| *** TxGirlGeek has quit IRC | 00:01 | |
| *** sdake has joined #openstack-ironic | 00:05 | |
| *** dustinc has quit IRC | 00:12 | |
| *** sdake has quit IRC | 00:29 | |
| *** sdake has joined #openstack-ironic | 00:49 | |
| *** mjturek has quit IRC | 00:50 | |
| *** sdake has quit IRC | 01:09 | |
| *** jcoufal has quit IRC | 01:22 | |
| *** gyee has quit IRC | 01:33 | |
| *** _fragatina has quit IRC | 01:56 | |
| *** jcoufal has joined #openstack-ironic | 02:01 | |
| *** TxGirlGeek has joined #openstack-ironic | 02:02 | |
| *** hwoarang has quit IRC | 02:37 | |
| *** hwoarang has joined #openstack-ironic | 02:38 | |
| *** dims has quit IRC | 02:38 | |
| *** dims has joined #openstack-ironic | 02:55 | |
| *** sdake has joined #openstack-ironic | 03:00 | |
| *** mmethot has quit IRC | 03:14 | |
| *** mmethot has joined #openstack-ironic | 03:21 | |
| *** dnuka has joined #openstack-ironic | 03:29 | |
| dnuka | good morning ironic | 03:30 |
|---|---|---|
| *** gkadam-bmgr has joined #openstack-ironic | 03:34 | |
| *** _fragatina has joined #openstack-ironic | 04:18 | |
| *** TxGirlGeek has quit IRC | 04:18 | |
| *** _fragatina has quit IRC | 04:19 | |
| *** _fragatina has joined #openstack-ironic | 04:19 | |
| *** _fragatina has quit IRC | 04:21 | |
| *** _fragatina has joined #openstack-ironic | 04:21 | |
| *** _fragatina has quit IRC | 04:27 | |
| *** _fragatina has joined #openstack-ironic | 04:28 | |
| *** _fragatina has quit IRC | 04:31 | |
| *** _fragatina has joined #openstack-ironic | 04:31 | |
| *** _fragatina has quit IRC | 04:32 | |
| *** _fragatina has joined #openstack-ironic | 04:33 | |
| *** _fragatina has quit IRC | 04:41 | |
| *** _fragatina has joined #openstack-ironic | 04:42 | |
| *** _fragatina has quit IRC | 04:42 | |
| *** _fragatina has joined #openstack-ironic | 04:43 | |
| *** _fragatina has quit IRC | 04:44 | |
| *** _fragatina has joined #openstack-ironic | 04:44 | |
| *** rh-jelabarre has quit IRC | 05:07 | |
| *** _fragatina has quit IRC | 05:11 | |
| *** _fragatina has joined #openstack-ironic | 05:11 | |
| *** _fragatina has quit IRC | 05:16 | |
| *** _fragatina has joined #openstack-ironic | 05:16 | |
| *** _fragatina has quit IRC | 05:19 | |
| *** _fragatina has joined #openstack-ironic | 05:20 | |
| *** _fragatina has quit IRC | 05:38 | |
| *** _fragatina has joined #openstack-ironic | 05:39 | |
| *** e0ne has joined #openstack-ironic | 06:12 | |
| *** zzzeek has quit IRC | 06:33 | |
| *** zzzeek has joined #openstack-ironic | 06:37 | |
| *** moshele has joined #openstack-ironic | 06:58 | |
| *** moshele has quit IRC | 07:02 | |
| *** e0ne has quit IRC | 07:05 | |
| *** openstackgerrit has joined #openstack-ironic | 07:15 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic master: Removing deprecated drac_host property https://review.openstack.org/633902 | 07:15 |
| *** hamdyk has joined #openstack-ironic | 07:16 | |
| TheJulia | Good morning! | 07:17 |
| dnuka | good morning TheJulia | 07:18 |
| TheJulia | manjeets: from what your describing, it sounds like your version of neutron is out of date. | 07:18 |
| TheJulia | dtantsur|afk: My asessment is that you are sane. I'm sorry. | 07:21 |
| *** dnuka is now known as dnuka|brb | 07:22 | |
| arne_wiebalck | good morning, Ironic! | 07:24 |
| *** dnuka|brb is now known as dnuka | 07:25 | |
| dnuka | hi arne_wiebalck o/ | 07:25 |
| arne_wiebalck | hey dnuka o/ | 07:26 |
| arne_wiebalck | TheJulia: Welcome to the Ironic earlybrids :) | 07:27 |
| TheJulia | :) | 07:27 |
| TheJulia | I'm on a train from Brno to Praha at the moment :) | 07:28 |
| dnuka | :) | 07:29 |
| *** jiapei has quit IRC | 07:35 | |
| *** jiapei has joined #openstack-ironic | 07:35 | |
| jiapei | Good afternoon everybody | 07:38 |
| *** jcoufal has quit IRC | 07:39 | |
| dnuka | good morning jiapei | 07:39 |
| jiapei | Hey morning dnuka | 07:39 |
| TheJulia | good morning jiapei | 07:39 |
| jiapei | Good morning TheJulia | 07:39 |
| jiapei | Praha is really a beautiful city, it's on my wish list :) | 07:40 |
| *** hwoarang has quit IRC | 07:41 | |
| *** hwoarang has joined #openstack-ironic | 07:44 | |
| arne_wiebalck | good morninf jiapei o/ | 07:44 |
| arne_wiebalck | s/morninf/morning/ | 07:44 |
| arne_wiebalck | TheJulia: Holidays or work? | 07:45 |
| jiapei | o/ Good morning arne_wiebalck | 07:45 |
| TheJulia | arne_wiebalck: work | 07:48 |
| TheJulia | jiapei: sadly it will be a depart train, get on bus to the airport kind of visit. | 07:48 |
| etingof | TheJulia, don't feel bad about it, Praha is not that welcoming in the winter time... better come again in Spring o/ | 07:49 |
| dnuka | good morning etingof o/ | 07:49 |
| etingof | o/ dnuka | 07:50 |
| jiapei | Morning etingof o/ | 07:51 |
| arne_wiebalck | Hey etingof o/ | 07:52 |
| etingof | hey-hey ironic \o | 07:52 |
| TheJulia | etingof: I will say the view of the countryside from the train is beautiful so far | 07:53 |
| etingof | TheJulia, it is, though it's better with all the chlorophyll cells up and running | 07:55 |
| *** jcoufal has joined #openstack-ironic | 07:56 | |
| TheJulia | :) | 07:56 |
| TheJulia | So before I go searching for all the things.... is anyone aware of things that require my immediate attention? | 08:04 |
| *** rpittau has joined #openstack-ironic | 08:07 | |
| TheJulia | o/ rpittau | 08:08 |
| rpittau | good morning ironic! o/ | 08:08 |
| rpittau | hey TheJulia :) | 08:08 |
| dnuka | hi rpittau o/ | 08:08 |
| arne_wiebalck | good morning rpittau o/ | 08:08 |
| rpittau | hi dnuka :) | 08:08 |
| rpittau | hey arne_wiebalck :) | 08:08 |
| *** dnuka is now known as dnuka|brb | 08:09 | |
| openstackgerrit | Merged openstack/ironic-specs master: Better explain how do we know that the Smart NIC is ready to be programmed https://review.openstack.org/632971 | 08:14 |
| *** sburette has joined #openstack-ironic | 08:36 | |
| *** amoralej|off is now known as amoralej | 08:49 | |
| *** dsneddon has quit IRC | 08:51 | |
| *** dsneddon has joined #openstack-ironic | 08:54 | |
| *** dnuka|brb is now known as dnuka | 08:57 | |
| TheJulia | If anyone is bored, adding "owner" field support to python-ironicclient would be awesome | 09:02 |
| *** iurygregory has joined #openstack-ironic | 09:05 | |
| etingof | that's not the 'description' field, right? | 09:05 |
| iurygregory | good morning everyone o/ | 09:05 |
| rpittau | hi iurygregory :) | 09:05 |
| dnuka | TheJulia, I'm interested :), but I'm not sure what you mean by "owner" field. | 09:06 |
| dnuka | morning iurygregory o/ | 09:07 |
| TheJulia | dnuka: I added an "owner" field to the node object | 09:07 |
| TheJulia | but never got to the client side of adding the ability to get/set in the library becasue of $TIME | 09:07 |
| TheJulia | or lack their of time | 09:07 |
| dnuka | TheJulia, #link :) | 09:08 |
| TheJulia | etingof: yeah, description is different. Owner is a dual purposed field right now but we only added information storage | 09:08 |
| etingof | for reference -- https://review.openstack.org/#/c/632935 | 09:08 |
| TheJulia | dnuka: the ironic change which merged? | 09:08 |
| patchbot | patch 632935 - python-ironicclient - Support node description - 1 patch set | 09:08 |
| TheJulia | rpittau: I looked at the tinycore change for IPA. It looks like it is still running in the ramdisk with python2.7. see: http://logs.openstack.org/79/633479/5/check/ipa-tempest-wholedisk-bios-ipmi-direct-tinyipa-src/a878a2f/controller/logs/ironic-bm-logs/node-0_no_ansi_2019-01-29-12:56:40_log.txt.gz | 09:12 |
| *** S4ren has joined #openstack-ironic | 09:13 | |
| TheJulia | rpittau: looking at the commit message again, I see you intend to make it a separate patch, that the patch is just to prepare.... so I just workflow+1'ed it. :) | 09:14 |
| rpittau | TheJulia, that patch doesn't include pyuthon 3 :) | 09:14 |
| rpittau | TheJulia, yes, I ahve the py3 patch in the working, it should be ok but I was waiting for this one to merge first, didnt want to mixup too many things | 09:14 |
| *** e0ne has joined #openstack-ironic | 09:15 | |
| *** S4ren has quit IRC | 09:16 | |
| *** S4ren has joined #openstack-ironic | 09:16 | |
| rpittau | TheJulia, thanks! | 09:18 |
| dnuka | etingof, I need to create a same kind of patch like this one - https://review.openstack.org/#/c/632935, but for the "owner" field right? :) | 09:19 |
| patchbot | patch 632935 - python-ironicclient - Support node description - 1 patch set | 09:19 |
| TheJulia | dnuka: exactly, that is if your up for it | 09:20 |
| TheJulia | hmm... why has train slowed to a crawl | 09:21 |
| *** tssurya has joined #openstack-ironic | 09:21 | |
| *** dtantsur|afk is now known as dtantsur | 09:21 | |
| dtantsur | morning ironic | 09:21 |
| dnuka | TheJulia, I'll give it a try :) | 09:21 |
| TheJulia | dnuka: awesome | 09:22 |
| dnuka | morning dtantsur o/ | 09:22 |
| TheJulia | dtantsur: Sorry your sane ;) | 09:22 |
| dtantsur | WUT? >_< | 09:22 |
| * TheJulia giggles | 09:22 | |
| dtantsur | hehe, good morning TheJulia. how's metaltubing going? | 09:22 |
| TheJulia | It goes, approaching Praha | 09:22 |
| rpittau | hi dtantsur :) | 09:22 |
| dtantsur | \o | 09:22 |
| iurygregory | morning dtantsur and TheJulia o/ | 09:23 |
| TheJulia | which means my laptopping is over until i park myself someplace after airport security | 09:23 |
| TheJulia | ugh, looks like https://review.openstack.org/#/c/628223/ got a transitory py27 failure | 09:28 |
| patchbot | patch 628223 - ironic - Allocation API: conductor API (without HA and take... - 10 patch sets | 09:28 |
| dtantsur | :( | 09:29 |
| TheJulia | Its still in the gate queue :( | 09:29 |
| dtantsur | okie, I'll recheck later | 09:29 |
| TheJulia | I'm hoping we can get the smartnic stuff queued up next week if we can get allocation merged, since we need to get api things merging | 09:30 |
| dtantsur | well, we need a 2nd +2 for the API patch *quick glance at mgoddard* | 09:31 |
| * mgoddard pops up head | 09:31 | |
| TheJulia | Okay, closing laptop and putting it away. I doub;t I'll have any sort of inflight wifi... so might not get on until later :( | 09:31 |
| TheJulia | At least this is a short flight | 09:32 |
| mgoddard | dtantsur: I'll take a look at some point today. Knee deep in Tenks right now | 09:32 |
| dtantsur | cool! | 09:36 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/python-ironicclient master: Move to zuulv3 https://review.openstack.org/633010 | 09:42 |
| *** trungnv has quit IRC | 09:47 | |
| *** derekh has joined #openstack-ironic | 09:48 | |
| openstackgerrit | Merged openstack/ironic master: Remove uses of logger name "oslo_messaging" https://review.openstack.org/633253 | 09:55 |
| *** dougsz has joined #openstack-ironic | 09:57 | |
| openstackgerrit | Merged openstack/ironic master: Add is_smartnic to Port data model https://review.openstack.org/629640 | 10:24 |
| openstackgerrit | Merged openstack/python-ironicclient master: Support for conductors exposed from API https://review.openstack.org/624032 | 10:24 |
| openstackgerrit | Merged openstack/ironic-python-agent master: Build tinyipa with tinycore 8.x https://review.openstack.org/633479 | 10:24 |
| dnuka | congrats rpittau :) | 10:24 |
| rpittau | oh nice :) | 10:25 |
| rpittau | thanks dnuka | 10:25 |
| openstackgerrit | M V P Nitesh proposed openstack/ironic master: Provides mount point as cinder requires it to attach volume https://review.openstack.org/633459 | 10:25 |
| dnuka | :) | 10:25 |
| rpittau | time to go for py3 | 10:25 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Support using JSON-RPC instead of oslo.messaging https://review.openstack.org/633052 | 10:26 |
| openstackgerrit | Mark Goddard proposed openstack/tenks master: Allow setting node and volume name prefixes per-spec https://review.openstack.org/633900 | 10:26 |
| dtantsur | hjensas: congrats on the neutron patch finally merging :) I guess we need it backported? | 10:28 |
| *** jistr is now known as jistr|chat | 10:31 | |
| hjensas | dtantsur: I pushed backports to rocky and queens upstream. It did'nt rebase cleanly on Pike. | 10:34 |
| *** rnoriega has quit IRC | 10:35 | |
| hjensas | dtantsur: I think we might not need it for pre-Queens. We did'nt have routed networks support in Ironic at that point (if I remember correctly). And updating the MAC is kind of only Ironic usecase. | 10:36 |
| *** rnoriega has joined #openstack-ironic | 10:36 | |
| dtantsur | cool, thanks! | 10:36 |
| hjensas | dtantsur: Will I see you at the beer event tomorrow? | 10:39 |
| dtantsur | hjensas: what's the beer event? | 10:40 |
| dtantsur | we planned on just hanging out downtown, to be honest | 10:40 |
| hjensas | Fosdem, there is usually somthing at Dillerium. | 10:40 |
| etingof | almost delirium ^ | 10:42 |
| * iurygregory : delerium beer? <3 | 10:42 | |
| etingof | perhaps delirium comes in Dillerium | 10:42 |
| hjensas | etingof: thanks, there is a bigger chance I'll find the place again now. :D | 10:43 |
| *** e0ne has quit IRC | 10:46 | |
| dtantsur | heh | 10:47 |
| * etingof can't help but wonder, for how long ironic is being used in the wild? | 10:48 | |
| etingof | initial git commit dates back to 2010 | 10:48 |
| dtantsur | etingof: from Kilo in Red Hat, from Juno in Yahoo (now Oath) and Rackspace OnMetal (JayF may confirm) | 10:51 |
| dtantsur | basically, Juno was the first working release :) | 10:51 |
| etingof | 2014, give or take | 10:51 |
| dtantsur | yeah, late 2014 | 10:52 |
| *** e0ne has joined #openstack-ironic | 10:52 | |
| dtantsur | hmm, actually I seem to remember that Yahoo used nova-baremetal in Juno, not Ironic. maybe I'm wrong, let's ask rloo when she's here. | 10:52 |
| *** priteau has joined #openstack-ironic | 10:53 | |
| etingof | that's close enough for code archaeology purpose | 10:56 |
| * etingof could have done some more of the carbon dating in git commit log | 10:58 | |
| dnuka | folks, a merge conflict occured even before submtting the very first patch to gerrit. Is this possible :) | 11:03 |
| *** jistr|chat is now known as jistr | 11:05 | |
| TheJulia | dtantsur: I think there was a migration. I know they did lots of downstream patches for $things | 11:06 |
| dtantsur | yeah, definitely not pure Juno :) And I don't think it was pure Kilo in our case. | 11:07 |
| dtantsur | dnuka: it can happen if you're based on an old commit, and the head of master has diverged too much. | 11:07 |
| TheJulia | etingof: that initial commit is from initial repo states, since ironic started as the nova codebase which was massively re-written | 11:07 |
| rpittau | dnuka, yes that's possible :) | 11:07 |
| dnuka | dtantsur, rpittau thanks :) | 11:08 |
| TheJulia | I think rackspace was icehouse timeframe, but it would have been around around when they were trying to stay as close to master as possible | 11:09 |
| dtantsur | history :) | 11:09 |
| TheJulia | A long time ago in a universe far far away... the bear metal bears came forth and deployed servers. They said it was Ironic. | 11:10 |
| dtantsur | TheJulia: I cannot stop boasting: look, a CI run without rabbitmq http://logs.openstack.org/52/633052/24/check/ironic-tempest-functional-python3/dbdd424/controller/logs/ :) | 11:11 |
| dtantsur | lol | 11:11 |
| TheJulia | dtantsur: enjoying that I see? :) | 11:11 |
| dtantsur | a lot :) | 11:11 |
| dtantsur | keystone does not need rabbit as well, apparently? this means we can switch bifrost (unless neutron is enabled) | 11:12 |
| TheJulia | Yeah, keystone is a direct API service | 11:13 |
| TheJulia | That would actually be kind of cool | 11:13 |
| * TheJulia is excited now | 11:13 | |
| * dtantsur ponders a similar patch to ironic-inspector, now that we're planning API-worker split there | 11:13 | |
| TheJulia | looks like the linchpin for the events work is going to be https://review.openstack.org/#/c/345934/8 | 11:21 |
| patchbot | patch 345934 - python-ironicclient - Add Events support - 8 patch sets | 11:21 |
| TheJulia | I guess that needs a reno | 11:21 |
| *** e0ne has quit IRC | 11:24 | |
| openstackgerrit | Mark Goddard proposed openstack/tenks master: Allow setting node and volume name prefixes per-spec https://review.openstack.org/633900 | 11:25 |
| *** e0ne has joined #openstack-ironic | 11:34 | |
| *** rpittau has quit IRC | 11:34 | |
| *** rpittau has joined #openstack-ironic | 11:34 | |
| *** sdake has quit IRC | 11:41 | |
| *** sdake has joined #openstack-ironic | 11:48 | |
| openstackgerrit | Hamdy Khader proposed openstack/ironic master: Add support for Smart NICs https://review.openstack.org/583573 | 11:50 |
| openstackgerrit | Hamdy Khader proposed openstack/ironic master: Expose is_smartnic in port API https://review.openstack.org/630508 | 11:50 |
| TheJulia | dtantsur: ugh http://logs.openstack.org/23/628223/10/gate/openstack-tox-py27/922cd7b/job-output.txt.gz#_2019-01-31_08_32_50_955893 <-- re allocations | 11:51 |
| dtantsur | eeeehhhmmm | 11:51 |
| dtantsur | we don't modify migrations in this patch, do we? | 11:52 |
| TheJulia | I don't think so | 11:52 |
| TheJulia | dtantsur: nope | 11:53 |
| TheJulia | dtantsur: the rpc schema ver is bumped, but nothing db side. I wonder if it was just a fluke race... or if py27 will start to fail more often | 11:53 |
| dtantsur | TheJulia: wdyt about https://github.com/metalkube/dev-scripts/pull/17#discussion_r252634671 | 11:58 |
| TheJulia | dtantsur: the two patches in the gate right now have passed py27 without issues | 11:58 |
| * dtantsur suspects his bad luck.. | 11:59 | |
| dtantsur | I can try rebasing and seeing if it helps anything | 11:59 |
| openstackgerrit | Hamdy Khader proposed openstack/python-ironicclient master: Add is-smartnic port attribute to port command https://review.openstack.org/629449 | 11:59 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allocation API: conductor API (without HA and take over) https://review.openstack.org/628223 | 12:01 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Allocation API: REST API implementation https://review.openstack.org/629260 | 12:01 |
| TheJulia | dtantsur: sounds like an ironic issue. That being said, the enabled order should be ipxe,pxe based on preference loading. | 12:01 |
| TheJulia | I've seen something super similar with iscsi deploy and fedora filesystem permissions | 12:02 |
| TheJulia | but I can't place it right now in my memory | 12:02 |
| *** amoralej is now known as amoralej|lunch | 12:02 | |
| * TheJulia goes and finds foods | 12:02 | |
| dtantsur | derekh: could you check if the nodes really end up with boot_interface=ipxe (not pxe)? | 12:02 |
| TheJulia | derekh: also, what is the deploy_interface ? | 12:03 |
| dtantsur | I think he does not change it, so iscsi? | 12:04 |
| openstackgerrit | Hamdy Khader proposed openstack/python-ironicclient master: Add 'hostname' to port's local link connection https://review.openstack.org/628773 | 12:05 |
| derekh | dtantsur: TheJulia node show has boot_interface=ipxe and deploy_interface=iscsi | 12:16 |
| openstackgerrit | Merged openstack/ironic master: Make versioned notifications topics configurable https://review.openstack.org/629818 | 12:16 |
| TheJulia | derekh: try direct please :) | 12:17 |
| derekh | TheJulia: will do | 12:18 |
| TheJulia | Thanks! | 12:18 |
| dtantsur | dunno if it will help. I rather suspect that we forgot to pass ipxe_enabled=True somewhere from the ipxe interface. | 12:18 |
| dtantsur | derekh: could you file an issue on the storyboard about your discovery? | 12:19 |
| derekh | dtantsur: yup, no prob | 12:19 |
| TheJulia | dtantsur: possible, but iscsi image download is slightly behavior in terms of image cache interaction | 12:20 |
| TheJulia | At least, that is what my memory thinks | 12:21 |
| dtantsur | TheJulia: you're right, but the failure happens on the ramdisk download, and it should be (?) the same for both deploy interfaces. | 12:21 |
| dtantsur | aaaand, another transient failure :( but the next patch passed all UT jobs | 12:21 |
| TheJulia | Ugh | 12:21 |
| openstackgerrit | Iury Gregory Melo Ferreira proposed openstack/python-ironicclient master: Move to zuulv3 https://review.openstack.org/633010 | 12:22 |
| TheJulia | Yeah, should be the same path... | 12:22 |
| derekh | dtantsur: TheJulia http://paste.openstack.org/show/744301/ | 12:22 |
| TheJulia | Mmm food | 12:22 |
| dtantsur | TheJulia: enjoy! | 12:24 |
| dtantsur | thanks derekh. yeah, something is wrong here, looking at the code now. | 12:26 |
| *** rh-jelabarre has joined #openstack-ironic | 12:26 | |
| dtantsur | aha, found it! | 12:27 |
| openstackgerrit | Dhanuka Warusadura proposed openstack/python-ironicclient master: Add node owner https://review.openstack.org/634215 | 12:28 |
| dnuka | no more merge conflicts :) | 12:29 |
| TheJulia | \o/ | 12:29 |
| TheJulia | *2 | 12:30 |
| dnuka | :) | 12:30 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: [WIP] Fix iPXE boot interface with ipxe_enabled=True https://review.openstack.org/634216 | 12:30 |
| dtantsur | derekh: do you think you could try ^^^? | 12:30 |
| dtantsur | TheJulia: ^^ | 12:30 |
| derekh | dtantsur: will do https://storyboard.openstack.org/#!/story/2004905 | 12:32 |
| TheJulia | I guess that would have different logic that far down for ipxe... | 12:34 |
| TheJulia | :( | 12:34 |
| dtantsur | I think the root of the issue is that we still use PXE paths for downloading images in this case | 12:36 |
| *** bfournie has quit IRC | 12:45 | |
| openstackgerrit | M V P Nitesh proposed openstack/ironic master: Provides mount point as cinder requires it to attach volume https://review.openstack.org/633459 | 12:58 |
| *** e0ne has quit IRC | 13:01 | |
| derekh | dtantsur: the patch appears to work, in that my node now deploys, but then after it powers back on I get this http://paste.openstack.org/show/744305/ and it powers off again, | 13:02 |
| derekh | dtantsur: its getting a lot further so this problem may be unrelated | 13:03 |
| derekh | bbiab | 13:03 |
| *** e0ne has joined #openstack-ironic | 13:04 | |
| openstackgerrit | Merged openstack/ironic master: Removing deprecated drac_host property https://review.openstack.org/633902 | 13:15 |
| *** gkadam-bmgr is now known as gkadam | 13:15 | |
| dtantsur | derekh: it's conductor crashing, we need to see why. | 13:16 |
| dtantsur | hmmmmm, wait | 13:16 |
| dtantsur | we call from the conductor to the same conductor.. and it tells us there's no conductors?? WUT? | 13:17 |
| dtantsur | can it be some race with creating a database? | 13:18 |
| *** gkadam has quit IRC | 13:20 | |
| *** amoralej|lunch is now known as amoralej | 13:23 | |
| *** e0ne has quit IRC | 13:25 | |
| *** jistr is now known as jistr|call | 13:25 | |
| *** e0ne has joined #openstack-ironic | 13:29 | |
| dnuka | good night folks o/ | 13:30 |
| *** dnuka has quit IRC | 13:30 | |
| *** e0ne has quit IRC | 13:30 | |
| *** jistr|call is now known as jistr | 13:31 | |
| *** rh-jelabarre has quit IRC | 13:31 | |
| *** e0ne has joined #openstack-ironic | 13:32 | |
| *** e0ne has quit IRC | 13:34 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic master: Fix iPXE boot interface with ipxe_enabled=False https://review.openstack.org/634216 | 13:34 |
| dtantsur | the complete patch ^^\ | 13:35 |
| *** rpittau has quit IRC | 13:35 | |
| *** e0ne has joined #openstack-ironic | 13:35 | |
| *** jcoufal has quit IRC | 13:37 | |
| arne_wiebalck | Root device hints FTW! Used them twice this week to boot machines which did not manage to boot after deployment: one with h/w RAID controller trying to expose the individual drives, the other with an attached disk array. \o/ | 13:45 |
| *** rpittau has joined #openstack-ironic | 13:46 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent master: Build tinyipa with python3 https://review.openstack.org/634232 | 13:47 |
| rpittau | wifi gave up exactly when I pressed enter after git review :/ | 13:47 |
| *** pcaruana has quit IRC | 13:52 | |
| *** pcaruana has joined #openstack-ironic | 14:02 | |
| *** rloo has joined #openstack-ironic | 14:07 | |
| *** sdake has quit IRC | 14:08 | |
| *** e0ne has quit IRC | 14:13 | |
| *** diga has joined #openstack-ironic | 14:16 | |
| openstackgerrit | Hamdy Khader proposed openstack/ironic master: Add support for Smart NICs https://review.openstack.org/583573 | 14:19 |
| openstackgerrit | Hamdy Khader proposed openstack/ironic master: Expose is_smartnic in port API https://review.openstack.org/630508 | 14:19 |
| *** baha has joined #openstack-ironic | 14:26 | |
| *** zul has joined #openstack-ironic | 14:31 | |
| *** mjturek has joined #openstack-ironic | 14:33 | |
| mjturek | anyone familiar with the baremetal basic ops test, after the node gets verified by SSH, right before cleaning starts, I see another neutron port come into existence.. Anyone know what that port might be for? It seems to be a virtual mac address | 14:35 |
| *** sdake has joined #openstack-ironic | 14:39 | |
| *** e0ne has joined #openstack-ironic | 14:45 | |
| mjturek | hmmm maybe should ask on the ML | 14:45 |
| *** jcoufal has joined #openstack-ironic | 14:53 | |
| *** rh-jelabarre has joined #openstack-ironic | 14:54 | |
| openstackgerrit | Mark Goddard proposed openstack/tenks master: Configure VM interfaces in macvtap passthrough mode https://review.openstack.org/634249 | 15:01 |
| openstackgerrit | Mark Goddard proposed openstack/tenks master: Use node state rather than 'cmd' https://review.openstack.org/634250 | 15:01 |
| openstackgerrit | Mark Goddard proposed openstack/tenks master: Default Ironic node properties are bogus https://review.openstack.org/634251 | 15:01 |
| *** hamdyk has quit IRC | 15:05 | |
| *** TxGirlGeek has joined #openstack-ironic | 15:18 | |
| *** priteau has quit IRC | 15:19 | |
| mjturek | \ | 15:22 |
| *** sdake has quit IRC | 15:24 | |
| openstackgerrit | Riccardo Pittau proposed openstack/ironic-python-agent master: Build tinyipa with python3 https://review.openstack.org/634232 | 15:28 |
| *** baha has quit IRC | 15:30 | |
| *** mjturek has quit IRC | 15:33 | |
| *** trown|outtypewww is now known as trown | 15:38 | |
| *** mjturek has joined #openstack-ironic | 15:41 | |
| *** priteau has joined #openstack-ironic | 15:43 | |
| *** robbbe has joined #openstack-ironic | 15:52 | |
| *** baha has joined #openstack-ironic | 15:58 | |
| *** _fragatina has quit IRC | 16:03 | |
| *** _fragatina has joined #openstack-ironic | 16:04 | |
| *** _fragatina has quit IRC | 16:04 | |
| *** _fragatina has joined #openstack-ironic | 16:05 | |
| *** arne_wiebalck_ has joined #openstack-ironic | 16:07 | |
| *** arne_wiebalck_ has quit IRC | 16:07 | |
| *** _fragatina has quit IRC | 16:10 | |
| *** robbbe has quit IRC | 16:19 | |
| *** sdake has joined #openstack-ironic | 16:20 | |
| *** robbbe has joined #openstack-ironic | 16:35 | |
| *** TxGirlGeek has quit IRC | 16:35 | |
| *** sthussey has joined #openstack-ironic | 16:39 | |
| *** etingof has left #openstack-ironic | 16:45 | |
| *** e0ne has quit IRC | 16:46 | |
| *** etingof has joined #openstack-ironic | 16:46 | |
| etingof | do we have a timestamp field on the node noting last power state reading? I do not see such thing so far | 16:47 |
| dtantsur | I don't think we do (I'd like to see it btw) | 16:48 |
| * arne_wiebalck thinks this was discussed recently and the answer was no | 16:48 | |
| *** priteau has quit IRC | 16:50 | |
| etingof | I am thinking to add such field and use it right away for prioritizing power status calls for the upcoming status (TheJulia's suggestion) | 16:58 |
| etingof | I am not sure if that would hurt performance... locking...? | 16:59 |
| *** mjturek has quit IRC | 17:00 | |
| *** mjturek has joined #openstack-ironic | 17:00 | |
| dtantsur | mmm, right, it will require going back to exclusive locks by default | 17:02 |
| dtantsur | which is probably why we don't do it | 17:02 |
| *** TxGirlGeek has joined #openstack-ironic | 17:03 | |
| *** iurygregory has quit IRC | 17:06 | |
| *** _fragatina has joined #openstack-ironic | 17:08 | |
| etingof | aha, but we do have in-memory dict holding retry counters - we could probably just sort nodes by its contents | 17:10 |
| etingof | for power sync ordering | 17:10 |
| *** sburette has quit IRC | 17:12 | |
| NobodyCam | Good Morning Ironic'ers | 17:12 |
| TheJulia | Eh, but an exclusive lock is a good thing I guess for BMC power polling... But that is the only way to force last_updated to change between checks. | 17:12 |
| * TheJulia looks out the window of the hotel lounge... Sees only darkness and feels confused by NobodyCam :) | 17:13 | |
| *** mjturek has quit IRC | 17:14 | |
| TheJulia | etingof: in memory really only seems the same path unless we have good reason to lock | 17:14 |
| NobodyCam | hehehehe | 17:14 |
| NobodyCam | Gota luv UGT | 17:14 |
| TheJulia | Which would update the DB record..... | 17:14 |
| TheJulia | Indeed. | 17:14 |
| *** mjturek has joined #openstack-ironic | 17:15 | |
| *** tssurya has quit IRC | 17:17 | |
| * etingof is not sure how to interpret TheJulia's thoughts... let's blame NobodyCam's early morning for that o/ | 17:18 | |
| *** dustinc has joined #openstack-ironic | 17:18 | |
| TheJulia | etingof: sorry, numberous threads trying to run that suggest I should go explore, exercise, or sleep.. likely all three. | 17:19 |
| *** gyee has joined #openstack-ironic | 17:19 | |
| rpittau | hi NobodyCam :) | 17:20 |
| NobodyCam | Good Morning rpittau :) | 17:22 |
| dtantsur | morning NobodyCam | 17:25 |
| dtantsur | etingof: we use shared lock for checking the power state, only acquire the exclusive lock if we need to update it in the database (or force it on the node) | 17:26 |
| etingof | dtantsur, right, so if we need to update the timestamp on the node in the DB we'd have to upgrade the lock frequently... | 17:28 |
| *** sdake has quit IRC | 17:28 | |
| *** dsneddon has quit IRC | 17:29 | |
| etingof | dtantsur, is the in-memory power failure states suitable for ordering the nodes? would not it be lighter to use that then? | 17:29 |
| dtantsur | etingof: I'm not sure I get the last question. What ordering do you mean? | 17:30 |
| NobodyCam | Hey hey dtantsur Good Morning | 17:30 |
| etingof | dtantsur, so the idea is to ponder previously failed nodes (on power status) to fail them faster and get them out of the way | 17:31 |
| TheJulia | etingof: I think so, but only for over 300 or so nodes. If less than that then I say skip sorting | 17:31 |
| manjeets | TheJulia, I had a bit old neutron (along with mech driver for a hardware switch) its like month old I believe, but I'll rebase it to master today | 17:31 |
| manjeets | thanks for pointers | 17:32 |
| etingof | dtantsur, the root of the problem being nodecount may be too high for the greenlets we herd | 17:32 |
| *** robbbe has quit IRC | 17:33 | |
| TheJulia | manjeets: vnic_type has been a thing for a few years.... But I might not understand the entire isaue | 17:33 |
| TheJulia | Iasue | 17:33 |
| etingof | thus, attempt to optimize the ordering | 17:33 |
| * etingof is seeking the enzyme to destroy the rna of the devconf virus | 17:35 | |
| TheJulia | Heh | 17:36 |
| TheJulia | A few folks in my class got whatever it was... | 17:36 |
| manjeets | TheJulia, I have enabled mech driver in neutron for switch we are testing in my prior setup I enabled ovs,switch_mech and when i was boot nova instance with speical port it never always failed on binding provisioning port | 17:37 |
| manjeets | so it never reached the point where it would call bind_port from switch_mech | 17:37 |
| manjeets | then i tried new configuration which didn't have provisioning network and I was able to see bind_port being executed from switch mech but was throwing error I mentioned above | 17:38 |
| dtantsur | TheJulia: do we need presence at the ops meetup? https://etherpad.openstack.org/p/BER-ops-meetup | 17:38 |
| manjeets | then tried just with ovs and error was still there on server boot | 17:38 |
| * manjeets rebuilds devstack with ironic will post logs in a bit | 17:41 | |
| *** dsneddon has joined #openstack-ironic | 17:45 | |
| rpittau | bye all! good evening! o/ | 17:48 |
| *** rpittau has quit IRC | 17:48 | |
| TheJulia | manjeets: okay, still sounds weird, but logs will help | 17:50 |
| TheJulia | dtantsur: it would be good for there to be someone from ironic there to answer questions, listen to feedback | 17:52 |
| dtantsur | TheJulia: I can do it for obvious reasons :) | 17:52 |
| TheJulia | \o/ | 17:53 |
| *** dougsz has quit IRC | 17:53 | |
| *** derekh has quit IRC | 18:00 | |
| *** sdake has joined #openstack-ironic | 18:07 | |
| manjeets | http://paste.openstack.org/show/744330/ | 18:11 |
| manjeets | TheJulia, I see this a lot in logs ^^ with ovs and other mech | 18:11 |
| manjeets | note: I'm still using a month old neutron | 18:12 |
| manjeets | but ironic and everything else is master | 18:12 |
| *** bfournie has joined #openstack-ironic | 18:26 | |
| *** trown is now known as trown|lunch | 18:29 | |
| TheJulia | manjeets: I would suspect your ml2 driver is just not loading | 18:29 |
| TheJulia | Anything in your neutron logs verifying that all your plugins are loading? | 18:30 |
| manjeets | TheJulia, neutron logs are fine, I see bind_port getting executed from ml2 driver I have for hardware switch | 18:36 |
| *** mjturek has quit IRC | 18:36 | |
| * TheJulia raises an eyebrow | 18:36 | |
| manjeets | this happens with or without the hardware ml2 driver, I noticed this with ovs as well | 18:37 |
| TheJulia | Well, OVS would never know what to do with a vnic_type of baremetal | 18:37 |
| manjeets | wait TheJulia I'm building devstack with ovs and new neutron will verify logs again | 18:37 |
| TheJulia | OVS is just going to go "I don't know what to do with this" and return that error | 18:38 |
| *** mjturek has joined #openstack-ironic | 18:38 | |
| *** amoralej is now known as amoralej|off | 18:39 | |
| *** dtantsur is now known as dtantsur|afk | 18:39 | |
| *** e0ne has joined #openstack-ironic | 18:39 | |
| dtantsur|afk | on PTO starting tomorrow, see some of you on FOSDEM, the others - on Wednesday | 18:39 |
| *** moshele has joined #openstack-ironic | 18:44 | |
| *** mjturek has quit IRC | 18:44 | |
| TheJulia | dtantsur|afk: enjoy... Oh new sqlalchemy from the 25th.... | 18:47 |
| manjeets | ah ohk TheJulia, since there's ovs agent but no agent for this mech driver could that be an issue ? | 18:48 |
| *** e0ne has quit IRC | 18:48 | |
| TheJulia | manjeets: no, the ml2 drivers, as I understand it, register when they load into neutron-api... What they support | 18:50 |
| manjeets | exactly that's what my understanding is | 18:51 |
| *** TxGirlGeek has quit IRC | 18:51 | |
| TheJulia | So to get that error, I wonder what driver your loading. | 18:52 |
| manjeets | I am loading openvswitch,switch_mech | 18:52 |
| manjeets | https://github.com/openstack/neutron/blob/master/neutron/plugins/ml2/drivers/mech_agent.py#L91 | 18:52 |
| TheJulia | Oh | 18:53 |
| *** TxGirlGeek has joined #openstack-ironic | 18:53 | |
| TheJulia | switch_mech is? | 18:53 |
| manjeets | the driver I'm writing for hardware switch | 18:53 |
| manjeets | I have added baremetal in supported vnics | 18:54 |
| TheJulia | Oh! | 18:55 |
| TheJulia | Okay | 18:56 |
| *** mjturek has joined #openstack-ironic | 18:56 | |
| *** moshele has quit IRC | 19:08 | |
| *** mjturek has quit IRC | 19:09 | |
| *** mjturek has joined #openstack-ironic | 19:10 | |
| *** moshele has joined #openstack-ironic | 19:16 | |
| TheJulia | manjeets: take a look at what netwoeking-generic-switch does in https://github.com/openstack/networking-generic-switch/blob/master/networking_generic_switch/generic_switch_mech.py in terms of supporting the baremetal vnic type | 19:17 |
| *** mgoddard has quit IRC | 19:18 | |
| *** mgoddard has joined #openstack-ironic | 19:19 | |
| manjeets | TheJulia, so missing is_port_supported could be the reason ? | 19:19 |
| manjeets | I think that'll fix the issue ? | 19:20 |
| manjeets | thanks TheJulia | 19:20 |
| TheJulia | I suspect so :) | 19:20 |
| *** moshele has quit IRC | 19:20 | |
| openstackgerrit | Merged openstack/ironic master: Allocation API: conductor API (without HA and take over) https://review.openstack.org/628223 | 19:28 |
| *** pcaruana has quit IRC | 19:30 | |
| *** trown|lunch is now known as trown | 19:43 | |
| *** dsneddon has quit IRC | 19:46 | |
| *** TxGirlGeek has quit IRC | 19:48 | |
| *** jcoufal has quit IRC | 19:50 | |
| *** rnoriega has quit IRC | 19:57 | |
| *** rnoriega has joined #openstack-ironic | 19:57 | |
| *** dsneddon has joined #openstack-ironic | 20:02 | |
| *** arne_wiebalck_ has joined #openstack-ironic | 20:02 | |
| *** arne_wiebalck_ has quit IRC | 20:03 | |
| *** mjturek has quit IRC | 20:03 | |
| *** arne_wiebalck_ has joined #openstack-ironic | 20:05 | |
| *** mjturek has joined #openstack-ironic | 20:05 | |
| *** TxGirlGeek has joined #openstack-ironic | 20:05 | |
| *** mjturek has quit IRC | 20:09 | |
| *** mjturek has joined #openstack-ironic | 20:10 | |
| *** mjturek has quit IRC | 20:34 | |
| *** TxGirlGeek has quit IRC | 21:00 | |
| *** jcoufal has joined #openstack-ironic | 21:02 | |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic master: Fix test for 'force_persistent_boot_device' (i)PXE driver_info option https://review.openstack.org/634315 | 21:12 |
| openstackgerrit | Arne Wiebalck proposed openstack/ironic master: Fix test for 'force_persistent_boot_device' (i)PXE driver_info option https://review.openstack.org/634315 | 21:16 |
| *** arne_wiebalck_ has quit IRC | 21:17 | |
| *** TxGirlGeek has joined #openstack-ironic | 21:25 | |
| *** rh-jelabarre has quit IRC | 21:33 | |
| *** bfournie has quit IRC | 21:58 | |
| *** trown is now known as trown|outtypewww | 22:02 | |
| JayF | etingof: dtantsur|afk: we started deploying Ironic, with our to-be-upstreamed agent driver, before Icehouse, even... I don't remember what the "H" release was, but probably that one or before | 22:03 |
| *** baha has quit IRC | 22:08 | |
| *** jcoufal has quit IRC | 22:19 | |
| *** sdake has quit IRC | 22:30 | |
| *** S4ren has quit IRC | 22:34 | |
| tonyb | Is it normal for the ironic-neutron-agent-heartbeat.info message queue to have lots of un-ACKed messages? | 22:38 |
| tonyb | [root@director ~]# docker exec -it --user root rabbitmq rabbitmqctl list_queues | grep heartbeat | 22:38 |
| tonyb | ironic-neutron-agent-heartbeat.info 137237 | 22:38 |
| tonyb | that number is slowly growing | 22:39 |
| tonyb | That's rocky so it may be differnt on master I'll try to spin up master next week sometime if needed | 22:40 |
| eandersson | Anything in the baremetal agent logs? | 22:45 |
| *** sdake has joined #openstack-ironic | 22:45 | |
| tonyb | eandersson: What kind of anything? | 22:48 |
| eandersson | Like errors or similar | 22:48 |
| tonyb | eandersson: I'm pretty sure I saw it logging the heartbeat | 22:49 |
| tonyb | eandersson: not that I noticed | 22:49 |
| tonyb | eandersson: I can remove the node and introspect it again and save the logs | 22:49 |
| *** MattMan has quit IRC | 22:51 | |
| *** MattMan has joined #openstack-ironic | 22:52 | |
| *** bfournie has joined #openstack-ironic | 23:01 | |
| *** dustinc has quit IRC | 23:18 | |
| tonyb | eandersson: Nothing that looks like an error | 23:37 |
| eandersson | tonyb, are they all un'acked? or just not consumed? | 23:38 |
| * tonyb admits he can't even find the producer or comsumer for that queue | 23:38 | |
| tonyb | eandersson: Not sure | 23:38 |
| tonyb | eandersson: just that they keep growing which seemed strange | 23:38 |
| eandersson | Is there a consumer attached to the queue? | 23:39 |
| eandersson | I haven't actually used networking-baremetal yet myself | 23:39 |
| eandersson | but what happens if you restart the agent? | 23:39 |
| eandersson | because it's supposed to be listening to that queue | 23:40 |
| eandersson | afaik | 23:40 |
| tonyb | Yeah there aren't any consumers | 23:42 |
| tonyb | [root@director ~]# docker exec -it --user root rabbitmq rabbitmqctl list_queues name consumers | grep heartbeat | 23:42 |
| tonyb | ironic-neutron-agent-heartbeat.info 0 | 23:42 |
| *** diga has quit IRC | 23:43 | |
| eandersson | http://git.openstack.org/cgit/openstack/networking-baremetal/tree/networking_baremetal/agent/ironic_neutron_agent.py#n59 | 23:48 |
| eandersson | So this looks like it should be listening. | 23:48 |
| eandersson | It even looks like it is supposed to both listen and notify | 23:49 |
| tonyb | Okay, I suspect that's my problem I can't find that anywhere | 23:50 |
| * tonyb goes to look at the installer to workout where that's supposed to be/come from | 23:51 | |
| tonyb | eandersson: Thanks | 23:51 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!