| *** hongbin has quit IRC | 00:15 | |
| *** yedongcan has joined #openstack-kuryr | 00:57 | |
| *** hongbin has joined #openstack-kuryr | 01:11 | |
| *** kiennt has joined #openstack-kuryr | 01:18 | |
| openstackgerrit | Merged openstack/kuryr-tempest-plugin master: Updated from global requirements https://review.openstack.org/477159 | 01:30 |
|---|---|---|
| openstackgerrit | Merged openstack/kuryr-kubernetes master: Updated from global requirements https://review.openstack.org/469489 | 01:40 |
| *** longfei_zhang has joined #openstack-kuryr | 01:52 | |
| openstackgerrit | Merged openstack/fuxi master: Updated from global requirements https://review.openstack.org/454474 | 01:56 |
| *** kiennt has quit IRC | 01:58 | |
| openstackgerrit | Hongbin Lu proposed openstack/fuxi master: Upgrade from docker-py to docker https://review.openstack.org/475895 | 02:00 |
| *** caowei has joined #openstack-kuryr | 02:03 | |
| *** yamamoto has joined #openstack-kuryr | 02:49 | |
| *** reedip has quit IRC | 03:09 | |
| *** yedongcan has quit IRC | 03:20 | |
| *** yedongcan has joined #openstack-kuryr | 03:21 | |
| *** hongbin has quit IRC | 03:30 | |
| *** yamamoto has quit IRC | 03:50 | |
| *** yamamoto has joined #openstack-kuryr | 03:53 | |
| *** hongbin has joined #openstack-kuryr | 03:55 | |
| *** vikasc has joined #openstack-kuryr | 03:57 | |
| *** kiennt has joined #openstack-kuryr | 04:10 | |
| vikasc | apuimedo, kzaitsev1pi ltomasbo irenab ivc_ dmellado janonymous : so multus has also been converted to a controller, cni kind of plugin like kuryr and just started supporting tpr based multi-neworks, https://github.com/Intel-Corp/multus-cni#usage-with-kubernetes-tpr-based-network-objects | 04:11 |
| vikasc | multus network object also seems to be having similar details as i proposed here,https://review.openstack.org/#/c/477066/ . Though format seems to be a little different. | 04:15 |
| *** hongbin has quit IRC | 04:21 | |
| openstackgerrit | Kien Nguyen proposed openstack/kuryr master: Add Apache License Content in index.rst https://review.openstack.org/455229 | 04:26 |
| *** longfei_zhang has quit IRC | 04:30 | |
| janonymous | vikasc: +1 | 05:29 |
| openstackgerrit | Merged openstack/kuryr-kubernetes master: Clarify _is_pending function use https://review.openstack.org/477824 | 05:29 |
| *** kiennt has quit IRC | 05:40 | |
| openstackgerrit | Merged openstack/fuxi master: Upgrade from docker-py to docker https://review.openstack.org/475895 | 05:47 |
| *** kiennt has joined #openstack-kuryr | 06:06 | |
| *** reedip has joined #openstack-kuryr | 06:13 | |
| ltomasbo | thanks vikasc! nice finding! | 06:21 |
| vikasc | ltomasbo, noticed on sig-network so thought of sharing | 06:22 |
| ltomasbo | great! | 06:23 |
| *** sridhar_ram has joined #openstack-kuryr | 06:23 | |
| vikasc | ltomasbo, thanks! | 06:30 |
| ltomasbo | apuimedo, FYI: https://review.openstack.org/#/c/442496/ | 06:35 |
| *** yboaron has joined #openstack-kuryr | 06:41 | |
| openstackgerrit | Merged openstack/kuryr-tempest-plugin master: Skip test_list_pods https://review.openstack.org/478524 | 06:43 |
| *** kzaitsev_ws has joined #openstack-kuryr | 06:55 | |
| openstackgerrit | Kirill Zaitsev proposed openstack/kuryr master: Use openstackdocstheme over oslosphinx https://review.openstack.org/478868 | 06:57 |
| *** pcaruana has joined #openstack-kuryr | 06:59 | |
| *** pcaruana has quit IRC | 07:03 | |
| *** pcaruana has joined #openstack-kuryr | 07:05 | |
| *** lihi has joined #openstack-kuryr | 07:21 | |
| *** aojea has joined #openstack-kuryr | 07:24 | |
| *** openstackgerrit has quit IRC | 07:48 | |
| ltomasbo | hi again vikasc | 07:57 |
| ltomasbo | I just read your multi-network support spec | 07:57 |
| vikasc | hi ltomasbo | 07:57 |
| vikasc | does it make sense a bit :) | 07:58 |
| ltomasbo | just one to double check with you | 07:58 |
| ltomasbo | yes, it does! | 07:58 |
| vikasc | sure, please go ahead | 07:58 |
| ltomasbo | so, your idea is to use a network object to create the networks in neutron from kuberentes, right? | 07:58 |
| ltomasbo | and then, something similar to my PoC patch to allow pods to choose those networks, right? | 07:59 |
| vikasc | no, network object represents a nic on k8s side | 07:59 |
| vikasc | so IMO, subnet will be the entity on neutron side | 07:59 |
| vikasc | thats how i was thinking | 08:00 |
| ltomasbo | so, then, the idea is to have a pod spec | 08:00 |
| ltomasbo | which includes that Network object too? | 08:00 |
| ltomasbo | or you create that in a two step process? | 08:00 |
| vikasc | since k8s scheduler does not understand networkobjects | 08:01 |
| ltomasbo | I just want to make sure I got the step by step actions (create network, create container in that network) | 08:01 |
| vikasc | so for now, i am thinking to represent/request network object by OIR in pod spec | 08:01 |
| ltomasbo | yes, that I get, that for sr-iov and these type of resources, you need to account where they are | 08:02 |
| ltomasbo | but, for standard neutron subnets/networks | 08:02 |
| ltomasbo | where we can take for granted they are present in any node (so, no special scheduling) | 08:02 |
| ltomasbo | what will be the steps? | 08:02 |
| ltomasbo | I understood from your spec that you use the network objects to create the networks | 08:03 |
| vikasc | can it be a case when even for standard neutron subnets, all nodes might not have all the subnets? | 08:04 |
| vikasc | is this a valid case? | 08:04 |
| vikasc | i am just thinking | 08:04 |
| ltomasbo | well, I guess there may be such a case | 08:05 |
| ltomasbo | I was just interested in the other part now, not focusing on the OIR | 08:05 |
| vikasc | i am sorry if i was not clear enough in spec and gave impression that network objects are used to create neutron networks. | 08:05 |
| ltomasbo | but I agree the OIR has a role here (and it is being used already by the SR-IOV supporT) | 08:05 |
| ltomasbo | so, network objects only create the network from the kuberentes point of view, right? | 08:06 |
| ltomasbo | and the kuryr controller will be in charge of creating them (if they do not exist already) | 08:06 |
| vikasc | yes | 08:06 |
| vikasc | creating subnets | 08:06 |
| *** openstackgerrit has joined #openstack-kuryr | 08:06 | |
| openstackgerrit | Kirill Zaitsev proposed openstack/kuryr-libnetwork master: Use openstackdocstheme over oslosphinx https://review.openstack.org/479308 | 08:07 |
| vikasc | if matching cidr subnet is not present | 08:07 |
| vikasc | so your optional subnet driver will come into picture | 08:07 |
| kzaitsev_ws | ltomasbo: vikasc: folks, can you please explain me the role of OIR there? =) | 08:07 |
| ltomasbo | but, a pod creation will just have an annotation of the kuberentes networks, right? like the name | 08:07 |
| ltomasbo | and then, the subnet kuryr driver need to map that with the network objects (handler) and then use/create the neutron subnet | 08:08 |
| vikasc | kzaitsev_ws, assuming you asking about standar subnets, there OIR max available can be a neutron port quota. still thinking on this | 08:09 |
| ltomasbo | kzaitsev_ws, as I understood it, it is needed to perform the scheduling, as for instance, for the SR-IOV case, not all the nodes may support it, and you need to schedule the pods where there are VFs available | 08:09 |
| vikasc | ltomasbo, yes | 08:09 |
| ltomasbo | vikasc, now that you mention, that could be used in conjuntion with ports pools | 08:09 |
| kzaitsev_ws | ltomasbo: yes, but OIR is quantitive, i.e. there are X OIRs | 08:10 |
| ltomasbo | to allocate the pods where there is already existing ports in the pool | 08:10 |
| kzaitsev_ws | so it feels very much like per-node quota, yes | 08:10 |
| vikasc | ltomasbo, i am thinking on same lines but dont have refines thoughts yet | 08:10 |
| vikasc | s/refined | 08:10 |
| ltomasbo | kzaitsev_ws, yep, it is qunatitative, but if you define to be 0 VFs, then you will not schedule pods there, right? | 08:10 |
| vikasc | ltomasbo, makes senses to me | 08:11 |
| ltomasbo | I still need to give it a second thought, but just wanted to double check with you vikasc that I understood waht you suggested on the spec | 08:12 |
| vikasc | ltomasbo, ports pool can be mapped to OIR counts and this will be applicable to all vif types | 08:12 |
| kzaitsev_ws | ltomasbo: the workflow is the following from k8s side. You advertise that node-1 has 4 OIR-x. and node-2 has 3 OIR-x | 08:12 |
| kzaitsev_ws | then a Pod requests 2 OIR-x and get's scheduled to node-2 (for example.) | 08:12 |
| kzaitsev_ws | so node-2 only has 1 OIR-x left and another Pod, that wants 2 OIR-xs would not get scheduled there | 08:13 |
| vikasc | and the controller will figure out network object naame from OIR on the node | 08:13 |
| vikasc | s/then | 08:13 |
| kzaitsev_ws | so yeah if we call it a per-node quota that fits | 08:13 |
| kzaitsev_ws | ltomasbo: (not sure what you mean by define 0VFs, that's in the request or on the node?) | 08:14 |
| vikasc | and controller will fetch network object from apiserver to get network metadata and accordingly other handlers will be invoked | 08:14 |
| kzaitsev_ws | afk for 2 mins, gotta get a delivery | 08:14 |
| ltomasbo | kzaitsev_ws, node I assume | 08:15 |
| *** yboaron_ has joined #openstack-kuryr | 08:15 | |
| *** aojea has quit IRC | 08:15 | |
| *** aojea has joined #openstack-kuryr | 08:16 | |
| vikasc | multi-network seems a misnomer. It is multi-nic to be precise. But somehow community around sig-network calls is multi-network :) | 08:21 |
| kzaitsev_ws | back ) | 08:21 |
| kzaitsev_ws | ltomasbo: then the node would not have these OIRs and the Pods, that request these OIRs would never get scheduled on it | 08:22 |
| kzaitsev_ws | my point is that OIRs map real nice to SRIOV, since you only have that much VFs on a node (say your nic supports 63 VFs — you only have 63 physnet-x OIRs on that node) | 08:23 |
| kzaitsev_ws | but for general admission/scheduling it looks like a misuse of a feature, and taints/toleraqtions would be more fit | 08:24 |
| kzaitsev_ws | vikasc: sorry for whining about the thing again ) | 08:24 |
| kzaitsev_ws | vikasc: true about multi-network. multi-interface would be better maybe? | 08:25 |
| apuimedo | https://review.openstack.org/#/c/442496/ is really good news | 08:25 |
| apuimedo | I'm sick and tired of the hook hack | 08:25 |
| kzaitsev_ws | however some of the docs you've shared mention, that multi-network does not always mean multiple interfaces. they might mean multiple addresses, routes for the same interface, so in the end it might be ok-ish | 08:26 |
| ltomasbo | apuimedo, I knew you would like it! | 08:27 |
| ltomasbo | kzaitsev_ws, yep, you are right | 08:28 |
| kzaitsev_ws | about OIR? =) | 08:28 |
| ltomasbo | it feels right for SR-IOV as it is actually a node feature | 08:28 |
| ltomasbo | yep | 08:28 |
| ltomasbo | I was mainly asking vikasc about the other parts of the spec | 08:28 |
| vikasc | kzaitsev_ws, even in standard subnets case, OIR will be used to get k8s scheduler working roght | 08:28 |
| vikasc | s/right | 08:28 |
| ltomasbo | but need to get a little bit more information about taints/tolerations to set my mind about this | 08:29 |
| vikasc | kzaitsev_ws, in case a node does not have a particular subnet requested by pod | 08:29 |
| kzaitsev_ws | vikasc: I'm ok with using them there, I just want to understand what they would represent | 08:29 |
| vikasc | kzaitsev_ws, neutron port pool quota per node | 08:30 |
| kzaitsev_ws | yeah, that fits | 08:30 |
| vikasc | kzaitsev_ws, and ultimately sriov will also come through port pools | 08:30 |
| kzaitsev_ws | a bit annoying, that you can't set -1 (unlimited), but probably ok | 08:30 |
| vikasc | kzaitsev_ws, that if needed, in future, we can implement by keep patching OIR for such case | 08:31 |
| vikasc | and dont let it decrease | 08:31 |
| kzaitsev_ws | vikasc: agree. basically it is a port-pool-quota anyway | 08:32 |
| vikasc | kzaitsev_ws, i might be saying absurd and it might not be needed at all | 08:32 |
| vikasc | kzaitsev_ws, i mean we have ways to dealt with it | 08:32 |
| apuimedo | vikasc: I'd rather not use OIR for normal subnets | 08:35 |
| apuimedo | unless the sig-net implementations want to use it | 08:35 |
| vikasc | apuimedo, suppose a pod wants to have two nic in two normal neutron subnets and in the cluster not all nodes have both of these subnets. Either OIR can be used to request each network or requested networks can be marked as pod annotations and taints/tolerations will be used | 08:39 |
| vikasc | Are you suggesting the later one or any other option? | 08:39 |
| apuimedo | vikasc: you're talking about physical networks? | 08:40 |
| *** garyloug has joined #openstack-kuryr | 08:41 | |
| vikasc | apuimedo, i meant network objects(TPR) that a pod can attach to. And network object would map to a neutron subnet | 08:42 |
| vikasc | apuimedo, talking in context to this write-up, https://review.openstack.org/#/c/477066/ | 08:46 |
| apuimedo | vikasc: where's the upstream doc on these attachable network objects? | 08:47 |
| apuimedo | as in, how do you attach them? | 08:47 |
| vikasc | apuimedo, you mean how to refer these in pod object? | 08:48 |
| vikasc | apuimedo, not sure if i understood what you meant by 'attach'? | 08:49 |
| vikasc | apuimedo, upstream dont support these objects natively and multi-nic is to supported using CRD (TPR successor) | 08:52 |
| vikasc | apuimedo, plus annotations | 08:52 |
| *** yboaron_ has quit IRC | 08:58 | |
| openstackgerrit | vikas choudhary proposed openstack/kuryr-kubernetes master: Add support to install Kuryr as a network addon https://review.openstack.org/466675 | 09:03 |
| vikasc | apuimedo, ^ | 09:03 |
| apuimedo | you said "attachable" | 09:03 |
| vikasc | apuimedo, sorry, not able to find where i used attachable. nevermind | 09:06 |
| vikasc | apuimedo, does my reply answer your question? | 09:06 |
| vikasc | apuimedo, ok "attach to" i used | 09:07 |
| apuimedo | vikasc: right | 09:07 |
| *** dimak_ has joined #openstack-kuryr | 09:08 | |
| vikasc | apuimedo, so to attach them, those will be requested as OIR in pod spec | 09:08 |
| vikasc | apuimedo, thats the approach in the patch. Alternate approach would be to use annotations and tanits/tolerations(to create dedicated clusters since k8s scheduler doesnt understand network objects and cant schedule for them). | 09:10 |
| vikasc | apuimedo, i prefered first one because: | 09:10 |
| vikasc | 1. OIR already accepeted with sriov vif | 09:11 |
| vikasc | 2. OIR approach is resource efficient compared with taints/tolerations | 09:11 |
| *** vikasc is now known as vikasc|lunch | 09:13 | |
| openstackgerrit | Danil Golov proposed openstack/kuryr-kubernetes master: Allow setting specific ports for SRIOV handler https://review.openstack.org/478494 | 09:16 |
| *** dimak_ has quit IRC | 09:17 | |
| openstackgerrit | Danil Golov proposed openstack/kuryr-kubernetes master: Allow setting specific ports for SRIOV handler https://review.openstack.org/478494 | 09:17 |
| *** openstackgerrit has quit IRC | 09:18 | |
| *** yamamoto has quit IRC | 09:20 | |
| *** sridhar_ram has quit IRC | 09:20 | |
| *** yamamoto has joined #openstack-kuryr | 09:22 | |
| *** vikasc|lunch is now known as vikasc | 09:31 | |
| *** garyloug has quit IRC | 09:33 | |
| *** garyloug has joined #openstack-kuryr | 09:33 | |
| *** kural has joined #openstack-kuryr | 09:36 | |
| apuimedo | vikasc: which approach is the community taking? | 09:55 |
| vikasc | apuimedo, there are couple of pocs at different stages of development which i observed on sig-network googlr-group channel. multus cni is latest one to use tpr. Still need to look deeper to see handling of specific use cases | 09:59 |
| apuimedo | ok | 10:00 |
| apuimedo | I'd rather we go for the same approach | 10:01 |
| apuimedo | and I doubt that's going to be OIR | 10:01 |
| apuimedo | I'd probably even rather have a scheduler extender | 10:01 |
| vikasc | yes, scheduler extender can help avoiding OIRs | 10:02 |
| vikasc | i was thinking how devicePlugins/ResourceClasses effort at sig-node would relate to this | 10:03 |
| vikasc | this new functionality would replace OIRs | 10:04 |
| vikasc | and it could be possible to have rich metadata about devices natively (which is being solved by network objects currently) | 10:06 |
| vikasc | ofcourse these functionalities will take a bit , ~1.9 or 1.0 | 10:07 |
| vikasc | 1.10 | 10:07 |
| apuimedo | yeah | 10:07 |
| apuimedo | that's why I said that we should check closely what others are doing | 10:07 |
| vikasc | for the interim, we can have scheduler externder | 10:07 |
| vikasc | i am part of this effort | 10:07 |
| vikasc | resourceclasses specifically and keeping an eye on deviceplugin progress | 10:08 |
| irenab | vikasc, OIR? | 10:09 |
| vikasc | opaque-integer-resources | 10:09 |
| *** atoth has quit IRC | 10:09 | |
| irenab | vikasc, I will try to catch up on your spec by today's meeting | 10:10 |
| vikasc | irenab, thanks! | 10:10 |
| vikasc | irenab, intention behind pushing spec was only to have a placeholder for discussions. | 10:11 |
| irenab | vikasc, cool, this is better than keeping it on the channel only | 10:12 |
| vikasc | irenab, exactly, thats what i was thinking | 10:12 |
| vikasc | irenab, so i just quickly wrote-up whatever i had in mind without bothering much on format or validation. So not strict on anything which i proposed there. | 10:14 |
| *** dims has quit IRC | 10:14 | |
| *** dims has joined #openstack-kuryr | 10:15 | |
| irenab | vikasc, should we proceed with usual review process or you plan to abandon it once discussion if over? | 10:15 |
| vikasc | irenab, lets start with review process and see if that spec makes sense a bit and how we all feel about it. | 10:16 |
| vikasc | irenab, if finally we feel that overall direction is fine | 10:16 |
| vikasc | irenab, i will reiterate to improve it otherwise abandon it | 10:17 |
| irenab | vikasc, sounds good | 10:18 |
| vikasc | irenab, thanks! | 10:19 |
| *** kiennt has quit IRC | 10:23 | |
| *** yamamoto has quit IRC | 10:45 | |
| *** yamamoto has joined #openstack-kuryr | 10:46 | |
| *** yamamoto has quit IRC | 10:46 | |
| *** caowei has quit IRC | 10:47 | |
| *** atoth has joined #openstack-kuryr | 10:57 | |
| *** kural has quit IRC | 11:14 | |
| *** kural has joined #openstack-kuryr | 11:21 | |
| *** mestery has quit IRC | 11:22 | |
| *** mestery has joined #openstack-kuryr | 11:26 | |
| *** aojea_ has joined #openstack-kuryr | 11:31 | |
| *** aojea has quit IRC | 11:34 | |
| *** openstackgerrit has joined #openstack-kuryr | 11:37 | |
| openstackgerrit | Kirill Zaitsev proposed openstack/kuryr master: Use openstackdocstheme over oslosphinx https://review.openstack.org/478868 | 11:37 |
| openstackgerrit | Kirill Zaitsev proposed openstack/kuryr-libnetwork master: Use openstackdocstheme over oslosphinx https://review.openstack.org/479308 | 11:38 |
| openstackgerrit | Kirill Zaitsev proposed openstack/kuryr-kubernetes master: Use openstackdocstheme over oslosphinx https://review.openstack.org/478856 | 11:42 |
| kzaitsev_ws | irenab: sorry for the annoying docs patches %) | 11:43 |
| irenab | kzaitsev_ws, its great that you are keeping kuryr aligned with the rest of the openstack projects | 11:43 |
| kzaitsev_ws | those are still usually annoying =) | 11:44 |
| irenab | yea.. | 11:44 |
| *** yamamoto has joined #openstack-kuryr | 11:46 | |
| kzaitsev_ws | one of these evenings I'll prbably do the same for fuxi | 11:47 |
| kzaitsev_ws | irenab: how was your trip to China? =) | 11:47 |
| openstackgerrit | Merged openstack/kuryr-libnetwork master: Support tagging existing subnetpool https://review.openstack.org/470640 | 11:47 |
| irenab | kzaitsev_ws, had very intensive agenda, but overall was good | 11:48 |
| *** yamamoto has quit IRC | 11:50 | |
| *** yamamoto has joined #openstack-kuryr | 11:50 | |
| *** yamamoto has quit IRC | 11:53 | |
| kzaitsev_ws | heh, nice that you're all back | 12:00 |
| kzaitsev_ws | and I can start pinging you with review requests =)) | 12:00 |
| *** aojea_ has quit IRC | 12:05 | |
| *** aojea has joined #openstack-kuryr | 12:05 | |
| *** aojea has quit IRC | 12:09 | |
| *** yamamoto has joined #openstack-kuryr | 12:10 | |
| *** yamamoto has quit IRC | 12:19 | |
| *** yamamoto has joined #openstack-kuryr | 12:23 | |
| *** yamamoto has quit IRC | 12:28 | |
| irenab | kzaitsev_ws, :-) you always can | 12:28 |
| *** rwallner has joined #openstack-kuryr | 12:31 | |
| *** ApheleiaS has joined #openstack-kuryr | 12:38 | |
| *** yamamoto has joined #openstack-kuryr | 12:59 | |
| *** aojea has joined #openstack-kuryr | 13:06 | |
| *** mattmceuen has joined #openstack-kuryr | 13:08 | |
| *** aojea has quit IRC | 13:10 | |
| *** kzaitsev_ws has quit IRC | 13:24 | |
| *** kzaitsev_ws has joined #openstack-kuryr | 13:24 | |
| kzaitsev_ws | actually would appreciate some reviews to https://review.openstack.org/#/c/471012/ all of my sriov things are based on it, so before I'll jump into updating any of those, I'd love to get some feedback here | 13:37 |
| kzaitsev_ws | apuimedo: btw. we ran a meeting last mon with dmellado %) | 13:45 |
| kzaitsev_ws | wanted to ping and warn you when you're back from Boston, but <insert excuse here> | 13:46 |
| irenab | kzaitsev_ws: I will review the set of patches related to multi vifs, sr-iov by tomorrow | 13:48 |
| kzaitsev_ws | hope it's ok (= we had garyloug come by and ask about DPDK work he is doing | 13:48 |
| kzaitsev_ws | irenab: awesome =) looking forward to it ) | 13:48 |
| irenab | apuimedo: sorry, cannot attend weekly today, will follow up afterwards | 13:59 |
| apuimedo | irenab: very well | 13:59 |
| *** aojea has joined #openstack-kuryr | 14:00 | |
| *** mchiappe1o is now known as mchiappero | 14:07 | |
| *** zengchen1 has joined #openstack-kuryr | 14:11 | |
| openstackgerrit | Merged openstack/kuryr-kubernetes master: Use openstackdocstheme over oslosphinx https://review.openstack.org/478856 | 14:24 |
| *** yedongcan has left #openstack-kuryr | 14:30 | |
| openstackgerrit | Merged openstack/kuryr-libnetwork master: Use openstackdocstheme over oslosphinx https://review.openstack.org/479308 | 14:33 |
| openstackgerrit | Merged openstack/kuryr-kubernetes master: Enable some off-by-default checks https://review.openstack.org/476424 | 14:37 |
| *** zengchen1 has quit IRC | 14:47 | |
| *** atoth has quit IRC | 14:50 | |
| *** hongbin has joined #openstack-kuryr | 15:09 | |
| *** yboaron has quit IRC | 15:27 | |
| *** yamamoto has quit IRC | 15:28 | |
| *** yamamoto has joined #openstack-kuryr | 15:31 | |
| *** yamamoto has quit IRC | 15:36 | |
| *** aojea has quit IRC | 15:48 | |
| kzaitsev_ws | vikasc: about https://review.openstack.org/#/c/466675/14/k8s_client.patch can we use six.b there instead? | 15:56 |
| kzaitsev_ws | I'm commenting on the patch, but gerrit seems to hang | 15:56 |
| *** hongbin_ has joined #openstack-kuryr | 16:19 | |
| *** kzaitsev_ws has quit IRC | 16:19 | |
| *** hongbin has quit IRC | 16:21 | |
| *** pc_m has quit IRC | 16:29 | |
| *** pc_m has joined #openstack-kuryr | 16:30 | |
| *** atoth has joined #openstack-kuryr | 16:33 | |
| *** yamamoto has joined #openstack-kuryr | 16:33 | |
| *** pcaruana has quit IRC | 16:35 | |
| *** pc_m has quit IRC | 16:37 | |
| *** yamamoto has quit IRC | 16:39 | |
| *** garyloug has quit IRC | 17:23 | |
| *** pc_m has joined #openstack-kuryr | 17:32 | |
| *** hongbin_ has quit IRC | 18:06 | |
| *** rwallner has quit IRC | 18:13 | |
| *** kural has quit IRC | 18:41 | |
| *** aojea has joined #openstack-kuryr | 18:41 | |
| *** rwallner has joined #openstack-kuryr | 19:18 | |
| *** aojea has quit IRC | 19:29 | |
| *** aojea has joined #openstack-kuryr | 19:33 | |
| *** hongbin has joined #openstack-kuryr | 20:11 | |
| *** aojea has quit IRC | 20:13 | |
| *** pc_m has quit IRC | 20:20 | |
| *** aojea has joined #openstack-kuryr | 20:21 | |
| *** aojea has quit IRC | 20:25 | |
| *** kzaitsev_mb has joined #openstack-kuryr | 20:27 | |
| *** kzaitsev_mb has quit IRC | 20:33 | |
| *** aojea has joined #openstack-kuryr | 20:33 | |
| *** kzaitsev_mb has joined #openstack-kuryr | 20:37 | |
| *** kzaitsev_mb has quit IRC | 20:54 | |
| *** kzaitsev_mb has joined #openstack-kuryr | 21:03 | |
| *** pc_m has joined #openstack-kuryr | 21:03 | |
| *** kzaitsev_mb has quit IRC | 21:22 | |
| *** kural has joined #openstack-kuryr | 21:22 | |
| *** kural_ has joined #openstack-kuryr | 21:22 | |
| *** kzaitsev_mb has joined #openstack-kuryr | 21:22 | |
| *** pmannidi has joined #openstack-kuryr | 21:39 | |
| *** neiljerram has quit IRC | 21:41 | |
| *** rwallner has quit IRC | 21:45 | |
| *** aojea has quit IRC | 22:03 | |
| *** pmannidi has quit IRC | 22:19 | |
| *** hongbin has quit IRC | 22:27 | |
| *** hongbin has joined #openstack-kuryr | 22:28 | |
| *** neiljerram has joined #openstack-kuryr | 22:30 | |
| *** pmannidi has joined #openstack-kuryr | 22:32 | |
| *** neiljerram has quit IRC | 22:51 | |
| *** kzaitsev_mb has quit IRC | 22:57 | |
| *** kural_ has quit IRC | 22:58 | |
| *** kural has quit IRC | 22:59 | |
| *** kural has joined #openstack-kuryr | 23:00 | |
| *** kural_ has joined #openstack-kuryr | 23:00 | |
| *** kural_ has quit IRC | 23:05 | |
| *** kural_ has joined #openstack-kuryr | 23:05 | |
| *** hongbin has quit IRC | 23:17 | |
| *** kural_ has quit IRC | 23:27 | |
| *** kural_ has joined #openstack-kuryr | 23:27 | |
| *** s1061123 has quit IRC | 23:57 | |
| *** s1061123 has joined #openstack-kuryr | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!