| *** liyifeng has quit IRC | 00:03 | |
| *** salv-orlando has joined #openstack-kolla | 00:04 | |
| *** salv-orlando has quit IRC | 00:08 | |
| *** lamt has joined #openstack-kolla | 00:10 | |
| *** yingjun has joined #openstack-kolla | 00:12 | |
| *** dave-mccowan has quit IRC | 00:13 | |
| *** liyifeng has joined #openstack-kolla | 00:15 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 00:17 |
|---|---|---|
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: WIP fencing pod https://review.openstack.org/383922 | 00:22 |
| sbezverk | kfox1111: around? | 00:28 |
| kfox1111 | yeah. whats up? | 00:28 |
| *** rkrum has joined #openstack-kolla | 00:36 | |
| sbezverk | kfox1111: sorry got distructed some threads on ML | 00:38 |
| sbezverk | kfox1111: I got ceph locks clearing part, but I notced one issue | 00:39 |
| sbezverk | it seems when I have tight timers for node failure I noticed false positive from the node which was running but was busy atm since it was getting all those containers from failed node | 00:41 |
| sbezverk | I would appreciate if you could go through the latest PS and let me know what you think. | 00:42 |
| sbezverk | for timers I had to revert to 10/30 seconds 10 seconds keepalive from kubelet to confirm liveness and declaring failure by controller-manager after 30 seconds | 00:43 |
| *** dwalsh has joined #openstack-kolla | 00:43 | |
| *** tovin07 has joined #openstack-kolla | 00:45 | |
| kfox1111 | sbezverk: thats very normal for failover systems :/ | 00:46 |
| *** tovin07 has quit IRC | 00:46 | |
| kfox1111 | its unknowable if a system is dead or busy. | 00:46 |
| kfox1111 | so you have to trade off fast failover for more false positives. | 00:47 |
| *** zhubingbing has joined #openstack-kolla | 00:48 | |
| zhubingbing | hello | 00:48 |
| zhubingbing | guys | 00:48 |
| kfox1111 | hi | 00:48 |
| sbezverk | kfox1111: right so I backed off a little timers.. but I should tell you I am very impressed by kubernetes itself, I kill nodes today maybe 30-40 times and it always recovered. | 00:48 |
| zhubingbing | hi | 00:48 |
| kfox1111 | sbezverk: yeah. I haven't played with it in a production enough setting yet to know for sure how well it will behave. | 00:49 |
| kfox1111 | Its usually not an issue that a ha supporting system will fail to pickup a dead node. usually its the problem you've seen, where it is too aggressive at killing a live one. | 00:50 |
| kfox1111 | It can be especially bad when the load is such that one node gets slow, the cluster namager shoots it, moves the load to a second node, then it gets slow, then killed, then it moves on, etc. | 00:50 |
| kfox1111 | it can end up killing off all your nodes. which is worse then having some slow nodes. | 00:51 |
| sbezverk | kfox1111: :-) that could be a good movie title , kubernetes, the serial killer | 00:52 |
| * kfox1111 chuckles | 00:52 | |
| kfox1111 | yeah. :) | 00:52 |
| kfox1111 | its especially funny when you have a 2 node setup, with ipmi and | 00:53 |
| kfox1111 | each node decides the other is dead. then they shoot each other and both die. :) | 00:53 |
| kfox1111 | well, not so funny in the middle of the night when it happens and your on call.... :) | 00:53 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Choose node with largest seqno number for mariadb recovery https://review.openstack.org/384247 | 00:55 |
| kfox1111 | in our experience ha is good and bad. there are a lot of risks with it as well as benifits. so we use it where we really need it, and avoid it otherwise. as the extra complexity/risk can actually reduce stability rather then help. :/ | 00:55 |
| kfox1111 | the problem is, unless its constantly tested, then it often breaks when your not watching. then its worse then it not being there. | 00:55 |
| kfox1111 | thats why Netflix came up with ChaosMonkey. :) | 00:56 |
| kfox1111 | fasinating idea, if you havent looked at it. I'd highly recommend reading about it. :) | 00:56 |
| *** liyifeng has quit IRC | 00:58 | |
| sbezverk | kfox1111: :-) | 00:59 |
| sbezverk | there are a couple of places where even I see a need to optimize, but so far I could not find how to do it more efficiently, so if you think of something please let me know. | 01:00 |
| kfox1111 | k | 01:02 |
| openstackgerrit | zhubingbing proposed openstack/kolla: add grafana log in heka https://review.openstack.org/384250 | 01:03 |
| *** david-lyle_ has joined #openstack-kolla | 01:04 | |
| *** salv-orlando has joined #openstack-kolla | 01:05 | |
| *** Jeffrey4l has quit IRC | 01:06 | |
| *** david-lyle has quit IRC | 01:08 | |
| *** salv-orlando has quit IRC | 01:09 | |
| *** david-lyle_ is now known as david-lyle | 01:10 | |
| *** Jeffrey4l has joined #openstack-kolla | 01:12 | |
| *** duonghq has joined #openstack-kolla | 01:14 | |
| *** Jeffrey4l has quit IRC | 01:16 | |
| *** mandre has quit IRC | 01:26 | |
| *** mandre has joined #openstack-kolla | 01:26 | |
| *** dwalsh has quit IRC | 01:26 | |
| *** zhubingbing_ has joined #openstack-kolla | 01:28 | |
| *** janem has joined #openstack-kolla | 01:28 | |
| *** eaguilar has joined #openstack-kolla | 01:29 | |
| *** zhubingbing has quit IRC | 01:29 | |
| *** janem has quit IRC | 01:33 | |
| *** awidders_ has joined #openstack-kolla | 01:37 | |
| *** awiddersheim has quit IRC | 01:37 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 01:40 |
| *** liyifeng has joined #openstack-kolla | 02:00 | |
| *** tovin07 has joined #openstack-kolla | 02:03 | |
| *** salv-orlando has joined #openstack-kolla | 02:06 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 02:07 |
| *** caisan_ has left #openstack-kolla | 02:07 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: add influxdb log in heka https://review.openstack.org/384257 | 02:08 |
| *** ssurana has quit IRC | 02:09 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: add grafana log in heka https://review.openstack.org/384250 | 02:10 |
| *** salv-orlando has quit IRC | 02:10 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 02:20 |
| zhubingbing_ | ping sdake | 02:28 |
| zhubingbing_ | can u look it bug ? https://bugs.launchpad.net/kolla/+bug/1622119 | 02:28 |
| openstack | Launchpad bug 1622119 in kolla "rax.org gate is broken yet again" [High,Confirmed] | 02:28 |
| *** haplo37_ has quit IRC | 02:29 | |
| *** haplo37_ has joined #openstack-kolla | 02:31 | |
| *** zhubingbing_ has quit IRC | 02:36 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 02:39 |
| *** sdake has joined #openstack-kolla | 02:46 | |
| *** zhubingbing has joined #openstack-kolla | 02:46 | |
| *** zhubingbing has quit IRC | 02:49 | |
| sdake | yo peeps | 02:50 |
| duonghq | zhubingbing: how many physical nodes do you have in your hands? | 02:52 |
| *** coolsvap has joined #openstack-kolla | 02:56 | |
| openstackgerrit | Li Yingjun proposed openstack/kolla: Use rootwrap in sahara config https://review.openstack.org/384266 | 02:57 |
| *** zhubingbing has joined #openstack-kolla | 02:58 | |
| *** rhallisey has quit IRC | 03:06 | |
| *** david-lyle has quit IRC | 03:09 | |
| zhubingbing | coolsvap | 03:15 |
| *** eaguilar has quit IRC | 03:16 | |
| sdake | so tired | 03:26 |
| sdake | time for bed - nightfolks | 03:27 |
| coolsvap | wassup sdake | 03:27 |
| coolsvap | good night | 03:27 |
| sdake | coolsvap not much -tc results are in - came in mdidle of pack | 03:28 |
| sdake | which is typical ;) | 03:28 |
| coolsvap | sdake: yes i saw :) | 03:28 |
| sdake | coolsvap 131 1 spot spots is pretty solid turnout i think | 03:33 |
| sdake | so pretty happy with that | 03:33 |
| sdake | to serve ont he nc, need ot win another 40-50 people over ;) | 03:33 |
| sdake | thats alot of pps | 03:33 |
| coolsvap | yes | 03:34 |
| *** haplo37 has quit IRC | 03:36 | |
| *** g3ek has quit IRC | 03:36 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 03:37 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MITAKA BRANCH https://review.openstack.org/299182 | 03:37 |
| *** daneyon has joined #openstack-kolla | 03:38 | |
| *** haplo37 has joined #openstack-kolla | 03:39 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: integrate gnocchi with cloudkitty https://review.openstack.org/382916 | 03:40 |
| *** sdake has quit IRC | 03:41 | |
| *** dave-mccowan has joined #openstack-kolla | 03:42 | |
| *** g3ek has joined #openstack-kolla | 03:43 | |
| *** daneyon has quit IRC | 03:43 | |
| *** liyifeng has quit IRC | 03:47 | |
| openstackgerrit | Merged openstack/kolla: Use rootwrap in sahara config https://review.openstack.org/384266 | 03:48 |
| openstackgerrit | zhubingbing proposed openstack/kolla: add grafana log in heka https://review.openstack.org/384250 | 03:50 |
| openstackgerrit | Merged openstack/kolla: Clean up TODOs from live documentation https://review.openstack.org/374652 | 03:55 |
| *** zhubingbing has quit IRC | 03:56 | |
| *** mdnadeem_ has quit IRC | 03:56 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 03:59 |
| *** eaguilar has joined #openstack-kolla | 04:06 | |
| *** phuongnh has joined #openstack-kolla | 04:07 | |
| openstackgerrit | Li Yingjun proposed openstack/kolla: Make sahara api worker configurable https://review.openstack.org/384277 | 04:07 |
| *** zhubingbing has joined #openstack-kolla | 04:18 | |
| *** dmsimard is now known as dmsimard|pto | 04:23 | |
| *** haplo37_ has quit IRC | 04:53 | |
| *** haplo37_ has joined #openstack-kolla | 04:55 | |
| *** bmace has quit IRC | 04:56 | |
| *** bmace has joined #openstack-kolla | 04:56 | |
| *** salv-orlando has joined #openstack-kolla | 05:07 | |
| *** salv-orl_ has joined #openstack-kolla | 05:12 | |
| *** salv-orlando has quit IRC | 05:17 | |
| *** eaguilar has quit IRC | 05:39 | |
| *** eaguilar has joined #openstack-kolla | 05:43 | |
| *** salv-orl_ has quit IRC | 05:52 | |
| *** kuntumtarm has joined #openstack-kolla | 06:07 | |
| *** jimmygc has joined #openstack-kolla | 06:10 | |
| *** egonzalez90 has joined #openstack-kolla | 06:13 | |
| openstackgerrit | dupengfei proposed openstack/kolla: Support shared path for glance images and nova instances disk https://review.openstack.org/372420 | 06:17 |
| openstackgerrit | dupengfei proposed openstack/kolla: Fix keystone,database,memcache for k8s. https://review.openstack.org/372089 | 06:19 |
| *** Jeffrey4l has joined #openstack-kolla | 06:19 | |
| *** daneyon has joined #openstack-kolla | 06:21 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: [wip]Enable Gnocchi plugin in grafana https://review.openstack.org/384303 | 06:21 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Choose node with largest seqno number for mariadb recovery https://review.openstack.org/384247 | 06:24 |
| *** daneyon has quit IRC | 06:26 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Bump openstack-mitaka package version in Oracle image https://review.openstack.org/384304 | 06:26 |
| *** jimmygc has quit IRC | 06:28 | |
| *** yingjun_ has joined #openstack-kolla | 06:32 | |
| *** g3ek has quit IRC | 06:33 | |
| *** yingjun has quit IRC | 06:33 | |
| *** yingjun_ has quit IRC | 06:34 | |
| *** ssurana has joined #openstack-kolla | 06:34 | |
| *** duonghq_ has joined #openstack-kolla | 06:34 | |
| *** haplo37 has quit IRC | 06:34 | |
| *** kuntumtarm has quit IRC | 06:34 | |
| *** duonghq has quit IRC | 06:35 | |
| *** duonghq_ is now known as duonghq | 06:35 | |
| *** ssurana has quit IRC | 06:38 | |
| *** haplo37 has joined #openstack-kolla | 06:39 | |
| *** yingjun has joined #openstack-kolla | 06:39 | |
| *** g3ek has joined #openstack-kolla | 06:39 | |
| *** rkrum has quit IRC | 06:45 | |
| *** mewald has joined #openstack-kolla | 06:46 | |
| openstackgerrit | Merged openstack/kolla: integrate gnocchi with cloudkitty https://review.openstack.org/382916 | 06:47 |
| *** tonanhngo has quit IRC | 06:53 | |
| *** tonanhngo has joined #openstack-kolla | 07:01 | |
| *** tonanhngo has quit IRC | 07:03 | |
| *** salv-orlando has joined #openstack-kolla | 07:05 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: add grafana log in heka https://review.openstack.org/384250 | 07:07 |
| *** Serlex has joined #openstack-kolla | 07:11 | |
| *** b_bezak has joined #openstack-kolla | 07:12 | |
| *** matrohon has joined #openstack-kolla | 07:14 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: add influxdb log in heka https://review.openstack.org/384257 | 07:18 |
| openstackgerrit | zhubingbing proposed openstack/kolla: fix collectd dockerfile https://review.openstack.org/383997 | 07:21 |
| zhubingbing | Eduardo | 07:21 |
| zhubingbing | ping eduardo | 07:21 |
| openstackgerrit | zhubingbing proposed openstack/kolla: add influxdb log in heka https://review.openstack.org/384257 | 07:26 |
| openstackgerrit | zhubingbing proposed openstack/kolla: add grafana log in heka https://review.openstack.org/384250 | 07:27 |
| egonzalez90 | zhubingbing, are you calling me or other Eduardo? | 07:28 |
| zhubingbing | sorry | 07:28 |
| zhubingbing | ~~~ | 07:29 |
| *** ccesario has quit IRC | 07:31 | |
| *** ccesario has joined #openstack-kolla | 07:36 | |
| *** mdnadeem has joined #openstack-kolla | 07:44 | |
| *** mnasiadka has joined #openstack-kolla | 07:47 | |
| *** Jeffrey4l has quit IRC | 07:49 | |
| *** bjolo has quit IRC | 07:49 | |
| *** bjolo has joined #openstack-kolla | 07:50 | |
| *** Jeffrey4l has joined #openstack-kolla | 07:52 | |
| *** athomas has joined #openstack-kolla | 07:53 | |
| *** gfidente has joined #openstack-kolla | 07:54 | |
| *** gfidente has joined #openstack-kolla | 07:54 | |
| *** msimonin has joined #openstack-kolla | 07:54 | |
| *** athomas has joined #openstack-kolla | 07:54 | |
| *** Jeffrey4l has quit IRC | 07:55 | |
| *** mnasiadka has quit IRC | 07:55 | |
| *** mnasiadka has joined #openstack-kolla | 07:55 | |
| *** Jeffrey4l has joined #openstack-kolla | 07:55 | |
| *** strigazi1AFK is now known as strigazi | 07:56 | |
| *** bjolo_ has joined #openstack-kolla | 08:00 | |
| *** eaguilar has quit IRC | 08:10 | |
| *** eaguilar has joined #openstack-kolla | 08:17 | |
| *** rkrum has joined #openstack-kolla | 08:18 | |
| *** janem has joined #openstack-kolla | 08:21 | |
| duonghq | anybody test external networking with latest Kolla code? | 08:22 |
| *** rkrum has quit IRC | 08:23 | |
| *** berendt has joined #openstack-kolla | 08:39 | |
| *** salv-orl_ has joined #openstack-kolla | 08:43 | |
| *** sdake has joined #openstack-kolla | 08:43 | |
| sdake | hey peeps | 08:44 |
| *** williamc_ has joined #openstack-kolla | 08:45 | |
| *** salv-orlando has quit IRC | 08:46 | |
| *** sbezverk_ has joined #openstack-kolla | 08:46 | |
| *** haplo37- has joined #openstack-kolla | 08:48 | |
| *** quasar__ has joined #openstack-kolla | 08:49 | |
| *** strigazi_ has joined #openstack-kolla | 08:49 | |
| *** dotplus_ has joined #openstack-kolla | 08:49 | |
| *** cmceniry_ has joined #openstack-kolla | 08:49 | |
| *** rbergero1 has joined #openstack-kolla | 08:50 | |
| duonghq | sdake: still awake? | 08:50 |
| sdake | i wen to bed at 8pm | 08:51 |
| sdake | my son woke me up at midnight | 08:51 |
| sdake | so ya i'm awake | 08:51 |
| sdake | and wide awake | 08:51 |
| duonghq | you cannot fall in sleep again? | 08:52 |
| sdake | after sleepign 4 hours and being woken up? | 08:52 |
| sdake | no | 08:52 |
| *** dgonzalez_ has joined #openstack-kolla | 08:53 | |
| *** mnasiadka has quit IRC | 08:54 | |
| *** athomas has quit IRC | 08:54 | |
| *** haplo37_ has quit IRC | 08:54 | |
| *** tovin07 has quit IRC | 08:54 | |
| *** lamt has quit IRC | 08:54 | |
| *** jmccarthy has quit IRC | 08:54 | |
| *** williamcaban has quit IRC | 08:54 | |
| *** strigazi has quit IRC | 08:54 | |
| *** cmceniry has quit IRC | 08:54 | |
| *** rbergeron has quit IRC | 08:54 | |
| *** dotplus has quit IRC | 08:54 | |
| *** Mech422 has quit IRC | 08:54 | |
| *** quasar has quit IRC | 08:54 | |
| *** HyperJohnGraham has quit IRC | 08:54 | |
| *** sbezverk has quit IRC | 08:54 | |
| *** dgonzalez has quit IRC | 08:54 | |
| *** dgonzalez_ is now known as dgonzalez | 08:54 | |
| *** Mech422 has joined #openstack-kolla | 08:54 | |
| *** mnasiadka has joined #openstack-kolla | 08:55 | |
| *** HyperJohnGraham has joined #openstack-kolla | 08:55 | |
| *** hieulq has quit IRC | 08:57 | |
| duonghq | do I need any special configuration for allow VM connect to Internet? | 08:58 |
| duonghq | with Kolla on master branch | 08:58 |
| sdake | duonghq you need to configure oenstack properly | 08:58 |
| sdake | duonghq and if your in a vm, your vm needs to be setup properly | 08:58 |
| *** jmccarthy has joined #openstack-kolla | 08:59 | |
| *** athomas has joined #openstack-kolla | 09:01 | |
| *** hieulq has joined #openstack-kolla | 09:03 | |
| *** mnasiadka has quit IRC | 09:04 | |
| *** flaper87 has joined #openstack-kolla | 09:19 | |
| sdake | Jeffrey4l around? | 09:21 |
| Jeffrey4l | sdake, yep | 09:21 |
| sdake | Jeffrey4l let me push a patch - i want you to see it. | 09:21 |
| denaitre | hello guys | 09:21 |
| Jeffrey4l | sdake, OK> | 09:21 |
| denaitre | I'm trying a deploy inside ubuntu 16.04 with kolla 3.0.0 | 09:21 |
| denaitre | I'm having an issue with openvswitch-db-derver container | 09:22 |
| denaitre | the logs returns "/usr/local/bin/kolla_start: line 24: exec: start-ovsdb-server: not found" | 09:22 |
| sdake | denaitre pip show kolla plz | 09:23 |
| Jeffrey4l | denaitre, try to re-build the openvswitch images. | 09:23 |
| denaitre | I found someone else having the same issue but no answer https://ask.openstack.org/en/question/96303/kolla-all-in-one-failed-with-openvswitch-db-container/ | 09:23 |
| sdake | Jeffrey4l nah he has 2.0.2 pip isntalled | 09:23 |
| *** Guest10079 is now known as zigo | 09:23 | |
| sdake | denaitre dont' rebuild openvswitch images - isntead run pip show kolla | 09:24 |
| Jeffrey4l | hmm denaitre, sdake i think you are using kolla 3.0 to deploy images from kolla 2.0? | 09:24 |
| sdake | Jeffrey4l yup that is what is going on here | 09:24 |
| sdake | Jeffrey4l because kolla is pip installed | 09:24 |
| Jeffrey4l | got. | 09:24 |
| denaitre | sdake: http://pastebin.com/mQyMeREh | 09:24 |
| sdake | Jeffrey4l wow 271 patches in rc2.. lol | 09:25 |
| denaitre | the container tag is 3.0.0 | 09:25 |
| sdake | denaitre can you paste your globals.yml pls | 09:25 |
| sdake | denaitre our install guide never says pip install ./ for kolla - yo ushould be working out of a git tree atm | 09:26 |
| sdake | it used to say that though | 09:26 |
| sdake | about 6 eeks ago | 09:26 |
| sdake | but lets check out your globals.yml too | 09:26 |
| Jeffrey4l | sdake, we have good progress in rc2 :) | 09:27 |
| sdake | Jeffrey4l no doubt | 09:27 |
| Jeffrey4l | when we will create the rc2 tag? | 09:27 |
| openstackgerrit | Steven Dake proposed openstack/kolla: [WIP] Fix upgrades for rabbitmq https://review.openstack.org/384029 | 09:28 |
| openstackgerrit | Steven Dake proposed openstack/kolla: Make Kuryr upgrade not crater the upgrade https://review.openstack.org/384383 | 09:28 |
| sdake | Jeffrey4l the 12th at 23;45 utc | 09:28 |
| Jeffrey4l | cool | 09:28 |
| sdake | the wip patch Jeffrey4l | 09:29 |
| sdake | this is inc0's approach | 09:29 |
| denaitre | sdake: http://pastebin.com/J3H1gA0w | 09:29 |
| * Jeffrey4l is checking | 09:29 | |
| sdake | the problem ehre is ther eis no shell expansion of * | 09:29 |
| sdake | and I need shell expansin for chown to work for nova | 09:29 |
| sdake | denaitre ok - here is what i suggeset - not sure if it will fix problem | 09:30 |
| sdake | denaitre sudo pip uninstall kolla | 09:30 |
| sdake | cd | 09:30 |
| sdake | git clone http://github.com/openstack/kolla | 09:30 |
| sdake | cd koll | 09:30 |
| sdake | cd kolla | 09:30 |
| sdake | cd tools | 09:30 |
| sdake | ls | 09:30 |
| sdake | ./build -options.... for openvswitch | 09:30 |
| sdake | denaitre in other words, follow the developer workflow in the QSG | 09:31 |
| sdake | in other words DONT pip install kolla ;) | 09:31 |
| denaitre | ok thanks :) | 09:31 |
| sdake | Jeffrey4l inc0 spent a couple hours bikeshedding the design of this upgrade problem.... | 09:32 |
| sdake | Jeffrey4l and that is what he came up with after the 2 hrs we bikeshedded it yesterday | 09:32 |
| sdake | and the 2 hrs the day before | 09:32 |
| Jeffrey4l | sdake, what about ubuntu upgrade? | 09:33 |
| sdake | Jeffrey4l ya - no doubt | 09:33 |
| sdake | Jeffrey4l he wants to tackle ubuntu upgrade after this one is solved | 09:33 |
| sdake | the issue is ubuntu upgrade with this technique will not work without shell epansion | 09:33 |
| sdake | chown -R /var/lib/kolla/* doesn't work for example | 09:33 |
| Jeffrey4l | sdake, how about create rabbitmq uid and gid with 500, a more meaningful id. | 09:34 |
| sdake | Jeffrey4l yes of course - t hat will be done | 09:34 |
| sdake | Jeffrey4l i need shell expanstion o make inc0's technique work | 09:34 |
| Jeffrey4l | sdake, why need *? isn't chown -R /var/lib/kolla OK? | 09:35 |
| sdake | how do i get shell expansion there? | 09:35 |
| sdake | Jeffrey4l it may or may not be - not sure | 09:35 |
| sdake | I tried chown /var/log/rabbitmq* | 09:35 |
| sdake | andt hat didn't work | 09:35 |
| sdake | i am not sure if we need shell expansion or not | 09:35 |
| sdake | but i think we will | 09:35 |
| *** caowei has quit IRC | 09:36 | |
| *** shardy has joined #openstack-kolla | 09:37 | |
| *** caowei has joined #openstack-kolla | 09:37 | |
| sdake | sup shardy | 09:37 |
| shardy | Hey sdake | 09:38 |
| Jeffrey4l | a simple command is not enough. for example, when need handle the /var/lib/nova/ files. some need chown to nova: maybe. and others may need libvirt: | 09:38 |
| sdake | Jeffrey4l i hear ya | 09:38 |
| *** shardy has quit IRC | 09:38 | |
| sdake | Jeffrey4l nova case can be handled by multiple container starts | 09:38 |
| sdake | but seems clunky | 09:38 |
| *** shardy has joined #openstack-kolla | 09:39 | |
| sdake | Jeffrey4l inc0 also doesn't want a central mappingin base | 09:43 |
| sdake | Jeffrey4l but i do - so he can deal with that ;) | 09:43 |
| Jeffrey4l | sdake, but will he create all users in base? | 09:44 |
| sdake | thati sh wat i'm doing | 09:44 |
| sdake | he wants each uid created in leaf containers | 09:44 |
| sdake | such as nova-base | 09:44 |
| sdake | so uids maybe reused | 09:44 |
| Jeffrey4l | hmm. i'd like to create all users in base. then nova-libvirt no need based on nova-base image. | 09:45 |
| Jeffrey4l | but. | 09:45 |
| sdake | Jeffrey4l right- i agree | 09:45 |
| Jeffrey4l | how about the unified uid? | 09:45 |
| sdake | creating in base has many advantaged | 09:45 |
| *** dwalsh has joined #openstack-kolla | 09:45 | |
| sdake | mainly it creates a global passwd file | 09:45 |
| sdake | and grop file | 09:45 |
| sdake | group file | 09:46 |
| Jeffrey4l | will we still have unified uid? if so, that is no need to create all users in base, actually. | 09:46 |
| sdake | i want unified uid | 09:46 |
| sdake | inc0 ants uid reuse | 09:46 |
| Jeffrey4l | what inc0 think? | 09:46 |
| Jeffrey4l | unified uid != global uid ( create all users in base image ) | 09:47 |
| sdake | in this case, i'm writing the patch so what i think is what matters ;) | 09:47 |
| sdake | unified uid != uid reuse either | 09:47 |
| Jeffrey4l | create nova and glance with the same uid is bad. | 09:47 |
| sdake | agree | 09:47 |
| sdake | terrible ;) | 09:47 |
| sdake | this late in teh game this patch doesn't need to be perfect it needs to work | 09:48 |
| Jeffrey4l | if we create nova with uid 500 and glance with uid 501 all the time. then it is OK to not create all users in base image. | 09:48 |
| sdake | and uid reuse isn't my idea of wokring | 09:48 |
| sdake | Jeffrey4l i guess, but then where does the mapping go | 09:48 |
| sdake | in the leaf containers? | 09:48 |
| sdake | there are 40+ base iles... | 09:49 |
| sdake | yo uwant ot grep those for a fresh uid to use? | 09:49 |
| *** yingjun has quit IRC | 09:49 | |
| Jeffrey4l | sdake, it is OK in any container. root or leaf. | 09:49 |
| sdake | well root is where its going | 09:49 |
| Jeffrey4l | OK. | 09:50 |
| sdake | if somone wants to change that after newton wfm ;) | 09:50 |
| Jeffrey4l | agree. | 09:50 |
| sdake | re one command | 09:50 |
| sdake | lets focus on that part of the patch | 09:50 |
| Jeffrey4l | ok. leaved 3 comments. | 09:51 |
| sdake | Jeffrey4l do yo uthin kthis approach will work though? | 09:53 |
| Jeffrey4l | sdake, yes. | 09:54 |
| sdake | Jeffrey4l what about shell expasnion? | 09:55 |
| *** duonghq has quit IRC | 09:55 | |
| sdake | that s not needed? | 09:55 |
| Jeffrey4l | sdake, hmm if we can solve the nova case. Then we do not need care about the expansion. nova case is the hardest. shell expansion can not solve it. | 09:56 |
| sdake | how do I get etach:false? | 10:00 |
| sdake | juste add as a container option? | 10:00 |
| Jeffrey4l | sdake, check this https://github.com/openstack/kolla/blob/master/ansible/roles/nova/tasks/bootstrap_service.yml#L6 | 10:01 |
| Jeffrey4l | with `detach: False` the container will be remove automatically when the command run completed. | 10:02 |
| *** tonanhngo has joined #openstack-kolla | 10:02 | |
| sdake | got it hanks | 10:03 |
| *** tonanhngo has quit IRC | 10:03 | |
| otavio | hello | 10:04 |
| sdake | sup otavio | 10:05 |
| otavio | Just got in office | 10:05 |
| openstackgerrit | Steven Dake proposed openstack/kolla: [WIP] Fix upgrades for rabbitmq https://review.openstack.org/384029 | 10:09 |
| sdake | Jeffrey4l have another quick scan | 10:09 |
| sdake | Jeffrey4l btw whats going on with that other issue the deletion of routes problem | 10:10 |
| Jeffrey4l | cool. checking | 10:10 |
| Jeffrey4l | sdake, no progress :( | 10:10 |
| sdake | Jeffrey4l lasat i saw there was a simple reproducer outside of openstack | 10:12 |
| Jeffrey4l | sdake, i tried without success. | 10:12 |
| sdake | Jeffrey4l that said - we need it fixed for inside openstack without relying on a fix from docker | 10:12 |
| sdake | the reproducer? | 10:12 |
| Jeffrey4l | yes. | 10:12 |
| * sdake groans | 10:12 | |
| Jeffrey4l | rabbitmq:x:996:995:RabbitMQ messaging server:/var/lib/rabbitmq:/sbin/nologin | 10:13 |
| Jeffrey4l | i think we need add the home folder during creation. | 10:13 |
| sdake | Jeffrey4l possible | 10:13 |
| Jeffrey4l | sdake, i am thinking. how about extend our set_config feature , and set the proper permission for the volume during start? | 10:14 |
| sdake | Jeffrey4l if that works - seems better then chowns :) | 10:15 |
| *** zhubingbing has quit IRC | 10:15 | |
| sdake | Jeffrey4l what is epmd home dir | 10:15 |
| sdake | and i'll pus hthis new patch | 10:15 |
| Jeffrey4l | sdake, epmd:x:997:996:Erlang Port Mapper Daemon:/tmp:/sbin/nologin | 10:16 |
| openstackgerrit | Steven Dake proposed openstack/kolla: [WIP] Fix upgrades for rabbitmq https://review.openstack.org/384029 | 10:16 |
| Jeffrey4l | sdake, check this http://paste.openstack.org/show/585163/ | 10:17 |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Add upgrade tasks for Kibana https://review.openstack.org/384401 | 10:18 |
| sdake | Jeffrey4l interesting | 10:19 |
| sdake | Jeffrey4l give me minut eto think if there are any holes in this method | 10:19 |
| sdake | what about if the dir doesn't exist on deploy? | 10:19 |
| Jeffrey4l | ok. we can use regexp to check the path. | 10:19 |
| Jeffrey4l | if not exist, do nothing. | 10:19 |
| Jeffrey4l | we are only changing exist files. | 10:20 |
| bjolo | morning | 10:20 |
| sdake | ya i guess json is the place to put this too | 10:20 |
| sdake | i had originally thoguht not | 10:20 |
| Jeffrey4l | or we can add more parameter to it to control this. | 10:20 |
| sdake | not keen on more parameters if what we have can handle it | 10:21 |
| *** eaguilar has quit IRC | 10:21 | |
| sdake | the cool thing about hte json change is it carries with the containers | 10:22 |
| sdake | rather then the deployment engine | 10:22 |
| sdake | i don't know how to write such regex unfotuantnely | 10:22 |
| *** tonanhngo has joined #openstack-kolla | 10:22 | |
| Jeffrey4l | hmm. no better idea too. but we need it. | 10:23 |
| *** tonanhngo has quit IRC | 10:23 | |
| Jeffrey4l | sdake, or use this https://docs.python.org/2/library/fnmatch.html | 10:23 |
| Jeffrey4l | this is not regexp | 10:23 |
| sdake | i am familair with fnmatch | 10:24 |
| Jeffrey4l | cool | 10:24 |
| sdake | you mean add a wildcard chown somewhere in our json? | 10:24 |
| Jeffrey4l | sdake, or this http://paste.openstack.org/show/585164/ | 10:24 |
| Jeffrey4l | sdake, we need use python fnmatch to return all the files it match. | 10:25 |
| sdake | Jeffrey4l got what your saying | 10:25 |
| Jeffrey4l | the later one add a regexp parameter. it will be more flexible. | 10:25 |
| sdake | use regex to find the matching files | 10:25 |
| Jeffrey4l | correct | 10:26 |
| Jeffrey4l | it will be used for nova case. | 10:26 |
| sdake | ya - I get that - I dont knwo how to write the regex ;) | 10:26 |
| sdake | i am not a regex wizard | 10:26 |
| Jeffrey4l | there are three file in the nova folder : disk, console.log, libvirt.xml | 10:26 |
| Jeffrey4l | we can use regexp to match any one file | 10:27 |
| Jeffrey4l | or use fnmatch: path: /var/lib/nova/instance/*/disk | 10:27 |
| Jeffrey4l | hmm. i think fnmatch should be enough now. :) | 10:27 |
| sdake | ya fnmatch seems better | 10:28 |
| sdake | fnmatch gives shell expansion ;-) | 10:28 |
| Jeffrey4l | correct. | 10:28 |
| sdake | Jeffrey4l are you tlaking about calling set_config.py once? | 10:31 |
| sdake | or per son blob related to each container | 10:31 |
| Jeffrey4l | it will be called all the time when container is started. | 10:31 |
| sdake | i know when it is called now | 10:32 |
| *** eaguilar has joined #openstack-kolla | 10:32 | |
| sdake | i mean in the future with the approach suggested | 10:32 |
| sdake | your talking about integration across the various son files | 10:32 |
| Jeffrey4l | hmm. change all the json files? yes. | 10:32 |
| *** eaguilar has quit IRC | 10:33 | |
| Jeffrey4l | sdake, sorry, i think we need use glob rather than fnmatch. chown | 10:33 |
| janem | Hi all.. getting error with deploy of AIO "Fetcing Ceph keyrings - permission denied publickey" - is there a way to define a pubkey? | 10:33 |
| *** eaguilar has joined #openstack-kolla | 10:33 | |
| sdake | janem in your ~rroot dir, make sure uathorized_keys = your public ssh key | 10:34 |
| sdake | in .ssh dir | 10:34 |
| sdake | ~root dir that is | 10:34 |
| *** neilus has joined #openstack-kolla | 10:35 | |
| janem | sdake: ah.. yeah.. I have ssh setup for a different user.. thanks | 10:35 |
| *** caowei has quit IRC | 10:38 | |
| *** neilus has quit IRC | 10:40 | |
| *** salv-orl_ has quit IRC | 10:46 | |
| janem | sdake: actually, I dont get it. So I need to add the containers pubkey to host root autorized_keys? | 10:47 |
| *** Jeffrey4l has quit IRC | 10:47 | |
| janem | sdake: this is all-in-one install on host | 10:49 |
| *** phuongnh has quit IRC | 10:49 | |
| *** eaguilar has quit IRC | 10:51 | |
| *** Jeffrey4l has joined #openstack-kolla | 10:53 | |
| *** salv-orlando has joined #openstack-kolla | 10:54 | |
| *** haplo37- has quit IRC | 10:55 | |
| *** haplo37_ has joined #openstack-kolla | 10:56 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Add upgrade tasks for Mongodb https://review.openstack.org/384413 | 10:57 |
| janem | sdake: sorry, got it. Thanks | 11:00 |
| *** rhallisey has joined #openstack-kolla | 11:14 | |
| otavio | sdake: how is uid fix going? | 11:18 |
| otavio | Jeffrey4l: I replied your review | 11:22 |
| Jeffrey4l | otavio, which one? | 11:23 |
| otavio | vagrant | 11:23 |
| Jeffrey4l | got. | 11:23 |
| Jeffrey4l | ok. fair enough. +2 otavio | 11:25 |
| *** awidders_ has quit IRC | 11:25 | |
| *** awiddersheim has joined #openstack-kolla | 11:27 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Add netaddr to requirements file https://review.openstack.org/382892 | 11:27 |
| *** awiddersheim has quit IRC | 11:28 | |
| *** awiddersheim has joined #openstack-kolla | 11:28 | |
| *** mliima has joined #openstack-kolla | 11:35 | |
| *** mliima has quit IRC | 11:41 | |
| *** sdake has quit IRC | 11:41 | |
| *** mliima has joined #openstack-kolla | 11:42 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Fix the wrong setting for Grafana session https://review.openstack.org/384448 | 11:52 |
| mliima | morning guys | 11:56 |
| *** rodrigo_pereira has joined #openstack-kolla | 12:02 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Make number of keystone processes configurable https://review.openstack.org/384458 | 12:06 |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Make number of aodh-api processes configurable https://review.openstack.org/384459 | 12:08 |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Make number of gnocchi-api processes configurable https://review.openstack.org/384460 | 12:09 |
| *** liyifeng has joined #openstack-kolla | 12:10 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Make number of ceilometer-api processes configurable https://review.openstack.org/384461 | 12:10 |
| *** shardy is now known as shardy_mtg | 12:12 | |
| *** shardy_mtg is now known as shardy | 12:12 | |
| *** rbergero1 is now known as rbergeron | 12:13 | |
| *** rbergeron has joined #openstack-kolla | 12:13 | |
| *** tonanhngo has joined #openstack-kolla | 12:17 | |
| *** tonanhngo has quit IRC | 12:18 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools and Docs https://review.openstack.org/381041 | 12:23 |
| bjolo | Jeffrey4l, still up_ | 12:30 |
| bjolo | ? | 12:30 |
| *** JoseMello has joined #openstack-kolla | 12:30 | |
| *** g3ek has quit IRC | 12:32 | |
| *** haplo37 has quit IRC | 12:33 | |
| *** shardy is now known as shardy_mtg | 12:37 | |
| *** g3ek has joined #openstack-kolla | 12:38 | |
| *** haplo37 has joined #openstack-kolla | 12:39 | |
| *** inc0 has joined #openstack-kolla | 12:42 | |
| inc0 | good morning | 12:43 |
| *** shardy_mtg is now known as shardy | 12:49 | |
| mliima | morning inc0 | 12:51 |
| *** yingjun has joined #openstack-kolla | 12:57 | |
| Jeffrey4l | bjolo, yep | 12:57 |
| bjolo | Jeffrey4l, https://bugs.launchpad.net/kolla/+bug/1629246 | 12:59 |
| openstack | Launchpad bug 1623013 in kolla "duplicate for #1629246 keystone-fernet: rsync: Failed to exec ssh: No such file or directory (2)" [Critical,In progress] - Assigned to Christian Berendt (berendt) | 12:59 |
| *** mark-casey1 has joined #openstack-kolla | 12:59 | |
| bjolo | i dont understand the status? | 12:59 |
| bjolo | change abandoned? | 13:00 |
| Jeffrey4l | bjolo, could u check this fix? https://review.openstack.org/#/c/369418 | 13:00 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools, Tests, and Docs https://review.openstack.org/381041 | 13:00 |
| inc0 | kfox1111, you're going for a record as number of ps go right? | 13:01 |
| bjolo | will do | 13:01 |
| Jeffrey4l | bjolo, my ps is abandoned and we are using Christian Berendt'd PS. | 13:02 |
| Jeffrey4l | he fixed lots of thing. | 13:02 |
| *** shardy is now known as shardy_mtg | 13:03 | |
| sbezverk_ | inc0: kfox1111: there might be hickup with using ipmi. Since almost always ipmi ip address is not host's management ip address, getting it becomes somewhat problematic. | 13:03 |
| inc0 | duh, true | 13:04 |
| openstackgerrit | Merged openstack/kolla: Mark performance monitoring stack experimental https://review.openstack.org/384237 | 13:04 |
| sbezverk_ | kfox1111: are you using any specific model in your environment we could follow for now for calculating ipmi ip addresses? | 13:05 |
| sbezverk_ | or maybe some naming convention? | 13:05 |
| *** salv-orlando has quit IRC | 13:06 | |
| inc0 | sbezverk_, it's totally arbitrary | 13:06 |
| inc0 | but what we could reqiore is hostname resolvable to ipmi | 13:07 |
| inc0 | at least for now | 13:07 |
| inc0 | until we figure out better solution | 13:07 |
| inc0 | k8s uses IPs not hosntames right? | 13:07 |
| sbezverk_ | inc0: hostnames | 13:07 |
| inc0 | ok so what we can expect is that {{ hostname }}-ipmi will resolve to ipmi address | 13:08 |
| inc0 | I think thats fair for now | 13:08 |
| sbezverk_ | inc0: http://paste.openstack.org/show/585175/ | 13:08 |
| sbezverk_ | inc0: for now I guess that is the only way to move forward.. | 13:09 |
| sbezverk_ | last night I made ceph lock remval working, if you could take a look at PS that would be great | 13:09 |
| Jeffrey4l | inc0, about uid upgrade issue. do u like this idea? http://paste.openstack.org/show/585176/ | 13:09 |
| *** shardy_mtg is now known as shardy | 13:09 | |
| inc0 | also make ipmi fencing possible to disable in conf plz | 13:09 |
| Jeffrey4l | we can extend the set_config function to set the proper permission for folders. | 13:10 |
| sbezverk_ | inc0: that was done at the start, each part is configurable | 13:10 |
| inc0 | Jeffrey4l, and chowning in start? I think that's best solution so far | 13:10 |
| inc0 | we don't need forced UID across project any more | 13:11 |
| inc0 | yeah, I like that one | 13:11 |
| Jeffrey4l | inc0, the only change we need to do is: 1. extend set_config.py 2. add permission section in all config.json file | 13:11 |
| inc0 | we might need to ensure that group for certain multi-user packages is assigned correctly | 13:11 |
| Jeffrey4l | there is no need change the ansible tasks. | 13:12 |
| inc0 | like epmd user for rabbitmq | 13:12 |
| inc0 | but that's a 1 liner in dockerfile | 13:12 |
| Jeffrey4l | multi-user packages means? | 13:12 |
| inc0 | rabbitmq creates several users | 13:12 |
| Jeffrey4l | i think we event no need create the users with specified uid/gid. | 13:12 |
| *** Liuqing has joined #openstack-kolla | 13:12 | |
| inc0 | no, if for example epmd needs access to file, we can create groiup kolla in dockerfile | 13:13 |
| inc0 | and assign this user group kolla | 13:13 |
| inc0 | and modify permissions accordingly | 13:13 |
| inc0 | I like that. | 13:13 |
| Jeffrey4l | inc0, do not like the idea about epmd. why we need kolla group for files? ( except for logs ) | 13:14 |
| *** zhubingbing has joined #openstack-kolla | 13:14 | |
| inc0 | Jeffrey4l, I just don't know if user epmd needs any access to files | 13:15 |
| inc0 | but if he does, we can solve it with group, all I'm saying | 13:15 |
| *** dwalsh has quit IRC | 13:15 | |
| inc0 | Jeffrey4l, but hold on | 13:15 |
| inc0 | kolla_start will be run as rabbitmq user right? | 13:16 |
| inc0 | so it might not be able to change perms | 13:16 |
| *** mnasiadka has joined #openstack-kolla | 13:16 | |
| *** coolsvap has quit IRC | 13:16 | |
| Jeffrey4l | set_config is ran by `sudo -E set_config` | 13:20 |
| Jeffrey4l | inc0, https://github.com/openstack/kolla/blob/master/docker/base/start.sh#L14 | 13:20 |
| inc0 | so we'd need to change sudoers to ensure chown perms if it doesnt have it yet | 13:21 |
| inc0 | but that's cool | 13:21 |
| *** dotplus_ is now known as dotplus | 13:21 | |
| openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla: Version check for udev is too low https://review.openstack.org/384496 | 13:21 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools, Tests, and Docs https://review.openstack.org/381041 | 13:21 |
| Jeffrey4l | yes. the sudoers is already included. | 13:21 |
| Jeffrey4l | inc0, https://github.com/openstack/kolla/blob/master/docker/base/sudoers#L16 | 13:22 |
| inc0 | ok, cool, let's give it a try | 13:22 |
| inc0 | do you need any help? | 13:22 |
| Jeffrey4l | ya | 13:22 |
| inc0 | Jeffrey4l, btw mind checking out https://review.openstack.org/#/c/381293/ ? | 13:23 |
| inc0 | thanks guys, just push +A somebody please;) | 13:25 |
| Jeffrey4l | inc0, done. | 13:26 |
| inc0 | thank you | 13:26 |
| Jeffrey4l | inc0, i will push a PS later. | 13:26 |
| openstackgerrit | Merged openstack/kolla: Heka deprecation https://review.openstack.org/381293 | 13:28 |
| *** salv-orlando has joined #openstack-kolla | 13:29 | |
| inc0 | rhallisey, around? | 13:32 |
| inc0 | nvm, got it;) | 13:32 |
| *** tonanhngo has joined #openstack-kolla | 13:34 | |
| *** salv-orlando has quit IRC | 13:34 | |
| *** tonanhngo has quit IRC | 13:35 | |
| *** lamt has joined #openstack-kolla | 13:38 | |
| *** b_bezak has quit IRC | 13:41 | |
| *** sdake has joined #openstack-kolla | 13:49 | |
| rhallisey | inc0, ya | 13:51 |
| rhallisey | ok | 13:51 |
| *** haplo37_ has quit IRC | 13:51 | |
| *** bjolo_ has quit IRC | 13:53 | |
| *** haplo37_ has joined #openstack-kolla | 13:54 | |
| *** sdake has quit IRC | 13:54 | |
| *** sdake has joined #openstack-kolla | 13:55 | |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools, Tests, and Docs https://review.openstack.org/381041 | 13:55 |
| kfox1111 | I'm so ready to be done with this ps... :/ | 13:55 |
| sdake | kfox1111 116 patchstes - most i've seen :) | 13:55 |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Document how to remove a service https://review.openstack.org/380032 | 13:56 |
| rhallisey | kfox1111, you already broke your other record | 13:57 |
| rhallisey | o.o | 13:57 |
| kfox1111 | yup. | 13:58 |
| *** tonanhngo has joined #openstack-kolla | 14:00 | |
| sbezverk_ | good thing that infra do not charge fr each PS attempt ;-) would be a good business for them | 14:01 |
| sbezverk_ | kfox1111: how do you feel if for ipmi we combine hostname with ipmi suffics, otherwise there is no way to get ipmi address for the host? | 14:02 |
| *** huikang has joined #openstack-kolla | 14:02 | |
| kfox1111 | sbezverk_: probably fine. | 14:05 |
| sdake | sbezverk_ i think we have had about 20k commits tested by infra this cycle | 14:06 |
| sdake | sbezverk_ rather about 20k patches tested | 14:06 |
| sdake | there are about 1100 or so commits for the cycle | 14:07 |
| sdake | times 15 or so jobs we have | 14:07 |
| sbezverk_ | kfox1111: do you have any servers supporting ipmi? I do not have any so it is hard for me to test? | 14:08 |
| kfox1111 | sbezverk_: all of them. :) | 14:08 |
| sbezverk_ | and it seems it does not work with VM | 14:08 |
| kfox1111 | Its pretty hard to live without. :) | 14:08 |
| kfox1111 | sbezverk_: look at what use to be called quantipleo. | 14:08 |
| kfox1111 | they have a ipmi emulator that hooks up to nova. | 14:09 |
| kfox1111 | so you can throw ipmi at it and reboot vm's. :) | 14:09 |
| kfox1111 | the other option, | 14:09 |
| sbezverk_ | kfoax1111: oh nice! | 14:09 |
| kfox1111 | redhat has a bunch of fencing tools. | 14:09 |
| kfox1111 | I think they abstract the power management stuff behind a driver interface. it may be useful to use that, as it has an ipmi driver, as well as a bunch of others. | 14:10 |
| sbezverk_ | kfox1111: in this case I can just stop kubelet, it will stop sending keep alives and see it gets powerd off | 14:11 |
| kfox1111 | that would test kubelet failures, but won't actually fence the machine. | 14:12 |
| sbezverk_ | kfox1111: could you check quantipleo name, my google search comes back with nothing | 14:12 |
| *** ayoung has joined #openstack-kolla | 14:13 | |
| kfox1111 | http://blog.nemebean.com/content/quintupleo-status-update | 14:14 |
| sbezverk_ | kfox1111: got it, thanks | 14:14 |
| kfox1111 | np. :) | 14:14 |
| *** mark-casey1 has quit IRC | 14:18 | |
| kfox1111 | yay... permission denied.... :) | 14:18 |
| kfox1111 | almost.... | 14:19 |
| inc0 | kfox1111, I still wait for tripleo-on-kubernates | 14:21 |
| inc0 | 4OK | 14:21 |
| kfox1111 | +1 | 14:21 |
| kfox1111 | quintupleo for baremetal emulation, tripleo for baremetal to k8s, kolla-kubernetes for cloud. all in the gate... :) | 14:22 |
| openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: External Ceph Tools, Tests, and Docs https://review.openstack.org/381041 | 14:24 |
| kfox1111 | just gota verify the file made it from vm to vm and unwind a bit of debugging and its done. | 14:25 |
| *** Marcellin__ has joined #openstack-kolla | 14:28 | |
| *** daneyon has joined #openstack-kolla | 14:28 | |
| *** salv-orlando has joined #openstack-kolla | 14:30 | |
| *** daneyon has quit IRC | 14:33 | |
| *** salv-orlando has quit IRC | 14:35 | |
| *** openstack has joined #openstack-kolla | 18:23 | |
| *** jascott1 has quit IRC | 18:24 | |
| *** jascott1- is now known as jascott1 | 18:24 | |
| *** jascott1_ has joined #openstack-kolla | 18:24 | |
| *** jascott1_ has quit IRC | 18:25 | |
| *** jascott1_ has joined #openstack-kolla | 18:25 | |
| sdake | rhallisey where is the schedule for summit? (the etherpad) | 18:25 |
| *** openstack has joined #openstack-kolla | 19:04 | |
| *** mliima_ has joined #openstack-kolla | 19:04 | |
| *** openstackstatus has joined #openstack-kolla | 19:06 | |
| *** ChanServ sets mode: +v openstackstatus | 19:06 | |
| *** jascott1_ has quit IRC | 19:06 | |
| *** neilus has quit IRC | 19:07 | |
| *** mliima has quit IRC | 19:07 | |
| sdake | daneyon aobut? | 19:07 |
| sdake | daneyon meeting | 19:07 |
| *** pbourke has quit IRC | 19:10 | |
| *** pbourke has joined #openstack-kolla | 19:10 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Use option api_paste_config in group wsgi in nova configuration https://review.openstack.org/384663 | 19:12 |
| *** bjolo_ has quit IRC | 19:14 | |
| *** matrohon has joined #openstack-kolla | 19:15 | |
| *** salv-orlando has quit IRC | 19:17 | |
| *** daneyon has quit IRC | 19:20 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Update keystone-paste file for Newton https://review.openstack.org/384666 | 19:20 |
| openstackgerrit | Merged openstack/kolla: Fix the wrong setting for Grafana session https://review.openstack.org/384448 | 19:22 |
| *** portdirect has quit IRC | 19:28 | |
| sdake | berendt is heat not working for you without that change | 19:28 |
| sdake | MarMat has got magnum to work as is - and magnum depends heavily on heat and wait conditions | 19:28 |
| berendt | sdake it is working, i am only cleaning up some deprecation warnings introduced with newton | 19:29 |
| openstackgerrit | Merged openstack/kolla: Make sahara api worker configurable https://review.openstack.org/384277 | 19:31 |
| openstackgerrit | Merged openstack/kolla: Install python-mysqldb for Ubuntu in openstack-base container https://review.openstack.org/382398 | 19:31 |
| sdake | berendt sweet thanks | 19:33 |
| *** shardy has joined #openstack-kolla | 19:33 | |
| *** srwilkers has joined #openstack-kolla | 19:35 | |
| openstackgerrit | Merged openstack/kolla: Add upgrade tasks for Kibana https://review.openstack.org/384401 | 19:42 |
| kfox1111 | sdake: inc0_: could you try and get access to a bare metal node from them? My next ps will be to break up the gate into a few finer grained peices and try and share as much as possible between the gate and minikube. so I'd like to get minikube into the gate too. | 19:45 |
| inc0_ | kfox1111, whom "them"? | 19:45 |
| sdake | kfox1111 that isn't possible | 19:46 |
| kfox1111 | then we can say, if your an op or a dev, and want to just kick the tires a bit, here, use this " minikube start; kubectl create -f http://.....foo.yaml;' done. :) | 19:46 |
| kfox1111 | infra | 19:46 |
| inc0_ | kfox1111, infra doesn't have bare metal | 19:46 |
| kfox1111 | none at all? | 19:46 |
| sdake | none | 19:46 |
| kfox1111 | :/ | 19:46 |
| sdake | https://www.youtube.com/watch?v=z6ftW7fUdp4&feature=youtu.be | 19:46 |
| inc0_ | they have tenants on clouds donated | 19:46 |
| *** srwilkers has quit IRC | 19:46 | |
| kfox1111 | know anyone at <insert company here> that has an external gate that would be willing to share? :) | 19:47 |
| sdake | my 3 children were in that video ;-) | 19:48 |
| inc0_ | kfox1111, will be hard | 19:48 |
| kfox1111 | nice. :) | 19:48 |
| inc0_ | I like new horizon | 19:48 |
| inc0_ | sdake, heat, magnum and...? | 19:50 |
| sdake | kolla.. | 19:50 |
| inc0_ | where? | 19:51 |
| sdake | the deployment | 19:51 |
| inc0_ | ok...this was kolla-deployed? | 19:51 |
| sdake | pretty sure ;) | 19:51 |
| inc0_ | I wish they'd mention it... | 19:51 |
| kfox1111 | I'm torn wit the new horizon workflows. | 19:52 |
| *** jascott1 has quit IRC | 19:52 | |
| kfox1111 | its more configurable, but less intuitive to brand new users. | 19:53 |
| kfox1111 | I like the ajax though a lot. much faster. | 19:53 |
| kfox1111 | does infra have any nested virtualization support at least? | 19:54 |
| inc0_ | kfox1111, 8 gig of ram says no | 19:54 |
| kfox1111 | I can launch in that. :) | 19:54 |
| kfox1111 | its more the cpu virutalization thats at issue. | 19:55 |
| sdake | kfox1111 no | 19:55 |
| kfox1111 | I could probably do it in 6gig. | 19:55 |
| inc0_ | well that depends on donated openstacks conf | 19:55 |
| sdake | kfox1111 to do what you need to do, you need to use two vms | 19:55 |
| inc0_ | assume not | 19:55 |
| kfox1111 | 2vm's? | 19:56 |
| kfox1111 | how would that help? | 19:56 |
| sdake | you can run nova on one vm | 19:56 |
| sdake | nova-compute | 19:56 |
| sdake | and control plane on another | 19:56 |
| sdake | this allows more memory | 19:56 |
| kfox1111 | I'd like to test running minikube. so that be running a vm in a vm in a vm. | 19:56 |
| kfox1111 | vm in vm works fine for minikube on bare metal for testing. | 19:56 |
| kfox1111 | but thats probably too much nesting without some hardware acceleration. | 19:57 |
| kfox1111 | as its only cirros in the nested vm, which is very light weight. | 19:57 |
| sdake | inc0_ -> check out this video: https://www.youtube.com/watch?v=je_84YoVCsk | 19:57 |
| kfox1111 | but all of kolla-kubernetes in a nested vm would be too much I think. | 19:57 |
| inc0_ | kfox1111, just use qemu | 19:57 |
| *** haplo37_ has quit IRC | 19:57 | |
| kfox1111 | inc0_: I do. like I said, I've done nested qemu a lot, but if the thing in the vm is cpu heavy, it tends to slow to a crawl. | 19:58 |
| kfox1111 | cirros works fine. not so sure about all of kubernetes+ceph+kolla-kubernetes. I kind of doubt it. | 19:58 |
| inc0_ | yeah, so in OSIC region you could do nested | 19:58 |
| inc0_ | but don't know about other regions | 19:58 |
| kfox1111 | do they have the nested intel or nested amd flags on there? | 19:59 |
| inc0_ | kfox1111, infra doesn't deploy openstack, they make use of tenants donated | 19:59 |
| inc0_ | yes, and trust me, it's intel;) | 19:59 |
| kfox1111 | cool. so we would just need a job to land only there then. | 20:00 |
| *** haplo37_ has joined #openstack-kolla | 20:00 | |
| inc0_ | not possible, nobody would go to bluebox if that would be possible;) | 20:00 |
| sdake | kfox1111 you can't force a job to land anywhere in infra | 20:00 |
| sdake | i have asked | 20:00 |
| kfox1111 | oh, nothings impossible. just gota figure out a way. :) | 20:01 |
| sdake | its impossible because its not desired by infra team | 20:01 |
| kfox1111 | well, looping over and over until you get the rigth node type wouldn't be desireable to them either. which would htey hate more? :) | 20:01 |
| sdake | they would rather you type recheck ;-) | 20:01 |
| kfox1111 | I can automate that. ;) | 20:02 |
| sdake | or alternively make the gate work properly :) | 20:02 |
| sdake | PLZ WATCH: https://www.youtube.com/watch?v=je_84YoVCsk | 20:02 |
| inc0_ | kfox1111, please don't violate infra, they're really cool people | 20:03 |
| kfox1111 | not planning on it. | 20:04 |
| kfox1111 | but do need to test stuff properly. | 20:04 |
| kfox1111 | so if they can help with that, that would be great. :) | 20:04 |
| *** HyperJohnGraham_ has joined #openstack-kolla | 20:06 | |
| sdake | kfox1111 i've been around and around with infra team on the memory issue | 20:06 |
| sdake | the bottom line is 8gb is the limit | 20:06 |
| sdake | and its not changing | 20:06 |
| sdake | the solution is multi-vms | 20:06 |
| sdake | if you can't get it to work with that deployment model i'm at a loss as ato how you gate without third party gating | 20:06 |
| inc0_ | but multi-vm CI isn't there yet right? | 20:07 |
| kfox1111 | 8gig is fine. | 20:07 |
| kfox1111 | thats not the issue. | 20:07 |
| kfox1111 | its simply the ability to use hardware virtualization from within a vm. | 20:07 |
| *** huikang has quit IRC | 20:08 | |
| kfox1111 | without it, it may take days to run rather then minutes. | 20:08 |
| *** salv-orlando has joined #openstack-kolla | 20:08 | |
| sdake | you cna't use hardware virt | 20:08 |
| sdake | but you can use qemu | 20:08 |
| kfox1111 | qemu will be too slow. | 20:08 |
| kfox1111 | :/ | 20:09 |
| sdake | teset it | 20:09 |
| sdake | days = too slow | 20:09 |
| kfox1111 | oh, I've done that... a lot. | 20:09 |
| sdake | 1-2 hrs = plausable | 20:09 |
| sdake | infra puts a hard limit of 60 minutes on gate jobs | 20:09 |
| sdake | but they will go higher if there is a good reason | 20:09 |
| sdake | ya - lots of constraints - making gating isn't as easy as it looks ;) | 20:09 |
| sdake | if the vms had 16gb, our gate in kolla would be op | 20:09 |
| kfox1111 | like I said, I think I can do it in 6gb. with the full gate tests I've got. but for minikube, I need that one extra level of virtualization. | 20:10 |
| sdake | right - but there is no hardware virt available | 20:10 |
| sdake | have to use software virt unfortunately - or no virt at all | 20:10 |
| sdake | this is problem with minikube, its introducing another layer of virt | 20:11 |
| sdake | and you need hardware to make that faster | 20:11 |
| sdake | anothe roption is to remove minikube | 20:11 |
| sdake | and then it runs without that extra layer of virt | 20:11 |
| *** jascott1 has joined #openstack-kolla | 20:12 | |
| *** dwalsh has quit IRC | 20:12 | |
| *** srwilkers has joined #openstack-kolla | 20:15 | |
| *** jascott1 has quit IRC | 20:16 | |
| *** jascott1 has joined #openstack-kolla | 20:16 | |
| *** HyperJohnGraham_ has quit IRC | 20:18 | |
| *** jascott1 has quit IRC | 20:20 | |
| *** jascott1 has joined #openstack-kolla | 20:20 | |
| *** HyperJohnGraham_ has joined #openstack-kolla | 20:21 | |
| *** srwilkers has quit IRC | 20:21 | |
| *** dwalsh has joined #openstack-kolla | 20:24 | |
| *** inc0_ has quit IRC | 20:24 | |
| sdake | folks -> roadmap -> http://www.openstack.org/assets/software/newton/Community-Generated-OpenStack-Roadmap-Newton-Release-v3.pdf | 20:24 |
| *** rhallisey has quit IRC | 20:29 | |
| *** rhallisey has joined #openstack-kolla | 20:29 | |
| *** rhallisey has quit IRC | 20:35 | |
| *** salv-orl_ has joined #openstack-kolla | 20:44 | |
| *** matrohon has quit IRC | 20:44 | |
| *** salv-orlando has quit IRC | 20:46 | |
| *** jascott1 has quit IRC | 21:00 | |
| *** jascott1 has joined #openstack-kolla | 21:01 | |
| *** jascott1 has quit IRC | 21:02 | |
| *** jascott1 has joined #openstack-kolla | 21:02 | |
| *** jascott1 has quit IRC | 21:06 | |
| sbezverk_ | kfox1111: ping | 21:09 |
| openstackgerrit | BorneMace proposed openstack/kolla: Added missing cleanup-images in setup.cfg https://review.openstack.org/384716 | 21:14 |
| otavio | sdake: Hello; is it possible for you to take a look in an error we are getting in a clean deploy of newton on vagrant? | 21:17 |
| sdake | otavio i dont know anythign about vagrant | 21:17 |
| otavio | rodrigo_pereira: o/ | 21:17 |
| sdake | pbourke is your man ;) | 21:17 |
| otavio | sdake: in fact it is not vagrant specific, it is horizon raising a template exception | 21:18 |
| otavio | rodrigo_pereira: please post the exception log in pastebin | 21:18 |
| sdake | i see - well horizon should not be raising exceptions | 21:18 |
| otavio | sdake: vagrant is just our test environment | 21:18 |
| otavio | sdake: one of | 21:18 |
| sdake | otavio when was the lst time you built images? | 21:18 |
| otavio | sdake: sat | 21:19 |
| sdake | ok | 21:19 |
| sdake | i merged a patch which put us on newton | 21:19 |
| sdake | prior we were on master | 21:19 |
| sdake | couple of patches in fact | 21:19 |
| sdake | you need those | 21:19 |
| otavio | sdake: let me update it than | 21:19 |
| sdake | if you built sat you probably dont have them | 21:19 |
| sdake | so you are deploying master of openstack - which is typically broken the first couple of weeks after release ;) | 21:19 |
| sdake | if you did a pip install | 21:20 |
| sdake | make sure to undo that | 21:20 |
| *** ppalacios has joined #openstack-kolla | 21:20 | |
| sdake | if your using our vagrant code, it may pip install | 21:20 |
| *** jascott1 has joined #openstack-kolla | 21:20 | |
| sdake | not sure how our vagrant implementation works - I don't use it | 21:20 |
| sdake | only pbourke and mandre | 21:20 |
| sdake | otavio and yo uwill need to reubild after getting new code | 21:21 |
| rodrigo_pereira | otavio: http://pastebin.com/BdR8C3t6 | 21:23 |
| otavio | sdake: I am doing it now. BTW, Rodrio and Pablo (local here) just told me that disabling localization make it work | 21:23 |
| sdake | interesting | 21:23 |
| sdake | how do you disable localization | 21:23 |
| ppalacios | sdake: setting USE_I18N = False on local_settings.py worked for us, but we are not sure if everything else is fine. | 21:24 |
| sdake | we have a localization patch in our review queue blocked on branchign of newton | 21:24 |
| *** jascott1_ has joined #openstack-kolla | 21:24 | |
| sdake | ppalacios we may just need that in kolla itself | 21:24 |
| sdake | kolla doesn't have translations implemented | 21:24 |
| sdake | rather in the ironic file | 21:24 |
| sdake | your env musst set that automatically | 21:24 |
| sdake | ppalacios since i'm in EN_US locale, I htink that it must not set that env variable | 21:25 |
| ppalacios | sdake: in fact, USE_I18N is set to True by default by django | 21:25 |
| ppalacios | sdake: got it | 21:25 |
| sdake | interesting | 21:25 |
| sdake | i haven't used master horizon for a week or more | 21:25 |
| ppalacios | sdake: I will make another test | 21:26 |
| sdake | maybe something changed that broke it | 21:26 |
| *** jascott1 has quit IRC | 21:26 | |
| sdake | still recommend using newton images :) | 21:26 |
| *** jheroux has quit IRC | 21:26 | |
| otavio | sdake: that may well be the case. I am cooking the images for test | 21:26 |
| sdake | if its broken with newton images we need a patch quickly to disable localization | 21:26 |
| sdake | sounds like a 1 liner | 21:26 |
| otavio | sdake: if it is the case, we can cook the patch for it | 21:26 |
| sdake | otavio rockin | 21:26 |
| kfox1111 | sdake: I think I can get a multinode gate up pretty easily. | 21:28 |
| sdake | kfox1111 if you figure that out - that would rock - its sort of hard ;) | 21:29 |
| sdake | as in I don't know how to do it hard | 21:29 |
| sdake | but openstack-infra may have pointers | 21:29 |
| kfox1111 | I talked with them. I think I can do it pretty easy. :) | 21:30 |
| kfox1111 | k8s will help here I think. as I just gota run one command on the othser side to get it all bound up. | 21:30 |
| sdake | kfox1111 cool | 21:30 |
| kfox1111 | wont know for sure till I try though. :) | 21:30 |
| sdake | kfox1111 i want same gating in kolla too - so if you figure it out i'll be picking your brain :) | 21:31 |
| sdake | that way we can gate more services | 21:31 |
| kfox1111 | sounds good. :) | 21:31 |
| sdake | kfox1111 see pm | 21:32 |
| ppalacios | sdake: enabling i18n on django and setting the browser to send Accept-Language header with en_us worked here. The accept-language is used by django to trigger translations if USE_I18N is true. So, if horizon will be not translated, it is safe to disable i18n, otherwise... | 21:33 |
| sdake | horizon is likely translated | 21:34 |
| *** g3ek has quit IRC | 21:37 | |
| *** g3ek has joined #openstack-kolla | 21:40 | |
| *** narasimha_SV_ has quit IRC | 21:40 | |
| sdake | ppalacios so even though the browser sets somethign wrong, I'd not expect horizon to backtrace as a result | 21:43 |
| ppalacios | sdake: I agree. | 21:43 |
| *** tonanhngo has quit IRC | 21:46 | |
| *** HyperJohnGraham_ has quit IRC | 21:46 | |
| sdake | ppalacios can you follow me to #horizon (or whoever has the backtrace+ppalacios:) | 21:47 |
| sdake | rather #openstack-horizon | 21:47 |
| otavio | sdake: the images are under build here. So we will shortly know if newton works or not | 21:47 |
| otavio | sdake: I think it is better to wait for the test result, no? | 21:47 |
| sdake | ok - but it may be master is broken :) | 21:47 |
| sdake | but ya we can wait for results | 21:47 |
| sdake | if newton is broken we definately must fix that asap | 21:47 |
| sdake | we being kolla adds a workaround or horizon fixes it | 21:48 |
| otavio | no problem for me. What about you ppalacios ? | 21:48 |
| ppalacios | ok for me too. | 21:48 |
| otavio | So folks. We are going to use kolla to deploy a small environment for initial use but we plan to short after move to a multinode environment for HA purpose. Is there any document about the upgrade process? | 21:50 |
| *** mdnadeem has quit IRC | 21:55 | |
| *** tonanhngo has joined #openstack-kolla | 21:55 | |
| *** tonanhngo_ has joined #openstack-kolla | 21:56 | |
| sdake | to upgrade you run kolla-ansible upgrade | 21:56 |
| sdake | and yes its documented :) | 21:56 |
| sdake | what you are talking about tho is adding controller nodes | 21:58 |
| sdake | i know you can do that if you ahve 3 already | 21:58 |
| sdake | not sure about going from 1->3 | 21:58 |
| otavio | the other doubt I have is about moving the swift objects from one node to others | 21:59 |
| otavio | as well as cinder (using cepth) | 21:59 |
| *** tonanhngo has quit IRC | 22:00 | |
| *** HyperJohnGraham_ has joined #openstack-kolla | 22:00 | |
| *** tonanhngo_ has quit IRC | 22:00 | |
| sdake | otavio really tbh - i'd just start with 3 nodes after you get used to the workflow :) | 22:02 |
| sdake | then adding compute nodes is super trivial from there | 22:02 |
| otavio | sdake: initially our use of compute nodes will be small. We will be using swift a lot | 22:04 |
| sdake | havne'eet tried adding a storage node before | 22:05 |
| sdake | its on our roadmap | 22:05 |
| sdake | it either needs docs or code or both | 22:05 |
| otavio | sdake: as we will be starting on a small infra, we don't want to put too many machines initially | 22:06 |
| *** absubram has quit IRC | 22:07 | |
| *** dwalsh has quit IRC | 22:08 | |
| jascott1_ | hi guys! Im a kk8s noob but am working with inc0 and trying to help for summit (following rhallisey's lead on ansible for kolla-kubernetes services) | 22:12 |
| jascott1_ | quick question about kolla-k8s install | 22:12 |
| jascott1_ | i have issue with this line in the install | 22:12 |
| jascott1_ | kubectl label node $ALLINONENODE kolla_controller=true | 22:12 |
| jascott1_ | Im not doing all in one and cant label the master | 22:12 |
| *** williamc_ has quit IRC | 22:13 | |
| *** williamcaban has joined #openstack-kolla | 22:13 | |
| jascott1_ | seems like it will be required. I can slog on but just wanted to check to see if anyone could shed any lighy | 22:14 |
| kfox1111 | yeah. | 22:14 |
| kfox1111 | nodes that will run the api services, etc, should be labeled kolla_controller=true. | 22:15 |
| kfox1111 | the nodes running pure nova computes should be kolla_compute=true | 22:15 |
| kfox1111 | if you want to break up your controllers farther, you can do that too. | 22:15 |
| jascott1_ | thanks kfox1111 i get it now. I was in a daydream that my controller stuff would run on my dedicated master but thats not the case. I knew Id feel dumb after asking but I still saved time :) | 22:16 |
| kfox1111 | no worries. :) | 22:17 |
| kfox1111 | with the labels, you can have them land wherever you want. :) | 22:17 |
| *** williamcaban has quit IRC | 22:19 | |
| *** absubram has joined #openstack-kolla | 22:19 | |
| kfox1111 | it kind of doesn't help that there's similar terms/names for things. | 22:19 |
| kfox1111 | k8s master and k8s compute | 22:19 |
| *** williamcaban has joined #openstack-kolla | 22:19 | |
| kfox1111 | kolla_controller and kolla_compute | 22:19 |
| *** portdirect has joined #openstack-kolla | 22:19 | |
| kfox1111 | but likely kolla_controller is on some of the k8s computes and | 22:20 |
| kfox1111 | and kolla_computes are on a different set of k8s computes. | 22:20 |
| kfox1111 | :/ | 22:20 |
| *** JoseMello has quit IRC | 22:21 | |
| *** absubram has quit IRC | 22:22 | |
| kfox1111 | jascott1_: if you need any more help, please let me know. we're all still learning. :) | 22:24 |
| *** williamcaban has quit IRC | 22:24 | |
| jascott1_ | kfox1111 thanks! will do. | 22:28 |
| ppalacios | sdake: We've just tested on newton and it is also broken | 22:35 |
| sdake | ppalacios not good | 22:36 |
| *** salv-orlando has joined #openstack-kolla | 22:42 | |
| *** salv-or__ has joined #openstack-kolla | 22:44 | |
| *** HyperJohnGraham_ has quit IRC | 22:44 | |
| *** sdake has quit IRC | 22:44 | |
| *** salv-orl_ has quit IRC | 22:47 | |
| *** salv-orlando has quit IRC | 22:48 | |
| *** papacz has quit IRC | 22:53 | |
| *** haplo37_ has quit IRC | 22:56 | |
| *** vhosakot has quit IRC | 22:58 | |
| *** haplo37_ has joined #openstack-kolla | 22:58 | |
| *** manjeets_ has quit IRC | 22:59 | |
| *** inc0 has joined #openstack-kolla | 23:02 | |
| *** gfidente has quit IRC | 23:08 | |
| *** sdake has joined #openstack-kolla | 23:12 | |
| *** tonanhngo has joined #openstack-kolla | 23:16 | |
| *** sdake_ has joined #openstack-kolla | 23:17 | |
| *** tonanhngo has quit IRC | 23:18 | |
| *** salv-or__ has quit IRC | 23:21 | |
| *** sdake has quit IRC | 23:21 | |
| *** salv-orlando has joined #openstack-kolla | 23:22 | |
| ppalacios | sdake_: done, my time feed my children (in this case, the children resumes to only me :) ) | 23:33 |
| *** ppalacios has quit IRC | 23:33 | |
| *** asalkeld has joined #openstack-kolla | 23:38 | |
| *** berendt has quit IRC | 23:40 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: Fencing pod https://review.openstack.org/383922 | 23:41 |
| *** rodrigo_pereira has quit IRC | 23:42 | |
| *** inc0 has quit IRC | 23:47 | |
| *** salv-orlando has quit IRC | 23:51 | |
| *** tonanhngo has joined #openstack-kolla | 23:58 | |
| *** tonanhngo has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!