is now known as jistr|mtg | 15:13 | |
| * milan relocates | 15:13 | |
| *** milan has quit IRC | 15:13 | |
| *** M00nr41n has quit IRC | 15:14 | |
| *** sacharya has quit IRC | 15:14 | |
| *** baoli has joined #openstack-ironic | 15:17 | |
| *** cdearborn has quit IRC | 15:17 | |
| *** baoli has quit IRC | 15:18 | |
| *** baoli has joined #openstack-ironic | 15:18 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: Add missing translation marker to clear_node_target_power_state https://review.openstack.org/334969 | 15:19 |
| *** mgould|brb is now known as mgould | 15:20 | |
| *** M00nr41n has joined #openstack-ironic | 15:20 | |
| *** jtomasek_ has quit IRC | 15:20 | |
| *** jaybeale has joined #openstack-ironic | 15:23 | |
| dtantsur | folks, do we really have a compelling use case for setting network drivers per node rather than per deployment? | 15:23 |
| dtantsur | it would simplify the network patches substantially | 15:23 |
| dtantsur | rloo, TheJulia, JayF, sambetts, maybe you remember? | 15:23 |
| rloo | dtantsur: is this via the new node.network_interface? | 15:24 |
| dtantsur | yeah.. do we really need it? | 15:24 |
| *** M00nr41n has quit IRC | 15:24 | |
| dtantsur | like, do we expect half of nodes to have network separation, half - no? | 15:24 |
| JayF | I don't know one way or the other. I suspect real-world means it'd be not useful | 15:24 |
| dtantsur | how will nova distinguish between such nodes? | 15:25 |
| JayF | but I know we have folks with crazier use cases and environments than I have ever been able to imagine :) | 15:25 |
| dtantsur | because then a user will want to request a node with or without separation | 15:25 |
| *** baoli has quit IRC | 15:25 | |
| * dtantsur will probably write to a ML | 15:25 | |
| rloo | dtantsur: if we don't add the network interface now, we should at least make sure that we can add it later if it turns out that people want it. | 15:26 |
| dtantsur | rloo, it's easier to add than to remove ;) | 15:26 |
| rloo | dtantsur: we rarely remove. we haven't even removed chassis :) | 15:27 |
| rloo | dtantsur: worth asking though. it seems like we have it cuz someone asked. but it would mean going back and looking at comments in previous revisions in who-knows-which-patch. | 15:28 |
| *** PollyZ has quit IRC | 15:28 | |
| rloo | and who knows, maybe i asked to make it more flexible. i cannot recall now. although i doubt that i would have asked for a NetworkInterface. | 15:28 |
| *** M00nr41n has joined #openstack-ironic | 15:29 | |
| NobodyCam | good morning rloo :) | 15:30 |
| rloo | dtantsur: according to this spec (I'm assuming we changed 'network_provider' to 'network_interface', it is per node: http://specs.openstack.org/openstack/ironic-specs/specs/approved/network-provider.html | 15:31 |
| rloo | morning NobodyCam! | 15:31 |
| NobodyCam | :) | 15:31 |
| mgould | morning NobodyCam rloo | 15:31 |
| NobodyCam | morning mgould :) | 15:32 |
| dtantsur | thanks rloo, but it's still unclear to me. I'm writing to the ML | 15:32 |
| *** PollyZ has joined #openstack-ironic | 15:32 | |
| *** jistr|mtg is now known as jistr | 15:32 | |
| *** Nisha_away has joined #openstack-ironic | 15:32 | |
| rloo | dtantsur: ok. that spec seems ... out of date... i'm not sure how much of that is going to be true. | 15:33 |
| dtantsur | http://lists.openstack.org/pipermail/openstack-dev/2016-June/098361.html | 15:33 |
| rloo | dtantsur: just wondering. what if you imported a bunch of nodes ('adopted' I guess) into ironic/openstack, and they haven't been migrated yet (whatever that entails) to use neutron. | 15:36 |
| dtantsur | rloo, this sounds troublesome, I agree. still I think "migrate asap" is an answer here | 15:37 |
| dtantsur | with "migrate" only meaning providing port information, I guess | 15:37 |
| rloo | dtantsur: need operator input I guess. I'm so useless... | 15:37 |
| dtantsur | no more than I am.. | 15:38 |
| rloo | dtantsur: so jroll wrote that spec, so i'd assume he knew of a usecase or he wouldn't have proposed per node. but i know, i shouldn't assume. | 15:39 |
| rloo | dtantsur: now i wonder if i should bother reviewing the patches until this is answered. | 15:40 |
| dtantsur | yeah, lets ask around. sometimes I also overdo with flexibility and then someone is like "but who would use that"? :) | 15:40 |
| *** alex_xu has quit IRC | 15:40 | |
| dtantsur | I won't; I need to understand "why" before "how" | 15:40 |
| rloo | dtantsur: i like flexibility. | 15:40 |
| rloo | dtantsur: cuz you never know peoples' use cases. | 15:40 |
| dtantsur | me too but not at expense of e.g. upgrade problems | 15:41 |
| dtantsur | and I don't quite understand how we define the default value in this case | 15:41 |
| dtantsur | like, do we need a configuration option as well? | 15:42 |
| *** dmk0202 has quit IRC | 15:42 | |
| rloo | dtantsur: i assumed it would become clear as i went through the patches. really, the spec should describe it but ... :-( | 15:42 |
| dtantsur | maybe we should not proceed with the spec so outdated | 15:43 |
| TheJulia | dtantsur: I know of several users who use ironic for users and managing their pool of hardware, so concievably some may want none, others may want flat, or neutron per node in that case | 15:43 |
| rloo | dtantsur: maybe it was part of the previous mid-cycle, where they changed this. | 15:43 |
| *** alex_xu has joined #openstack-ironic | 15:44 | |
| dtantsur | TheJulia, all 3 in one cluster? so they have neutron with and without separation, and also their own DHCP server? | 15:44 |
| TheJulia | I can see it, I know some tht use their own dhcp server but want the capability for isolation if they have configured the network as such | 15:45 |
| TheJulia | flat, maybe not so much in that scenario | 15:45 |
| sean-k-mooney | TheJulia: hi. looking at the bifrost readme it support ubuntu 14.04 but i noticed it does not create initv service files for ironic* services. is 14.04 still tested? | 15:45 |
| TheJulia | NobodyCam: your thoughts? | 15:46 |
| TheJulia | sean-k-mooney: we test 14.04 in the gate, but we write upstart configs | 15:46 |
| dtantsur | TheJulia, I wonder how well neutron works with another dhcp. we had some fun making inspector work with neutron.. | 15:46 |
| cinerama | sean-k-mooney, i also use 14.04 to do my general purpose testing | 15:46 |
| * NobodyCam reads | 15:47 | |
| TheJulia | dtantsur: I would think of it as different distinct pools of hardware. I might have some conductors wired into multiple network fabrics | 15:47 |
| JayF | I mean, the question not only needs to be "do people want us to support this" but also "does it make sense to support this" | 15:47 |
| TheJulia | JayF: ++ | 15:47 |
| sean-k-mooney | cinerama: TheJulia are the upstart configs only created when you run in testing mode or should they always be created | 15:47 |
| JayF | I share dtantsur's concern that having such a core difference per node without having it be schedulable in nova is very ill advised | 15:47 |
| TheJulia | sean-k-mooney: always from what I remember | 15:47 |
| TheJulia | I think we have two distinct users, those that use ironic directly that want to assert something, and then via nova, from nova, I do agree, it is problematic although I could still see a deployments wanting to offer both for any number of reasons | 15:48 |
| mat128 | JayF, dtantsur: I know I would be fine having that multitenancy yes/no per ironic installation rather than per node | 15:48 |
| sean-k-mooney | ok im working on the kolla intergration and the ironic services are not being started as part of the bifrost ironic install playbooks on 14.04. on centos it works fine but that uses systemd | 15:49 |
| TheJulia | then that forces separate installations if I have an HPC compute cluster in one corner that I don't want to involve neutron in the networking for | 15:49 |
| mat128 | TheJulia: you can always use a driver that does not involve neutron | 15:49 |
| mat128 | a node driver | 15:49 |
| TheJulia | mat128: but the discussion seems to be removal of that | 15:49 |
| TheJulia | unless I'm on another planet, which is always a possibility | 15:50 |
| mat128 | removal of node drivers? hell no :) | 15:50 |
| mat128 | oh | 15:50 |
| TheJulia | sean-k-mooney: it should always create them, but I don't have the code in front of me at the moment | 15:50 |
| NobodyCam | I have seen this configuration, it was basically a pre production env | 15:50 |
| dtantsur | mat128, removal of proposed node.network_interface and replacing it with configuration only | 15:50 |
| cinerama | sean-k-mooney, hmm, that is weird. so do you get the upstart files? | 15:50 |
| mat128 | the removal of the "network driver" part? | 15:50 |
| dtantsur | yep | 15:50 |
| rloo | mat128: yeah, so all have to use the same configurable setting. | 15:50 |
| TheJulia | I think that is asking for people complaining | 15:50 |
| rloo | seems too restrictive to me but I don't have a use case myself. | 15:50 |
| JayF | from jim by proxy: "The main thing is it needs to be discoverable via api because nova needs to know | 15:51 |
| dtantsur | JayF, then ask him how nova interprets value of "None" which is valid per driver composition spec | 15:52 |
| dtantsur | TheJulia, do people already complain? | 15:52 |
| TheJulia | dtantsur: yes :) | 15:52 |
| dtantsur | interesting | 15:52 |
| dtantsur | lucasagomes, fwiw my fedora has iptables | 15:53 |
| TheJulia | I know of several internal deployments that essentially need the flexibility since they use ironic to manage their hardware pools in general | 15:53 |
| lucasagomes | dtantsur, you can install it, but it doesn't come installed by default AFAICT | 15:53 |
| lucasagomes | at least for the workstation image | 15:53 |
| dtantsur | I don't remember installing it, but maybe something pulled it | 15:53 |
| JayF | dtantsur: tl;dr for repsonse to that; take it to the ml | 15:53 |
| lucasagomes | (it comes with firewalld instead) | 15:53 |
| rloo | dtantsur: it seems to me that if TheJulia knows of some cases, then it is needed/can be useful. | 15:53 |
| lucasagomes | dtantsur, or maybe something changed... I'm still on f22 :-) | 15:54 |
| mat128 | dtantsur, TheJulia, rloo, JayF: Forget my comment, I did not correctly understand the question. | 15:54 |
| dtantsur | rloo, TheJulia, interesting. I won't block the patches based on this alone, but the default should be figured out. especially the handling of None/NULL | 15:54 |
| mat128 | network_interface is for the flat/lacp/active-backup choice, right? | 15:54 |
| *** sdake_ has joined #openstack-ironic | 15:54 | |
| dtantsur | mat128, no problem. I think so, I don't know what lacp means | 15:55 |
| *** hshiina has quit IRC | 15:55 | |
| TheJulia | dtantsur: yes, agree that a site default might be a useful thing as well, defaulting to none just seems to ask for headaches, although if we tested that, it could expose more issues | 15:55 |
| mat128 | dtantsur: bonding? | 15:56 |
| dtantsur | TheJulia, I mean Python None, not the "none" string | 15:56 |
| dtantsur | or database NULL if you want | 15:56 |
| mat128 | ok, just as a reference standpoint | 15:56 |
| *** tesseract- has quit IRC | 15:56 | |
| dtantsur | mat128, then no, it's about who manages network for the node: "nobody", "neutron, flat network", "neutron, tenant separation" | 15:56 |
| mat128 | oooh | 15:57 |
| *** sdake has quit IRC | 15:57 | |
| mat128 | that should belong to the node I believe | 15:57 |
| mat128 | example use case for it | 15:57 |
| TheJulia | dtantsur: yes, that :) | 15:57 |
| *** ifarkas has quit IRC | 15:58 | |
| mat128 | I have a bunch of servers using latest and greatest ToR which I have neutron ML2 support for | 15:58 |
| mat128 | these nodes use the "neutron-*" modes | 15:58 |
| mat128 | but I could also have a bunch of nodes that are on dumb switches, or manually configured switches outside of Openstack's control | 15:58 |
| mat128 | these nodes should be on the "nobody" mode | 15:58 |
| mat128 | all of these nodes could live happily in the same Ironic installation | 15:58 |
| mat128 | s/could/should | 15:58 |
| sean-k-mooney | cinerama: am where would i check for the upstart files? should they be in /etc/init.d/ or somewhere else? | 15:59 |
| dtantsur | mat128, TheJulia, thanks, please put your considerations on the ML thread, so that we're sure all folks are in sync | 15:59 |
| mat128 | as Jim pointed out via JayF proxy, Nova should be able to discover/know, because it might have implications (maybe integrated network is a nice feature we want to expose through a flavor), or maybe the configdrive needs to be configured differently | 15:59 |
| mat128 | writing a reply :) | 15:59 |
| dtantsur | mat128, yeah, but it's unclear what nova does on seeing None there.. | 16:00 |
| *** sukhdev has joined #openstack-ironic | 16:00 | |
| *** daemontool_ has joined #openstack-ironic | 16:00 | |
| dtantsur | which means "use the default" | 16:00 |
| mat128 | dtantsur: maybe it doesnt have to anything to do | 16:00 |
| mat128 | in a normal use case, what would nova do differently? | 16:00 |
| dtantsur | I'm not sure, that's what jroll said :) | 16:00 |
| mat128 | Unless there is something I don't see, nova doesnt need to do anything differently when the network is managed by nobody, neutron flat or neutron tenant separated | 16:01 |
| mat128 | nova interacts with neutron for the VIF parts | 16:01 |
| mat128 | nova will be impacted with vlan-aware instances and future work, but not with the simple network_interafce choice | 16:01 |
| *** xhku has quit IRC | 16:02 | |
| *** vishwanathj has joined #openstack-ironic | 16:02 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic-python-agent: Add a log extension https://review.openstack.org/248832 | 16:03 |
| *** daemontool has quit IRC | 16:03 | |
| *** Sukhdev_ has joined #openstack-ironic | 16:04 | |
| rloo | hi lucasagomes, wrt https://review.openstack.org/#/c/334431/, do you know what the oslo patch is? i want to find out what the delay might be | 16:05 |
| lucasagomes | rloo, https://review.openstack.org/#/c/308398/ (i've added it to the comment in the code) | 16:05 |
| lucasagomes | perhaps should put in the commit message too | 16:05 |
| rloo | lucasagomes: thx. I didn't actually look at the code change; just read the commit msg :) | 16:05 |
| lucasagomes | rloo, no worries :-) | 16:06 |
| *** xhku has joined #openstack-ironic | 16:06 | |
| *** gfhellma has joined #openstack-ironic | 16:09 | |
| *** xhku has quit IRC | 16:11 | |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add power state change notifications https://review.openstack.org/321865 | 16:14 |
| openstackgerrit | Mario Villaplana proposed openstack/ironic: Add notification base classes and docs https://review.openstack.org/298461 | 16:14 |
| sean-k-mooney | cinerama: looking at the code the upstart files are added to /etc/init which does contain the files. | 16:14 |
| *** klindgren has joined #openstack-ironic | 16:15 | |
| lucasagomes | rloo, so, I don't know if/when it's going to land in oslo | 16:16 |
| rloo | lucasagomes: right. esp since they seem to think that no one wants it. | 16:16 |
| *** appu103_ has joined #openstack-ironic | 16:16 | |
| openstackgerrit | Chris Krelle proposed openstack/ironic: Add support for the audit middleware https://review.openstack.org/272658 | 16:17 |
| rloo | lucasagomes: replied and will try to track down harlowja | 16:17 |
| *** apughazh has joined #openstack-ironic | 16:17 | |
| lucasagomes | rloo, thanks | 16:17 |
| *** rama_y has joined #openstack-ironic | 16:18 | |
| rloo | lucasagomes: it is delaying you though but you have lots of other things to do in the meantime? | 16:18 |
| cinerama | sean-k-mooney, what happens if you try to manually start the services? | 16:18 |
| sean-k-mooney | cinerama: if i use service or start the service is not found | 16:18 |
| openstackgerrit | Yossi Ovadia proposed openstack/ironic-python-agent: Handle diskless hardware connected to remote iscsi https://review.openstack.org/327807 | 16:18 |
| lucasagomes | rloo, right yeah... currently also working on the logs for IPA and a high prio bug about nodes getting stuck on power state transitions | 16:18 |
| sean-k-mooney | cinerama: for example service --status-all dose not show the service | 16:19 |
| lucasagomes | rloo, https://bugs.launchpad.net/ironic/+bug/1588901 | 16:19 |
| openstack | Launchpad bug 1588901 in Ironic "Nodes stuck on power state transitions" [High,In progress] - Assigned to Lucas Alvares Gomes (lucasagomes) | 16:19 |
| lucasagomes | I've added a plumbing patch for it, few reviews so far tho | 16:19 |
| rloo | lucasagomes: ok, lets see what happens first. will try to find out this week. | 16:19 |
| sean-k-mooney | service ironic-conductor start | 16:19 |
| sean-k-mooney | ironic-conductor: unrecognized service | 16:19 |
| lucasagomes | rloo, thanks | 16:19 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: another test for ironic networking https://review.openstack.org/330210 | 16:20 |
| rloo | lucasagomes: that reminds me. i was thinking it would be good to go through all high-priority bugs. [I just want to code...] | 16:20 |
| lucasagomes | rloo, if you want to take a look at the first patch https://review.openstack.org/#/c/327642/ :-) | 16:20 |
| lucasagomes | sean-k-mooney, openstack-ironic-conductor? | 16:20 |
| lucasagomes | depends on the distros packages (or if you manually installed ironic tho) | 16:21 |
| TheJulia | sean-k-mooney: upstart in a container? | 16:21 |
| rloo | lucasagomes: nope. i need to space out my reviewing. oh geez, i looked. dont' you have to add an rfe if the db is being modified? | 16:21 |
| sean-k-mooney | TheJulia: yes | 16:21 |
| sean-k-mooney | ()[root@bifrost playbooks]# service openstack-ironic-conductor status | 16:21 |
| sean-k-mooney | openstack-ironic-conductor: unrecognized service | 16:21 |
| TheJulia | sean-k-mooney: is it the special base "upstart is modified to work" container image? | 16:21 |
| lucasagomes | rloo, well, I could... tho I can only have a timeout for transition if we record the time the transition started | 16:21 |
| lucasagomes | so I thought that having only the bug was fine/obivous for that | 16:22 |
| sean-k-mooney | TheJulia: upstart is other wise working fine, e.g. it works for mysql and rabbitmq | 16:22 |
| lucasagomes | we do the same for inspection_started_at or provision_updated_at thing | 16:22 |
| rloo | lucasagomes: i wonder if we can add '[rfe]' to that bug. | 16:22 |
| cinerama | sean-k-mooney, on ubuntu the service name is different | 16:22 |
| TheJulia | sean-k-mooney: mysql and rabbit ship with sysv init scripts if memory serves | 16:22 |
| sean-k-mooney | TheJulia: yes they do | 16:22 |
| lucasagomes | rloo, we can yeah | 16:22 |
| * lucasagomes guess | 16:22 | |
| rloo | lucasagomes: you're adding a new feature, right? to fix that? | 16:23 |
| sean-k-mooney | cinerama: when i do a service list there are not openstakc or ironic services listed | 16:23 |
| cinerama | sean-k-mooney: how about "service ironic-conductor status"? | 16:23 |
| lucasagomes | rloo, I think you can look at it that way yes | 16:23 |
| TheJulia | sean-k-mooney: so, I guess it is just that we're lacking sysv init scripts for you in a container scenario | 16:23 |
| lucasagomes | at the end it's a bug, because we do not timeout power state transitions | 16:23 |
| lucasagomes | so it potentially can get stuck | 16:24 |
| lucasagomes | but yeah, adding a timeout may also count as a feature I guess | 16:24 |
| sean-k-mooney | well i am runing /sbin/init and if i do /sbin/init --version it is upstart | 16:24 |
| TheJulia | sean-k-mooney: hmmm... I wonder which wins use order wise | 16:24 |
| cinerama | interestingly, my startup scripts work but i do not see the ironic services in service --status-all | 16:25 |
| *** sabeen has quit IRC | 16:25 | |
| sean-k-mooney | this might help http://paste.openstack.org/show/523805/ | 16:25 |
| lucasagomes | rloo, if you prefer having it as an RFE, I'm OK with that too | 16:25 |
| rloo | lucasagomes: so with that spec proposing soft power & nmi, this change will also apply to the soft * stuff too, right? but not the nmi? | 16:26 |
| sean-k-mooney | TheJulia: i think initV scripts will be used if they are present and then upstart but in either case as long as only one version of the script exists the order "should" not matter | 16:27 |
| rloo | lucasagomes: so this is adding a new config option too? | 16:27 |
| lucasagomes | rloo, right, soft reboot does have a timeout tho I have to take a look at how it's implemented | 16:27 |
| lucasagomes | it would be good to keep it in sync | 16:27 |
| rloo | lucasagomes: yeah, i think it is an 'rfe' then :) | 16:27 |
| lucasagomes | rloo, yup | 16:28 |
| lucasagomes | rloo, ack, I will change it | 16:28 |
| cinerama | oh interesting, so i think service --status-all is using /etc/init.d? | 16:28 |
| rloo | lucasagomes: thx. i guess we are rockin' cuz we have lots of stuff we want to do / implement. But sigh, so many to review... | 16:28 |
| lucasagomes | yeah | 16:28 |
| sean-k-mooney | cinerama: yeah it looks like it is | 16:28 |
| cinerama | if i do a service ironic-api status though it looks OK | 16:29 |
| *** Jabadia has joined #openstack-ironic | 16:29 | |
| TheJulia | cinerama: but are you in a container at the moment? | 16:30 |
| rloo | lucasagomes: so I don't know if you should add a new rfe and keep the existing as a real bug, or combine. Personally, I don't care. | 16:30 |
| cinerama | sean-k-mooney, how about "initctl list"? | 16:30 |
| *** Nisha_away has quit IRC | 16:30 | |
| sean-k-mooney | cinerama: its empthy | 16:30 |
| lucasagomes | rloo, I think it's a mix of both | 16:30 |
| *** Nisha_away has joined #openstack-ironic | 16:30 | |
| lucasagomes | perhaps add the rfe tag but make sure people see that it's a bug as well | 16:30 |
| *** derekh has quit IRC | 16:31 | |
| Jabadia | Where is the right place to ask questions about about the 'overcloud-full.initrd' and images in general ? I need to add driver/module to initramfs of overcloud | 16:31 |
| sean-k-mooney | s/empthy/empty | 16:31 |
| rloo | lucasagomes: right, you need that feature to work to address the bug. i've seen in the past, where there has been an rfe that also ended up addressing a bug. again, i don't particularly care :) | 16:31 |
| lucasagomes | rloo, will check it and also talk to naohiro about how thye are planning to do it for soft p off | 16:31 |
| rloo | lucasagomes: good. thx. | 16:31 |
| lucasagomes | rloo, yeah same here, as long as it's fixed | 16:31 |
| lucasagomes | I don't care either :-) | 16:31 |
| lucasagomes | and I will have to call it a day | 16:32 |
| rloo | ciao lucasagomes! | 16:32 |
| lucasagomes | came to the office today for some presentation thingy | 16:32 |
| lucasagomes | and gotta run to catch my train back home | 16:32 |
| lucasagomes | rloo, thanks a lot | 16:32 |
| lucasagomes | see you, have a great evening all | 16:32 |
| *** apughazh has quit IRC | 16:32 | |
| NobodyCam | night lucasagomes | 16:34 |
| cinerama | sean-k-mooney: heh. i think i will need to get a container going etc to help more | 16:34 |
| *** daemontool_ has quit IRC | 16:35 | |
| sean-k-mooney | cinerama: i think its just somthing to do with the service file not being linking in to the right place or something along those lines | 16:35 |
| sean-k-mooney | i was trying some of the suggestions here http://askubuntu.com/questions/299371/correct-way-to-install-a-custom-upstart-service | 16:35 |
| sean-k-mooney | but they did not work | 16:35 |
| cinerama | sean-k-mooney: if you can give me some pointers on how i can duplicate your setup i can have a play | 16:37 |
| sean-k-mooney | im currently working off of https://review.openstack.org/#/c/319703/1/doc/bifrost.rst but the simplet thing to do would be to run a vanilla ubuntu container with /sbin/init as the application | 16:39 |
| sean-k-mooney | cinerama: let me get a command for you | 16:39 |
| *** yogi has quit IRC | 16:40 | |
| cinerama | sean-k-mooney, cool, thanks. this should make for a fun afternoon :) | 16:40 |
| sean-k-mooney | cinerama: docker run -it --net=host -v /dev:/dev -d --privileged --name bifrost ubuntu:14.04 /sbin/init | 16:40 |
| sean-k-mooney | cinerama: that will create a privalaged container with the network namesapce disabled from the ubuntu base image | 16:41 |
| sean-k-mooney | cinerama: you should then be able to run the bifrost install script normally | 16:41 |
| *** sabeen has joined #openstack-ironic | 16:41 | |
| sean-k-mooney | cinerama: well playbook | 16:41 |
| sean-k-mooney | cinerama: if you already have ironic/bifrost running then it will fail because of the --net=host | 16:42 |
| sean-k-mooney | so if you dont want to acctully clean down some machinces when you are done you can leave that out | 16:42 |
| sean-k-mooney | cinerama: kolla will be moving to ubuntu 16.04 with systemd this cycle so if i cant get it working it upstart it wont be an issue but as we currently use 14.04 for ubuntu im testing with it | 16:44 |
| cinerama | sean-k-mooney, well it probably won't be too hard to fix if i can poke at it a bit | 16:45 |
| *** xhku has joined #openstack-ironic | 16:45 | |
| TheJulia | sean-k-mooney: have you seen https://github.com/docker/docker/issues/2276 ? | 16:46 |
| *** apughazh has joined #openstack-ironic | 16:46 | |
| sean-k-mooney | TheJulia: no lets see how broken upstart is in a container. | 16:47 |
| *** Sukhdev_ has quit IRC | 16:47 | |
| *** apughazh has quit IRC | 16:47 | |
| *** Sukhdev_ has joined #openstack-ironic | 16:48 | |
| sean-k-mooney | i do know that there uses to be a special ubuntu-upstart docker file maybe i need to dig into it a little more | 16:48 |
| sean-k-mooney | having said that i may have found the issue | 16:49 |
| TheJulia | I seem to remember bcornec ran into it trying to containerize bifrost a long while back | 16:49 |
| sean-k-mooney | ()[root@bifrost playbooks]# init-checkconf /etc/init/ironic-conductor.conf | 16:49 |
| sean-k-mooney | ERROR: version of /sbin/initctl too old | 16:49 |
| TheJulia | \o | 16:49 |
| TheJulia | err | 16:49 |
| TheJulia | \o/ | 16:49 |
| *** dulek has quit IRC | 16:49 | |
| *** pcaruana has quit IRC | 16:49 | |
| sean-k-mooney | TheJulia: so the first time i tried to contaieries bifrost i use the ubuntu-upstart container with no issues | 16:50 |
| sean-k-mooney | i gues i need to look at what is different between ubuntu-upstart:14.04 and ubuntu:14.04 | 16:50 |
| *** rcernin has quit IRC | 16:51 | |
| sean-k-mooney | cinerama: rather then wasing your time on what might be an issue with the base container let me poke around and see if i change figure out what the difference is | 16:52 |
| sean-k-mooney | cinerama: if im still stuck tomorow ill let ye know. | 16:53 |
| *** baoli has joined #openstack-ironic | 16:53 | |
| *** baoli has quit IRC | 16:54 | |
| cinerama | sean-k-mooney, sounds good. a quick search turned up some possible issues with vanilla ubuntu containers but you're probably more aware of those than i am | 16:54 |
| *** baoli has joined #openstack-ironic | 16:54 | |
| sean-k-mooney | cinerama: ya looks like i need to add this to my docker image to fix the vanila ubuntu container https://github.com/tianon/dockerfiles/blob/4d24a12b54b75b3e0904d8a285900d88d3326361/sbin-init/ubuntu/upstart/14.04/Dockerfile#L12-L36 | 16:54 |
| *** klindgren has quit IRC | 16:55 | |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Introducing Inspector Grenade testing https://review.openstack.org/327667 | 16:59 |
| *** vishwanathj has quit IRC | 17:02 | |
| *** klindgren has joined #openstack-ironic | 17:02 | |
| *** caldw__ has joined #openstack-ironic | 17:03 | |
| *** ChubYann has joined #openstack-ironic | 17:04 | |
| *** aparnav has joined #openstack-ironic | 17:06 | |
| *** appu103_ has quit IRC | 17:06 | |
| *** baoli has quit IRC | 17:07 | |
| *** aparnav has quit IRC | 17:13 | |
| *** aparnav has joined #openstack-ironic | 17:13 | |
| *** praneshp has quit IRC | 17:16 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-lib: Use autospec in mocked objects https://review.openstack.org/335136 | 17:17 |
| *** mbound has quit IRC | 17:17 | |
| harlowja | rloo whats up | 17:18 |
| harlowja | i didn't do it | 17:18 |
| harlowja | lol | 17:18 |
| rloo | harlowja: oh. the oslo.utils patch. sec, let me find it. | 17:18 |
| rloo | harlowja: https://review.openstack.org/#/c/308398 | 17:18 |
| rloo | harlowja: nova extra_specs_ops. i see there are two patches, and some uncertainty as to whether to even put it in oslo.utils. | 17:19 |
| rloo | harlowja: ironic does want it. and lucasagomes has a patch to put it in ironic-libs cuz he doesn't know whether it will get into oslo.utils. | 17:19 |
| *** hoonetorg has quit IRC | 17:19 | |
| rloo | harlowja: i'd rather it just get put in oslo.utils. | 17:19 |
| rloo | harlowja: it seems like you or alexis doesn't want the actual nova code cuz it is ugly? | 17:20 |
| harlowja | ya, i don't like the ugly being in oslo | 17:20 |
| harlowja | oslo != uglystepson | 17:20 |
| harlowja | lol | 17:20 |
| harlowja | or is it uglycousin | 17:20 |
| harlowja | idk | 17:20 |
| rloo | harlowja: is there a way to put the ugly first, then your beautiful, or do you prefer only the beautiful? | 17:20 |
| harlowja | so i was under the impression that jroll said neither were needed | 17:21 |
| harlowja | :-P | 17:21 |
| *** baoli has joined #openstack-ironic | 17:21 | |
| rloo | harlowja: dunno where you got that impression, but i suspect for a period of time, it wasn't needed. but now it is needed again. | 17:21 |
| *** baoli has quit IRC | 17:21 | |
| rloo | harlowja: needed for a different reason i believe. but we want to use similar syntax to nova (unfortunately but they started the precedent) | 17:22 |
| *** xhku has quit IRC | 17:22 | |
| *** PollyZ has quit IRC | 17:22 | |
| *** cdearborn has joined #openstack-ironic | 17:22 | |
| *** baoli has joined #openstack-ironic | 17:22 | |
| *** klindgren has quit IRC | 17:23 | |
| *** gfhellma1 has joined #openstack-ironic | 17:23 | |
| harlowja | ya, i forget somewhere along the line i think jroll was like i'll just do it in sqlcode | 17:23 |
| rloo | harlowja: so, given that ironic wants it. should we just copy nova's version into our ironic-lib for ironic's internal use? | 17:24 |
| *** aNupoisc has joined #openstack-ironic | 17:24 | |
| *** gfhellma has quit IRC | 17:24 | |
| rloo | harlowja: or do you think it is worth putting something in oslo.utils? maybe premature to put in oslo.utils? | 17:24 |
| harlowja | what was wrong with https://review.openstack.org/#/c/313699/ :-P | 17:26 |
| *** xhku has joined #openstack-ironic | 17:26 | |
| harlowja | it at least formalized it somewhat | 17:26 |
| harlowja | that was the less ugly one | 17:26 |
| *** xhku has quit IRC | 17:26 | |
| *** klindgren has joined #openstack-ironic | 17:26 | |
| rloo | harlowja: nothing (I think, I actually haven't looked at any of the patches). | 17:26 |
| rloo | harlowja: the concern is whether your version works with nova. so we wanted to take a cautious approach. i think. (shouldn't say 'we') | 17:27 |
| *** xhku has joined #openstack-ironic | 17:27 | |
| rloo | harlowja: i suspect that if there were enough unit tests to test it out, it would be fine. | 17:27 |
| harlowja | ya, i think we built up enough knowledge while making that 313699 that we know where the diffs are (there is a tiny one) | 17:27 |
| rloo | harlowja: and i thought your patch relied on the first patch, which is why alexis or someone said to squash them? | 17:27 |
| harlowja | nah second patch ummm sorta just rewrote the first one, lol | 17:28 |
| rloo | harlowja: so i like having the original there, so you can see the history. | 17:28 |
| rloo | harlowja: but i don't know how oslo/libraries do things. | 17:28 |
| rloo | harlowja: eg, if it had been in that incubator dir, you'd see the original + changes to that. | 17:28 |
| harlowja | i'd be ok with that, merging the original and then replacing it | 17:29 |
| harlowja | for history reasons | 17:29 |
| *** caldw__ has quit IRC | 17:29 | |
| harlowja | just the whole theory of 'nova has scary code --> oslo' scares me | 17:29 |
| rloo | harlowja: ok, so basically, we need to make sure it works and there are sufficient unit tests. | 17:29 |
| harlowja | don't like oslo being a dump | 17:29 |
| harlowja | yup | 17:29 |
| harlowja | it has a bunch | 17:29 |
| rloo | harlowja: c'mon, let's keep scary code around. so ccccooool. | 17:29 |
| harlowja | :( | 17:29 |
| harlowja | cool if u like working in a dump | 17:30 |
| rloo | ha ha | 17:30 |
| rloo | geez. now i wonder if i have to review those patches. | 17:30 |
| rloo | harlowja: so i guess you're going to say that you think your patch is complete? | 17:30 |
| * dtantsur wants to gently remind that ugly working code is still better than beautiful non-working | 17:31 | |
| *** Jabadia has quit IRC | 17:31 | |
| harlowja | hey the new stuff works :-P | 17:32 |
| harlowja | passes all the same unit tests ;) | 17:32 |
| rloo | dtantsur: yup. that's why i want to make sure there are sufficient unit tests! | 17:32 |
| harlowja | its stricter than the old stuff, but imho that's ok | 17:32 |
| harlowja | thats the diff, this new one is strict, not letting random crap in, lol | 17:32 |
| rloo | harlowja: if it is stricter, it is possible it won't work for nova. | 17:32 |
| rloo | harlowja: in case random crap already got in, in nova... | 17:33 |
| dtantsur | harlowja, my definitions of works for libraries usually include "merged" ;) | 17:33 |
| harlowja | rloo ya, the nova people ... no comment, lol | 17:33 |
| harlowja | dtantsur oh | 17:33 |
| harlowja | why u have so high of expections | 17:33 |
| dtantsur | lol | 17:33 |
| harlowja | expectations | 17:33 |
| rloo | harlowja: i think we are fine with a stricter version, as long as the nova operators work similar to what nova expects. | 17:33 |
| harlowja | afaik it should | 17:33 |
| dtantsur | yeah, when I just propose a patch it's usually brilliant, then people make ugly sh*t of it with their reviews.. | 17:34 |
| * dtantsur is kidding of course :D | 17:34 | |
| harlowja | ya i know that feeling | 17:34 |
| harlowja | i'm not kidding | 17:34 |
| harlowja | lol | 17:34 |
| harlowja | damn humans | 17:34 |
| harlowja | like rloo | 17:34 |
| harlowja | lol | 17:34 |
| rloo | harlowja: really? why would you take sh*t from folks? | 17:34 |
| harlowja | lol | 17:34 |
| harlowja | sometimes the shit is convered in gol | 17:34 |
| harlowja | *gold | 17:34 |
| * dtantsur checks if it's Friday already | 17:34 | |
| harlowja | so u don't realize its shit till after the fact | 17:34 |
| harlowja | when u get around to dusting it , then u realize its not gold | 17:35 |
| harlowja | lol | 17:35 |
| * rloo worries that this is rapidly going downhill or down the sewer or something... | 17:35 | |
| harlowja | lol | 17:35 |
| harlowja | so ya, just review those things | 17:35 |
| harlowja | :-P | 17:35 |
| rloo | harlowja: ok thx. i'll add a comment to that effect. and try to review them this week. | 17:35 |
| harlowja | thx boss | 17:36 |
| * rloo is glad that harlowja knows who is boss | 17:36 | |
| harlowja | i'm not stupid | 17:37 |
| harlowja | u don't mess with the ruby | 17:37 |
| harlowja | lol | 17:37 |
| * dtantsur is pretty sure it's Friday now | 17:37 | |
| harlowja | lol | 17:37 |
| *** [1]cdearborn has quit IRC | 17:37 | |
| *** d0ugal has quit IRC | 17:37 | |
| TheJulia | lol | 17:37 |
| *** d0ugal has joined #openstack-ironic | 17:37 | |
| *** d0ugal has quit IRC | 17:37 | |
| *** d0ugal has joined #openstack-ironic | 17:37 | |
| * rloo is going to copy what harlowja says, for whatever performance thing i might need. | 17:38 | |
| harlowja | performance thing? | 17:38 |
| harlowja | lol | 17:38 |
| rloo | harlowja: who knows. these companies all make you do performance things. this is going to be "rloo's influence upstream". | 17:38 |
| rloo | harlowja: performance review. sorry, not performance of hw/sw :) | 17:39 |
| harlowja | wfm | 17:39 |
| harlowja | bb | 17:39 |
| rloo | thx larsks | 17:40 |
| rloo | oops, larsks, that was a mistake. please ignore. | 17:40 |
| rloo | thx harlowja | 17:40 |
| *** moshele has joined #openstack-ironic | 17:41 | |
| *** d0ugal has quit IRC | 17:41 | |
| *** d0ugal has joined #openstack-ironic | 17:42 | |
| *** d0ugal has quit IRC | 17:42 | |
| *** d0ugal has joined #openstack-ironic | 17:42 | |
| *** sabeen has quit IRC | 17:42 | |
| *** gfhellma1 has quit IRC | 17:42 | |
| *** sabeen has joined #openstack-ironic | 17:43 | |
| *** sabeen has quit IRC | 17:43 | |
| *** praneshp has joined #openstack-ironic | 17:43 | |
| *** e0ne has joined #openstack-ironic | 17:44 | |
| *** PollyZ has joined #openstack-ironic | 17:44 | |
| *** rcernin has joined #openstack-ironic | 17:45 | |
| *** aparnav has quit IRC | 17:46 | |
| *** appu103_ has joined #openstack-ironic | 17:48 | |
| *** PollyZ has quit IRC | 17:50 | |
| *** trown is now known as trown|lunch | 17:50 | |
| *** dtantsur is now known as dtantsur|afk | 17:51 | |
| *** piet has quit IRC | 17:52 | |
| *** gfhellma has joined #openstack-ironic | 17:55 | |
| *** fragatin_ has joined #openstack-ironic | 17:55 | |
| *** fragatina has quit IRC | 17:55 | |
| harlowja | np | 17:55 |
| sean-k-mooney | cinerama: TheJulia: just as an fyi when i rebuild the container with the changes from the ubunut-upstart container everything works fine. | 17:55 |
| JayF | dtantsur|afk: re: my vote on that about not landing things with broken dpes | 17:56 |
| JayF | dtantsur|afk: I really don't want us to start playing the ipa-ramdisk-feature-matrix game | 17:56 |
| *** e0ne has quit IRC | 17:56 | |
| *** Goneri has joined #openstack-ironic | 17:57 | |
| sean-k-mooney | cinerama: TheJulia: the ironic service still dont show up in service --status-all but there in initctl list just as cinerama saw out side of a container | 17:57 |
| cinerama | sean-k-mooney, that's probably okay - status-all only works with sysvinit (it says so in the man page) | 18:02 |
| *** electrofelix is now known as electrofelix|afk | 18:03 | |
| *** sukhdev has quit IRC | 18:03 | |
| sean-k-mooney | cinerama: yep well all the service seam to be working. im going to try and enroll the vms i create with the create vms script and check if it can deploy | 18:04 |
| sean-k-mooney | im failly sure everything is now working correctly. | 18:05 |
| sambetts | dtantsur|afk: the reasoning behind having network interfaces per node is because different nodes have different capabilties in regards to networking, for example my hardware can do things that others can't, and a big ask of customers is to support mixed hardare environments but not be limited to the feature set of the node with the least features | 18:05 |
| sambetts | dtantsur|afk: at least thats my use case for having them | 18:06 |
| cinerama | sean-k-mooney: great, i'm glad things are working now | 18:06 |
| *** sambetts is now known as sambetts|afk | 18:06 | |
| *** Sukhdev_ has quit IRC | 18:09 | |
| *** apughazh has joined #openstack-ironic | 18:14 | |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Unify test playbooks https://review.openstack.org/327912 | 18:15 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: Introducing Inspector Grenade testing https://review.openstack.org/327667 | 18:16 |
| *** mbound has joined #openstack-ironic | 18:18 | |
| *** mbound has quit IRC | 18:24 | |
| *** pcaruana has joined #openstack-ironic | 18:25 | |
| *** PollyZ has joined #openstack-ironic | 18:25 | |
| *** sabeen has joined #openstack-ironic | 18:30 | |
| *** PollyZ has quit IRC | 18:30 | |
| *** PollyZ has joined #openstack-ironic | 18:35 | |
| *** baoli has quit IRC | 18:40 | |
| *** baoli has joined #openstack-ironic | 18:40 | |
| *** Sukhdev has joined #openstack-ironic | 18:40 | |
| *** athomas has quit IRC | 18:41 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Disable flat network driver https://review.openstack.org/335060 | 18:41 |
| *** baoli has quit IRC | 18:41 | |
| *** baoli has joined #openstack-ironic | 18:42 | |
| *** davidlenwell has quit IRC | 18:42 | |
| *** gfhellma has quit IRC | 18:43 | |
| *** rbudden has quit IRC | 18:43 | |
| *** Nisha_away has quit IRC | 18:48 | |
| *** davidlenwell has joined #openstack-ironic | 18:52 | |
| *** baoli has quit IRC | 18:55 | |
| *** trown|lunch is now known as trown | 18:56 | |
| *** baoli has joined #openstack-ironic | 19:00 | |
| *** baoli has quit IRC | 19:01 | |
| *** sdake has joined #openstack-ironic | 19:02 | |
| *** sdake_ has quit IRC | 19:03 | |
| *** baoli has joined #openstack-ironic | 19:03 | |
| *** rook_ has quit IRC | 19:03 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Do Not Merge: another test for ironic networking https://review.openstack.org/330210 | 19:07 |
| *** klindgren has quit IRC | 19:08 | |
| *** rook_ has joined #openstack-ironic | 19:11 | |
| *** jcoufal has quit IRC | 19:12 | |
| *** baoli has quit IRC | 19:12 | |
| *** appu103_ has quit IRC | 19:13 | |
| *** baoli has joined #openstack-ironic | 19:13 | |
| *** baoli has quit IRC | 19:13 | |
| *** jaybeale has quit IRC | 19:14 | |
| *** baoli has joined #openstack-ironic | 19:14 | |
| *** fellypefca has joined #openstack-ironic | 19:15 | |
| *** sdake_ has joined #openstack-ironic | 19:16 | |
| *** terje has quit IRC | 19:17 | |
| *** baoli has quit IRC | 19:18 | |
| *** athomas has joined #openstack-ironic | 19:19 | |
| *** sdake has quit IRC | 19:20 | |
| *** _vsaienko has joined #openstack-ironic | 19:21 | |
| PollyZ | I have a question, if I would like to get a list of headers, what the argument would be, I tried "flask.request.headers()", "flask.request.headers.get()" and something else, but none of these works | 19:22 |
| *** baoli has joined #openstack-ironic | 19:28 | |
| *** baoli has quit IRC | 19:28 | |
| *** baoli has joined #openstack-ironic | 19:29 | |
| *** aNupoisc has quit IRC | 19:30 | |
| *** moshele has quit IRC | 19:32 | |
| *** ayoung has quit IRC | 19:35 | |
| *** fragatin_ has quit IRC | 19:36 | |
| *** moshele has joined #openstack-ironic | 19:37 | |
| *** ametts has quit IRC | 19:40 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-lib: Add reno to test-requirements https://review.openstack.org/335205 | 19:42 |
| *** [2]cdearborn has joined #openstack-ironic | 19:43 | |
| *** xhku has quit IRC | 19:44 | |
| *** fragatina has joined #openstack-ironic | 19:48 | |
| *** moshele has quit IRC | 19:49 | |
| *** klindgren has joined #openstack-ironic | 19:53 | |
| *** hoonetorg has joined #openstack-ironic | 19:54 | |
| *** ametts has joined #openstack-ironic | 19:56 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-lib: Use autospec in mocked objects https://review.openstack.org/335136 | 19:57 |
| *** cdearborn has quit IRC | 19:57 | |
| *** gfhellma has joined #openstack-ironic | 19:58 | |
| *** athomas has quit IRC | 20:02 | |
| mariojv | PollyZ: http://stackoverflow.com/questions/29386995/how-to-get-http-headers-in-flask might help | 20:03 |
| *** PollyZ_ has joined #openstack-ironic | 20:05 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-lib: Add keyword arg 'log_stdout' to utils.execute() https://review.openstack.org/335211 | 20:07 |
| *** PollyZ has quit IRC | 20:07 | |
| *** _vsaienko has quit IRC | 20:08 | |
| *** rook_ has quit IRC | 20:17 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-lib: Add reno to test-requirements https://review.openstack.org/335205 | 20:18 |
| *** makowals has joined #openstack-ironic | 20:18 | |
| *** makowals has quit IRC | 20:18 | |
| *** jcoufal has joined #openstack-ironic | 20:21 | |
| *** livelace has quit IRC | 20:22 | |
| *** caldw__ has joined #openstack-ironic | 20:23 | |
| openstackgerrit | Ruby Loo proposed openstack/ironic-lib: Add keyword arg 'log_stdout' to utils.execute() https://review.openstack.org/335211 | 20:23 |
| *** ayoung has joined #openstack-ironic | 20:29 | |
| *** jcoufal has quit IRC | 20:30 | |
| *** rbudden has joined #openstack-ironic | 20:30 | |
| *** fellypefca has quit IRC | 20:38 | |
| *** jcoufal has joined #openstack-ironic | 20:38 | |
| *** caldw__ has quit IRC | 20:40 | |
| *** jtomasek_ has joined #openstack-ironic | 20:41 | |
| *** jjohnson2 has quit IRC | 20:43 | |
| *** e0ne has joined #openstack-ironic | 20:45 | |
| openstackgerrit | Pradeep Chandrasekar proposed openstack/ironic: Fix typo in inspection.rst https://review.openstack.org/335219 | 20:45 |
| *** ametts has quit IRC | 20:49 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Implement Keystone Policy Support https://review.openstack.org/327437 | 20:52 |
| *** Goneri has quit IRC | 20:53 | |
| *** gfhellma1 has joined #openstack-ironic | 20:54 | |
| *** fragatina has quit IRC | 20:54 | |
| *** fragatina has joined #openstack-ironic | 20:54 | |
| *** aNupoisc has joined #openstack-ironic | 20:55 | |
| *** gfhellma has quit IRC | 20:55 | |
| *** sdake has joined #openstack-ironic | 21:01 | |
| JayF | someone who knows a thing or three about spec formatting mind helping me try and suss out why I'm getting bad parsing on that Ironic spec? http://docs-draft.openstack.org/37/327437/5/check/gate-ironic-specs-docs/66d07b3//doc/build/html/specs/not-implemented/keystone-policy-support.html | 21:02 |
| JayF | from https://review.openstack.org/327437 | 21:03 |
| JayF | my syntax matches every rst doc I found | 21:03 |
| JayF | and I can't find a working example in ironic history | 21:03 |
| *** sdake_ has quit IRC | 21:04 | |
| *** blakec has joined #openstack-ironic | 21:05 | |
| *** trown is now known as trown|outtypewww | 21:08 | |
| *** dmk0202 has joined #openstack-ironic | 21:11 | |
| *** e0ne has quit IRC | 21:12 | |
| *** blakec1 has joined #openstack-ironic | 21:14 | |
| JayF | aha finally got it | 21:16 |
| JayF | tl;dr blah[0]_ doesn't render, blah [0]_ does | 21:16 |
| *** blakec has quit IRC | 21:17 | |
| *** harlowja has quit IRC | 21:19 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Implement Keystone Policy Support https://review.openstack.org/327437 | 21:19 |
| *** piet has joined #openstack-ironic | 21:21 | |
| *** blakec has joined #openstack-ironic | 21:22 | |
| *** blakec1 has quit IRC | 21:25 | |
| *** baoli has quit IRC | 21:30 | |
| *** baoli has joined #openstack-ironic | 21:33 | |
| *** baoli has quit IRC | 21:34 | |
| *** baoli has joined #openstack-ironic | 21:34 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Implement rescue mode https://review.openstack.org/171878 | 21:39 |
| *** sdake has quit IRC | 21:39 | |
| *** dmk0202 has quit IRC | 21:42 | |
| *** piet has quit IRC | 21:48 | |
| *** piet has joined #openstack-ironic | 21:48 | |
| *** thiagop has quit IRC | 21:53 | |
| *** jtomasek_ has quit IRC | 21:56 | |
| *** blakec has quit IRC | 22:01 | |
| *** davidlenwell has quit IRC | 22:02 | |
| *** piet has quit IRC | 22:03 | |
| *** baoli has quit IRC | 22:08 | |
| *** gfhellma1 has quit IRC | 22:10 | |
| *** davidlenwell has joined #openstack-ironic | 22:11 | |
| *** aNupoisc has quit IRC | 22:15 | |
| *** harlowja has joined #openstack-ironic | 22:16 | |
| *** harlowja has quit IRC | 22:23 | |
| *** [2]cdearborn has quit IRC | 22:27 | |
| *** wajdi_ has joined #openstack-ironic | 22:28 | |
| *** wajdi__ has joined #openstack-ironic | 22:29 | |
| *** wajdi_ has quit IRC | 22:29 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Implement Keystone Policy Support https://review.openstack.org/327437 | 22:31 |
| *** wajdi has quit IRC | 22:31 | |
| JayF | mariojv: ^ thanks, fixed the typos | 22:31 |
| JayF | mariojv: and responded to the other comment | 22:32 |
| *** mbound_ has joined #openstack-ironic | 22:33 | |
| *** wajdi has joined #openstack-ironic | 22:34 | |
| *** aNupoisc has joined #openstack-ironic | 22:35 | |
| *** rama_y has quit IRC | 22:35 | |
| mariojv | JayF: thanks, still lgtm | 22:36 |
| mariojv | the more granular operations against nodes thing would certainly be interesting | 22:37 |
| *** wajdi__ has quit IRC | 22:37 | |
| JayF | yeah; IDK if it's a technical thing or a "take bite sized pieces" that is why we're not limiting it | 22:38 |
| mariojv | potentially lots of use cases; it might make it easier for someone to run a control plane on top of ironic or something | 22:38 |
| JayF | I think it gets a lot more expensive to decide things other than based on endpoint | 22:38 |
| JayF | it's def something for us to look at when splitting the v2 api, to make/allow for saner policies | 22:38 |
| *** PollyZ_ has quit IRC | 22:38 | |
| mariojv | maybe, probably depends on how big the deployment and what you're filtering on | 22:39 |
| mariojv | if tags are used and indexed in the db or something, not that expensive, but idk | 22:39 |
| JayF | I mean, think about it this way: | 22:39 |
| JayF | if in routes you block based on policy, DoS attack requires a lot more hits | 22:39 |
| *** wajdi has quit IRC | 22:39 | |
| JayF | than if you get down to parsing the body, looking up the tags on a node, etc etc | 22:39 |
| JayF | not saying we should never do it, just that it's not as much of a slam dunk as all the other pieces are | 22:39 |
| mariojv | for sure, it would be more complicated | 22:40 |
| mrda | Morning all | 22:41 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Make agent driver always PXE before cleaning https://review.openstack.org/263842 | 22:47 |
| *** Sukhdev has quit IRC | 22:47 | |
| *** Sukhdev has joined #openstack-ironic | 22:55 | |
| *** baoli_ has joined #openstack-ironic | 22:57 | |
| *** rcernin has quit IRC | 23:03 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic: Make agent driver always PXE before cleaning https://review.openstack.org/263842 | 23:05 |
| *** praneshp has quit IRC | 23:05 | |
| *** harlowja has joined #openstack-ironic | 23:12 | |
| *** klindgren has quit IRC | 23:25 | |
| *** gfhellma has joined #openstack-ironic | 23:25 | |
| *** [2]cdearborn has joined #openstack-ironic | 23:25 | |
| *** sabeen has quit IRC | 23:28 | |
| *** cdearborn has joined #openstack-ironic | 23:31 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-python-agent: Use latest CoreOS stable when building https://review.openstack.org/327891 | 23:40 |
| *** mbound_ has quit IRC | 23:44 | |
| *** adn has joined #openstack-ironic | 23:44 | |
| *** aNupoisc has quit IRC | 23:45 | |
| *** aNupoisc has joined #openstack-ironic | 23:46 | |
| *** [2]cdearborn has quit IRC | 23:48 | |
| *** piet has joined #openstack-ironic | 23:48 | |
| *** adn has quit IRC | 23:49 | |
| *** cdearborn has quit IRC | 23:51 | |
| *** aNupoisc has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!