| *** tonanhngo has joined #openstack-kuryr | 00:00 | |
| *** shashank_hegde has quit IRC | 00:20 | |
| *** banix has joined #openstack-kuryr | 00:28 | |
| *** diogogmt has quit IRC | 00:28 | |
| *** limao_ has joined #openstack-kuryr | 00:46 | |
| limao_ | Thanks apuimedo, get it | 00:49 |
|---|---|---|
| *** tonanhngo has quit IRC | 00:59 | |
| *** shashank_hegde has joined #openstack-kuryr | 01:13 | |
| *** banix has quit IRC | 01:20 | |
| *** banix has joined #openstack-kuryr | 01:21 | |
| *** diogogmt has joined #openstack-kuryr | 01:25 | |
| *** shashank_hegde has quit IRC | 01:26 | |
| *** shashank_hegde has joined #openstack-kuryr | 01:27 | |
| *** shashank_hegde has quit IRC | 01:32 | |
| *** huikang has joined #openstack-kuryr | 01:33 | |
| *** banix has quit IRC | 01:37 | |
| *** sdake has quit IRC | 02:13 | |
| *** sdake has joined #openstack-kuryr | 02:13 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr-libnetwork: Fix hard-code listen ip "0.0.0.0" https://review.openstack.org/354563 | 02:20 |
| *** huikang has quit IRC | 02:21 | |
| *** huikang has joined #openstack-kuryr | 02:40 | |
| *** yuanying has quit IRC | 02:50 | |
| *** limao has quit IRC | 02:52 | |
| *** tonanhngo has joined #openstack-kuryr | 03:00 | |
| *** tonanhngo has quit IRC | 03:01 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr-libnetwork: Fix hard-code listen ip "0.0.0.0" https://review.openstack.org/354563 | 03:11 |
| *** limao_ has quit IRC | 03:12 | |
| *** diogogmt has quit IRC | 03:28 | |
| *** sdake has quit IRC | 03:34 | |
| *** oanson has joined #openstack-kuryr | 03:41 | |
| *** yuanying has joined #openstack-kuryr | 03:49 | |
| *** limao has joined #openstack-kuryr | 03:54 | |
| *** yuanying has quit IRC | 03:55 | |
| *** oanson has quit IRC | 03:55 | |
| *** huikang has quit IRC | 04:01 | |
| *** huikang has joined #openstack-kuryr | 04:11 | |
| *** limao_ has joined #openstack-kuryr | 04:13 | |
| *** yuanying has joined #openstack-kuryr | 04:16 | |
| *** limao_ has quit IRC | 04:17 | |
| *** shashank_hegde has joined #openstack-kuryr | 04:19 | |
| *** tonanhngo has joined #openstack-kuryr | 04:21 | |
| *** oanson has joined #openstack-kuryr | 04:47 | |
| *** huikang has quit IRC | 05:03 | |
| *** limao_ has joined #openstack-kuryr | 05:23 | |
| *** limao_ has quit IRC | 05:25 | |
| *** limao_ has joined #openstack-kuryr | 05:26 | |
| *** yuanying_ has joined #openstack-kuryr | 05:29 | |
| *** yuanying has quit IRC | 05:29 | |
| *** janki|afk has joined #openstack-kuryr | 05:35 | |
| *** limao has quit IRC | 06:02 | |
| *** tonanhngo has quit IRC | 06:25 | |
| *** janki|afk has quit IRC | 06:27 | |
| openstackgerrit | Zhigang Li proposed openstack/kuryr-libnetwork: Fix URL in README.rst https://review.openstack.org/356258 | 06:40 |
| *** shashank_hegde has quit IRC | 06:58 | |
| *** tonanhngo has joined #openstack-kuryr | 07:01 | |
| *** tonanhngo has quit IRC | 07:02 | |
| *** shashank_hegde has joined #openstack-kuryr | 07:17 | |
| *** shashank_hegde has quit IRC | 07:17 | |
| *** devvesa has joined #openstack-kuryr | 07:23 | |
| *** irenab_ has joined #openstack-kuryr | 07:42 | |
| vikasc | devvesa, hey morning :) | 07:44 |
| vikasc | devvesa, i found a bug in raven | 07:44 |
| *** irenab has quit IRC | 07:44 | |
| *** irenab_ is now known as irenab | 07:44 | |
| vikasc | devvesa, responsible for yesterday's 'unbound' problem | 07:45 |
| vikasc | devvesa, https://github.com/midonet/kuryr/blob/k8s/kuryr/raven/watchers.py#L462 here this 'binding:host_id': should also be set | 07:48 |
| irenab | vikasc: it is not possible at this point | 07:48 |
| vikasc | irenab, i understand limitation | 07:48 |
| irenab | racven does not know the host _id | 07:48 |
| vikasc | irenab, scheduler will determine host | 07:49 |
| vikasc | irenab, but if we dont pass it "binding:vif_type" is coming as 'unbound' from neutron | 07:49 |
| irenab | I am not sure one can enforce if raven runs before or ater the scheduler | 07:49 |
| vikasc | irenab, and same is passed to cni driver | 07:50 |
| irenab | vikasc: will be back to you later, in the meeting | 07:50 |
| vikasc | irenab, and there it is not able to pick binding executable | 07:50 |
| vikasc | irenab, sure | 07:50 |
| vikasc | irenab, cant we wait in watcher and keep querying pod until node info is not available? | 08:03 |
| vikasc | irenab, i mean querying pod info from api server | 08:04 |
| irenab | I wonder how it works for us now … | 08:05 |
| irenab | vikasc: : jut for ‘hacking’ option, if you refer some arbitrary host, does it work? | 08:06 |
| vikasc | irenab, i set it to same master node only for haking purpose.. after that getting this error on driver | 08:07 |
| vikasc | E0817 13:11:25.784556 25673 pod_workers.go:138] Error syncing pod f803828f-644d-11e6-94a6-08002778068c, skipping: failed to "SetupNetwork" for "alpha.nginx_default" with SetupNetworkError: "Failed to setup network for pod \" | 08:08 |
| vikasc | irenab, still debugging | 08:08 |
| irenab | I would expect it to fail on ovs agent | 08:08 |
| vikasc | irenab, i have one node only.. i will also try hard coding this node | 08:08 |
| irenab | any neurron l2 agent errors? | 08:10 |
| vikasc | irenab, thats what i am looking at .. could not find so far | 08:10 |
| irenab | then I am not sure its binding problem, at least not neutorn | 08:11 |
| vikasc | irenab, now that from raven port info is coming with correct binding type.. will debug on driver side where exactly its failing | 08:12 |
| *** garyloug has joined #openstack-kuryr | 08:28 | |
| openstackgerrit | Merged openstack/kuryr: Update info in the configuration file https://review.openstack.org/347591 | 08:30 |
| openstackgerrit | Merged openstack/kuryr-libnetwork: Fix URL in README.rst https://review.openstack.org/356258 | 08:45 |
| janonymous | Hi, While running kuryr tests i am getting error of 404 Not found from client .... same thing i am getting when i dont use sudo for docker command although when i unset DOCKER_HOST variable error is resolved on Cli but not with test cases ... | 08:47 |
| janonymous | vagrant@devstack:~/devstack$ docker network ls | 08:47 |
| janonymous | Error response from daemon: { | 08:47 |
| janonymous | "Err": "404: Not Found" | 08:47 |
| janonymous | } | 08:47 |
| janonymous | vagrant@devstack:~/devstack$ sudo docker network ls | 08:47 |
| janonymous | NETWORK ID NAME DRIVER SCOPE | 08:47 |
| janonymous | 6e96f6bb1388 bridge bridge local | 08:47 |
| janonymous | d1f88e2c8036 host host local | 08:47 |
| janonymous | 2e50c82d4f16 net2 kuryr local | 08:47 |
| janonymous | 22708717b6bc none null local | 08:47 |
| janonymous | 4ec0a54ed180 test_net1 kuryr local | 08:47 |
| janonymous | fc82333796ce test_net2 kuryr local | 08:47 |
| *** limao_ has quit IRC | 08:48 | |
| *** limao has joined #openstack-kuryr | 08:56 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr-libnetwork: Remove pyroute2 in kuryr-libnetwork requirement.txt https://review.openstack.org/356342 | 09:42 |
| *** openstack has joined #openstack-kuryr | 10:15 | |
| *** sdake_ has joined #openstack-kuryr | 10:16 | |
| *** sdake_ has quit IRC | 10:18 | |
| *** sdake_ has joined #openstack-kuryr | 10:19 | |
| *** sdake has quit IRC | 10:19 | |
| *** limao has quit IRC | 10:48 | |
| *** yamamoto has joined #openstack-kuryr | 10:50 | |
| vikasc | devvesa, ping | 10:53 |
| *** yamamoto has quit IRC | 10:56 | |
| *** yamamoto has joined #openstack-kuryr | 10:57 | |
| *** yamamoto has quit IRC | 11:02 | |
| *** yamamoto has joined #openstack-kuryr | 11:03 | |
| *** yamamoto has quit IRC | 11:07 | |
| *** yamamoto has joined #openstack-kuryr | 11:07 | |
| *** yamamoto has quit IRC | 11:09 | |
| *** yamamoto has joined #openstack-kuryr | 11:09 | |
| *** sdake_ is now known as sdake | 11:24 | |
| *** limao has joined #openstack-kuryr | 11:25 | |
| *** yamamoto has quit IRC | 11:40 | |
| *** yamamoto has joined #openstack-kuryr | 11:46 | |
| *** garyloug has quit IRC | 11:58 | |
| *** sdake_ has joined #openstack-kuryr | 12:01 | |
| *** tonanhngo has joined #openstack-kuryr | 12:04 | |
| *** tonanhngo has quit IRC | 12:05 | |
| *** sdake has quit IRC | 12:05 | |
| *** sdake_ is now known as sdake | 12:05 | |
| *** garyloug has joined #openstack-kuryr | 12:07 | |
| *** yamamoto has quit IRC | 12:15 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Set up MTU docker interface https://review.openstack.org/355712 | 12:26 |
| *** yamamoto has joined #openstack-kuryr | 12:28 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Set up MTU docker interface https://review.openstack.org/355712 | 12:28 |
| *** yamamoto has quit IRC | 12:32 | |
| devvesa | vikasc: pong | 12:39 |
| devvesa | (sorry, late) | 12:39 |
| vikasc | devvesa, hi | 12:39 |
| vikasc | devvesa, no worries :) | 12:40 |
| devvesa | :) what's up? | 12:40 |
| vikasc | devvesa, how do you guys debug driver :) | 12:40 |
| devvesa | I use remote-pdb | 12:40 |
| vikasc | devvesa, thanks for the pointer, will google about it | 12:40 |
| devvesa | kind of "from remote_pdb import RemotePdb; RemotePdb('127.0.0.1', 4444).set_trace()" | 12:41 |
| devvesa | Then process waits untils a tcp connection at port 4444 | 12:41 |
| vikasc | devvesa, so found reason for yesterdays issue.. while creating a port raven should set host also | 12:41 |
| devvesa | you can connect to it using telnet: "telnet foo-host 4444" | 12:41 |
| devvesa | what's the issue? | 12:42 |
| vikasc | devvesa, if 'binding:host_id' is not sent to neutron in port creation request it sets vif_type as unbound | 12:42 |
| vikasc | devvesa, and then in driver it fails to pick correct executable using 'unbound' | 12:43 |
| vikasc | devvesa, to workaround this , i am harcoding my node hostname for now | 12:43 |
| devvesa | Uhm, and that's a Raven issue? Never experienced it... I wonder why | 12:43 |
| devvesa | maybe irenab knows | 12:43 |
| *** pablochacin has joined #openstack-kuryr | 12:44 | |
| vikasc | devvesa, I talked to her | 12:44 |
| devvesa | What she says? | 12:44 |
| vikasc | devvesa, she says its not possible at the moment to set host id | 12:44 |
| irenab | devvesa: I think its becouse we use it with MidoNet plugin, Our port binding is different | 12:45 |
| vikasc | devvesa, because raven cant be guaranteed to run after schedular has figured out host | 12:45 |
| vikasc | devvesa, yes | 12:45 |
| devvesa | Oh, that means MidoNet plugin does not consume this attribute | 12:45 |
| irenab | for ML2 ihost_id t is needed to bind the port | 12:45 |
| devvesa | understood | 12:45 |
| devvesa | So I suggest you to force it by now, next week I'll start the patch submitting | 12:46 |
| devvesa | (i will be on PTO till Tuesday) | 12:46 |
| devvesa | And see if we can fix it upstream soon | 12:46 |
| vikasc | devvesa, thanks for pdb help | 12:46 |
| vikasc | devvesa, i still not able to launch pod | 12:46 |
| devvesa | What I suggest make sense to you? | 12:46 |
| irenab | vikasc: fornow, please just hack, we should figure out if some sort of wait till k8sscheduler populates the node info is possible | 12:47 |
| irenab | in the raven watcher | 12:47 |
| vikasc | irenab, i was also thinking on same lines | 12:47 |
| irenab | devvesa: do you think we can change raven naming while restructure the code? | 12:47 |
| vikasc | devvesa, yes, makes sense tome | 12:48 |
| vikasc | *to me | 12:48 |
| devvesa | irenab: Yes. What do you suggest | 12:48 |
| devvesa | I am very bad ad namings. | 12:48 |
| irenab | devvesa: I am terrible with names | 12:48 |
| devvesa | kuryr-k8s? | 12:48 |
| devvesa | it is just the service name at all | 12:48 |
| irenab | kuryr-k8s-controller? | 12:49 |
| irenab | kuryr-k8s-watcher | 12:49 |
| devvesa | It is ok to me as long as we don't have to answer again "what's Raven?" | 12:49 |
| irenab | just something more self explanatory | 12:49 |
| devvesa | I have to be AFK next hour. See you later! | 12:50 |
| devvesa | vikasc: any doubt, ask irenab :) | 12:50 |
| *** devvesa has quit IRC | 12:50 | |
| vikasc | irenab, anything appended with 'watcher' should be fine :) | 12:50 |
| irenab | vikasc: agree | 12:52 |
| irenab | vikasc: I will try to look into this scheduler/raven dependency. Really do not want us to add CNI to call neutron to set the host | 12:53 |
| apuimedo | I propose 'oko' | 12:54 |
| vikasc | apuimedo, sorry i dint get 'oko' | 12:54 |
| apuimedo | instead of raven | 12:54 |
| vikasc | ohh | 12:54 |
| vikasc | k | 12:54 |
| apuimedo | :P | 12:54 |
| vikasc | :) | 12:55 |
| pablochacin | apuimedo: just when I started thinking names couldn't go worse, you came in. | 12:55 |
| vikasc | :D | 12:55 |
| apuimedo | :D | 12:55 |
| *** yamamoto has joined #openstack-kuryr | 12:55 | |
| apuimedo | hey, I kept it easier | 12:55 |
| irenab | apuimedo: is good on none intuitive names :-) | 12:55 |
| apuimedo | I first though dalekohled | 12:55 |
| apuimedo | xD | 12:55 |
| *** tonanhngo has joined #openstack-kuryr | 12:59 | |
| apuimedo | ok, ok | 13:01 |
| apuimedo | I got it now | 13:01 |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Set up MTU docker interface https://review.openstack.org/355712 | 13:02 |
| apuimedo | irenab: vikasc: pablochacin: 'spotter' | 13:03 |
| irenab | apuimedo: nice, getting better | 13:14 |
| apuimedo | irenab: :'( | 13:14 |
| irenab | do we need prefix k8s/kuryr? | 13:14 |
| apuimedo | I was hoping for "that's it! You are not chosing crazy names anymore" | 13:14 |
| apuimedo | irenab: no. I don't think so | 13:15 |
| apuimedo | let me check | 13:15 |
| irenab | I like it | 13:15 |
| apuimedo | I only found https://github.com/bcrowell/spotter | 13:16 |
| apuimedo | and it's completely unrelated (finding math and science questions) | 13:16 |
| apuimedo | so I think it is safe to use | 13:16 |
| irenab | I vote +1 | 13:17 |
| apuimedo | oops http://www.spotter.com/ | 13:17 |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Add UT for binding https://review.openstack.org/356454 | 13:17 |
| irenab | hmm | 13:17 |
| irenab | need to drop for a while | 13:17 |
| apuimedo | ok | 13:18 |
| apuimedo | irenab: let's just call it kspotter | 13:19 |
| apuimedo | and be done with it | 13:19 |
| *** limao has quit IRC | 13:19 | |
| pablochacin | are we serious with the spotter name? | 13:21 |
| pablochacin | i meant, you, are you serious? | 13:21 |
| apuimedo | https://media.makeameme.org/created/YEAH-I-AM-n5trg3.jpg | 13:24 |
| apuimedo | pablochacin: ^^ | 13:24 |
| *** yamamoto has quit IRC | 13:29 | |
| *** yamamoto has joined #openstack-kuryr | 13:30 | |
| pablochacin | http://scifi.stackexchange.com/questions/2013/which-episode-is-the-double-facepalm-image-macro-from | 13:41 |
| pablochacin | apuimedo: ^^ | 13:41 |
| apuimedo | :-) | 13:42 |
| pablochacin | There should be a PEP for naming, seriously. | 13:42 |
| *** sdake has quit IRC | 13:50 | |
| vikasc | apuimedo, irenab pablochacin how about including 'ipam' in name | 13:56 |
| apuimedo | vikasc: I don't think so, it does many other things | 13:57 |
| vikasc | apuimedo, agree.. | 13:58 |
| *** huikang has joined #openstack-kuryr | 13:58 | |
| vikasc | apuimedo, i am wishing some name which is a bit more than just watcher or looker | 13:59 |
| vikasc | apuimedo, some name which suggest that it take action also after watching :) | 14:00 |
| * vikasc thinking | 14:00 | |
| *** garyloug has quit IRC | 14:01 | |
| apuimedo | sentinel? | 14:01 |
| *** sdake_ has joined #openstack-kuryr | 14:01 | |
| vikasc | apuimedo, how about "sniper" .. watcher and shooter | 14:15 |
| apuimedo | too bellicose for my taste | 14:15 |
| vikasc | :D | 14:15 |
| *** hongbin has joined #openstack-kuryr | 14:17 | |
| *** vikasc has quit IRC | 14:21 | |
| *** garyloug has joined #openstack-kuryr | 14:23 | |
| *** yamamoto has quit IRC | 14:27 | |
| *** yamamoto has joined #openstack-kuryr | 14:29 | |
| *** tonanhngo has quit IRC | 14:33 | |
| *** vikasc has joined #openstack-kuryr | 14:33 | |
| *** yamamoto has quit IRC | 14:34 | |
| *** limao has joined #openstack-kuryr | 14:37 | |
| *** limao_ has joined #openstack-kuryr | 14:39 | |
| *** limao has quit IRC | 14:41 | |
| *** diogogmt has joined #openstack-kuryr | 14:47 | |
| *** limao_ has quit IRC | 14:49 | |
| *** limao has joined #openstack-kuryr | 14:49 | |
| *** limao has quit IRC | 15:00 | |
| *** limao has joined #openstack-kuryr | 15:00 | |
| *** openstackgerrit has quit IRC | 15:03 | |
| *** openstackgerrit has joined #openstack-kuryr | 15:04 | |
| *** limao has quit IRC | 15:05 | |
| *** oanson has quit IRC | 15:17 | |
| *** huikang has quit IRC | 15:21 | |
| *** huikang has joined #openstack-kuryr | 15:21 | |
| *** huikang has quit IRC | 15:26 | |
| *** yamamoto has joined #openstack-kuryr | 15:32 | |
| *** yamamoto has quit IRC | 15:37 | |
| apuimedo | vikasc: ping | 15:46 |
| apuimedo | HenryG: We were discussing the other day about our kuryr->kuryr-lib dependency | 15:52 |
| apuimedo | and I wanted to ask your take, since you have a similar situation in neutron with neutron->neutron-lib | 15:53 |
| apuimedo | I saw that you keep neutron master pinned to a specific neutron-lib release | 15:55 |
| apuimedo | so if a new neutron change required additions to neutron-lib | 15:55 |
| apuimedo | the timeline would be | 15:55 |
| apuimedo | send patch to neutron-lib | 15:55 |
| apuimedo | request a new release | 15:55 |
| apuimedo | and in the patch to neutron, update the requirements line | 15:55 |
| apuimedo | HenryG: isn't that a very long process for those cases? | 15:56 |
| apuimedo | I suspect that neutron-lib is so mature and feature full that changes there are not that common | 15:56 |
| *** sdake_ is now known as sdake | 16:00 | |
| *** sdake is now known as sdake_ | 16:00 | |
| openstackgerrit | Merged openstack/kuryr-libnetwork: Fix hard-code listen ip "0.0.0.0" https://review.openstack.org/354563 | 16:10 |
| *** pablochacin has quit IRC | 16:15 | |
| *** shashank_hegde has joined #openstack-kuryr | 16:39 | |
| apuimedo | irenab: gsagie: can you r+2 w+1 https://review.openstack.org/#/c/345921/ | 16:42 |
| apuimedo | I'm trying to reduce the queue | 16:42 |
| openstackgerrit | Merged openstack/kuryr: Updated from global requirements https://review.openstack.org/348633 | 16:47 |
| *** huikang has joined #openstack-kuryr | 16:49 | |
| apuimedo | btw, irenab: there's already https://wiki.openstack.org/wiki/Watcher | 16:52 |
| apuimedo | so I'd rather not use watcher | 16:52 |
| *** garyloug has quit IRC | 16:56 | |
| *** shashank_hegde has quit IRC | 17:04 | |
| apuimedo | gsagie: irenab this one should be merged too https://review.openstack.org/#/c/354600/2 | 17:06 |
| *** yamamoto has joined #openstack-kuryr | 17:34 | |
| *** yamamoto has quit IRC | 17:38 | |
| *** shashank_hegde has joined #openstack-kuryr | 17:50 | |
| *** tonanhngo has joined #openstack-kuryr | 17:51 | |
| *** tonanhngo has quit IRC | 18:51 | |
| *** tonanhngo has joined #openstack-kuryr | 19:12 | |
| *** vikasc has quit IRC | 19:22 | |
| *** vikasc has joined #openstack-kuryr | 19:34 | |
| *** tonanhngo has quit IRC | 20:22 | |
| *** tonanhngo has joined #openstack-kuryr | 20:46 | |
| *** pablochacin has joined #openstack-kuryr | 20:54 | |
| *** huikang has quit IRC | 21:49 | |
| *** huikang has joined #openstack-kuryr | 21:50 | |
| *** huikang has quit IRC | 21:55 | |
| *** shashank_hegde has quit IRC | 22:01 | |
| *** sdake_ has quit IRC | 22:06 | |
| *** shashank_hegde has joined #openstack-kuryr | 22:11 | |
| *** huikang has joined #openstack-kuryr | 22:16 | |
| *** sdake has joined #openstack-kuryr | 22:21 | |
| *** huikang has quit IRC | 22:29 | |
| *** huikang has joined #openstack-kuryr | 22:29 | |
| *** yamamoto has joined #openstack-kuryr | 22:30 | |
| *** huikang has quit IRC | 22:34 | |
| *** sdake has quit IRC | 22:38 | |
| *** limao has joined #openstack-kuryr | 22:46 | |
| *** limao has quit IRC | 22:57 | |
| *** limao has joined #openstack-kuryr | 22:57 | |
| *** hongbin has quit IRC | 22:58 | |
| *** limao has quit IRC | 23:03 | |
| *** limao has joined #openstack-kuryr | 23:04 | |
| *** limao has quit IRC | 23:15 | |
| *** limao has joined #openstack-kuryr | 23:15 | |
| *** pablochacin has quit IRC | 23:17 | |
| *** limao has quit IRC | 23:26 | |
| *** limao has joined #openstack-kuryr | 23:27 | |
| openstackgerrit | Liping Mao proposed openstack/kuryr: Add UT for binding https://review.openstack.org/356454 | 23:27 |
| *** shashank_hegde has quit IRC | 23:29 | |
| *** limao_ has joined #openstack-kuryr | 23:38 | |
| *** limao has quit IRC | 23:43 | |
| *** limao_ has quit IRC | 23:49 | |
| *** shashank_hegde has joined #openstack-kuryr | 23:54 | |
| *** diogogmt has quit IRC | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
