*** kiennt26 has joined #openstack-kuryr | 01:02 | |
*** wangbo has joined #openstack-kuryr | 01:24 | |
*** caowei has joined #openstack-kuryr | 01:35 | |
*** lakerzhou1 has joined #openstack-kuryr | 01:48 | |
*** yamamoto has joined #openstack-kuryr | 01:54 | |
*** hongbin has joined #openstack-kuryr | 02:00 | |
*** wangbo has quit IRC | 02:15 | |
openstackgerrit | Merged openstack/kuryr-libnetwork master: Enable debug if ENABLE_DEBUG_LOG_LEVEL is set https://review.openstack.org/505852 | 02:38 |
---|---|---|
*** lakerzhou1 has quit IRC | 02:52 | |
*** wangbo has joined #openstack-kuryr | 02:53 | |
*** wangbo has quit IRC | 02:54 | |
*** wangbo has joined #openstack-kuryr | 02:55 | |
*** wangbo has quit IRC | 03:16 | |
*** wangbo has joined #openstack-kuryr | 03:44 | |
*** janki has joined #openstack-kuryr | 03:47 | |
*** hongbin has quit IRC | 04:04 | |
janonymous | Guest71207: i guess u can use : https://review.openstack.org/#/c/505187/1/devstack/local.conf.sample changes | 04:11 |
*** wangbo has quit IRC | 04:32 | |
openstackgerrit | Kien Nguyen proposed openstack/kuryr-libnetwork master: Fix coverage test https://review.openstack.org/506985 | 04:39 |
openstackgerrit | Kien Nguyen proposed openstack/kuryr-libnetwork master: Fix coverage test https://review.openstack.org/506985 | 04:51 |
*** caowei has quit IRC | 04:56 | |
openstackgerrit | Kien Nguyen proposed openstack/kuryr-libnetwork master: Fix coverage test https://review.openstack.org/506985 | 04:56 |
*** livelace has joined #openstack-kuryr | 05:03 | |
livelace | apuimedo, ltomasbo Hello guys, I have one question :) Could one kuryr-controller work for different types of vif-binding simultaneous ? | 05:07 |
livelace | nested-vlan && neutron-vif on one kuryr-controller ? | 05:07 |
*** wangbo has joined #openstack-kuryr | 05:09 | |
*** deepika08 has joined #openstack-kuryr | 05:45 | |
*** caowei has joined #openstack-kuryr | 05:50 | |
*** caowei has quit IRC | 05:53 | |
*** caowei has joined #openstack-kuryr | 05:56 | |
ltomasbo | livelace, unfortunately, I don't think so | 05:56 |
*** yamamoto_ has joined #openstack-kuryr | 06:36 | |
*** yamamoto has quit IRC | 06:36 | |
openstackgerrit | Jaivish Kothari(janonymous) proposed openstack/kuryr-kubernetes master: CNI Daemon template https://review.openstack.org/480028 | 06:37 |
ltomasbo | apuimedo, https://review.openstack.org/#/c/504915/ (I lost your previous +2 when adding the blueprint) | 06:56 |
*** kiennt26 has quit IRC | 07:05 | |
*** kiennt26 has joined #openstack-kuryr | 07:05 | |
*** pcaruana has joined #openstack-kuryr | 07:06 | |
*** yamamoto_ has quit IRC | 07:14 | |
*** yamamoto has joined #openstack-kuryr | 07:16 | |
openstackgerrit | Merged openstack/kuryr-kubernetes master: Avoid port update neutron call during pods boot up https://review.openstack.org/504915 | 07:28 |
*** yboaron has joined #openstack-kuryr | 07:29 | |
*** yamamoto_ has joined #openstack-kuryr | 07:33 | |
*** yamamoto has quit IRC | 07:36 | |
*** yboaron has quit IRC | 07:38 | |
*** yamamoto_ has quit IRC | 07:41 | |
openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Optimize return_to_pool actions at Pool drivers https://review.openstack.org/505605 | 07:44 |
openstackgerrit | Merged openstack/kuryr-libnetwork master: Fix code coverage ci job https://review.openstack.org/506842 | 07:46 |
*** egonzalez has joined #openstack-kuryr | 07:47 | |
*** kiennt26 has quit IRC | 08:01 | |
*** deepika08 has quit IRC | 08:05 | |
*** yamamoto has joined #openstack-kuryr | 08:10 | |
*** yamamoto has quit IRC | 08:32 | |
*** yboaron has joined #openstack-kuryr | 08:34 | |
*** wangbo has quit IRC | 09:12 | |
openstackgerrit | Jaivish Kothari(janonymous) proposed openstack/kuryr-kubernetes master: CNI Daemon template https://review.openstack.org/480028 | 09:30 |
openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Add Pool Manager to handle subports https://review.openstack.org/498698 | 09:39 |
*** kiennt26 has joined #openstack-kuryr | 09:39 | |
*** yamamoto has joined #openstack-kuryr | 09:40 | |
*** edisonxiang_ is now known as edisonxiang | 09:42 | |
*** wangbo has joined #openstack-kuryr | 09:48 | |
*** wangbo has quit IRC | 09:50 | |
janonymous | dulek: Something like : http://paste.openstack.org/show/621832/ | 09:58 |
janonymous | dulek: for https://review.openstack.org/#/c/480028/18/kuryr_kubernetes/cni/api.py | 10:00 |
dulek | janonymous: Yeah, this is much better! | 10:00 |
*** yamamoto has quit IRC | 10:00 | |
janonymous | dulek: But i think these are modifications, which could be added in follow ups, along with logging... what do you say? | 10:01 |
dulek | janonymous: I'm not sure what apuimedo will be requiring before merging. I find this small patch pretty straightforward. Proper logging can probably be added in follow up. | 10:02 |
dulek | But I would insist on getting initial unit tests with the patch. | 10:02 |
*** livelace has quit IRC | 10:03 | |
janonymous | dulek: Thanks a ton for your reviews... | 10:08 |
dulek | janonymous: Thanks for working on this! :) | 10:09 |
openstackgerrit | Jaivish Kothari(janonymous) proposed openstack/kuryr-kubernetes master: CNI Daemon template https://review.openstack.org/480028 | 10:10 |
janonymous | dulek: ^^ changed, ut i will add in follow up as i mentioned in commit msg | 10:10 |
*** kiennt26 has quit IRC | 10:10 | |
openstackgerrit | Kien Nguyen proposed openstack/kuryr-libnetwork master: Fix coverage test https://review.openstack.org/506985 | 10:10 |
*** kiennt26 has joined #openstack-kuryr | 10:11 | |
dulek | I'm not sure if that's a good practice, but I guess it's up to core reviewers to decide. :) | 10:11 |
janonymous | dulek: i will try to add those soon :) | 10:11 |
openstackgerrit | Kien Nguyen proposed openstack/kuryr-libnetwork master: Fix coverage test https://review.openstack.org/506985 | 10:22 |
*** kiennt26 has quit IRC | 10:23 | |
*** edisonxiang has quit IRC | 10:25 | |
*** edisonxiang has joined #openstack-kuryr | 10:26 | |
*** caowei has quit IRC | 10:29 | |
*** wangbo has joined #openstack-kuryr | 10:33 | |
*** wangbo has quit IRC | 10:40 | |
*** wangbo has joined #openstack-kuryr | 10:42 | |
*** wangbo has quit IRC | 10:43 | |
*** wangbo has joined #openstack-kuryr | 10:45 | |
*** edisonxiang has quit IRC | 10:45 | |
*** edisonxiang has joined #openstack-kuryr | 10:45 | |
*** wangbo has quit IRC | 10:58 | |
*** livelace has joined #openstack-kuryr | 11:06 | |
janonymous | If anyone want to know about CNI working i prepared it sometime back to get the flow: https://etherpad.openstack.org/p/kuryr_flow/export/pdf | 11:06 |
janonymous | thought that might be useful | 11:06 |
*** yamamoto has joined #openstack-kuryr | 11:30 | |
*** wangbo has joined #openstack-kuryr | 11:38 | |
*** wangbo_ has joined #openstack-kuryr | 11:42 | |
*** wangbo has quit IRC | 11:42 | |
*** alishaaneja has joined #openstack-kuryr | 11:43 | |
*** yamamoto has quit IRC | 12:00 | |
*** yamamoto has joined #openstack-kuryr | 12:18 | |
openstackgerrit | Yossi Boaron proposed openstack/kuryr-kubernetes master: Add support for service type=LoadBalancer https://review.openstack.org/498253 | 12:29 |
*** yamamoto has quit IRC | 12:33 | |
openstackgerrit | Yossi Boaron proposed openstack/kuryr-kubernetes master: Add support for service type=LoadBalancer https://review.openstack.org/498253 | 12:35 |
openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Add list and show pool commands to Pool Manager https://review.openstack.org/504410 | 12:39 |
*** yamamoto has joined #openstack-kuryr | 12:52 | |
*** edisonxiang_ has joined #openstack-kuryr | 12:57 | |
openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Add Pool Manager to handle subports https://review.openstack.org/498698 | 12:59 |
*** edisonxiang has quit IRC | 13:00 | |
*** lakerzhou has joined #openstack-kuryr | 13:03 | |
openstackgerrit | Luis Tomas Bolivar proposed openstack/kuryr-kubernetes master: Add list and show pool commands to Pool Manager https://review.openstack.org/504410 | 13:06 |
*** wangbo_ has quit IRC | 13:16 | |
*** alishaaneja has quit IRC | 13:17 | |
*** gouthamr has joined #openstack-kuryr | 13:27 | |
*** kiennt26 has joined #openstack-kuryr | 13:29 | |
*** gouthamr has quit IRC | 13:29 | |
*** gouthamr has joined #openstack-kuryr | 13:30 | |
*** lakerzhou has quit IRC | 13:37 | |
*** dougbtv has quit IRC | 13:58 | |
dulek | Do we have the meeting today? | 14:00 |
*** zengchen1 has joined #openstack-kuryr | 14:04 | |
janonymous | apuimedo: ^^ | 14:04 |
*** garyloug has joined #openstack-kuryr | 14:07 | |
zengchen1 | is the meeting canceled? | 14:07 |
janonymous | dmellado: #startmeeting tag? :P | 14:08 |
dmellado | janonymous: yeah, actually I was waiting to see more attendees | 14:09 |
dmellado | xD | 14:09 |
janonymous | :D | 14:09 |
*** hongbin has joined #openstack-kuryr | 14:20 | |
*** livelace2 has joined #openstack-kuryr | 14:20 | |
livelace2 | ltomasbo, Is there possibility to run several controller at the same time ? Do they conflict with each other ? | 14:21 |
ivc | livelace2 kuryr-libnetwork or kuryr-kubernetes? | 14:25 |
livelace | ivc, kuryr-kubernetes | 14:29 |
livelace | I want to use both, VM and baremetal with one Openshift cluster | 14:29 |
ltomasbo | livelace, with just one OpenShift cluster? | 14:29 |
ltomasbo | or with 2 openshift clusters? | 14:29 |
livelace | with one | 14:30 |
livelace | It will be cool, if nodes can send to controller the type of network interface binding | 14:31 |
ltomasbo | then I'm not sure, probably not | 14:31 |
ivc | not yet (tm) :) | 14:31 |
ltomasbo | livelace, how openshift computes will know if they are running inside a VM or not? | 14:32 |
ltomasbo | should be trnasparent for them,right? | 14:33 |
livelace | Through config file ? | 14:33 |
ltomasbo | though kuryr-cni does know about it | 14:33 |
livelace | Yes, but controller doesn't work with two or more types of binding | 14:34 |
openstackgerrit | Merged openstack/kuryr-kubernetes master: Fix Incorrect Module Reference https://review.openstack.org/491669 | 14:34 |
livelace | I think it is the good direction to support different types of nodes across the cluster | 14:34 |
janonymous | thanks ivc | 14:34 |
ivc | ltomasbo livelace this is related to the cfg handling we were discussing some time ago | 14:34 |
ivc | janonymous yw | 14:35 |
*** yboaron has quit IRC | 14:35 | |
ltomasbo | ivc, yep, sounds familiar | 14:35 |
*** livelace2 has quit IRC | 14:35 | |
ivc | ltomasbo livelace there was a discussion about handling heterogeneous envs like one livelace is describing | 14:36 |
ivc | there are some comments/TODOs left in code atm | 14:36 |
livelace | It means that it's in demand :) | 14:36 |
ivc | and multi-controller is not rly related to that. main purpose of multi-controller is to enable HA | 14:37 |
*** zengchen1 has quit IRC | 14:37 | |
ivc | but you are right that multi-controller could also help | 14:38 |
ivc | ltomasbo apuimedo livelace probably we can make a smaller step in that direction and introduce 'filter' concept so that we can specify a configurable filter per handler | 14:39 |
livelace | Cool | 14:40 |
ivc | so you can have 2 controllers with different cfg to handle different 'parts' of the env (with non-intersecting filters) | 14:40 |
ivc | might actually make an interesting use-case | 14:40 |
ltomasbo | ivc, then you may want to have a controller per compute? | 14:41 |
dmellado | yeah, and having it just as filter shouldn't make it that hard to implement | 14:41 |
ltomasbo | ivc, not only for HA but also for performance | 14:41 |
ivc | ltomasbo i'd not 'want' that but that would be certainly possible. would not solve HA (as it does not provide failover) | 14:42 |
ltomasbo | ivc, why not? will not the kubernetes replica controller schedule new pods to new servers if one of the compute nodes is down? | 14:43 |
ivc | ltomasbo tho it is close to the HA concept we had in mind (tagging containers/nodes with per-controller hash/id and rebalancing it in the event of controller failure) | 14:44 |
ivc | ltomasbo we should not forget about network access (i.e. controller requires connectivity to neutron) and it was one of the initial requirements to not rely on k8s nodes having access to neutron admin network | 14:46 |
ltomasbo | ivc, ahh, true | 14:47 |
ivc | ltomasbo thats the main reason we have dedicated controller. the idea to run controller along with cni-daemon on the same k8s nodes floated around about a year ago :) but we discarded that | 14:47 |
ltomasbo | ivc, so, the idea for the HA is to have something similar to neutron HA, with, let's say, 3 kuryr-controllers? | 14:48 |
*** livelace2 has joined #openstack-kuryr | 14:49 | |
ivc | ltomasbo most likely but we dont have a proper blueprint/spec yet | 14:49 |
ltomasbo | ok | 14:50 |
ivc | ltomasbo i've posted some early 'draft' about a year ago that also covered some HA concepts but i've abandoned it long ago | 14:51 |
ltomasbo | ivc, do you have the link handy so that I can take a look (just curious)? | 14:51 |
dmellado | ivc: a year ago looks like we might need to totaly refactor it | 14:52 |
dmellado | I'd be also curious on the link xD | 14:52 |
ivc | ltomasbo https://review.openstack.org/#/c/369105/1/doc/source/devref/event-processing.rst | 14:53 |
ivc | dmellado ltomasbo look for "Event concurrency across the nodes" section | 14:53 |
dmellado | ivc: I'll review it and check, thanks | 14:54 |
ltomasbo | ivc, thanks! | 14:54 |
ivc | dmellado ltomasbo if you plan to revive that one, please consider also including "Events as state" and "Declarative event handlers" sections too | 14:56 |
ivc | those are rly the core concept of our handlers/drivers and we do not have it properly documented anywhere | 14:57 |
ivc | concepts* | 14:57 |
ltomasbo | ivc, ok thanks! | 14:57 |
dmellado | sure! ltomasbo let's have a look at it and discuss tomorrow | 14:57 |
ltomasbo | sure | 14:57 |
*** kiennt26 has quit IRC | 15:21 | |
*** apuimedo has quit IRC | 15:32 | |
*** pcaruana has quit IRC | 15:33 | |
*** pcaruana has joined #openstack-kuryr | 15:34 | |
*** janki has quit IRC | 15:52 | |
*** egonzalez has quit IRC | 16:01 | |
*** pcaruana has quit IRC | 16:15 | |
*** garyloug has quit IRC | 16:45 | |
*** livelace2 has quit IRC | 16:47 | |
*** livelace has quit IRC | 16:59 | |
*** livelace2 has joined #openstack-kuryr | 17:01 | |
*** egonzalez has joined #openstack-kuryr | 17:33 | |
*** alraddarla has joined #openstack-kuryr | 18:06 | |
*** livelace2 has quit IRC | 18:12 | |
*** livelace2 has joined #openstack-kuryr | 18:25 | |
*** livelace2 has quit IRC | 18:30 | |
*** aojea has quit IRC | 18:34 | |
*** ltomasbo has quit IRC | 18:35 | |
*** ltomasbo has joined #openstack-kuryr | 18:37 | |
*** livelace2 has joined #openstack-kuryr | 18:38 | |
*** livelace2 has quit IRC | 18:44 | |
*** aojea has joined #openstack-kuryr | 18:45 | |
*** aojea has quit IRC | 18:49 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/fuxi master: Updated from global requirements https://review.openstack.org/495179 | 18:55 |
*** livelace2 has joined #openstack-kuryr | 18:56 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/kuryr master: Updated from global requirements https://review.openstack.org/496048 | 18:57 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/kuryr-libnetwork master: Updated from global requirements https://review.openstack.org/502663 | 18:57 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/kuryr-kubernetes master: Updated from global requirements https://review.openstack.org/500060 | 18:57 |
*** aojea has joined #openstack-kuryr | 19:08 | |
*** egonzalez has quit IRC | 20:05 | |
*** ltomasbo has quit IRC | 20:14 | |
*** ltomasbo has joined #openstack-kuryr | 20:17 | |
*** c00281451_ has joined #openstack-kuryr | 20:48 | |
*** zengchen has quit IRC | 20:51 | |
*** gouthamr has quit IRC | 21:04 | |
*** aojea has quit IRC | 21:36 | |
*** gouthamr has joined #openstack-kuryr | 22:12 | |
*** hongbin has quit IRC | 23:29 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!