| Tango|2 | gus: Hi Angus, so glad to see you pop up :) | 00:03 |
|---|---|---|
| *** david-lyle has quit IRC | 00:03 | |
| openstackgerrit | Bradley Jones proposed openstack/magnum-ui: Initial skeleton repo https://review.openstack.org/211750 | 00:04 |
| Tango|2 | gus: One of the essential blueprint we have is to get the external load balancer to work with the k8s cluster as deployed by Magnum | 00:04 |
| -openstackstatus- NOTICE: Zuul was restarted due to an error; events (such as approvals or new patchsets) since 23:01 UTC have been lost and affected changes will need to be rechecked | 00:04 | |
| Tango|2 | gus: From looking at your k8s openstack driver, all we need to do is to configure correctly since we already have all the networking | 00:05 |
| gus | Tango|2: yes, modulo possible bugs. | 00:06 |
| gus | I haven't actually used recently k8s versions, and I think some of the details haven't kept up with the recent k8s networking rearchs. | 00:07 |
| Tango|2 | gus: I was trying different ways but have not been able to get the load balancer part to come up correctly. | 00:07 |
| gus | Tango|2: yeah that ;) | 00:07 |
| gus | there's been a very recent PR or two that I was pulled in to review, so I think those are probably necessary. | 00:08 |
| gus | my real problem is that I don't have a current k8s + openstack test environment so I made a few recent changes blind, which is poor form. | 00:08 |
| *** mikal has joined #openstack-containers | 00:08 | |
| Tango|2 | gus: ok, I can look for those. But maybe what can really help is to synch up with you on how you set up your dev environment | 00:09 |
| *** sthillma_ has joined #openstack-containers | 00:09 | |
| Tango|2 | gus: oh ok, then would it help it I run in my environment based on your suggestion, and provide you the observation and logs? | 00:10 |
| *** sthillma has quit IRC | 00:11 | |
| *** sthillma_ is now known as sthillma | 00:11 | |
| *** yuanying-alt has joined #openstack-containers | 00:11 | |
| gus | Tango|2: yes, both of those things ;) I'll learn about Magnum today and work on getting my own test setup going. | 00:13 |
| *** yuanying-alt has quit IRC | 00:15 | |
| gus | Tango|2: https://github.com/GoogleCloudPlatform/kubernetes/pull/12203 | 00:15 |
| gus | Tango|2: https://github.com/GoogleCloudPlatform/kubernetes/pull/12262 | 00:15 |
| *** sthillma_ has joined #openstack-containers | 00:16 | |
| gus | Tango|2: https://github.com/GoogleCloudPlatform/kubernetes/pull/12288 | 00:16 |
| gus | I think that's the important ones.. | 00:16 |
| *** sthillma has quit IRC | 00:17 | |
| *** sthillma_ is now known as sthillma | 00:17 | |
| Tango|2 | gus: Let me email you my current config option to the kube-controller-manager and the parameters for the k8s driver. Can you take a look to see if they are correct, or if I am missing anything? | 00:18 |
| *** jruano has quit IRC | 00:23 | |
| gus | Tango|2: yep. -> gus@inodes.org | 00:24 |
| Tango|2 | gus: Thanks so much. I will check to see if it's possible for me to set up a dev environment in public so you can log in as well | 00:26 |
| Tango|2 | gus: So do you usually hang out on google-containers, in case I want to ping? | 00:27 |
| *** dane_leblanc has quit IRC | 00:27 | |
| gus | Tango|2: I'm not machine constrained, so assuming it's easy to set up my own test env, don't put devote too much effort to making yours available. | 00:28 |
| gus | Tango|2: I /joined right before posting, but my irc client will lurk here and catch references to "gus" from now on. | 00:28 |
| gus | (and I'm in Sydney, so an hour or so ago is the start of my workday) | 00:31 |
| Tango|2 | gus: If you want to set up Magnum, you would want a physical server. A VM will be too slow. You can find the instruction for devstack here: https://github.com/openstack/magnum/blob/master/doc/source/dev/dev-quickstart.rst | 00:31 |
| Tango|2 | gus: Basically, Magnum will create VM for the master and the nodes and install all the services | 00:32 |
| gus | oh right, since this all runs in the guest (see how much I've learned already! ;) | 00:32 |
| Tango|2 | gus: It sets up a private neutron subnet for the cluster, and a router to get to the public subnet | 00:33 |
| Tango|2 | gus: So the load balancer is the only way to publish a service from the k8s cluster to the external network | 00:34 |
| gus | ack, that seems sensible - and I can see why it's important to get external-lb working ;) | 00:35 |
| *** achanda_ has joined #openstack-containers | 00:35 | |
| Tango|2 | gus: Yep, we can't go to production without this | 00:35 |
| Tango|2 | gus: and we are trying to be somewhat production ready in Liberty | 00:36 |
| openstackgerrit | Bradley Jones proposed openstack/magnum-ui: Initial skeleton repo https://review.openstack.org/211750 | 00:36 |
| *** manjeets has quit IRC | 00:36 | |
| gus | Tango|2: and do you need any fixes to be in a k8s release, or can you build+deploy a k8s from git? | 00:36 |
| *** achanda has quit IRC | 00:38 | |
| *** Drago has quit IRC | 00:39 | |
| Tango|2 | gus: We would pick up any k8s release that's required. With the way we currently build our image, we have to pick up an actual release from k8s | 00:39 |
| *** achanda_ has quit IRC | 00:39 | |
| gus | Tango|2: ack. | 00:39 |
| Tango|2 | gus: I have built k8s from git before and copied over the binaries, but that would be for development only | 00:40 |
| openstackgerrit | Daneyon Hansen proposed openstack/magnum: Magnum Container Networking Spec https://review.openstack.org/204686 | 00:51 |
| *** david-lyle has joined #openstack-containers | 00:54 | |
| vilobhmm | sdake : ping | 00:56 |
| eghobo | gus: ping? | 00:58 |
| gus | eghobo: ? | 00:58 |
| eghobo | just curious, do you know anything about Rackspace driver for kub? | 00:59 |
| eghobo | https://github.com/GoogleCloudPlatform/kubernetes/pull/11904 | 00:59 |
| eghobo | I couldn't figure out how to test/debug these drivers, any suggestions? | 01:00 |
| *** harshs has joined #openstack-containers | 01:00 | |
| gus | eghobo: a bit, yes. It was basically a cut+paste of my openstack driver from some time ago (unfortunately - I wrote the openstack driver originally with the intention of better code sharing with a future rackspace variant - and I haven't gone back to refactor it into that original vision) | 01:01 |
| *** sthillma has quit IRC | 01:02 | |
| gus | eghobo: yeah, that doesn't seem right. In what way does it fail with --cloud_provider=rackspace ? | 01:03 |
| *** harshs_ has joined #openstack-containers | 01:03 | |
| eghobo | 'kube-apiserver.service: Main process exited, code=exited, status=2/INVALIDARGUMENT' | 01:04 |
| gus | that's it, no additional logs or errors? | 01:04 |
| eghobo | no | 01:05 |
| gus | basically: removing that flag completely disables any cloud-provider logic - so k8s won't be able to create load-balancers, find nodes automatically, etc. | 01:05 |
| *** harshs has quit IRC | 01:05 | |
| *** harshs_ is now known as harshs | 01:05 | |
| gus | So: if that makes the tests pass, then the tests aren't actually testing anything rackspace-specific ;) | 01:06 |
| eghobo | yes, I got it (I found driver later ;)) | 01:07 |
| *** gangil has quit IRC | 01:07 | |
| gus | So, without having investigated any further yet: I think that failure is uncovering a real problem - and removing the option is just papering over whatever the real problem is. | 01:07 |
| *** eliqiao has quit IRC | 01:07 | |
| eghobo | I agree, my problem was I was interested to investigate but don't know how | 01:08 |
| gus | (and we need more comprehensive testing in k8s for both openstack and rackspace drivers) | 01:08 |
| eghobo | could you point me to any test which I can use as example? | 01:09 |
| openstackgerrit | Bradley Jones proposed openstack/magnum-ui: Initial skeleton repo https://review.openstack.org/211750 | 01:10 |
| gus | eghobo: ah. I'd start by finding the (stderr?) logs of the failing test job. | 01:10 |
| gus | eghobo: or if it really does just exit immediately, then try to run kube-apiserver with that flag locally and just chase through where it goes wrong. | 01:11 |
| *** yuanying-alt has joined #openstack-containers | 01:12 | |
| *** adrian_otto has joined #openstack-containers | 01:13 | |
| eghobo | ok let me play with it ;) | 01:15 |
| *** yuanying-alt has quit IRC | 01:16 | |
| *** zhenguo has joined #openstack-containers | 01:16 | |
| openstackgerrit | Bradley Jones proposed openstack/magnum-ui: Initial skeleton repo https://review.openstack.org/211750 | 01:17 |
| *** Marga__ has joined #openstack-containers | 01:19 | |
| *** Marga_ has quit IRC | 01:20 | |
| *** moravec has joined #openstack-containers | 01:20 | |
| *** wanghua has joined #openstack-containers | 01:20 | |
| *** agireud has joined #openstack-containers | 01:24 | |
| *** suro-patz has quit IRC | 01:26 | |
| sdake | vilobhmm i'm alive | 01:26 |
| *** humble_ has joined #openstack-containers | 01:26 | |
| *** wanghua has quit IRC | 01:29 | |
| vilobhmm | sdake : is it still a good time to talk | 01:30 |
| vilobhmm | will hardly take 5 min | 01:30 |
| sdake | wound me | 01:30 |
| vilobhmm | https://blueprints.launchpad.net/magnum/+spec/objects-from-bay | 01:30 |
| vilobhmm | right now [1] https://github.com/openstack/magnum/blob/stable/kilo/magnum/objects/pod.py | 01:30 |
| vilobhmm | [2] https://github.com/openstack/magnum/blob/stable/kilo/magnum/objects/replicationcontroller.py | 01:30 |
| vilobhmm | [3] https://github.com/openstack/magnum/blob/stable/kilo/magnum/objects/service.py | 01:30 |
| vilobhmm | objects for pod, rc, service | 01:31 |
| vilobhmm | and we have similar for bay | 01:31 |
| sdake | yes i am familiar with that code | 01:31 |
| vilobhmm | so what is expected is for native clients like k8s | 01:31 |
| sdake | basically read operations need to happen via the rest endpoint | 01:31 |
| vilobhmm | whats is the current state | 01:31 |
| sdake | write operations need to happen via the rest endpoint instead of the conductor | 01:31 |
| sdake | read operations happen in the api service | 01:32 |
| sdake | the read operation needs to be sent to ht econductor via rpc | 01:32 |
| sdake | i'd focus on read first, then write | 01:32 |
| vilobhmm | ok | 01:32 |
| *** erkules has joined #openstack-containers | 01:32 | |
| sdake | the read operation needs to be sent to the conductor via rpc | 01:32 |
| sdake | the conductor has information about the endpoints and hsa a python native cient to access the ssytem with | 01:32 |
| sdake | one requirement is it has to work whether the folks are using native clients or not | 01:33 |
| sdake | so a mix of the two is expected | 01:33 |
| sdake | it is necessary in some way to convert the k8s id into a uuid | 01:33 |
| sdake | that will remain consistent for the run of the system | 01:34 |
| vilobhmm | consider pod object to understand better, api->object pod-> db for native client does nativeclient->object pod->db ? | 01:34 |
| sdake | the id of the bay/pod/rc | 01:34 |
| sdake | we never hit the db in this new model | 01:34 |
| sdake | reads happen entirely from the rest endpoint | 01:34 |
| sdake | the rest endpoint hsa a db in memory (etcd) | 01:35 |
| *** erkules_ has quit IRC | 01:35 | |
| vilobhmm | can you point to some code for example ? | 01:35 |
| sdake | i had some no idea where it went | 01:35 |
| vilobhmm | ok | 01:35 |
| *** hongbin has joined #openstack-containers | 01:35 | |
| sdake | i should have pushed up a review but never did | 01:35 |
| sdake | fail on my part | 01:35 |
| sdake | i was not very far along | 01:35 |
| vilobhmm | i thought one source of truth will be info stored in magnum db as table pod where as other will be stored in kubernates | 01:36 |
| sdake | i got as far as getting the reads into the conductor | 01:36 |
| sdake | nope, that is two sources of truth | 01:36 |
| sdake | one source of truth = all stored in k8s db | 01:36 |
| sdake | if we have to use the db to make it work i think that is ok | 01:36 |
| sdake | but I wasn't sure if this woudl be possible with native clients, since they dont store to the db | 01:37 |
| sdake | lets say python-magnumclient stores to the db via api->conductor | 01:37 |
| vilobhmm | ok | 01:37 |
| sdake | then native client reads from kubernetes | 01:37 |
| sdake | that would work | 01:37 |
| vilobhmm | agreed | 01:38 |
| sdake | lets say natie-client stores to etcd via api->conductor->rest endpoint->k8s api handler | 01:38 |
| sdake | python-magnumclient reads from where? the mysql db? There is nothing stored in it ;) | 01:38 |
| *** tobe_ has joined #openstack-containers | 01:39 | |
| *** jruano has joined #openstack-containers | 01:40 | |
| sdake | rather | 01:40 |
| sdake | native client stores to etcd via native client->rest endpoint -> etcd | 01:40 |
| vilobhmm | hmmm….makes sense..but can't even the python magnum client via the magnum api talk -> conductor->rest-endpoint-k8s api handler or that is not architecturally possible ? | 01:40 |
| sdake | yes that is exactly what needs to happen | 01:40 |
| sdake | that is not what happens today | 01:40 |
| sdake | atm reads happen as follows | 01:41 |
| *** moravec has quit IRC | 01:41 | |
| sdake | python-magnumclient->rest api->mysql | 01:41 |
| vilobhmm | ok | 01:41 |
| sdake | i think its a couple weeks of work if someone had complete 8 hours a day time to focus on it | 01:42 |
| vilobhmm | how they should happen is python-magnumclient->rest api->conductor->rest-endpoint-k8s-api handler ? | 01:42 |
| sdake | maybe less if familiar with architecture | 01:42 |
| sdake | yup | 01:42 |
| vilobhmm | sure…seems feasible | 01:42 |
| sdake | another requirement is that data returned by magnum should be the same whether its created by native client or python-magnumclient | 01:43 |
| vilobhmm | but in that case will we still need the pod, rc, service objects ? | 01:43 |
| sdake | for native client? | 01:43 |
| sdake | nope | 01:43 |
| vilobhmm | or the respective tables associated with them | 01:43 |
| vilobhmm | makes sense | 01:43 |
| vilobhmm | for native client | 01:43 |
| sdake | i am not clear on this point but i dont think so | 01:43 |
| sdake | the db tables have to go | 01:44 |
| vilobhmm | yup | 01:44 |
| sdake | for native or python client | 01:44 |
| vilobhmm | agree | 01:44 |
| vilobhmm | so i think i am clear now | 01:44 |
| vilobhmm | i will just illustrate what needs to be done | 01:44 |
| sdake | you could do write first if you want too | 01:44 |
| sdake | just dont do both at same time, think it will make it dramtically harder | 01:44 |
| vilobhmm | ok | 01:44 |
| sdake | but read more likely to be successful | 01:45 |
| sdake | with write if you remove writes to db, where to python-magnumclient reads come from | 01:45 |
| vilobhmm | 1. read path as of today is python-magnumclient->rest api->mysql should be python-magnumclient->rest api->conductor->rest-endpoint-k8s-api handler for both native and python clients | 01:45 |
| sdake | for native client should be native client->rest-endpoint-k8s-api | 01:45 |
| vilobhmm | sure…my bad | 01:45 |
| vilobhmm | 2. validate if #1 is achieved do we really need pod, service, rc objects and respective magnum db tables | 01:46 |
| sdake | we need to prserve the api | 01:46 |
| sdake | the main reason being people want to use heat with magnum | 01:46 |
| vilobhmm | what do you mean by preserve the api ? | 01:47 |
| vilobhmm | sure | 01:47 |
| sdake | pod api object needs to stay as ell as service/rc | 01:47 |
| vilobhmm | ok | 01:47 |
| *** harshs has quit IRC | 01:48 | |
| vilobhmm | api object can stay but the storage where they gets stored needs to be the same which happens to be done via k8s rest api ? | 01:48 |
| vilobhmm | sdake : ^^ | 01:48 |
| sdake | right | 01:48 |
| vilobhmm | rather than tight coupling with db | 01:48 |
| vilobhmm | got it | 01:49 |
| vilobhmm | now the next question is how do you know that the attirbutes of api obj are exactly similar to what is there is k8s ? | 01:49 |
| vilobhmm | hopefully they should be .... | 01:49 |
| sdake | we pass a manifest | 01:50 |
| vilobhmm | ok | 01:50 |
| sdake | one reason this work was so hard in the past is the damn api to k8s kept changing | 01:50 |
| vilobhmm | alrite | 01:50 |
| vilobhmm | ok | 01:50 |
| sdake | and we had no python client in magnum as well | 01:50 |
| vilobhmm | ok | 01:50 |
| sdake | both those problems are solved now | 01:50 |
| vilobhmm | :) | 01:50 |
| vilobhmm | i think i am clear with the problem and the approach to solve it | 01:51 |
| vilobhmm | thanks :) | 01:51 |
| vilobhmm | i will drop an email to the mailing list ; and will jot down the details as we discussed here | 01:51 |
| vilobhmm | so that everyeone is aware of the approach proposed | 01:52 |
| vilobhmm | and start work on implementing may be will start with read…will see | 01:52 |
| *** erkules has quit IRC | 01:52 | |
| sdake | the real challenge is going to be to get the data to look the same coming out of python client when created by native or python client | 01:52 |
| vilobhmm | sdake : does that sound fair to you ? | 01:52 |
| sdake | wfm make it happen i'll love you forever ;-) | 01:52 |
| vilobhmm | :) | 01:53 |
| vilobhmm | thank you again! | 01:53 |
| vilobhmm | sdake : ^^ | 01:53 |
| sdake | enjoy | 01:53 |
| *** erkules has joined #openstack-containers | 01:54 | |
| *** alejandrito has quit IRC | 01:55 | |
| *** vilobhmm has quit IRC | 02:05 | |
| *** eliqiao has joined #openstack-containers | 02:11 | |
| *** y_sawai has joined #openstack-containers | 02:14 | |
| *** y_sawai has quit IRC | 02:14 | |
| *** wanghua has joined #openstack-containers | 02:20 | |
| *** humble_ has quit IRC | 02:21 | |
| *** mathspanda has joined #openstack-containers | 02:21 | |
| *** jwcroppe has joined #openstack-containers | 02:23 | |
| *** Marga__ has quit IRC | 02:30 | |
| *** Marga_ has joined #openstack-containers | 02:30 | |
| *** david-lyle has quit IRC | 02:33 | |
| *** jruano has quit IRC | 02:34 | |
| *** adrian_otto has quit IRC | 02:34 | |
| *** adrian_otto has joined #openstack-containers | 02:35 | |
| *** funzo has joined #openstack-containers | 02:36 | |
| *** funzo has quit IRC | 02:40 | |
| *** agireud has quit IRC | 02:42 | |
| *** Tennyson has joined #openstack-containers | 02:45 | |
| *** eghobo has quit IRC | 02:47 | |
| *** agireud has joined #openstack-containers | 03:00 | |
| *** yuanying-alt has joined #openstack-containers | 03:01 | |
| *** adrian_otto has quit IRC | 03:02 | |
| *** humble_ has joined #openstack-containers | 03:02 | |
| *** vilobhmm has joined #openstack-containers | 03:02 | |
| *** jruano has joined #openstack-containers | 03:03 | |
| *** vilobhmm1 has joined #openstack-containers | 03:04 | |
| *** yuanying-alt has quit IRC | 03:05 | |
| *** wanghua has quit IRC | 03:05 | |
| *** vilobhmm has quit IRC | 03:07 | |
| *** banix has quit IRC | 03:08 | |
| *** agireud has quit IRC | 03:10 | |
| *** jwcroppe has quit IRC | 03:15 | |
| *** tobe_ has quit IRC | 03:15 | |
| *** tobe_ has joined #openstack-containers | 03:16 | |
| *** fawadkhaliq has joined #openstack-containers | 03:22 | |
| *** humble__ has joined #openstack-containers | 03:24 | |
| *** vahidh has quit IRC | 03:25 | |
| *** vahidh has joined #openstack-containers | 03:26 | |
| *** achanda has joined #openstack-containers | 03:27 | |
| *** david-lyle has joined #openstack-containers | 03:30 | |
| *** sdake_ has joined #openstack-containers | 03:35 | |
| *** sdake has quit IRC | 03:39 | |
| *** sdake_ is now known as sdake | 03:46 | |
| *** hongbin has quit IRC | 03:49 | |
| *** yuanying-alt has joined #openstack-containers | 04:01 | |
| *** jruano has quit IRC | 04:03 | |
| *** vilobhmm has joined #openstack-containers | 04:03 | |
| *** vilobhmm1 has quit IRC | 04:03 | |
| *** yuanying-alt has quit IRC | 04:06 | |
| *** manjeets has joined #openstack-containers | 04:06 | |
| *** funzo has joined #openstack-containers | 04:06 | |
| *** unicell has joined #openstack-containers | 04:06 | |
| manjeets | i added unit tests and when i run tox it says py27 commands failed | 04:09 |
| manjeets | any suggestion | 04:09 |
| *** funzo has quit IRC | 04:11 | |
| manjeets | is there any to way log tox output ? | 04:24 |
| yuanying | tox only said "py27 commands failed"? | 04:24 |
| *** madhuri has joined #openstack-containers | 04:34 | |
| *** sthillma has joined #openstack-containers | 04:36 | |
| madhuri | sdake: ping | 04:36 |
| sdake | madhuri shoot | 04:36 |
| madhuri | Want to discuss TLS? | 04:37 |
| sdake | do you need to discuss something with me? | 04:37 |
| madhuri | Ahh now I am implementing Barbican support to store certs in it | 04:39 |
| madhuri | sdake: But may be you can advise on this patch https://review.openstack.org/#/c/207324/ | 04:39 |
| *** sthillma_ has joined #openstack-containers | 04:39 | |
| sdake | you crit me with a wall of text on that patch review | 04:40 |
| sdake | 10pm need to sleep | 04:40 |
| sdake | didn't look at code in question but we don't want temp files for storing certs if at all possible to avoid | 04:41 |
| sdake | https://review.openstack.org/#/c/211817/1 | 04:42 |
| madhuri | Ok then we can discuss tomorrow | 04:42 |
| sdake | woops wrong link | 04:42 |
| *** sthillma has quit IRC | 04:42 | |
| *** sthillma_ is now known as sthillma | 04:42 | |
| sdake | http://paste.openstack.org/raw/412638/ | 04:42 |
| sdake | check that out madhuri | 04:42 |
| sdake | the part that is interesting is down at the bottom | 04:42 |
| madhuri | Sorry but I didn't get it | 04:44 |
| madhuri | sdake: Andrew wants to discuss something? | 04:45 |
| sdake | roger he does | 04:45 |
| sdake | madhuri my 3 children (magnum heat kolla) are all diverse ;-) | 04:45 |
| madhuri | :) cool | 04:46 |
| madhuri | Ok I will ping him | 04:46 |
| madhuri | apmelton: Hi | 04:47 |
| *** tobe_ has quit IRC | 04:53 | |
| yuanying | I also want to discuss about it | 04:54 |
| *** eghobo has joined #openstack-containers | 05:00 | |
| *** tobe_ has joined #openstack-containers | 05:04 | |
| *** sthillma has quit IRC | 05:12 | |
| *** yuanying-alt has joined #openstack-containers | 05:17 | |
| sdake | apmelton is east coast iirc probaby asleep | 05:18 |
| *** kmARC has joined #openstack-containers | 05:20 | |
| *** kmARC has quit IRC | 05:21 | |
| *** adrian_otto has joined #openstack-containers | 05:22 | |
| *** kmARC has joined #openstack-containers | 05:22 | |
| *** yuanying-alt has quit IRC | 05:23 | |
| *** adrian_otto has quit IRC | 05:26 | |
| *** vilobhmm has quit IRC | 05:32 | |
| *** fawadkhaliq has quit IRC | 05:38 | |
| *** eghobo_ has joined #openstack-containers | 05:42 | |
| *** eghobo has quit IRC | 05:45 | |
| *** humble__ has quit IRC | 05:48 | |
| *** diga_ has joined #openstack-containers | 05:50 | |
| *** madhuri has quit IRC | 05:52 | |
| *** shunliz_ has joined #openstack-containers | 05:52 | |
| *** eghobo has joined #openstack-containers | 05:53 | |
| *** eghobo_ has quit IRC | 05:54 | |
| *** madhuri has joined #openstack-containers | 05:54 | |
| *** fawadkhaliq has joined #openstack-containers | 05:59 | |
| *** zhenguo has quit IRC | 06:13 | |
| *** irenab has quit IRC | 06:14 | |
| *** yuanying-alt has joined #openstack-containers | 06:19 | |
| *** suro-patz has joined #openstack-containers | 06:19 | |
| *** yuanying-alt has quit IRC | 06:23 | |
| *** chandankumar has joined #openstack-containers | 06:23 | |
| *** suro-patz has quit IRC | 06:24 | |
| *** zz_dimtruck is now known as dimtruck | 06:26 | |
| *** unicell1 has joined #openstack-containers | 06:28 | |
| *** unicell has quit IRC | 06:28 | |
| *** suro-patz has joined #openstack-containers | 06:29 | |
| *** sdake_ has joined #openstack-containers | 06:35 | |
| *** sdake has quit IRC | 06:39 | |
| *** irenab has joined #openstack-containers | 06:43 | |
| *** suro-patz has quit IRC | 06:54 | |
| *** fangfenghua has joined #openstack-containers | 07:00 | |
| *** alex_klimov has joined #openstack-containers | 07:03 | |
| *** alex_klimov has quit IRC | 07:03 | |
| *** alex_klimov has joined #openstack-containers | 07:04 | |
| *** humble__ has joined #openstack-containers | 07:04 | |
| *** xek has quit IRC | 07:04 | |
| *** xek has joined #openstack-containers | 07:05 | |
| *** humble_ has quit IRC | 07:08 | |
| *** diga_ has quit IRC | 07:09 | |
| *** BhavyaM has joined #openstack-containers | 07:11 | |
| *** fawadkhaliq has quit IRC | 07:12 | |
| *** fawadk has joined #openstack-containers | 07:13 | |
| *** ig0r_ has joined #openstack-containers | 07:15 | |
| *** ishant has joined #openstack-containers | 07:16 | |
| *** eghobo has quit IRC | 07:18 | |
| *** achanda has quit IRC | 07:31 | |
| *** Marga__ has joined #openstack-containers | 07:33 | |
| *** liudong has joined #openstack-containers | 07:34 | |
| *** achanda has joined #openstack-containers | 07:36 | |
| *** zhenguo has joined #openstack-containers | 07:36 | |
| *** Marga_ has quit IRC | 07:37 | |
| *** ig0r__ has joined #openstack-containers | 07:39 | |
| *** ig0r_ has quit IRC | 07:41 | |
| *** tobe_ has quit IRC | 07:44 | |
| *** achanda has quit IRC | 07:46 | |
| *** dimtruck is now known as zz_dimtruck | 07:50 | |
| *** tobe_ has joined #openstack-containers | 07:50 | |
| openstackgerrit | Merged openstack/python-magnumclient: Updated from global requirements https://review.openstack.org/210924 | 07:58 |
| *** tobe_ has quit IRC | 08:00 | |
| *** sdake_ has quit IRC | 08:04 | |
| *** tobe_ has joined #openstack-containers | 08:05 | |
| *** yuanying-alt has joined #openstack-containers | 08:07 | |
| *** yuanying-alt has quit IRC | 08:12 | |
| *** ig0r__ has quit IRC | 08:13 | |
| *** diga_ has joined #openstack-containers | 08:14 | |
| *** ig0r_ has joined #openstack-containers | 08:17 | |
| *** Marga__ has quit IRC | 08:17 | |
| *** Marga_ has joined #openstack-containers | 08:18 | |
| *** ig0r_ has quit IRC | 08:19 | |
| *** manjeets has quit IRC | 08:23 | |
| *** humble_ has joined #openstack-containers | 08:26 | |
| *** humble__ has quit IRC | 08:27 | |
| *** humble_ has quit IRC | 08:31 | |
| *** humble_ has joined #openstack-containers | 08:31 | |
| Kennan2 | ping yuanying: | 08:31 |
| yuanying | hi | 08:31 |
| Kennan2 | for this https://review.openstack.org/#/c/205431/ I am ok with that, but I did not if ok to integrate to code now | 08:32 |
| *** humble_ has quit IRC | 08:32 | |
| Kennan2 | do you know who is TLS owner (expert) Sdake ? | 08:32 |
| *** humble_ has joined #openstack-containers | 08:33 | |
| yuanying | expert? | 08:33 |
| Kennan2 | I checked you mail in magnum | 08:34 |
| Kennan2 | asked about related implementaion for libiary | 08:34 |
| Kennan2 | does it impact this one ? | 08:34 |
| yuanying | No | 08:34 |
| yuanying | Only about anchor | 08:35 |
| yuanying | Barbican is necessary | 08:35 |
| Kennan2 | I think this one just add barbican client options, it is OK | 08:35 |
| yuanying | So, This patch is needed for us | 08:35 |
| Kennan2 | BTW, I checked your design guide, it also have another implementions option | 08:36 |
| Kennan2 | not use barbican | 08:36 |
| Kennan2 | is it in this liberty ? | 08:36 |
| yuanying | Actually, tls support has two part, one is for key storage, other is CA | 08:36 |
| yuanying | barbican is only implementation about key storage part (if we don't use db to store key) | 08:37 |
| yuanying | Anchor is needed for CA, and barbican also support CA | 08:38 |
| Kennan2 | ok. yuanying: | 08:47 |
| Kennan2 | more clear now | 08:47 |
| *** ddepaoli has joined #openstack-containers | 08:47 | |
| openstackgerrit | Merged openstack/magnum: Add barbicanclient support in Magnum. https://review.openstack.org/205431 | 09:03 |
| *** apuimedo has joined #openstack-containers | 09:06 | |
| *** kmARC has quit IRC | 09:28 | |
| *** Tennyson has quit IRC | 09:32 | |
| *** ishant_ has joined #openstack-containers | 09:34 | |
| *** ishant has quit IRC | 09:36 | |
| *** fangfenghua has quit IRC | 09:41 | |
| *** Kennan2 is now known as Kennan | 09:42 | |
| *** mathspanda has quit IRC | 09:46 | |
| Kennan | it seems some of our patched failed for jenkins issue | 09:49 |
| Kennan | NoValidHost: No valid host was found. There are not enough hosts available. | 09:49 |
| tcammann | Yeah its happenining a lot Kennan :( | 09:49 |
| Kennan | I asked it in infra irc, but not get response now | 09:49 |
| *** jwcroppe has joined #openstack-containers | 09:49 | |
| tcammann | Might need to bump memory on that system, couldn't get the root cause of it yet | 09:49 |
| Kennan | yes, it seems nova related issue, maybe infra run out of resouce | 09:50 |
| *** Marga__ has joined #openstack-containers | 09:50 | |
| tcammann | Aha! 2015-08-12 09:12:00.301 INFO nova.filters [req-201acad1-d2b3-4adc-aa0a-ae6237d7d743 admin admin] Filter DiskFilter returned 0 hosts | 09:51 |
| tcammann | ran out of disk space and failed in that filter it seems | 09:52 |
| *** Marga_ has quit IRC | 09:52 | |
| tcammann | This is from the nova scheduler log FYI | 09:52 |
| Kennan | ok, I checked nova condutor logs, not check that. :) | 09:53 |
| *** erkules has quit IRC | 09:54 | |
| *** erkules has joined #openstack-containers | 09:54 | |
| *** yuanying-alt has joined #openstack-containers | 09:56 | |
| *** yuanying-alt has quit IRC | 10:01 | |
| *** ig0r_ has joined #openstack-containers | 10:10 | |
| *** kmARC has joined #openstack-containers | 10:13 | |
| *** diga_ has quit IRC | 10:19 | |
| *** jkraj has joined #openstack-containers | 10:19 | |
| *** fangfenghua has joined #openstack-containers | 10:22 | |
| *** yuanying has quit IRC | 10:30 | |
| *** yuanying has joined #openstack-containers | 10:30 | |
| *** jwcroppe has quit IRC | 10:35 | |
| *** jwcroppe has joined #openstack-containers | 10:36 | |
| *** jwcroppe has quit IRC | 10:40 | |
| *** fawadk has quit IRC | 10:41 | |
| *** ig0r_ has quit IRC | 10:46 | |
| *** humble_ has quit IRC | 10:52 | |
| *** ishant_ has quit IRC | 10:54 | |
| *** ishant has joined #openstack-containers | 10:56 | |
| *** ishant has quit IRC | 10:56 | |
| *** tobe_ has quit IRC | 11:03 | |
| *** zhenguo has quit IRC | 11:03 | |
| *** tobe_ has joined #openstack-containers | 11:04 | |
| *** tobe_ has quit IRC | 11:09 | |
| ddepaoli | hi all! | 11:09 |
| ddepaoli | Does anyone ever install magnum on an environment different from devstack? | 11:10 |
| ddepaoli | I'm using juno installed with mirantis fuel 6.1 | 11:12 |
| *** juggler has quit IRC | 11:13 | |
| *** BhavyaM has quit IRC | 11:18 | |
| *** fangfenghua has quit IRC | 11:20 | |
| *** madhuri has quit IRC | 11:26 | |
| *** tobe_ has joined #openstack-containers | 11:33 | |
| *** julim has joined #openstack-containers | 11:38 | |
| *** tobe_ has quit IRC | 11:39 | |
| *** Marga__ has quit IRC | 11:40 | |
| *** Marga_ has joined #openstack-containers | 11:41 | |
| *** yuanying-alt has joined #openstack-containers | 11:45 | |
| *** yuanying-alt has quit IRC | 11:49 | |
| tcammann | ddepaoli: it should be possible | 12:06 |
| tcammann | Can use this guide to help https://github.com/openstack/magnum/blob/master/doc/source/dev/dev-manual-devstack.rst | 12:07 |
| tcammann | Kennan: did you get anywhere with fixing the functional tests? | 12:08 |
| ddepaoli | ok thanks tcammann | 12:08 |
| *** dave-mccowan has quit IRC | 12:12 | |
| *** shunliz_ has quit IRC | 12:15 | |
| *** funzo has joined #openstack-containers | 12:22 | |
| *** funzo has quit IRC | 12:27 | |
| *** jruano has joined #openstack-containers | 12:38 | |
| *** yuanying-alt has joined #openstack-containers | 12:46 | |
| *** jjlehr has joined #openstack-containers | 12:51 | |
| *** yuanying-alt has quit IRC | 12:53 | |
| *** irenab has quit IRC | 12:53 | |
| *** zz_dimtruck is now known as dimtruck | 12:55 | |
| *** mathspanda has joined #openstack-containers | 13:05 | |
| *** jwcroppe has joined #openstack-containers | 13:10 | |
| *** apuimedo has quit IRC | 13:16 | |
| *** jjlehr has quit IRC | 13:16 | |
| *** irenab has joined #openstack-containers | 13:21 | |
| *** jwcroppe has quit IRC | 13:23 | |
| *** jwcroppe has joined #openstack-containers | 13:23 | |
| *** jwcroppe has quit IRC | 13:27 | |
| *** funzo has joined #openstack-containers | 13:38 | |
| *** ctrath has joined #openstack-containers | 13:40 | |
| *** funzo has quit IRC | 13:43 | |
| *** dane_leblanc has joined #openstack-containers | 13:43 | |
| *** pradk has joined #openstack-containers | 13:44 | |
| *** pradk has quit IRC | 13:44 | |
| *** jjlehr has joined #openstack-containers | 13:51 | |
| *** ctrath has quit IRC | 13:58 | |
| openstackgerrit | Bradley Jones proposed openstack/magnum-ui: Add Bay Model Table View https://review.openstack.org/212039 | 14:00 |
| *** sdake_ has joined #openstack-containers | 14:00 | |
| *** dave-mccowan has joined #openstack-containers | 14:01 | |
| *** absubram has joined #openstack-containers | 14:04 | |
| *** muralia has joined #openstack-containers | 14:04 | |
| *** muralia has quit IRC | 14:16 | |
| *** muralia has joined #openstack-containers | 14:16 | |
| *** jruano has quit IRC | 14:17 | |
| *** banix has joined #openstack-containers | 14:17 | |
| *** mathspanda has quit IRC | 14:21 | |
| *** wznoinsk has joined #openstack-containers | 14:22 | |
| *** jjfreric has joined #openstack-containers | 14:31 | |
| *** jjfreric has left #openstack-containers | 14:31 | |
| *** jjfreric has joined #openstack-containers | 14:38 | |
| *** banix has quit IRC | 14:39 | |
| *** jjfreric has left #openstack-containers | 14:42 | |
| *** jjfreric_ has joined #openstack-containers | 14:43 | |
| *** EricGonczer_ has joined #openstack-containers | 14:46 | |
| *** adrian_otto has joined #openstack-containers | 14:47 | |
| *** vilobhmm has joined #openstack-containers | 14:54 | |
| *** chandankumar has left #openstack-containers | 14:58 | |
| *** adrian_otto has quit IRC | 15:00 | |
| *** banix has joined #openstack-containers | 15:03 | |
| *** gus has quit IRC | 15:06 | |
| *** gus has joined #openstack-containers | 15:08 | |
| *** funzo has joined #openstack-containers | 15:09 | |
| *** dave-mccowan has quit IRC | 15:09 | |
| *** ctrath has joined #openstack-containers | 15:09 | |
| *** jjfreric_ has quit IRC | 15:10 | |
| *** funzo has quit IRC | 15:13 | |
| *** jjfreric has joined #openstack-containers | 15:17 | |
| *** vilobhmm has quit IRC | 15:23 | |
| *** ddepaoli has quit IRC | 15:23 | |
| *** jjfreric has quit IRC | 15:25 | |
| *** BertrandN has joined #openstack-containers | 15:26 | |
| *** Drago has joined #openstack-containers | 15:30 | |
| *** Drago has quit IRC | 15:31 | |
| *** Drago has joined #openstack-containers | 15:31 | |
| *** suro-patz has joined #openstack-containers | 15:34 | |
| *** suro-patz1 has joined #openstack-containers | 15:36 | |
| *** zhenguo has joined #openstack-containers | 15:36 | |
| *** suro-patz has quit IRC | 15:38 | |
| *** agireud has joined #openstack-containers | 15:38 | |
| *** zhenguo has quit IRC | 15:42 | |
| *** banix has quit IRC | 15:48 | |
| *** suro-patz1 has quit IRC | 15:50 | |
| *** suro-patz has joined #openstack-containers | 15:51 | |
| *** hongbin has joined #openstack-containers | 15:54 | |
| *** jjfreric has joined #openstack-containers | 15:54 | |
| *** jjlehr has quit IRC | 15:54 | |
| *** suro-patz has quit IRC | 15:55 | |
| *** suro-patz has joined #openstack-containers | 15:55 | |
| *** jjfreric1 has joined #openstack-containers | 15:57 | |
| *** suro-patz has quit IRC | 15:59 | |
| *** jjfreric has quit IRC | 16:00 | |
| *** vilobhmm has joined #openstack-containers | 16:01 | |
| *** jjfreric has joined #openstack-containers | 16:02 | |
| *** funzo has joined #openstack-containers | 16:09 | |
| *** sankarshan is now known as sankarshan_away | 16:11 | |
| *** jkraj has quit IRC | 16:11 | |
| *** alex_klimov has quit IRC | 16:11 | |
| *** sdake_ is now known as sdake | 16:13 | |
| *** eghobo has joined #openstack-containers | 16:13 | |
| *** funzo has quit IRC | 16:14 | |
| *** jjfreric has quit IRC | 16:20 | |
| *** jjfreric has joined #openstack-containers | 16:21 | |
| *** jjfreric has quit IRC | 16:23 | |
| *** jjfreric has joined #openstack-containers | 16:23 | |
| *** unicell1 has quit IRC | 16:25 | |
| *** Marga_ has quit IRC | 16:30 | |
| *** Marga_ has joined #openstack-containers | 16:31 | |
| *** sthillma has joined #openstack-containers | 16:34 | |
| vilobhmm | sdake : ping | 16:36 |
| sdake | shoot | 16:36 |
| vilobhmm | I didn't get the etcd part we discussed yest | 16:36 |
| sdake | ya I gotta jet for a eye appointment | 16:37 |
| sdake | i hsould be back in 1-2 hours | 16:37 |
| vilobhmm | ok | 16:37 |
| *** dave-mccowan has joined #openstack-containers | 16:37 | |
| vilobhmm | later | 16:37 |
| *** diga_ has joined #openstack-containers | 16:40 | |
| *** pradk has joined #openstack-containers | 16:42 | |
| *** Marga_ has quit IRC | 16:42 | |
| *** Marga_ has joined #openstack-containers | 16:43 | |
| *** kmARC has quit IRC | 16:44 | |
| *** diga_ has quit IRC | 16:46 | |
| *** EricGonc_ has joined #openstack-containers | 16:47 | |
| *** unicell has joined #openstack-containers | 16:48 | |
| *** EricGonczer_ has quit IRC | 16:50 | |
| *** sankarshan_away is now known as sankarshan | 16:50 | |
| *** suro-patz has joined #openstack-containers | 16:54 | |
| *** sthillma has quit IRC | 16:58 | |
| *** mfalatic has quit IRC | 16:58 | |
| *** mfalatic has joined #openstack-containers | 16:59 | |
| *** sthillma has joined #openstack-containers | 16:59 | |
| *** eliqiao has quit IRC | 17:00 | |
| *** mfalatic has quit IRC | 17:02 | |
| *** mfalatic has joined #openstack-containers | 17:03 | |
| *** eliqiao has joined #openstack-containers | 17:03 | |
| *** dane_leblanc_ has joined #openstack-containers | 17:05 | |
| *** suro-patz has quit IRC | 17:05 | |
| *** dane_leblanc has quit IRC | 17:07 | |
| *** xek has quit IRC | 17:08 | |
| *** jjlehr has joined #openstack-containers | 17:11 | |
| *** Marga_ has quit IRC | 17:12 | |
| *** suro-patz has joined #openstack-containers | 17:13 | |
| *** achanda has joined #openstack-containers | 17:13 | |
| *** xek has joined #openstack-containers | 17:13 | |
| *** chandankumar has joined #openstack-containers | 17:16 | |
| *** belmoreira has joined #openstack-containers | 17:19 | |
| *** belmoreira has quit IRC | 17:21 | |
| *** dave-mcc_ has joined #openstack-containers | 17:28 | |
| *** dane_leblanc__ has joined #openstack-containers | 17:28 | |
| *** adrian_otto has joined #openstack-containers | 17:30 | |
| *** dane_leblanc_ has quit IRC | 17:30 | |
| *** jwcroppe has joined #openstack-containers | 17:30 | |
| *** unicell has left #openstack-containers | 17:31 | |
| *** dave-mccowan has quit IRC | 17:31 | |
| *** rpothier has joined #openstack-containers | 17:32 | |
| *** banix has joined #openstack-containers | 17:34 | |
| *** kmARC has joined #openstack-containers | 17:34 | |
| *** dane_leblanc__ has quit IRC | 17:36 | |
| *** rpothier has quit IRC | 17:36 | |
| *** dane_leblanc__ has joined #openstack-containers | 17:37 | |
| *** dflorea has joined #openstack-containers | 17:37 | |
| *** dave-mccowan has joined #openstack-containers | 17:38 | |
| *** dave-mcc_ has quit IRC | 17:40 | |
| *** rpothier has joined #openstack-containers | 17:41 | |
| *** harshs has joined #openstack-containers | 17:46 | |
| *** jjlehr has quit IRC | 17:46 | |
| *** absubram has quit IRC | 17:55 | |
| *** absubram has joined #openstack-containers | 17:57 | |
| *** Marga_ has joined #openstack-containers | 18:01 | |
| *** banix has quit IRC | 18:01 | |
| *** dflorea has quit IRC | 18:06 | |
| *** dflorea has joined #openstack-containers | 18:07 | |
| *** banix has joined #openstack-containers | 18:22 | |
| *** eghobo_ has joined #openstack-containers | 18:27 | |
| *** eghobo has quit IRC | 18:29 | |
| *** absubram has quit IRC | 18:35 | |
| *** absubram has joined #openstack-containers | 18:38 | |
| *** kmARC has quit IRC | 18:42 | |
| apmelton | tcammann: ping | 18:43 |
| *** EricGonc_ has quit IRC | 18:48 | |
| *** absubram has quit IRC | 18:48 | |
| *** EricGonczer_ has joined #openstack-containers | 18:51 | |
| *** absubram has joined #openstack-containers | 18:54 | |
| *** sthillma has quit IRC | 19:02 | |
| *** dflorea has quit IRC | 19:03 | |
| *** dflorea has joined #openstack-containers | 19:04 | |
| *** dflorea has quit IRC | 19:08 | |
| *** dflorea has joined #openstack-containers | 19:08 | |
| *** dflorea has quit IRC | 19:10 | |
| *** funzo has joined #openstack-containers | 19:25 | |
| *** funzo has quit IRC | 19:30 | |
| *** chandankumar has quit IRC | 19:36 | |
| *** kmARC has joined #openstack-containers | 19:39 | |
| *** EricGonc_ has joined #openstack-containers | 19:45 | |
| *** EricGonczer_ has quit IRC | 19:49 | |
| *** dflorea has joined #openstack-containers | 19:59 | |
| *** kmARC has quit IRC | 20:01 | |
| *** achanda has quit IRC | 20:15 | |
| *** jruano has joined #openstack-containers | 20:28 | |
| *** apuimedo has joined #openstack-containers | 20:30 | |
| *** jjfreric has quit IRC | 20:30 | |
| *** Marga_ has quit IRC | 20:39 | |
| *** Marga_ has joined #openstack-containers | 20:39 | |
| *** dflorea has quit IRC | 20:41 | |
| *** sthillma has joined #openstack-containers | 20:41 | |
| *** navneet has joined #openstack-containers | 20:56 | |
| *** dflorea has joined #openstack-containers | 20:58 | |
| *** jjfreric has joined #openstack-containers | 20:58 | |
| *** jruano has quit IRC | 21:00 | |
| *** achanda has joined #openstack-containers | 21:01 | |
| *** navneet has quit IRC | 21:03 | |
| *** jjfreric has quit IRC | 21:06 | |
| *** alex_klimov has joined #openstack-containers | 21:11 | |
| *** adrian_otto has quit IRC | 21:15 | |
| *** zul has quit IRC | 21:21 | |
| *** jjlehr has joined #openstack-containers | 21:25 | |
| *** rpothier has quit IRC | 21:25 | |
| *** dflorea has quit IRC | 21:33 | |
| *** dflorea has joined #openstack-containers | 21:34 | |
| *** zul has joined #openstack-containers | 21:35 | |
| *** gangil has joined #openstack-containers | 21:35 | |
| *** adrian_otto has joined #openstack-containers | 21:37 | |
| *** muralia has quit IRC | 21:37 | |
| *** muralia has joined #openstack-containers | 21:38 | |
| *** dflorea has quit IRC | 21:39 | |
| *** eghobo_ has quit IRC | 21:39 | |
| *** funzo has joined #openstack-containers | 21:41 | |
| vilobhmm | sdake : ping | 21:43 |
| *** fawadkhaliq has joined #openstack-containers | 21:43 | |
| *** eghobo has joined #openstack-containers | 21:45 | |
| *** funzo has quit IRC | 21:45 | |
| *** fawadkhaliq has quit IRC | 21:50 | |
| *** ctrath has quit IRC | 21:50 | |
| sdake | shoot vilobhmm | 21:54 |
| *** jjfreric1 has quit IRC | 21:55 | |
| vilobhmm | trying to understand better lets start with simple case of rc_show | 21:57 |
| vilobhmm | https://github.com/openstack/magnum/blob/master/magnum/conductor/api.py#L116 | 21:57 |
| vilobhmm | it doesn't have any implementation in https://github.com/openstack/magnum/blob/master/magnum/conductor/handlers/k8s_conductor.py | 21:57 |
| *** agireud has quit IRC | 21:57 | |
| vilobhmm | since the request gets satisfied from db | 21:58 |
| vilobhmm | which is righ | 21:58 |
| vilobhmm | magnumclient->restapi->db | 21:58 |
| vilobhmm | now for the way we want it "python-magnum client->rest api->conductor->rest-endpoint-k8s-api handler " | 21:58 |
| *** jjlehr has quit IRC | 21:59 | |
| *** dflorea has joined #openstack-containers | 21:59 | |
| vilobhmm | there will be a need not to fetch details using https://github.com/openstack/magnum/blob/master/magnum/conductor/api.py#L116 but use rest-endpoint-k8s-api handler | 21:59 |
| vilobhmm | at present there seems to be none of the api's exposed at https://github.com/openstack/magnum/blob/master/magnum/conductor/handlers/k8s_conductor.py which can be used…something like as its done here https://github.com/openstack/magnum/blob/master/magnum/conductor/handlers/k8s_conductor.py#L149 | 22:01 |
| vilobhmm | sdake : so the question is will need to introduce new api's from k8s to serve our purpose ? | 22:02 |
| vilobhmm | and if so I could not get a valid k8s api reference manual used by magnum | 22:02 |
| vilobhmm | sdake : ^^ | 22:08 |
| openstackgerrit | Merged openstack/magnum: Add a new field 'master_count' to bay in API https://review.openstack.org/207246 | 22:09 |
| *** pradk has quit IRC | 22:18 | |
| openstackgerrit | Merged openstack/magnum: Make simultaneous bay deletion workable https://review.openstack.org/210957 | 22:22 |
| *** dflorea has quit IRC | 22:26 | |
| *** achanda has quit IRC | 22:40 | |
| sdake | vilobhmm sorry got distracted | 22:41 |
| sdake | so we may or may not end up needing a way to obtain a certificate so the k8s endpoint can be communicated with | 22:41 |
| sdake | that could be api or via a show operation | 22:42 |
| sdake | i have preference to show persoanlly | 22:42 |
| sdake | let me look at the source links you provided - 5 minutes | 22:42 |
| *** devlaps has joined #openstack-containers | 22:44 | |
| vilobhmm | sdake : np | 22:45 |
| *** dane_leblanc__ has quit IRC | 22:45 | |
| sdake | vilobhmm if I read your quetion properly what you are basically saying is you need to get UUID from the k8s rest api endpoint | 22:47 |
| vilobhmm | sdake : yes | 22:47 |
| sdake | I think instead of getting it (because it provides no such thing) you should generat it | 22:47 |
| vilobhmm | ok | 22:47 |
| sdake | but it does provide some kind of unique id | 22:47 |
| sdake | I forget exactly what it is | 22:47 |
| sdake | if you login to the machine you can run kubectl get services | 22:48 |
| sdake | and see the unique ids it comes up with i htink | 22:48 |
| vilobhmm | ok | 22:48 |
| sdake | now, createreplicationctonroller, totally correct the objects it creates are not documented | 22:48 |
| sdake | that is because the source is autogenerated using swagger (GROAN) | 22:48 |
| vilobhmm | ok | 22:48 |
| sdake | you can look at the swagger objects to see what k8s exports | 22:48 |
| vilobhmm | the point i am coming to | 22:49 |
| vilobhmm | https://github.com/openstack/magnum/blob/master/magnum/conductor/api.py#L116 | 22:49 |
| sdake | possibly https://review.openstack.org/#/c/212194/1/jenkins/jobs/projects.yaml | 22:50 |
| vilobhmm | replace objects.ReplicationController.get_by_uuid(context, uuid) with the respective rest call in k8s which can be part of this module https://github.com/openstack/magnum/blob/master/magnum/common/pythonk8sclient/client/ApivbetaApi.py | 22:50 |
| sdake | rather possibly https://github.com/openstack/magnum/blob/master/magnum/common/pythonk8sclient/client/models/V1beta3_Pod.py#L43 | 22:50 |
| vilobhmm | need not be objects.ReplicationController.get_by_uuid(context, uuid) specifically but an example since it is getting details from db | 22:50 |
| sdake | yup but whatever the generation name is or whatever is unique in kubernetes (not sure on the unique id) needs to be used to create a consistent UUID | 22:51 |
| vilobhmm | does that make sense sdake ? Is the approach proper ^^ | 22:51 |
| sdake | is which approach proper? the one you and I just came up with? | 22:51 |
| sdake | if yes, then yes | 22:52 |
| vilobhmm | in short replace object->db interaction with object->conductor/handlers/k8s_conductor.py->magnum/common/pythonk8sclient/client/ApivbetaApi.py | 22:53 |
| *** dimtruck is now known as zz_dimtruck | 22:53 | |
| vilobhmm | object->k8s restapi | 22:53 |
| vilobhmm | you approach seems ok to me | 22:54 |
| vilobhmm | i would think more on it | 22:54 |
| vilobhmm | sdake : swagger seems to be useful thanks | 22:56 |
| *** dflorea has joined #openstack-containers | 22:56 | |
| sdake | right | 22:57 |
| sdake | that is the only way this can be done imo :) | 22:58 |
| sdake | it will slow things down a bit I think | 22:58 |
| sdake | but really creation and deletion of containers is not a super time sensitive operation | 22:58 |
| vilobhmm | ok cool | 22:58 |
| vilobhmm | will submit something atleast one of the api changes for read path and we can see how it goes… | 22:59 |
| vilobhmm | thanks sdake | 22:59 |
| *** hongbin has quit IRC | 23:07 | |
| *** EricGonczer_ has joined #openstack-containers | 23:07 | |
| *** EricGonc_ has quit IRC | 23:11 | |
| *** alex_klimov has quit IRC | 23:11 | |
| *** achanda has joined #openstack-containers | 23:16 | |
| *** gangil has quit IRC | 23:17 | |
| *** zul has quit IRC | 23:17 | |
| *** agireud has joined #openstack-containers | 23:17 | |
| *** banix has quit IRC | 23:36 | |
| *** adrian_otto has quit IRC | 23:45 | |
| *** hongbin has joined #openstack-containers | 23:47 | |
| *** Drago has quit IRC | 23:50 | |
| *** zz_dimtruck is now known as dimtruck | 23:50 | |
| *** Marga_ has quit IRC | 23:52 | |
| Kennan | tcammann: it passed now strange for jenkins | 23:52 |
| *** dflorea has quit IRC | 23:54 | |
| *** achanda has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!