| *** pmannidi has joined #openstack-kuryr | 00:13 | |
| *** limao has joined #openstack-kuryr | 00:51 | |
| *** yedongcan has joined #openstack-kuryr | 00:53 | |
| *** ajo has quit IRC | 00:54 | |
| *** leifmadsen has quit IRC | 00:55 | |
| *** pmannidi has quit IRC | 00:57 | |
| *** leifmadsen has joined #openstack-kuryr | 00:57 | |
| *** ajo has joined #openstack-kuryr | 01:00 | |
| *** dimak_ has quit IRC | 01:07 | |
| *** pmannidi has joined #openstack-kuryr | 01:13 | |
| *** pmannidi has quit IRC | 01:21 | |
| *** dimak_ has joined #openstack-kuryr | 01:21 | |
| *** aojea has joined #openstack-kuryr | 01:30 | |
| *** limao has quit IRC | 01:36 | |
| *** aojea has quit IRC | 01:36 | |
| *** limao has joined #openstack-kuryr | 01:36 | |
| *** pmannidi has joined #openstack-kuryr | 01:37 | |
| *** dimak_ has quit IRC | 01:41 | |
| *** hongbin has joined #openstack-kuryr | 02:16 | |
| *** jayahn has joined #openstack-kuryr | 02:26 | |
| *** hongbin has quit IRC | 03:20 | |
| *** aojea has joined #openstack-kuryr | 03:32 | |
| *** aojea has quit IRC | 03:37 | |
| *** limao has quit IRC | 04:11 | |
| *** limao has joined #openstack-kuryr | 04:11 | |
| *** limao has quit IRC | 04:16 | |
| *** salv-orlando has joined #openstack-kuryr | 04:23 | |
| *** salv-orlando has quit IRC | 04:28 | |
| *** aojea has joined #openstack-kuryr | 04:33 | |
| *** aojea has quit IRC | 04:38 | |
| *** limao has joined #openstack-kuryr | 04:47 | |
| *** salv-orlando has joined #openstack-kuryr | 04:56 | |
| *** janki has joined #openstack-kuryr | 05:22 | |
| *** ltomasbo|away is now known as ltomasbo | 05:26 | |
| *** aojea has joined #openstack-kuryr | 05:32 | |
| *** dimak_ has joined #openstack-kuryr | 05:35 | |
| *** reedip_afk is now known as reedip | 05:40 | |
| *** kzaitsev_ws has joined #openstack-kuryr | 06:04 | |
| *** salv-orlando has quit IRC | 06:04 | |
| *** aojea has quit IRC | 06:07 | |
| *** aojea has joined #openstack-kuryr | 06:07 | |
| *** aojea has quit IRC | 06:12 | |
| irenab | apuimedo, ivc : ping | 06:19 |
|---|---|---|
| *** reedip has quit IRC | 06:23 | |
| *** reedip has joined #openstack-kuryr | 06:24 | |
| kzaitsev_ws | have anyone here tried node-feature-discovery project? (just curious) | 06:29 |
| *** janki has quit IRC | 06:30 | |
| *** pcaruana has joined #openstack-kuryr | 06:33 | |
| *** janki has joined #openstack-kuryr | 06:49 | |
| *** jayahn has quit IRC | 07:00 | |
| *** limao has quit IRC | 07:10 | |
| *** limao has joined #openstack-kuryr | 07:12 | |
| yedongcan | apuimedo, irenab, ivc: ping | 07:15 |
| irenab | yedongcan, pong | 07:15 |
| yedongcan | irenab: hi, does kuryr-k8s support multi-tenancy now? | 07:16 |
| *** salv-orlando has joined #openstack-kuryr | 07:17 | |
| yedongcan | Or any mapping between k8s namespace into OpenStack Keystone tenant? | 07:18 |
| *** aojea has joined #openstack-kuryr | 07:24 | |
| irenab | yedongcan, not yet | 07:28 |
| irenab | yedongcan, now that basic mode (native k8s like) is supported, we can add other drivers, such as one for multu tenancy | 07:33 |
| *** egonzalez has joined #openstack-kuryr | 07:38 | |
| irenab | kzaitsev_ws, I didn't try it, but saw cool demo at CNCF conference | 07:41 |
| apuimedo | irenab: kzaitsev_ws: what's that? | 07:52 |
| kzaitsev_ws | irenab: i'm trying to make it work on 1.6.1 cluster, but seems that new RBAC auth broke things | 07:52 |
| kzaitsev_ws | node-feature-discovery runs a job on each of k8s nodes and attempts to detect some features of hardware available on that node | 07:53 |
| kzaitsev_ws | it then reports the result back to k8s by adding labels to the node object | 07:54 |
| kzaitsev_ws | that was the reason I hacked kubernetes-api service into lbaas handled by kuryr-k8s earlier | 07:55 |
| yedongcan | irenab: got it , thanks | 07:57 |
| irenab | apuimedo, it is useful for scheduling pods with special requirements, like SRIOV | 07:58 |
| kzaitsev_ws | irenab: yeah =) that's my strategy ) | 07:59 |
| apuimedo | right | 07:59 |
| kzaitsev_ws | only I want to modify it to use Opaque Integer Resources instead of Labels. | 07:59 |
| irenab | I think there is a SIG for node discovery, checking .. | 08:00 |
| kzaitsev_ws | this way Pods would be able to request 2+ VFs and k8s would handle scheduling correctly | 08:00 |
| apuimedo | that's cool! | 08:01 |
| apuimedo | but how often do they do the feature discovery? | 08:01 |
| apuimedo | (for VF availability management) | 08:01 |
| kzaitsev_ws | for NFD (node feature disc) it is a 1-time Job (k8s Job I mean). | 08:02 |
| apuimedo | ah | 08:02 |
| apuimedo | that complicates things | 08:02 |
| *** yedongcan has quit IRC | 08:02 | |
| kzaitsev_ws | apuimedo: you mean bookkeeping the amount of free resources? | 08:03 |
| kzaitsev_ws | that's where Opaque Integer Resources come it. it's alpha in 1.5 k8s and only accounting has been implemented. but for VFs that's all I need | 08:04 |
| apuimedo | right | 08:04 |
| kzaitsev_ws | https://kubernetes.io/docs/concepts/configuration/manage-compute-resources-container/#opaque-integer-resources-alpha-feature | 08:04 |
| irenab | apuimedo, kzaitsev_ws : in additions thare should be scheduling piece to do the dynamic resource counting | 08:04 |
| kzaitsev_ws | you can tell a node that it has 10 bananas. and then a Pod would request 4 bananas. | 08:05 |
| kzaitsev_ws | if there aren't free nodes with enough bananas — the Pod would stay in Pending state | 08:05 |
| irenab | kzaitsev_ws, I think its the part of the node discovery, tht it checks how many resource are there | 08:05 |
| kzaitsev_ws | irenab: well node discovery just sets labels actually | 08:06 |
| irenab | kzaitsev_ws, I mean in addtion to the capability it should have the quantity as well | 08:06 |
| irenab | sig-node is the group I meant | 08:07 |
| apuimedo | there is a sig for everything | 08:07 |
| irenab | sounds like openstack :-) | 08:08 |
| kzaitsev_ws | what's sig? =) | 08:09 |
| apuimedo | special interest group | 08:09 |
| irenab | kzaitsev_ws, mailing list and weekly meetings to discuss related topics | 08:10 |
| irenab | https://groups.google.com/forum/#!forum/kubernetes-sig-node | 08:11 |
| dmellado | yeah, it's like openstack projects | 08:14 |
| dmellado | when I was at the PTG I was shocked by the amount of projects I had no idea about (and pets xD) | 08:15 |
| kzaitsev_ws | got to join that one =) | 08:16 |
| kzaitsev_ws | and probably a bunch of other SIGs ) | 08:17 |
| irenab | dmellado, I am still recovering from the exposure to the enormous number of projects at CNCF :-) | 08:20 |
| *** garyloug has joined #openstack-kuryr | 08:24 | |
| *** yedongcan has joined #openstack-kuryr | 08:30 | |
| *** janki has quit IRC | 08:41 | |
| *** janki has joined #openstack-kuryr | 08:49 | |
| irenab | apuimedo, ivc : I have some last minute scheduled meeting that overlaps with the actors proposal discussion | 08:51 |
| irenab | I am not able to reschedule it. Any chance we can move actors discussion for tomorrow? | 08:52 |
| *** janki has quit IRC | 08:56 | |
| *** janki has joined #openstack-kuryr | 08:57 | |
| dmellado | irenab: no problem from my side | 08:57 |
| dmellado | how about you, folks? | 08:58 |
| kzaitsev_ws | should be no problem for me too ) | 09:02 |
| kzaitsev_ws | thanks for inviting me btw ) | 09:02 |
| kzaitsev_ws | and yeah. I've made node-feature discovery work on 1.6 Had to give some permissions to system:serviceaccount:default:default user though | 09:03 |
| kzaitsev_ws | is there a way to run a pod/job with specific user passed to it? | 09:05 |
| janonymous | No problem with me too | 09:06 |
| mchiappero | hey folks, I would appreciate a feedback on https://review.openstack.org/#/c/440669/ | 09:18 |
| mchiappero | it's been sitting there for some weeks now | 09:18 |
| apuimedo | irenab: fine for me. Up to ivc | 09:20 |
| apuimedo | kzaitsev_ws: no idea | 09:20 |
| kzaitsev_ws | seems that I can do it with spec.serviceAccountName =) | 09:25 |
| kzaitsev_ws | mchiappero: I've looked at it several times and each time was afraid to start reviewing )) | 09:29 |
| dmellado | mchiappero: I'll take a look, even if it scares me xD | 09:32 |
| apuimedo | mchiappero: sorry. It was holidays | 09:34 |
| mchiappero | kzaitsev_ws: LOL | 09:34 |
| mchiappero | apuimedo: I know :) I'm just trying to see if someone is willing to have a look again before the next kuryr meeting :) | 09:35 |
| apuimedo | I will :-) | 09:37 |
| mchiappero | thanks :) | 09:41 |
| ltomasbo | mchiappero, I'll take a look later today too | 09:42 |
| mchiappero | thanks everyone! | 09:42 |
| irenab | mchiappero, I will check too | 09:46 |
| dmellado | btw apuimedo and folks | 09:47 |
| dmellado | are you all ok with moving the meeting as irenab requested? | 09:47 |
| irenab | mchiappero, any advice on the setup env to verify? | 09:47 |
| dmellado | if so, let's change the meeting so I can be a happy dude and have an empty slot in my schedule too | 09:47 |
| dmellado | xD | 09:47 |
| irenab | dmellado, so easy to make you happy :-) | 09:48 |
| *** salv-orl_ has joined #openstack-kuryr | 09:48 | |
| dmellado | irenab: heh | 09:48 |
| dmellado | it's like apuimedo, if he gets to meet a certain german guy from midokura, he'll be happy too | 09:48 |
| mchiappero | irenab: we usually run devstack stacks simply using the local.conf already upstream | 09:48 |
| apuimedo | dmellado: no. I'll be happy when I get Zelda BotW DLC | 09:49 |
| dmellado | apuimedo: did you buy it? | 09:49 |
| apuimedo | and FCBarcelona gets a proper coach | 09:49 |
| mchiappero | it shouldn't require any specific change | 09:49 |
| apuimedo | dmellado: nope. I never buy before reviews | 09:49 |
| apuimedo | :-) | 09:49 |
| dmellado | I was hoping to get your review | 09:49 |
| dmellado | on zelda, not on mchiappero patch xD | 09:49 |
| mchiappero | :D | 09:50 |
| *** salv-orlando has quit IRC | 09:51 | |
| irenab | apuimedo, dmellado I can check about Zelda with my son, remember him mentioning something over obviously not wasted school break he had :-) | 09:54 |
| irenab | mchiappero, I guess you need stack with nova for the undercloud and the one with kuryr for the overcloud. What image do you use for the overcloud VM? | 09:56 |
| irenab | apuimedo, this brings me back to the fullstack tests, actually the lack of them ... | 09:57 |
| dmellado | irenab: +1 on that we need that | 09:58 |
| dmellado | maybe we can chat about that in tomorrow's meeting | 09:58 |
| irenab | dmellado, +1 | 09:59 |
| *** janki has quit IRC | 10:05 | |
| *** limao has quit IRC | 10:09 | |
| *** limao has joined #openstack-kuryr | 10:09 | |
| *** limao has quit IRC | 10:14 | |
| *** janki has joined #openstack-kuryr | 10:16 | |
| mchiappero | irenab: yes, you can follow the instructions for nested containers using VLAN | 10:23 |
| mchiappero | however you need to skip the steps regarding the trunking configuration | 10:23 |
| irenab | cool, posted some comments meanwhile | 10:23 |
| mchiappero | just stack the undercloud, stack the overcloud, and there you go | 10:24 |
| irenab | mchiappero, +1 | 10:24 |
| mchiappero | irenab: thanks ;0 | 10:24 |
| mchiappero | sorry, off-by-1 position | 10:24 |
| mchiappero | :) | 10:24 |
| *** janki has quit IRC | 10:37 | |
| *** janki has joined #openstack-kuryr | 10:37 | |
| *** janki has quit IRC | 10:41 | |
| *** jayahn has joined #openstack-kuryr | 10:44 | |
| *** janki has joined #openstack-kuryr | 10:50 | |
| *** salv-orl_ has quit IRC | 10:52 | |
| *** janki has quit IRC | 10:53 | |
| *** yedongcan1 has joined #openstack-kuryr | 10:53 | |
| *** yedongcan has quit IRC | 10:54 | |
| *** yedongcan1 has left #openstack-kuryr | 10:55 | |
| dmellado | heh | 11:15 |
| dmellado | that happens to me now and then and then it looks like klingon | 11:15 |
| *** janki has joined #openstack-kuryr | 11:17 | |
| *** salv-orlando has joined #openstack-kuryr | 11:42 | |
| *** salv-orlando has quit IRC | 11:48 | |
| apuimedo | ivc: ping | 11:49 |
| ivc | apuimedo pong | 11:49 |
| apuimedo | did you see the request above to move the meeting to tomorrow? | 11:49 |
| apuimedo | (same time) | 11:49 |
| ivc | yup. was checking logs and just noticed that | 11:50 |
| ivc | i'm busy tomorrow unfortunately | 11:50 |
| apuimedo | ok then. Let's keep today then | 11:51 |
| ivc | apuimedo also i'm warning you its less about actors and more about driver composition and handler loading :) | 11:52 |
| vikasc | ivc, apuimedo : do we have a doc or reference link that i can go through quickly before meeting? | 11:54 |
| vikasc | and also while discussing | 11:54 |
| ivc | vikasc not rly. this is sort of 'brainstorm' meeting | 11:54 |
| vikasc | ivc, cool | 11:55 |
| vikasc | thanks | 11:55 |
| ivc | well i'm hoping it to be a 'brainstorm' instead of 'monologue' :) | 11:55 |
| ivc | irenab should we maybe postpone the meeting until next week? or will you check the recording and post your comments to ml/irc? | 11:58 |
| apuimedo | kzaitsev_ws: finally the meeting is now | 12:03 |
| *** jayahn has quit IRC | 12:16 | |
| *** alraddarla_ has joined #openstack-kuryr | 13:51 | |
| *** janki has quit IRC | 14:06 | |
| irenab | apuimedo: please post the recording of the meeting. Sorry, I could not make it | 14:12 |
| apuimedo | I will. Gotta transcode it :( | 14:14 |
| *** garyloug has quit IRC | 14:16 | |
| *** dimak_ has quit IRC | 14:17 | |
| *** garyloug has joined #openstack-kuryr | 14:19 | |
| *** hongbin has joined #openstack-kuryr | 14:19 | |
| *** alraddarla_ has quit IRC | 14:35 | |
| *** alraddarla_ has joined #openstack-kuryr | 14:48 | |
| ivc | mchiappero i've added some comments for https://review.openstack.org/#/c/440669/ | 15:11 |
| *** pcaruana has quit IRC | 15:28 | |
| *** alraddarla_ has quit IRC | 15:37 | |
| *** aojea has quit IRC | 15:51 | |
| *** alraddarla_ has joined #openstack-kuryr | 16:06 | |
| *** ltomasbo is now known as ltomasbo|away | 16:23 | |
| *** salv-orlando has joined #openstack-kuryr | 16:54 | |
| *** egonzalez has quit IRC | 17:00 | |
| *** garyloug has quit IRC | 17:03 | |
| *** tonanhngo has joined #openstack-kuryr | 17:20 | |
| *** tonanhngo has quit IRC | 17:24 | |
| *** tonanhngo has joined #openstack-kuryr | 17:28 | |
| *** aojea has joined #openstack-kuryr | 17:38 | |
| *** aojea has quit IRC | 18:13 | |
| *** aojea has joined #openstack-kuryr | 18:14 | |
| *** aojea has quit IRC | 18:18 | |
| *** aojea has joined #openstack-kuryr | 18:34 | |
| *** aojea_ has joined #openstack-kuryr | 18:35 | |
| *** aojea has quit IRC | 18:38 | |
| *** aojea_ has quit IRC | 18:41 | |
| *** aojea_ has joined #openstack-kuryr | 18:44 | |
| *** aojea has joined #openstack-kuryr | 18:47 | |
| *** aojea_ has quit IRC | 18:48 | |
| *** aojea_ has joined #openstack-kuryr | 18:50 | |
| *** aojea has quit IRC | 18:53 | |
| *** lihi has quit IRC | 19:45 | |
| *** lihi has joined #openstack-kuryr | 19:45 | |
| *** tonanhngo has quit IRC | 19:46 | |
| *** tonanhngo has joined #openstack-kuryr | 19:50 | |
| *** tonanhngo has quit IRC | 19:51 | |
| *** tonanhngo has joined #openstack-kuryr | 19:52 | |
| *** tonanhngo has quit IRC | 19:59 | |
| *** tonanhngo has joined #openstack-kuryr | 20:05 | |
| *** aojea has joined #openstack-kuryr | 20:18 | |
| *** aojea_ has quit IRC | 20:21 | |
| *** aojea_ has joined #openstack-kuryr | 20:25 | |
| *** aojea has quit IRC | 20:28 | |
| *** aojea_ has quit IRC | 21:06 | |
| *** tonanhngo has quit IRC | 21:46 | |
| *** salv-orl_ has joined #openstack-kuryr | 21:49 | |
| *** tonanhngo has joined #openstack-kuryr | 21:49 | |
| *** tonanhngo has quit IRC | 21:51 | |
| *** salv-orlando has quit IRC | 21:52 | |
| *** tonanhngo has joined #openstack-kuryr | 21:58 | |
| *** tonanhngo_ has joined #openstack-kuryr | 22:01 | |
| *** tonanhngo has quit IRC | 22:03 | |
| *** alraddarla_ has quit IRC | 22:03 | |
| *** tonanhngo_ has quit IRC | 22:05 | |
| *** tonanhngo has joined #openstack-kuryr | 22:11 | |
| *** limao has joined #openstack-kuryr | 22:37 | |
| *** salv-orl_ has quit IRC | 22:44 | |
| openstackgerrit | Hongbin Lu proposed openstack/kuryr-kubernetes master: [WIP] Spec for fuxi-kubernetes https://review.openstack.org/452554 | 23:06 |
| *** limao has quit IRC | 23:26 | |
| openstackgerrit | Hongbin Lu proposed openstack/kuryr-kubernetes master: [WIP] Spec for fuxi-kubernetes https://review.openstack.org/452554 | 23:44 |
| *** hongbin has quit IRC | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!