| *** pc_m has quit IRC | 00:36 | |
| *** yamamoto_ has quit IRC | 00:37 | |
| *** yamamoto has joined #openstack-containers | 00:39 | |
| *** pc_m has joined #openstack-containers | 00:44 | |
| *** mdnadeem has joined #openstack-containers | 00:46 | |
| *** tonanhngo_ has quit IRC | 00:48 | |
| *** tonanhngo has joined #openstack-containers | 00:48 | |
| *** tonanhngo has quit IRC | 00:53 | |
| *** tonanhngo has joined #openstack-containers | 00:54 | |
| *** tonanhngo_ has joined #openstack-containers | 00:56 | |
| *** EricGonczer_ has quit IRC | 00:57 | |
| *** tonanhngo has quit IRC | 00:58 | |
| *** tonanhngo_ has quit IRC | 01:01 | |
| *** vijaykc4 has joined #openstack-containers | 01:01 | |
| *** dave-mccowan has joined #openstack-containers | 01:08 | |
| *** vijaykc4 has quit IRC | 01:12 | |
| *** adrian_otto has joined #openstack-containers | 01:24 | |
| *** ramishra has quit IRC | 01:39 | |
| *** adrian_otto has quit IRC | 02:04 | |
| *** ricolin has joined #openstack-containers | 02:17 | |
| *** diga has quit IRC | 02:18 | |
| *** vijaykc4 has joined #openstack-containers | 02:20 | |
| *** kbyrne has quit IRC | 02:20 | |
| *** kbyrne has joined #openstack-containers | 02:23 | |
| *** dave-mccowan has quit IRC | 02:27 | |
| *** hongbin has joined #openstack-containers | 02:28 | |
| *** ramishra has joined #openstack-containers | 02:30 | |
| *** chhavi has quit IRC | 02:33 | |
| *** EricGonczer_ has joined #openstack-containers | 02:46 | |
| *** hongbin has quit IRC | 02:48 | |
| *** hongbin has joined #openstack-containers | 02:49 | |
| *** hongbin has quit IRC | 02:52 | |
| *** hongbin has joined #openstack-containers | 02:53 | |
| *** dave-mccowan has joined #openstack-containers | 02:54 | |
| *** vijaykc4 has quit IRC | 03:13 | |
| *** vijaykc4 has joined #openstack-containers | 03:13 | |
| *** vijaykc4 has quit IRC | 03:14 | |
| *** vijaykc4 has joined #openstack-containers | 03:24 | |
| *** ramishra has quit IRC | 03:28 | |
| *** ramishra has joined #openstack-containers | 03:29 | |
| *** yamamoto has quit IRC | 03:36 | |
| *** tianqing has joined #openstack-containers | 03:37 | |
| *** vijaykc4 has quit IRC | 03:38 | |
| *** EricGonczer_ has quit IRC | 03:41 | |
| *** ramishra has quit IRC | 03:42 | |
| *** hongbin has quit IRC | 03:43 | |
| *** yamamoto has joined #openstack-containers | 03:47 | |
| *** dave-mccowan has quit IRC | 03:48 | |
| *** yamamoto has quit IRC | 03:48 | |
| *** yamamoto has joined #openstack-containers | 03:49 | |
| *** ramishra has joined #openstack-containers | 03:50 | |
| *** adrian_otto has joined #openstack-containers | 03:50 | |
| *** tianqing has quit IRC | 03:57 | |
| *** chhavi has joined #openstack-containers | 04:13 | |
| *** vijaykc4 has joined #openstack-containers | 04:13 | |
| *** ykarel|away has joined #openstack-containers | 04:18 | |
| *** adrian_otto has quit IRC | 04:26 | |
| *** trinaths has joined #openstack-containers | 04:35 | |
| *** adisky__ has joined #openstack-containers | 04:37 | |
| *** mvpnitesh has joined #openstack-containers | 04:42 | |
| *** vijaykc4 has quit IRC | 05:09 | |
| mnaser | strigazi https://review.openstack.org/#/c/491800/ merged | 05:14 |
|---|---|---|
| mnaser | in an hour or so mirrors should be sync'd! | 05:14 |
| mnaser | it's 1:14 am, i'll be able to push up something that uses the mirrors tomorrow once everything syncs | 05:15 |
| *** janki has joined #openstack-containers | 05:29 | |
| *** adrian_otto has joined #openstack-containers | 05:42 | |
| *** adrian_otto has quit IRC | 05:46 | |
| *** vijaykc4 has joined #openstack-containers | 05:47 | |
| *** adrian_otto has joined #openstack-containers | 05:49 | |
| strigazi | mnaser I saw it, we also removed some extra jobs https://review.openstack.org/#/c/492177/ | 05:52 |
| *** ykarel_ has joined #openstack-containers | 05:54 | |
| *** adrian_otto has quit IRC | 05:55 | |
| *** adrian_otto has joined #openstack-containers | 05:56 | |
| *** ykarel|away has quit IRC | 05:57 | |
| *** vijaykc4 has quit IRC | 05:59 | |
| *** mjura has joined #openstack-containers | 05:59 | |
| *** adrian_otto has quit IRC | 05:59 | |
| *** yamamoto has quit IRC | 06:01 | |
| *** adrian_otto has joined #openstack-containers | 06:01 | |
| *** adrian_otto has quit IRC | 06:02 | |
| *** yamamoto has joined #openstack-containers | 06:05 | |
| *** rcernin has joined #openstack-containers | 06:06 | |
| *** yasemin has quit IRC | 06:11 | |
| *** Oku_OS-away is now known as Oku_OS | 06:12 | |
| *** yasemin has joined #openstack-containers | 06:12 | |
| *** hishh has joined #openstack-containers | 06:16 | |
| *** mjura has quit IRC | 06:21 | |
| openstackgerrit | Anandprakash proposed openstack/magnum master: Can anybody merge it? https://review.openstack.org/492389 | 06:25 |
| openstackgerrit | Spyros Trigazis (strigazi) proposed openstack/magnum master: tests: Use swarm-mode for api tests https://review.openstack.org/492390 | 06:28 |
| *** mjura has joined #openstack-containers | 06:34 | |
| *** mvpnitesh has quit IRC | 06:39 | |
| *** mvpnitesh has joined #openstack-containers | 06:41 | |
| *** ykarel_ is now known as ykarel | 06:48 | |
| openstackgerrit | Spyros Trigazis (strigazi) proposed openstack/magnum master: tests: Use swarm-mode for api tests https://review.openstack.org/492390 | 06:55 |
| strigazi | ykarel: ^^ What do you think? | 06:55 |
| strigazi | ykarel: swarm is fast. copylogs can be on demand. | 06:56 |
| strigazi | ykarel: swarm-mode is faster that legacy swarm. copylogs can be on demand. | 06:56 |
| strigazi | ykarel: swarm-mode is faster than legacy swarm. copylogs can be on demand. | 06:56 |
| strigazi | I finally got it | 06:56 |
| openstackgerrit | Spyros Trigazis (strigazi) proposed openstack/magnum master: Update to Fedora Atomic 26 https://review.openstack.org/491806 | 07:12 |
| openstackgerrit | Spyros Trigazis (strigazi) proposed openstack/magnum master: Set default iptables FORWARD policy to ACCEPT https://review.openstack.org/490478 | 07:13 |
| *** mdnadeem has quit IRC | 07:42 | |
| ykarel | strigazi, checking | 07:57 |
| *** mdnadeem has joined #openstack-containers | 08:01 | |
| *** mgoddard has joined #openstack-containers | 08:02 | |
| *** mdnadeem has quit IRC | 08:03 | |
| *** mdnadeem has joined #openstack-containers | 08:03 | |
| ykarel | strigazi, looks ok to me except, copy_logs default change to false | 08:16 |
| ykarel | strigazi, copy_logs can help in checking config/unwanted errors/warnings in logs while doing reviews | 08:17 |
| *** mvpnitesh has quit IRC | 08:20 | |
| *** dsariel has joined #openstack-containers | 08:20 | |
| *** jcjo has joined #openstack-containers | 08:24 | |
| *** yamamoto has quit IRC | 08:27 | |
| *** yamamoto has joined #openstack-containers | 08:28 | |
| *** jcjo has quit IRC | 08:30 | |
| strigazi | ykarel it does help but the kubernetes job times out on success because log copying takes to long. | 08:31 |
| *** yamamoto has quit IRC | 08:31 | |
| strigazi | ykarel it does help but the kubernetes job times out on success because log copying takes too long. | 08:31 |
| *** yamamoto has joined #openstack-containers | 08:32 | |
| ykarel | mmmm | 08:32 |
| *** armaan has quit IRC | 08:33 | |
| strigazi | We need a solution for that. With the current state of the ci (slow/very few nodes) I don't think we can afford it. | 08:33 |
| ykarel | strigazi, how frequent | 08:33 |
| *** armaan has joined #openstack-containers | 08:33 | |
| ykarel | strigazi, Ok then, later on we can reanable it if ci goes fast | 08:33 |
| strigazi | ykarel: that is the idea | 08:34 |
| ykarel | strigazi, BTW congrats for being PTL, just saw your nomination | 08:35 |
| strigazi | ykarel: thanks | 08:36 |
| *** ykarel is now known as ykarel|lunch | 08:46 | |
| *** vijaykc4 has joined #openstack-containers | 08:50 | |
| *** mgoddard has quit IRC | 08:55 | |
| *** mgoddard has joined #openstack-containers | 08:59 | |
| *** salmankhan has joined #openstack-containers | 09:12 | |
| *** jcjo has joined #openstack-containers | 09:13 | |
| *** jcjo has quit IRC | 09:17 | |
| *** ramishra has quit IRC | 09:18 | |
| *** ramishra has joined #openstack-containers | 09:20 | |
| *** mgoddard has quit IRC | 09:20 | |
| *** ykarel|lunch is now known as ykarel | 09:34 | |
| *** ricolin has quit IRC | 10:03 | |
| *** armaan has quit IRC | 10:07 | |
| *** armaan has joined #openstack-containers | 10:08 | |
| *** armaan has quit IRC | 10:10 | |
| *** armaan has joined #openstack-containers | 10:10 | |
| *** mgoddard has joined #openstack-containers | 10:13 | |
| *** mdnadeem has quit IRC | 10:16 | |
| *** yamamoto has quit IRC | 10:21 | |
| *** yamamoto has joined #openstack-containers | 10:25 | |
| *** armaan has quit IRC | 10:26 | |
| *** yamamoto has quit IRC | 10:28 | |
| *** yamamoto has joined #openstack-containers | 10:28 | |
| *** mdnadeem has joined #openstack-containers | 10:30 | |
| *** armaan has joined #openstack-containers | 10:35 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/magnum-ui master: Imported Translations from Zanata https://review.openstack.org/492087 | 10:37 |
| *** armaan has quit IRC | 10:39 | |
| *** yamamoto has quit IRC | 10:41 | |
| *** ykarel_ has joined #openstack-containers | 10:41 | |
| *** ykarel__ has joined #openstack-containers | 10:42 | |
| *** ykarel_ has quit IRC | 10:43 | |
| *** ykarel has quit IRC | 10:43 | |
| *** ykarel__ is now known as ykarel | 10:58 | |
| *** mdnadeem has quit IRC | 11:06 | |
| *** yamamoto has joined #openstack-containers | 11:09 | |
| *** armaan has joined #openstack-containers | 11:11 | |
| *** mdnadeem has joined #openstack-containers | 11:12 | |
| *** armaan has quit IRC | 11:16 | |
| *** mdnadeem has quit IRC | 11:19 | |
| *** yamamoto_ has joined #openstack-containers | 11:20 | |
| *** ykarel_ has joined #openstack-containers | 11:21 | |
| *** yamamoto has quit IRC | 11:24 | |
| *** ykarel has quit IRC | 11:25 | |
| *** vijaykc4 has quit IRC | 11:31 | |
| *** kzaitsev_ws has joined #openstack-containers | 11:36 | |
| *** dave-mccowan has joined #openstack-containers | 11:43 | |
| *** vijaykc4 has joined #openstack-containers | 11:43 | |
| *** mtanino has joined #openstack-containers | 11:47 | |
| strigazi | ykarel_: I think, I'll cut the Pike branch today to avaoid issue with requirements. We will just need to cherry-pick features until the release. Agreed? | 11:47 |
| strigazi | ykarel_: look with copy_logs disabled it run in 57 mins http://logs.openstack.org/78/490478/3/check/gate-functional-dsvm-magnum-k8s-ubuntu-xenial/cff48a1/logs/devstack-gate-post_test_hook.txt.gz#_2017-08-10_11_10_36_876 https://review.openstack.org/#/c/490478/ | 11:54 |
| strigazi | ykarel_: I'll fix my patch for the changes in api which uses swarm-mode | 11:54 |
| *** vijaykc4 has quit IRC | 11:54 | |
| strigazi | mnaser: here? | 11:56 |
| *** trinaths has quit IRC | 11:56 | |
| *** ykarel__ has joined #openstack-containers | 11:59 | |
| *** ykarel_ has quit IRC | 12:00 | |
| ykarel__ | strigazi, which requirements issues? | 12:00 |
| *** ykarel__ is now known as ykarel | 12:00 | |
| strigazi | http://lists.openstack.org/pipermail/openstack-dev/2017-August/120922.html | 12:01 |
| strigazi | the requiremetns team will open the requirements repo for pike and we will start receiving requirement updates from queens in the our master branch which will be pike | 12:02 |
| *** vijaykc4 has joined #openstack-containers | 12:02 | |
| strigazi | The consequence is that we need to push in two places if cut the branch | 12:03 |
| strigazi | And before cutting I propose to do the following change to ease our ci problem. Push in one patch these three chnages: https://review.openstack.org/#/c/491806/ https://review.openstack.org/#/c/490478/ https://review.openstack.org/#/c/492390/ | 12:04 |
| strigazi | The first one is to use swarm-mode in the api tests and disable copy_logs by default. the second two is to update the to fedora-26 from the fedora repos | 12:05 |
| strigazi | ykarel: thoughts? | 12:05 |
| strigazi | ykarel: https://etherpad.openstack.org/p/requirements_pike_thaw | 12:07 |
| *** zenirc369 has quit IRC | 12:10 | |
| *** ricolin has joined #openstack-containers | 12:12 | |
| ykarel | strigazi, checking, internet is slow on my side | 12:13 |
| strigazi | ykarel: ok | 12:13 |
| *** ykarel_ has joined #openstack-containers | 12:15 | |
| *** ykarel has quit IRC | 12:18 | |
| ykarel_ | strigazi, looks Ok we can go this way | 12:21 |
| ykarel_ | is it ok to go with fedora 26? | 12:22 |
| strigazi | ykarel_ testing locally | 12:22 |
| *** zenirc369 has joined #openstack-containers | 12:23 | |
| *** mtanino has quit IRC | 12:25 | |
| *** vijaykc4 has quit IRC | 12:32 | |
| *** vijaykc4 has joined #openstack-containers | 12:32 | |
| *** catintheroof has joined #openstack-containers | 12:45 | |
| *** EricGonczer_ has joined #openstack-containers | 12:47 | |
| mnaser | strigazi morning | 12:49 |
| mnaser | or evening for you :p | 12:49 |
| strigazi | mnaser 4th coffee | 12:50 |
| mnaser | strigazi dang, long day? https://review.openstack.org/#/c/490478/ this one came so close to merging | 12:51 |
| mnaser | i wonder if there's a problem with that api job | 12:51 |
| strigazi | mnaser fixing it. We'll use swarm-mode for the api tests. It was using the unstable swarm. | 12:52 |
| strigazi | mnaser As I said to ykarel_ I propose to do the following change to ease our ci problem. Push in one patch these three chnages: https://review.openstack.org/#/c/491806/ https://review.openstack.org/#/c/490478/ https://review.openstack.org/#/c/492390/ | 12:53 |
| *** mgoddard has quit IRC | 12:57 | |
| *** mdnadeem has joined #openstack-containers | 12:57 | |
| *** spiette has joined #openstack-containers | 12:59 | |
| *** EricGonczer_ has quit IRC | 13:02 | |
| strigazi | mnaser btw, the mirror didn't sync yet bu tthet is not a problem | 13:08 |
| strigazi | mnaser btw, the mirror didn't sync yet but that is not a problem | 13:08 |
| *** ramishra has quit IRC | 13:16 | |
| *** ramishra has joined #openstack-containers | 13:18 | |
| *** mgoddard has joined #openstack-containers | 13:28 | |
| *** mdnadeem has quit IRC | 13:30 | |
| *** vijaykc4 has quit IRC | 13:37 | |
| mnaser | strigazi oh nice, hmm i wonder why mirrors didnt sync yet let me follow up on that | 13:49 |
| mnaser | sorry afk i was in meeting zz | 13:49 |
| *** tianqing has joined #openstack-containers | 13:49 | |
| tianqing | i want to add one record when boot the k8s node? what i should do ? | 13:50 |
| tianqing | any one knows ? | 13:50 |
| strigazi | tianqing do what? I'm not sure what you mean | 13:51 |
| tianqing | ok.. | 13:52 |
| tianqing | i explain it.. | 13:52 |
| tianqing | I want to add some records to the coredns.... | 13:52 |
| tianqing | because k8s node needs to know the controller ... | 13:53 |
| *** marst_ has joined #openstack-containers | 13:54 | |
| strigazi | you want the node to be able to resolve the hostname of the master? | 13:54 |
| tianqing | not the k8s master, i want to it can resolve the openstack controller... | 13:55 |
| tianqing | Is it possible? | 13:55 |
| tianqing | i review doc, find that the coredns only used in cluster... | 13:55 |
| tianqing | I do not know how to let the coredns to resolve the openstack controller... | 13:56 |
| strigazi | tianqing depends on how you installed magnum. this is change that you need to do to the compute nodes. | 13:57 |
| tianqing | i do not understand... can you explain more about that? | 13:59 |
| tianqing | the magnum knows the controller, and can resolve the controller, i mean the k8s node. How the k8s node resolve the controller ip... | 14:01 |
| mnaser | strigazi the k8s cluster uses the DNS which you supply it when booting it | 14:01 |
| mnaser | so you should make sure that whatever DNS server you give it can resolve the controller | 14:01 |
| openstackgerrit | Mathieu Velten proposed openstack/magnum master: Launch kube-proxy as a system container https://review.openstack.org/488443 | 14:05 |
| openstackgerrit | Mathieu Velten proposed openstack/magnum master: Launch k8s scheduler & controller-manager as system containers https://review.openstack.org/487943 | 14:05 |
| openstackgerrit | Mathieu Velten proposed openstack/magnum master: Use atomic containers for kubelet & apiserver https://review.openstack.org/487357 | 14:05 |
| openstackgerrit | Mathieu Velten proposed openstack/magnum master: Add a kube_tag label to control the k8s containers to pull https://review.openstack.org/491740 | 14:05 |
| tianqing | yeah.@mnaser... | 14:06 |
| tianqing | i saw we add coredns three months ago... | 14:06 |
| mnaser | so whatever dns server you're suing it has to be able to resolve the controller | 14:06 |
| tianqing | so i think, can this be solved by coredns? | 14:07 |
| tianqing | for example, add the controller's record to coredns? | 14:07 |
| mnaser | coredns forwards to the local resolver | 14:07 |
| mnaser | im pretty sure you need to make sure whatever DNS you provide works | 14:08 |
| mnaser | strigazi looks like the job failed... "AttributeError: 'module' object has no attribute 'valid_swarm_mode_cluster_template'" ? | 14:08 |
| *** mjura has quit IRC | 14:09 | |
| tianqing | you mean that i should add the local dns resolver in to vm? the local dns resolver should be able to resolve controller... | 14:10 |
| openstackgerrit | Mohammed Naser proposed openstack/magnum master: tests: Use swarm-mode for api tests https://review.openstack.org/492390 | 14:10 |
| mnaser | strigazi found the issue and resolved, ill keep an eye out on it | 14:11 |
| tianqing | The original way i solve this problem is add one record into /etc/hosts.. | 14:11 |
| *** ramishra has quit IRC | 14:11 | |
| mnaser | thats a workaround you probably need to setup proper dns | 14:11 |
| tianqing | i find all master and minion will use network_service.sh, so i add one echo in it to add record to /etc/hosts... | 14:12 |
| tianqing | But this way is to ugly... | 14:12 |
| tianqing | too ugly... | 14:12 |
| *** ramishra has joined #openstack-containers | 14:12 | |
| strigazi | mnaser yeap, I tested locally | 14:13 |
| tianqing | our openstack system does not have dns resolver outside k8s.... | 14:14 |
| strigazi | mnaser this is new. The previous problem with api is that the api tests create a swarm cluster. with f26 etcd runs with the etcd user and the tls creds are root:root | 14:14 |
| mnaser | strigazi would running etcd in containers help? | 14:15 |
| * mnaser isn't too knowledgable of swarm | 14:15 | |
| tianqing | So, the only way is to setup one dns outside the k8s cluster? and when create cluster template, add that dns resolver? | 14:16 |
| strigazi | mnaser I just need to change the permissions. | 14:17 |
| strigazi | tianqing could work | 14:17 |
| tianqing | the coredns is used in the cluster, and can not resolve my problem? | 14:17 |
| mnaser | strigazi ok sounds good. i guess there's not much more we can do right now? | 14:18 |
| mnaser | other than hopefully things get through | 14:18 |
| strigazi | tianqing coredns is only for use inside the cluster | 14:18 |
| strigazi | mnaser (un)fortunately I'll push a fix all patch. | 14:19 |
| mnaser | yeah for f26 i think itll be a one unbreak-the-world fix and then we have the magnum-api swarm mode fixes too | 14:19 |
| strigazi | mnaser yes | 14:19 |
| strigazi | mnaser if we had PRs like in github would be better. | 14:20 |
| *** EricGonczer_ has joined #openstack-containers | 14:20 | |
| mnaser | strigazi ill try to get to the bottom of why mirrors are not appearing once infra folks are more available | 14:20 |
| mnaser | you can commit a chain of commits but usually each one will be tested | 14:21 |
| strigazi | mnaser it is impossible to merge many patches in one go right? | 14:21 |
| strigazi | mnaser AFAIK it is not | 14:21 |
| mnaser | you can but what will happen is that there will be a patch request for every single one | 14:21 |
| mnaser | if you make 1 commit, edit, commit again, git review will say "you are publishing 2 edits" | 14:21 |
| mnaser | and it will create to changesets | 14:21 |
| mnaser | s/to/two/ | 14:21 |
| strigazi | I could do that just for reviewers to see | 14:22 |
| mnaser | as much as this is nice though, it will slow us down a ton in the case of something like not having enough CI resources | 14:22 |
| mnaser | because every single one would have to go through testing | 14:22 |
| strigazi | mnaser I won't do it. | 14:22 |
| mnaser | so thats a lot more rechecks | 14:22 |
| mnaser | haha | 14:22 |
| tianqing | mnaster, help to merge this https://review.openstack.org/#/c/491663/2 | 14:22 |
| strigazi | tianqing I'll do my best | 14:23 |
| tianqing | go ahead... | 14:23 |
| *** EricGonc_ has joined #openstack-containers | 14:24 | |
| *** EricGonczer_ has quit IRC | 14:25 | |
| mnaser | strigazi looks like my mirroring script failed for some reason boo | 14:28 |
| *** tianqing has quit IRC | 14:32 | |
| mnaser | strigazi figured out the image problem and fix is in | 14:49 |
| *** yamamoto_ has quit IRC | 15:08 | |
| *** yamamoto has joined #openstack-containers | 15:10 | |
| strigazi | dims yt? | 15:10 |
| dims | strigazi : pong | 15:11 |
| *** yamamoto has quit IRC | 15:11 | |
| strigazi | dims I'm getting a strange error with kube 1.7.3 and cinder in the controller manager. http://paste.openstack.org/show/618064/ | 15:12 |
| strigazi | dims have you seen this before? | 15:13 |
| mnaser | strigazi i am going to guess that the cinder url is incorrect | 15:14 |
| mnaser | shouldnt it be http://137.138.120.8:<cinderport>/ | 15:15 |
| mnaser | its trying to access http://137.138.120.8/ instead | 15:15 |
| dims | nope. you may want to check logs for the 403 too | 15:15 |
| mnaser | openstack endpoint list for that cloud .. what do you see the cinder endpoint there as? | 15:15 |
| *** Oku_OS is now known as Oku_OS-away | 15:15 | |
| mnaser | dims missing port probably shows that error because it hits the default apache page (cause it's contacting the wrong place) | 15:16 |
| *** ramishra has quit IRC | 15:16 | |
| strigazi | in today's devstack the cinder endpoints don't have the port | 15:16 |
| mnaser | strigazi if they dont have a port, they must have a path like /cinder or /volume or something | 15:17 |
| strigazi | mnaser it has /volume | 15:17 |
| *** vijaykc4 has joined #openstack-containers | 15:18 | |
| mnaser | strigazi i see, and the openstack endpoint list shows it correctly configured at /volume ? | 15:18 |
| dims | strigazi : maybe something wrong in the catalog | 15:18 |
| mnaser | dims is it possible that version detection tries to make a request to the root, ignoring the path prefix? | 15:18 |
| mnaser | dims so when it wants to get the version of cinder currently running, instead of doing http://whatever/cinder/ .. it assumes that it never runs inside a path and always makes a request to the root | 15:19 |
| dims | y, there's some stuff for auto detection | 15:19 |
| dims | strigazi : try adding the following in your cloud conf | 15:19 |
| dims | [BlockStorage] | 15:19 |
| *** janki has quit IRC | 15:19 | |
| dims | bs-version=v2 | 15:19 |
| mnaser | dims fonud it | 15:20 |
| mnaser | https://github.com/gophercloud/gophercloud/blob/master/openstack/blockstorage/v1/apiversions/urls.go#L14-L18 | 15:20 |
| mnaser | blindly assumes that it is in / | 15:20 |
| dims | mnaser : cool | 15:21 |
| mnaser | i would assume that bs-version would short circuit the checking and it wouldn't run into that problem | 15:21 |
| strigazi | mnaser dims works with bs-version | 15:22 |
| strigazi | mnaser it looks like we won't revert it after all | 15:22 |
| strigazi | mnaser we will get crazy with this in the end | 15:23 |
| dims | strigazi : one of the reasons i am not touching that auto detect code | 15:23 |
| mnaser | dims im not strong in go but i think adopting this could fix it https://github.com/gophercloud/gophercloud/pull/423 | 15:23 |
| dims | see what they say there mnaser - "Each time I thought I've had this sorted out, I stumble upon another issue :)" | 15:24 |
| mnaser | i think the version detection is fundamentally wrong | 15:24 |
| mnaser | and i have backing for it in a second | 15:24 |
| mnaser | but afaik the openstack catalog will always contain versioned endpoints for cinder | 15:24 |
| strigazi | dims mnaser Since 423 is not merged we need to keel bs-version | 15:24 |
| mnaser | if i remember correctly, it's even documented somewhere | 15:24 |
| mnaser | strigazi dims http://specs.openstack.org/openstack/service-types-authority/ shows clearly that cinder has versioned names | 15:27 |
| mnaser | https://docs.openstack.org/cinder/latest/install/cinder-controller-install-rdo.html official install instructions say that you must add versioned endpoints | 15:27 |
| *** sdake_ is now known as sdake | 15:27 | |
| mnaser | https://github.com/openstack/puppet-cinder/blob/master/manifests/keystone/auth.pp puppet explicitly installs with versioned endpoints (so that includes tripleo based deployments) | 15:28 |
| mnaser | openstack ansible sets up versioned endpoints - https://github.com/openstack/openstack-ansible-os_cinder/blob/master/tasks/cinder_service_setup.yml | 15:28 |
| mnaser | dims imho that code should be removed. the best way to detect versions is through the catalog. if you find cinderv2, use it if you dont, look for cinder | 15:29 |
| *** vijaykc4 has quit IRC | 15:31 | |
| mnaser | strigazi https://review.openstack.org/#/c/489657/ bring this back? | 15:31 |
| mnaser | and drop https://review.openstack.org/#/c/491492/ ? | 15:31 |
| *** adrian_otto has joined #openstack-containers | 15:35 | |
| *** vijaykc4 has joined #openstack-containers | 15:37 | |
| strigazi | mnaser pike magnum will support 1.7+ we have it? | 15:38 |
| strigazi | mnaser pike magnum will support 1.7+ why have it? | 15:38 |
| mnaser | strigazi good point | 15:38 |
| strigazi | mnaser however we will need the revert to merge the fix using the f26 image :) it is a nightmare | 15:39 |
| mnaser | strigazi so we'll do k8s 1.7+ on f25? | 15:40 |
| strigazi | mnaser not the road is: go to f26 with 1.6.7 | 15:40 |
| strigazi | mnaser going to 1.6.7 requires remove the bs-version. | 15:40 |
| mnaser | yeah im starting to see the little dependency cycle we get into | 15:40 |
| strigazi | mnaser then we go to 1.7.x and we bring it back | 15:41 |
| mnaser | ok so | 15:41 |
| mnaser | lets do this | 15:41 |
| mnaser | https://review.openstack.org/#/c/492558/ | 15:41 |
| strigazi | mnaser we merge bs-version too fast | 15:41 |
| mnaser | just hope we can get this to merge | 15:41 |
| strigazi | mnaser we merged bs-version too fast | 15:41 |
| mnaser | yeah it shouldnt have been in there, or a test should have caught it | 15:41 |
| mnaser | its fine, it happens | 15:41 |
| mnaser | ok so now we're back to being a little stuck till that merges and we get our awesome images mirrored locally | 15:42 |
| strigazi | mnaser maybe I remove the reno? | 15:43 |
| mnaser | strigazi why not leave it and get 1.7 stuff merged? | 15:44 |
| strigazi | mnaser becasue with 1.7 in some case we need the bs-version. And anyway let's merge the reno when 1.7 are merged. | 15:45 |
| strigazi | mnaser becasue with 1.7 in some cases we need the bs-version. And anyway let's merge the reno when 1.7 stuff are merged. | 15:45 |
| strigazi | mnaser: what do you think? | 15:46 |
| mnaser | strigazi so continue with the revert till we get 1.7 in and then bring back the bs-version patch again? | 15:47 |
| strigazi | mnaser yes | 15:47 |
| mnaser | strigazi does we validate the k8s cluster at the gate by deploying something on it and checking that it deployed succesfully? | 15:47 |
| strigazi | mnaser no. we can't yet :( | 15:48 |
| mnaser | strigazi thats fine, so technically reverting this wont mean that the k8s gate will be broken forever | 15:48 |
| strigazi | mnaser: no | 15:48 |
| strigazi | mnaser: the gate won't be broken forever | 15:48 |
| openstackgerrit | Mohammed Naser proposed openstack/magnum master: Revert "Add Cinder-API-ver to k8s-cloud-provider config" https://review.openstack.org/491492 | 15:49 |
| mnaser | ^ :) | 15:49 |
| * mnaser has to go grab a quick bite, brb | 15:49 | |
| *** vijaykc4 has quit IRC | 16:03 | |
| *** vijaykc4 has joined #openstack-containers | 16:11 | |
| *** yamamoto has joined #openstack-containers | 16:12 | |
| *** adrian_otto has quit IRC | 16:12 | |
| *** mgoddard has quit IRC | 16:14 | |
| *** mgoddard has joined #openstack-containers | 16:15 | |
| *** kzaitsev_ws has quit IRC | 16:17 | |
| *** yamamoto has quit IRC | 16:17 | |
| openstackgerrit | Spyros Trigazis (strigazi) proposed openstack/magnum master: tests: Use swarm-mode for api tests https://review.openstack.org/492390 | 16:19 |
| *** vijaykc4 has quit IRC | 16:20 | |
| strigazi | mnaser: ykarel_ with this patch locally everything passes ^^ | 16:24 |
| ykarel_ | strigazi, Good | 16:25 |
| strigazi | and in the CI we won't have a problem puling the images | 16:25 |
| *** pcaruana has quit IRC | 16:25 | |
| *** vijaykc4 has joined #openstack-containers | 16:30 | |
| *** hongbin has joined #openstack-containers | 16:33 | |
| *** rcernin has quit IRC | 16:33 | |
| *** ykarel_ has quit IRC | 16:41 | |
| *** hishh has quit IRC | 16:42 | |
| *** mgoddard has quit IRC | 16:45 | |
| *** marst_ has quit IRC | 16:54 | |
| *** marst_ has joined #openstack-containers | 16:54 | |
| mnaser | strigazi awesome! | 17:01 |
| mnaser | strigazi i just requested for a promotion for the mirror patch | 17:03 |
| mnaser | hopefully infra lets us have it and then it'll jump to the top of the queue | 17:04 |
| strigazi | mnaser they do that? I think we are good for most of the cases now since we pull from fedora. Mirrored would be better of course. | 17:05 |
| mnaser | strigazi for special cases they can promote jobs to the top of the queue | 17:06 |
| mnaser | but i havent heard an answer yet i can ask again.. its been sitting in the queue for 2 hours :( | 17:06 |
| *** salmankhan has quit IRC | 17:07 | |
| strigazi | I ran the functional tests 10 times on my machine. One test failed once with: http://paste.openstack.org/show/618076/ i can't repro. I hope it was an outlier. After 10 hours in the office I think I can go home, See you tmr | 17:08 |
| mnaser | strigazi thanks for all your work today, have a good evening! :). see you later | 17:09 |
| mnaser | i'll hopefully get +1s everywhere for you when you're back today | 17:10 |
| mnaser | s/today/tomorrow | 17:10 |
| *** marst_ has quit IRC | 17:13 | |
| *** adrian_otto has joined #openstack-containers | 17:16 | |
| *** jberkus has quit IRC | 17:18 | |
| *** vijaykc4 has quit IRC | 17:19 | |
| *** adrian_otto has quit IRC | 17:20 | |
| *** ricolin has quit IRC | 17:25 | |
| *** vijaykc4 has joined #openstack-containers | 17:36 | |
| *** tonanhngo has joined #openstack-containers | 17:58 | |
| *** tonanhngo has quit IRC | 18:00 | |
| *** tonanhngo has joined #openstack-containers | 18:01 | |
| *** chhavi has quit IRC | 18:08 | |
| *** adrian_otto has joined #openstack-containers | 18:31 | |
| *** rcernin has joined #openstack-containers | 18:36 | |
| *** vijaykc4 has quit IRC | 18:50 | |
| *** rcernin has quit IRC | 18:59 | |
| *** adisky__ has quit IRC | 19:02 | |
| *** portdirect is now known as eteppete | 19:02 | |
| *** openstackgerrit has quit IRC | 19:03 | |
| *** eteppete is now known as portdirect | 19:03 | |
| *** adrian_otto has quit IRC | 19:05 | |
| *** salmankhan has joined #openstack-containers | 19:16 | |
| *** portdirect has quit IRC | 19:17 | |
| *** portdirect has joined #openstack-containers | 19:17 | |
| *** funzo has quit IRC | 20:14 | |
| *** openstackgerrit has joined #openstack-containers | 20:20 | |
| openstackgerrit | Mohammed Naser proposed openstack/magnum master: tests: Use swarm-mode for api tests https://review.openstack.org/492390 | 20:20 |
| *** sidx64 has joined #openstack-containers | 20:26 | |
| mnaser | strigazi mirrors merged, updated patch to use them :D | 20:31 |
| *** sdake is now known as k2so | 20:37 | |
| *** salmankhan has quit IRC | 20:57 | |
| *** funzo has joined #openstack-containers | 21:15 | |
| *** funzo has quit IRC | 21:20 | |
| *** tonanhngo has quit IRC | 21:32 | |
| *** tonanhngo has joined #openstack-containers | 21:32 | |
| *** tonanhngo has quit IRC | 21:40 | |
| *** jberkus has joined #openstack-containers | 21:41 | |
| *** salmankhan has joined #openstack-containers | 21:53 | |
| openstackgerrit | Mohammed Naser proposed openstack/magnum master: Revert "Add Cinder-API-ver to k8s-cloud-provider config" https://review.openstack.org/491492 | 21:55 |
| openstackgerrit | Mohammed Naser proposed openstack/magnum master: Remove /etc/ssl/certs in the controller manager pod https://review.openstack.org/488511 | 21:56 |
| *** sidx64 has quit IRC | 21:57 | |
| *** salmankhan has quit IRC | 22:00 | |
| *** tonanhngo has joined #openstack-containers | 22:06 | |
| *** tonanhngo has quit IRC | 22:11 | |
| *** tonanhngo has joined #openstack-containers | 22:12 | |
| *** tonanhngo has quit IRC | 22:17 | |
| *** openstackgerrit has quit IRC | 22:18 | |
| *** EricGonc_ has quit IRC | 22:33 | |
| *** catintheroof has quit IRC | 22:45 | |
| *** catintheroof has joined #openstack-containers | 22:45 | |
| *** catintheroof has quit IRC | 22:49 | |
| *** yamamoto has joined #openstack-containers | 22:51 | |
| *** funzo has joined #openstack-containers | 23:17 | |
| *** funzo has quit IRC | 23:21 | |
| *** hongbin has quit IRC | 23:21 | |
| *** adrian_otto has joined #openstack-containers | 23:23 | |
| *** yamamoto has quit IRC | 23:33 | |
| *** yamamoto has joined #openstack-containers | 23:40 | |
| *** kbyrne has quit IRC | 23:48 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!