| *** godleon has joined #openstack-kolla | 00:00 | |
| *** MarMat_ has joined #openstack-kolla | 00:08 | |
| *** mummer has quit IRC | 00:08 | |
| *** MarMat has quit IRC | 00:08 | |
| *** dwalsh has joined #openstack-kolla | 00:11 | |
| *** mummer has joined #openstack-kolla | 00:15 | |
| *** MarMat_ has quit IRC | 00:17 | |
| *** weiyu has joined #openstack-kolla | 00:25 | |
| *** salv-orlando has joined #openstack-kolla | 00:28 | |
| *** salv-orlando has quit IRC | 00:33 | |
| *** ccesario___ has joined #openstack-kolla | 00:40 | |
| *** mbound has quit IRC | 00:45 | |
| *** mummer has quit IRC | 00:45 | |
| *** thumpba has joined #openstack-kolla | 00:48 | |
| *** ozialien10 has quit IRC | 00:59 | |
| *** jeh has quit IRC | 01:01 | |
| *** MarMat has joined #openstack-kolla | 01:07 | |
| *** daneyon has joined #openstack-kolla | 01:09 | |
| *** ravig has joined #openstack-kolla | 01:10 | |
| *** daneyon has quit IRC | 01:14 | |
| *** jeh has joined #openstack-kolla | 01:16 | |
| *** jeh has joined #openstack-kolla | 01:16 | |
| *** dwalsh has quit IRC | 01:17 | |
| *** aNupoisc has quit IRC | 01:25 | |
| *** ravig has quit IRC | 01:31 | |
| *** MarMat has quit IRC | 01:31 | |
| *** MarMat has joined #openstack-kolla | 01:32 | |
| *** haplo37 has joined #openstack-kolla | 01:44 | |
| *** mbound has joined #openstack-kolla | 01:46 | |
| *** mbound has quit IRC | 01:51 | |
| *** haplo37 has quit IRC | 01:55 | |
| *** haplo37 has joined #openstack-kolla | 01:56 | |
| *** salv-orlando has joined #openstack-kolla | 01:58 | |
| *** salv-orlando has quit IRC | 02:05 | |
| *** unicell has quit IRC | 02:08 | |
| *** haplo37 has quit IRC | 02:14 | |
| *** MarMat has quit IRC | 02:20 | |
| *** stvnoyes has joined #openstack-kolla | 02:25 | |
| *** absubram has joined #openstack-kolla | 02:29 | |
| *** mummer has joined #openstack-kolla | 02:29 | |
| *** absubram_ has joined #openstack-kolla | 02:30 | |
| *** kjelly has quit IRC | 02:32 | |
| *** mummer has quit IRC | 02:33 | |
| *** absubram has quit IRC | 02:34 | |
| *** absubram_ is now known as absubram | 02:34 | |
| *** MarMat has joined #openstack-kolla | 02:52 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/kolla-kubernetes: Use configmap to mount files https://review.openstack.org/314838 | 02:53 |
|---|---|---|
| *** daneyon has joined #openstack-kolla | 02:57 | |
| *** daneyon has quit IRC | 03:02 | |
| *** thumpba has quit IRC | 03:04 | |
| *** absubram has quit IRC | 03:05 | |
| *** ccesario___ has quit IRC | 03:07 | |
| *** ccesario___ has joined #openstack-kolla | 03:07 | |
| *** absubram has joined #openstack-kolla | 03:08 | |
| *** erhudy has quit IRC | 03:21 | |
| *** salv-orlando has joined #openstack-kolla | 03:24 | |
| *** signed8bit_Zzz is now known as signed8bit | 03:26 | |
| *** MarMat has quit IRC | 03:26 | |
| *** ccesario___ has quit IRC | 03:30 | |
| *** MarMat has joined #openstack-kolla | 03:30 | |
| *** mbound has joined #openstack-kolla | 03:30 | |
| *** salv-orlando has quit IRC | 03:30 | |
| *** signed8bit has quit IRC | 03:32 | |
| *** ravig has joined #openstack-kolla | 03:32 | |
| *** haplo37 has joined #openstack-kolla | 03:34 | |
| *** Jeffrey4l has joined #openstack-kolla | 03:35 | |
| *** mbound has quit IRC | 03:35 | |
| *** ravig has quit IRC | 03:35 | |
| *** salv-orlando has joined #openstack-kolla | 03:36 | |
| *** ccesario___ has joined #openstack-kolla | 03:36 | |
| *** salv-orlando has quit IRC | 03:41 | |
| *** jtriley has quit IRC | 03:47 | |
| *** ccesario___ has quit IRC | 03:51 | |
| *** jtriley has joined #openstack-kolla | 04:02 | |
| *** ravig has joined #openstack-kolla | 04:05 | |
| *** MarMat_ has joined #openstack-kolla | 04:22 | |
| *** MarMat has quit IRC | 04:22 | |
| *** yuanying has quit IRC | 04:44 | |
| *** salv-orlando has joined #openstack-kolla | 04:47 | |
| *** salv-orlando has quit IRC | 04:52 | |
| *** ravig has quit IRC | 04:53 | |
| *** MarMat_ has quit IRC | 04:58 | |
| *** jtriley has quit IRC | 04:58 | |
| *** ravig has joined #openstack-kolla | 04:58 | |
| *** thumpba has joined #openstack-kolla | 05:05 | |
| *** sdake has quit IRC | 05:08 | |
| *** thumpba has quit IRC | 05:10 | |
| *** coolsvap has joined #openstack-kolla | 05:10 | |
| *** mark-casey has quit IRC | 05:12 | |
| *** ravig has quit IRC | 05:22 | |
| *** salv-orlando has joined #openstack-kolla | 05:28 | |
| *** unicell has joined #openstack-kolla | 05:29 | |
| *** ravig has joined #openstack-kolla | 05:33 | |
| *** daneyon has joined #openstack-kolla | 05:40 | |
| *** absubram has quit IRC | 05:40 | |
| *** absubram has joined #openstack-kolla | 05:41 | |
| *** daneyon has quit IRC | 05:44 | |
| *** vhosakot has quit IRC | 05:50 | |
| *** jtriley has joined #openstack-kolla | 05:55 | |
| *** jtriley has quit IRC | 06:00 | |
| *** yuanying has joined #openstack-kolla | 06:01 | |
| *** mikelk has joined #openstack-kolla | 06:19 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added unit test for kolla_docker https://review.openstack.org/285138 | 06:19 |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added unit test for kolla_docker https://review.openstack.org/285138 | 06:19 |
| *** hibob has quit IRC | 06:26 | |
| *** jmccarthy has joined #openstack-kolla | 06:30 | |
| *** haplo37 has quit IRC | 06:39 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added unit test for check container https://review.openstack.org/286513 | 06:44 |
| *** sacharya has quit IRC | 06:56 | |
| *** jtriley has joined #openstack-kolla | 06:56 | |
| *** fragatina has quit IRC | 07:00 | |
| *** fragatina has joined #openstack-kolla | 07:00 | |
| *** salv-orlando has quit IRC | 07:02 | |
| *** jtriley has quit IRC | 07:02 | |
| *** fragatin_ has joined #openstack-kolla | 07:05 | |
| *** fragatina has quit IRC | 07:06 | |
| -openstackstatus- NOTICE: pip 8.1.2 broke our local python mirror, some jobs will fail with "No matching distribution found". We're investigating. Do not "recheck" until the issue is solved | 07:06 | |
| *** ChanServ changes topic to "pip 8.1.2 broke our local python mirror, some jobs will fail with "No matching distribution found". We're investigating. Do not "recheck" until the issue is solved" | 07:06 | |
| *** thumpba has joined #openstack-kolla | 07:06 | |
| *** fragatin_ has quit IRC | 07:10 | |
| *** thumpba has quit IRC | 07:11 | |
| openstackgerrit | Sharat Sharma proposed openstack/kolla: Drop root for keystone. https://review.openstack.org/314891 | 07:12 |
| *** fragatina has joined #openstack-kolla | 07:12 | |
| *** fragatina has quit IRC | 07:17 | |
| *** yuanying has quit IRC | 07:18 | |
| *** ravig has quit IRC | 07:18 | |
| *** Mr_Broken has joined #openstack-kolla | 07:19 | |
| *** cristicalin has joined #openstack-kolla | 07:20 | |
| *** athomas has joined #openstack-kolla | 07:21 | |
| *** fragatina has joined #openstack-kolla | 07:24 | |
| *** b_bezak has joined #openstack-kolla | 07:26 | |
| *** fragatina has quit IRC | 07:28 | |
| *** fragatina has joined #openstack-kolla | 07:30 | |
| *** weiyu has quit IRC | 07:35 | |
| *** Mech422 has joined #openstack-kolla | 07:37 | |
| *** egonzalez has joined #openstack-kolla | 07:37 | |
| Mech422 | Morning - anyone know if the version requirement on jinja2 2.8 is a 'hard' requirement? Trusty only has jinja 2.7 packages and I'd prefer not to monkey with pip in production... | 07:37 |
| *** weiyu_ has joined #openstack-kolla | 07:38 | |
| *** unicell has quit IRC | 07:39 | |
| *** jmccarthy1 has joined #openstack-kolla | 07:46 | |
| *** berendt has joined #openstack-kolla | 07:47 | |
| *** jmccarthy1 has left #openstack-kolla | 07:47 | |
| *** gbraad has quit IRC | 07:50 | |
| *** pbourke has joined #openstack-kolla | 07:50 | |
| *** unicell has joined #openstack-kolla | 07:51 | |
| *** egonzalez90 has joined #openstack-kolla | 07:53 | |
| *** egonzalez has quit IRC | 07:53 | |
| *** egonzalez90 has quit IRC | 07:53 | |
| *** sacharya has joined #openstack-kolla | 07:56 | |
| *** jtriley has joined #openstack-kolla | 07:58 | |
| *** berendt has quit IRC | 08:01 | |
| *** sacharya has quit IRC | 08:01 | |
| *** jtriley has quit IRC | 08:03 | |
| *** dmk0202 has joined #openstack-kolla | 08:08 | |
| *** salv-orl_ has joined #openstack-kolla | 08:19 | |
| *** rmart04 has joined #openstack-kolla | 08:22 | |
| *** daneyon has joined #openstack-kolla | 08:22 | |
| *** papacz has joined #openstack-kolla | 08:23 | |
| *** daneyon has quit IRC | 08:27 | |
| *** yuanying has joined #openstack-kolla | 08:34 | |
| *** yuanying has quit IRC | 08:37 | |
| *** yuanying has joined #openstack-kolla | 08:38 | |
| *** yuanying has quit IRC | 08:41 | |
| *** shardy has joined #openstack-kolla | 08:43 | |
| *** yuanying has joined #openstack-kolla | 08:43 | |
| *** unicell has quit IRC | 08:44 | |
| *** gfidente has joined #openstack-kolla | 08:44 | |
| openstackgerrit | Paul Bourke proposed openstack/kolla: Debug oraclelinux/centos rabbitmq deploy issue https://review.openstack.org/300988 | 08:44 |
| *** yuanying has quit IRC | 08:47 | |
| *** diga has joined #openstack-kolla | 08:50 | |
| *** yuanying has joined #openstack-kolla | 08:52 | |
| *** yuanying has quit IRC | 08:57 | |
| *** jtriley has joined #openstack-kolla | 08:59 | |
| *** sacharya has joined #openstack-kolla | 08:59 | |
| *** prithiv has joined #openstack-kolla | 09:03 | |
| *** Serlex has joined #openstack-kolla | 09:03 | |
| *** jtriley has quit IRC | 09:04 | |
| *** sacharya has quit IRC | 09:04 | |
| *** prithiv has quit IRC | 09:06 | |
| *** thumpba has joined #openstack-kolla | 09:10 | |
| *** weiyu_ has quit IRC | 09:14 | |
| *** thumpba has quit IRC | 09:15 | |
| *** SiRiuS__ has joined #openstack-kolla | 09:15 | |
| *** weiyu_ has joined #openstack-kolla | 09:16 | |
| *** fragatina has quit IRC | 09:24 | |
| *** fragatina has joined #openstack-kolla | 09:25 | |
| *** rmart04 has quit IRC | 09:25 | |
| *** rmart04 has joined #openstack-kolla | 09:26 | |
| *** weiyu_ has quit IRC | 09:33 | |
| *** weiyu_ has joined #openstack-kolla | 09:35 | |
| *** weiyu_ has quit IRC | 09:36 | |
| *** prithiv has joined #openstack-kolla | 09:37 | |
| *** mdnadeem has joined #openstack-kolla | 09:42 | |
| *** mdnadeem_ has quit IRC | 09:44 | |
| *** weiyu_ has joined #openstack-kolla | 09:46 | |
| *** cristicalin has quit IRC | 09:51 | |
| *** berendt has joined #openstack-kolla | 09:51 | |
| *** mdnadeem_ has joined #openstack-kolla | 09:53 | |
| *** mdnadeem has quit IRC | 09:56 | |
| *** berendt has quit IRC | 09:57 | |
| *** jtriley has joined #openstack-kolla | 10:00 | |
| *** coolsvap_ has joined #openstack-kolla | 10:01 | |
| *** ccesario___ has joined #openstack-kolla | 10:03 | |
| *** athomas has quit IRC | 10:03 | |
| *** diga has quit IRC | 10:03 | |
| *** jtriley has quit IRC | 10:04 | |
| *** coolsvap has quit IRC | 10:04 | |
| openstackgerrit | Prithiv proposed openstack/kolla: Asks for ansible vault password https://review.openstack.org/314945 | 10:06 |
| *** SiRiuS__ has quit IRC | 10:07 | |
| *** SiRiuS__ has joined #openstack-kolla | 10:07 | |
| *** athomas has joined #openstack-kolla | 10:08 | |
| *** signed8bit has joined #openstack-kolla | 10:09 | |
| *** berendt has joined #openstack-kolla | 10:14 | |
| openstackgerrit | Paul Bourke proposed openstack/kolla: Debug oraclelinux/centos rabbitmq deploy issue https://review.openstack.org/300988 | 10:16 |
| openstackgerrit | Paul Bourke proposed openstack/kolla: Fix build for oraclelinux base https://review.openstack.org/310774 | 10:16 |
| *** ccesario___ has quit IRC | 10:19 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/kolla-kubernetes: Use configmap to mount files https://review.openstack.org/314838 | 10:22 |
| *** dwalsh has joined #openstack-kolla | 10:25 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/kolla-kubernetes: Add a memcached service https://review.openstack.org/314772 | 10:26 |
| openstackgerrit | Prithiv proposed openstack/kolla: Update doc with password encryption using vault https://review.openstack.org/314950 | 10:29 |
| *** ntpttr- has joined #openstack-kolla | 10:33 | |
| openstackgerrit | Prithiv proposed openstack/kolla: Update doc with password encryption using vault https://review.openstack.org/314950 | 10:34 |
| openstackgerrit | Merged openstack/kolla: Use updated wsgi file for Keystone 9.0.0 release https://review.openstack.org/313814 | 10:37 |
| openstackgerrit | Prithiv proposed openstack/kolla: Update doc with password encryption using vault https://review.openstack.org/314950 | 10:40 |
| *** asalkeld has quit IRC | 10:40 | |
| *** prithiv has quit IRC | 10:46 | |
| *** weiyu_ has quit IRC | 10:51 | |
| *** prithiv has joined #openstack-kolla | 10:51 | |
| openstackgerrit | Prithiv proposed openstack/kolla: Update doc with password encryption using vault https://review.openstack.org/314950 | 10:53 |
| openstackgerrit | Paul Bourke proposed openstack/kolla: Debug oraclelinux/centos rabbitmq deploy issue https://review.openstack.org/300988 | 10:55 |
| *** coolsvap_ has quit IRC | 10:57 | |
| *** prithiv has quit IRC | 10:58 | |
| *** rmart04 has quit IRC | 11:00 | |
| *** jtriley has joined #openstack-kolla | 11:00 | |
| *** ccesario___ has joined #openstack-kolla | 11:00 | |
| *** unicell has joined #openstack-kolla | 11:01 | |
| *** daneyon has joined #openstack-kolla | 11:05 | |
| *** jtriley has quit IRC | 11:05 | |
| *** godleon has quit IRC | 11:09 | |
| *** daneyon has quit IRC | 11:09 | |
| *** cristicalin has joined #openstack-kolla | 11:13 | |
| *** sacharya has joined #openstack-kolla | 11:31 | |
| *** sacharya has quit IRC | 11:37 | |
| *** cristicalin has quit IRC | 11:38 | |
| *** rhallisey has joined #openstack-kolla | 11:38 | |
| mandre | rhallisey: I've seen you've assigned yourself to the 3rd party plugins blueprint, do you mind if I take it? | 11:40 |
| *** berendt has quit IRC | 11:40 | |
| rhallisey | mandre, yessir | 11:40 |
| rhallisey | idc | 11:41 |
| rhallisey | I"m going to work on in soon | 11:41 |
| rhallisey | I have a partial wip | 11:41 |
| rhallisey | mandre, we can share it if you like | 11:41 |
| rhallisey | mandre, let me add some work items to it and we can split them | 11:41 |
| mandre | yeah, sure... I've been thinking about how to make it flexible recently and wanted to try it out | 11:42 |
| rhallisey | mandre, do you agree with adding it as a new container? | 11:42 |
| rhallisey | or do you have another method to try | 11:43 |
| *** salv-orl_ has quit IRC | 11:43 | |
| *** berendt has joined #openstack-kolla | 11:44 | |
| Mech422 | Morning - anyone know if the version requirement on jinja2 2.8 is a 'hard' requirement? Trusty only has jinja 2.7 packages and I'd prefer not to monkey with pip in production... | 11:44 |
| rhallisey | mandre, what services need the plugins besides neutron, horizon, nova, and keystone | 11:44 |
| Mech422 | especially just for a point release | 11:44 |
| rhallisey | Mech422, not positive on that, I'll see when that was changed. | 11:45 |
| Mech422 | rhallisey: oh thanks :-) | 11:45 |
| Mech422 | rhallisey: btw - dropping back to ansible 1.9 almost killed me :-P | 11:45 |
| Mech422 | rhallisey: took me 2 hours to figure out why my set_fact's suddenly stopped working | 11:45 |
| rhallisey | Mech422, ya we will be moving to ansible 2 soon | 11:45 |
| Mech422 | rhallisey: Yay! | 11:46 |
| sean-k-mooney | rhallisey: os-vif and cinder might | 11:46 |
| *** jamie_h has quit IRC | 11:46 | |
| sean-k-mooney | what is the scope of the plugin blueprint by the way? | 11:47 |
| mandre | sorry rhallisey, I was on the phone | 11:47 |
| rhallisey | Mech422, we've been on 2.8 since last November | 11:47 |
| sean-k-mooney | is it this https://etherpad.openstack.org/p/kolla-newton-summit-plugin-planning | 11:47 |
| rhallisey | Mech422, try it out, not sure if it will work | 11:47 |
| rhallisey | sean-k-mooney, line 50-53 | 11:48 |
| mandre | so yeah, actually I want to use advanced jinja2 and inject dockerfile snippets for plugins so that we can include plugins at build time | 11:48 |
| mandre | meaning no new container | 11:48 |
| rhallisey | mandre, agreed, I think this needs to be build into the Dockerfile with fancy J2 templating | 11:49 |
| rhallisey | mandre, I think you have to have a new container | 11:49 |
| *** berendt has quit IRC | 11:49 | |
| rhallisey | well | 11:49 |
| rhallisey | technically you don't | 11:49 |
| Mech422 | rhallisey: thanks for checking - Hopefully, I can try it tonight...I gotta backport my provision plays first :-/ | 11:49 |
| rhallisey | mandre, by why not layer on top of the service? | 11:49 |
| rhallisey | Mech422, no problem | 11:50 |
| sean-k-mooney | rhallisey mandre will you be able to supprot multiple plugins at the same time in one service? | 11:50 |
| rhallisey | sean-k-mooney, yes | 11:50 |
| *** berendt has joined #openstack-kolla | 11:50 | |
| sean-k-mooney | ok :) | 11:50 |
| Mech422 | rhallisey: any 'gotcha's you can think of for my first install? I'm shooting for Kolla/Mitaka on Ubuntu Trusty | 11:51 |
| mandre | sean-k-mooney, rhallisey: and you will also be able to do advanced manipulation of your plugins beside the 'pip install' | 11:51 |
| Mech422 | (hopefully tying midonet into the mix after I get 'base' installs working...) | 11:51 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/kolla-kubernetes: [WIP] Add Mariadb service https://review.openstack.org/314990 | 11:52 |
| rhallisey | Mech422, run the prechecks | 11:52 |
| rhallisey | Mech422, and if yo hace any issues feel free to ask | 11:52 |
| Mech422 | rhallisey: thanks! | 11:52 |
| mandre | rhallisey: I'll play a bit with my idea and throw a patch when I have the code working | 11:54 |
| sean-k-mooney | rhallisey: mandre: two things i would like to add to kolla this cycle are add ovs with dpdk support and add opendaylight support. | 11:54 |
| sean-k-mooney | should they build on top of the plugin work or be enabled directly? opendaylight itself would be a new conteiner but im thinkin about the neutron server change for the odl mechanisium driver | 11:55 |
| mandre | sean-k-mooney: we'll be able to handle any plugins for any service in a generic fashion | 11:55 |
| mandre | I plan to have code for the common plugins in kolla repo, then you'll only have to set "enabled_neutron_server_plugins=networking-odl" in your kolla-build.conf and voila | 11:56 |
| rhallisey | mandre, which service do you want to go with | 11:57 |
| rhallisey | mandre, so I start with a different one | 11:57 |
| mandre | rhallisey: I'll do neutron at first | 11:57 |
| rhallisey | k | 11:57 |
| *** ChanServ changes topic to "Kolla IRC meetngs on Wednesdays @ 16:30 UTC even weeks, 23:00 UTC odd weeks - see agenda @ https://goo.gl/OXB0DL - IRC channel is *LOGGED* @ http://goo.gl/3mzZ7b (old logs from #kolla http://goo.gl/VKpPzA)" | 11:57 | |
| -openstackstatus- NOTICE: We have a workaround for our mirrors to attempt to translate package names if a match isn't immediately obvious. A more complete fix is yet to come. It is now safe to 'recheck' any jobs that failed due to "No matching distribution found". Please join #openstack-infra if you discover more problems. | 11:57 | |
| sean-k-mooney | mandre: cool ill keep an eye out for that change :) | 11:58 |
| rhallisey | mandre, agreed I think this needs to be in the Dockerfiles | 11:58 |
| *** prithiv has joined #openstack-kolla | 11:58 | |
| *** salv-orlando has joined #openstack-kolla | 12:01 | |
| *** jtriley has joined #openstack-kolla | 12:01 | |
| mandre | rhallisey, sean-k-mooney: I should have the basic structure in place in a couple of hours | 12:01 |
| *** mlima has joined #openstack-kolla | 12:01 | |
| dims | rhallisey : did anyone else show up for k8s work? i see lots of names in https://etherpad.openstack.org/p/kolla-newton-summit-kolla-kubernetes-underlay and https://etherpad.openstack.org/p/AUS-ops-openstack-in-containers | 12:02 |
| dims | sean-k-mooney : i see your name there too :) (hint!) | 12:02 |
| rhallisey | dims, we're directing everyone to the spec at the moment | 12:02 |
| rhallisey | dims, but you already reviewed it :) | 12:02 |
| dims | rhallisey : i see | 12:02 |
| dims | :) | 12:03 |
| sean-k-mooney | dims: yes i am interested in that effort i have to clear a number of other items of my plate first though. ill check out the spec though | 12:03 |
| sbezverk_ | dims about my comment.. I just want to make sure you got what I meant | 12:03 |
| dims | sbezverk_ : the --from param one? | 12:03 |
| dims | sean-k-mooney : ack thanks | 12:03 |
| sbezverk_ | dims instead of using configmap with individial files, use configmap built from the folder | 12:04 |
| sean-k-mooney | ok got to run to a meetting... back in a while | 12:04 |
| sbezverk_ | dioms if it was your undersatnding then no more question.. | 12:04 |
| sbezverk_ | oops sorry dims | 12:04 |
| dims | sbezverk_ : from what i understood, kubectl iterates over the files in the directory. i want to log the list of files explicitly :) | 12:05 |
| dims | sbezverk_ : so we know when/if some files were not generated in that directory | 12:05 |
| sbezverk_ | dims got it.. valid point.. | 12:05 |
| dims | since it's handful...should be fine :) we can always optimize | 12:06 |
| *** jtriley has quit IRC | 12:06 | |
| dims | sbezverk_ : thanks for the reviews, it definitely helped me dig up (i am just learning k8s now) | 12:06 |
| sbezverk_ | dims in the same boat here, btw I got keystone container and service running in my 5 node k8s cluster, it does not do much but at least it stopped crashing :-) | 12:08 |
| rhallisey | sbezverk_, nice! | 12:08 |
| dims | sbezverk_ : lol ++ | 12:09 |
| sbezverk_ | rhallisey one thing bugs me, how do you plan to make services discover each other? dns? or something else? | 12:09 |
| sbezverk_ | rhallisey you cannot use static IP as then you would lose mibility of containers | 12:10 |
| rhallisey | good question, kubernetes can tell us, but that requires some ordering.. | 12:12 |
| rhallisey | sbezverk_, how many keystone pods did you bring up in your cluster? | 12:12 |
| sbezverk_ | rhallisey since I have 3 masters and 2 workers, I used replication controller to start 2 pods | 12:13 |
| rhallisey | sbezverk_, was there a single vip to access the keytone service? | 12:13 |
| sbezverk_ | rhallisey supposed to since I was using service object as well and it supposed to create "VIP" but I have not tested it.. | 12:15 |
| rhallisey | kubectl get pods | 12:18 |
| rhallisey | sbezverk_, I pretty sure using the replication controller they share an address | 12:18 |
| *** signed8bit is now known as signed8bit_Zzz | 12:19 | |
| sbezverk_ | rhallisey I noticed one issue, at least I think it is an issue, but I am not sure if it is misconfig on my side or it is "feature" of kubernetes. When I kill completely one worker I do not see anything reporting this event and all replicas on pod show as running.. | 12:21 |
| rhallisey | sbezverk_, it could be that one of your workers was not running any pods | 12:22 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MASTER BRANCH https://review.openstack.org/294906 | 12:24 |
| sbezverk_ | rhallisey no, I checked docker ps and I saw each pod was running on a separate worker.. | 12:25 |
| sbezverk_ | rhallisey anyway it is just observation, I will keep digging | 12:25 |
| rhallisey | sbezverk_, interesting. I think if you kill the pods running on the node with no worker it should mirgrate to the other node | 12:25 |
| rhallisey | try that out | 12:25 |
| sbezverk_ | rhallisey will do. | 12:26 |
| *** jtriley has joined #openstack-kolla | 12:29 | |
| sbezverk_ | dims if you have something already parsing jinja2 and preparing config files for containers, please let me know I will give it a spin, manually preparing config files is borring :-) | 12:30 |
| dims | sbezverk_ : it's already in the kolla-kubernetes repo | 12:30 |
| dims | sbezverk_ : i run "kolla-kubernetes -d run keystone" and "kolla-kubernetes -d kill keystone" | 12:31 |
| rhallisey | dims, the issue I was running into using your cli was I did `python setup.py install` | 12:31 |
| rhallisey | dims, then it looks for kolla-kubernetes in /usr/lib/python... | 12:32 |
| dims | rhallisey : y, i'll try to get that working today | 12:32 |
| dims | if you haven't already. | 12:32 |
| *** pbourke has quit IRC | 12:32 | |
| rhallisey | dims, cool, I haven't fixed it yet, no | 12:32 |
| *** openstackgerrit has quit IRC | 12:33 | |
| rhallisey | dims, I have a ton of patch maintenance that I need to finsiedh before fully working on kolla-kube | 12:33 |
| *** openstackgerrit has joined #openstack-kolla | 12:33 | |
| sbezverk_ | dims the host where you run kolla-kubernetes requires kubectl right? | 12:33 |
| rhallisey | so I'm really glad you're helping | 12:33 |
| dims | rhallisey : ack | 12:33 |
| dims | sbezverk_ : yes, i can switch over to python-k8sclient soon | 12:33 |
| rhallisey | dims, are you using hyperkube? | 12:34 |
| *** signed8bit_Zzz is now known as signed8bit | 12:34 | |
| dims | rhallisey : yep | 12:34 |
| rhallisey | gotcha | 12:34 |
| sbezverk_ | dims would be usefull to have an input parameter for api server ip address and ports | 12:35 |
| dims | sbezverk_ : yep | 12:35 |
| sbezverk_ | dims since you run aoi you do not need it I guess at this point, but for me I need to specify it in kubectl commands | 12:36 |
| *** ayoung has quit IRC | 12:36 | |
| *** Liuqing has joined #openstack-kolla | 12:37 | |
| *** stvnoyes has left #openstack-kolla | 12:38 | |
| *** sbezverk_ is now known as sbezverk | 12:40 | |
| wznoinsk | hi all | 12:44 |
| *** jtriley has quit IRC | 12:52 | |
| *** erhudy has joined #openstack-kolla | 12:54 | |
| *** sean-k-m1 has joined #openstack-kolla | 12:54 | |
| *** sdake has joined #openstack-kolla | 12:55 | |
| *** sean-k-mooney has quit IRC | 12:55 | |
| *** jamie_h has joined #openstack-kolla | 12:56 | |
| *** sdake_ has joined #openstack-kolla | 12:58 | |
| *** b_bezak has quit IRC | 13:00 | |
| *** sdake has quit IRC | 13:02 | |
| *** prithiv has quit IRC | 13:05 | |
| *** dave-mccowan has quit IRC | 13:09 | |
| *** SiRiuS__ has quit IRC | 13:10 | |
| *** berendt has quit IRC | 13:13 | |
| openstackgerrit | Martin André proposed openstack/kolla: [WIP] Add support for third party plugins https://review.openstack.org/315027 | 13:13 |
| *** prithiv has joined #openstack-kolla | 13:16 | |
| *** berendt has joined #openstack-kolla | 13:20 | |
| dims | sbezverk : ack | 13:21 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MASTER BRANCH https://review.openstack.org/294906 | 13:25 |
| *** jtriley has joined #openstack-kolla | 13:25 | |
| *** berendt has quit IRC | 13:26 | |
| *** dave-mccowan has joined #openstack-kolla | 13:28 | |
| *** esharao_ has joined #openstack-kolla | 13:32 | |
| rhallisey | mandre, I'll post what I have in a minute | 13:32 |
| mandre | rhallisey: cool, have you seen https://review.openstack.org/315027? | 13:33 |
| rhallisey | mandre, yes, mines a bit different | 13:33 |
| ccesario | morning guys! | 13:41 |
| *** absubram has quit IRC | 13:45 | |
| *** prithiv has quit IRC | 13:46 | |
| *** prithiv has joined #openstack-kolla | 13:47 | |
| *** daneyon has joined #openstack-kolla | 13:47 | |
| *** daneyon has quit IRC | 13:51 | |
| Mr_Broken | morning guys | 13:56 |
| Mr_Broken | I was able to deploy kolla stable/mitaka version on multinode setup | 13:56 |
| *** thumpba has joined #openstack-kolla | 13:57 | |
| Mr_Broken | I had to install cinder_volume on separate host(my compute node) - since i needed storage space and LVM Iscsi driver | 13:57 |
| Mr_Broken | volumes are being created successfully | 13:57 |
| Mr_Broken | but when trying to attach i get : | 13:57 |
| Mr_Broken | 2016-05-11 15:28:16.311 19698 ERROR cinder.volume.targets.iscsi [req-6993e02a-8786-4370-a07a-ba05663063f1 d4dd93a54f5e45abade9d9880f7630d8 0df6d47e88f040bcb92841d73fdb744f - - -] The volume driver requires the iSCSI initiator name in the connector. | 13:58 |
| Mr_Broken | 3:45:05 PM: 2016-05-11 15:28:16.316 19698 ERROR oslo_messaging._drivers.common [req-6993e02a-8786-4370-a07a-ba05663063f1 d4dd93a54f5e45abade9d9880f7630d8 0df6d47e88f040bcb92841d73fdb744f - - -] Returning exception Invalid input received: Connector doesn't have required information: initiator to caller | 13:58 |
| Mr_Broken | i have open-iscsi installed, tgt running, and /etc/iscsi/initiatorname.iscsi exists | 13:59 |
| Mr_Broken | any ideas ? | 13:59 |
| Mr_Broken | ps. controllers and compute are running Ubuntu | 14:00 |
| wznoinsk | could someone have a look and comment on my https://review.openstack.org/#/c/313595/5 , should we support mariadb_port only or make use of database_port only or what the patchset proposes? | 14:00 |
| patchbot | wznoinsk: patch 313595 - kolla - add support for non-default database_port in kolla... | 14:00 |
| *** prithiv has quit IRC | 14:01 | |
| *** ayoung has joined #openstack-kolla | 14:01 | |
| *** salv-orl_ has joined #openstack-kolla | 14:04 | |
| sbezverk | Mr_Broken are you trying to use iscsi patch from Master on stable mitaka, or you try to use external iscsi host? | 14:04 |
| *** SiRiuS__ has joined #openstack-kolla | 14:04 | |
| Mr_Broken | cinder installed externally on compute node | 14:05 |
| mag009_ | hey guys | 14:06 |
| mag009_ | I've tried to follow your cisco plugin for calico | 14:06 |
| *** salv-orlando has quit IRC | 14:06 | |
| mag009_ | the only problem that I have is if I do python import calico it cannot find the module | 14:07 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: WIP - Add support for third party plugins for Keystone https://review.openstack.org/315068 | 14:07 |
| mag009_ | in my docker I see a folder called /plugins/neutron-server-plugin-networking-calico-archive-master | 14:07 |
| rhallisey | mandre, ^ check that out | 14:07 |
| rhallisey | I do everything through the kolla-build.conf | 14:08 |
| rhallisey | after that I'll plug we could plug into anisble | 14:08 |
| mag009_ | the plugin is there tho : /var/lib/kolla/venv/local/lib/python2.7/site-packages/networking_calico | 14:08 |
| mag009_ | rhallisey: are you talking to me ? | 14:09 |
| rhallisey | mag009_, no, mandre and I are sketching out third party plugins as we speak :) | 14:09 |
| mag009_ | ah | 14:10 |
| rhallisey | I haven't used our current model for neutron | 14:10 |
| mag009_ | kinda related :) | 14:10 |
| rhallisey | but it's subject to change | 14:10 |
| rhallisey | mag009_, ya :) | 14:10 |
| mandre | rhallisey: looking now | 14:10 |
| mandre | mag009_: you came at the right time :) | 14:10 |
| sbezverk | Mr_Broken I see, we have not tested this scenario | 14:10 |
| rhallisey | ya great timing | 14:10 |
| mag009_ | yeah it seem :) | 14:11 |
| mag009_ | although I'd like to move on with this thats the only missing part before we can start using it | 14:11 |
| Mr_Broken | sbezverk: thing is i had this working a week ago ... :D no idea what happened to iscsi and tgt in the meantime :D | 14:11 |
| *** dwalsh has quit IRC | 14:11 | |
| *** prithiv has joined #openstack-kolla | 14:12 | |
| sbezverk | Mr_Broken you get this error when you try to launch instance or create cinder volume? | 14:13 |
| mag009_ | do you guys have a blueprint that I can look at ? | 14:13 |
| mag009_ | something to give me an idea how you are going to do this or your still brainstorming about it >? | 14:14 |
| *** Liuqing has quit IRC | 14:14 | |
| rhallisey | mag009_, https://blueprints.launchpad.net/kolla/+spec/third-party-plugin-support | 14:14 |
| mag009_ | ok so far you've only done the docker image | 14:15 |
| Mr_Broken | sbezverk: when i try to attach a volume to instance (or when trying to make an instance with volume from image) | 14:16 |
| mag009_ | for plugin in neutron we need to be able to set different config | 14:16 |
| Mr_Broken | sbezverk: i can create a volume itself - no problems (lvm creates an lv , and it looks good) | 14:16 |
| *** ravig has joined #openstack-kolla | 14:16 | |
| rhallisey | mag009_, what are some of the things you need? packages, config files, directories? Anything else? | 14:16 |
| sbezverk | Mr_Broken in ths case, please check if iscsd is running on compute node, nova uses iscsiadm for these activities but iscsiadm uses iscsd for discovered info storage.. | 14:17 |
| mag009_ | for calico i need to set these : mechanism_drivers=calico , typ_drivers= flat,local and tenant_network_types= flat,local | 14:17 |
| mag009_ | and finally : neutron.conf : dhcp_agents_per_network=9999 | 14:18 |
| mag009_ | as calico agent are running on the compute nodes | 14:18 |
| Mr_Broken | root 2428 0.0 0.0 5224 156 ? Ss 14:38 0:00 /sbin/iscsid | 14:18 |
| Mr_Broken | root 2429 0.0 0.0 14860 4644 ? S<Ls 14:38 0:00 /sbin/iscsid | 14:18 |
| Mr_Broken | sbezverk : iscsd is working, | 14:18 |
| rhallisey | mag009_, the first thing you mentioned are those config settings in a calico config file? | 14:19 |
| mag009_ | ml2.ini | 14:19 |
| rhallisey | right ml2.ini | 14:19 |
| rhallisey | ok | 14:19 |
| *** prithiv has quit IRC | 14:19 | |
| *** Liuqing has joined #openstack-kolla | 14:19 | |
| mag009_ | but right now my main problem is the import of calico module isn't working | 14:20 |
| sbezverk | Mr_Broken try to run this command: iscsiadm --mode discoverydb --type sendtargets --portal X.X.X.X --discover | 14:20 |
| sbezverk | where X.X.X.X is ip of your tgtd daemon host | 14:20 |
| Mr_Broken | yeah | 14:21 |
| Mr_Broken | iscsiadm: No portals found | 14:21 |
| Mr_Broken | getting there | 14:21 |
| Mr_Broken | ;] | 14:21 |
| Mr_Broken | I think i need to configure tgt target properly | 14:21 |
| Mr_Broken | ok getting there : | 14:23 |
| Mr_Broken | root@compute4:/var/lib/cinder/volumes# iscsiadm --mode discoverydb --type sendtargets --portal 10.136.212.159 --discover | 14:23 |
| Mr_Broken | 10.136.212.159:3260,1 iqn.2001-04.example.com:target.lvm.iscsi | 14:23 |
| Mr_Broken | needed to define the target in tgt | 14:23 |
| Mr_Broken | and enable all initiators to connect to it | 14:24 |
| sbezverk | Mr_Broken make sure in cinder.conf iscsi_helper=tgtadm | 14:24 |
| *** MarMat has joined #openstack-kolla | 14:24 | |
| Mr_Broken | sbezverk: yes thats done for sure | 14:25 |
| Mr_Broken | root@compute4:/var/lib/cinder/volumes# cat /etc/cinder/cinder.conf | grep iscsi_helper | 14:25 |
| Mr_Broken | iscsi_helper = tgtadm | 14:25 |
| sbezverk | Mr_Broken it is done automatically by cinder as long as you have right iscsi helper | 14:25 |
| sbezverk | cinder process will take care of creating targets | 14:25 |
| sbezverk | and then nova will use iscsiadm to discover them and attach | 14:26 |
| Mr_Broken | well it seems it doesnt :] | 14:27 |
| rhallisey | mag009_, not sure why it's not working I've never used it. Hopefully diving into a new solution now we'll find a answer | 14:27 |
| Mr_Broken | still the same problem : | 14:27 |
| Mr_Broken | "InvalidInput: Invalid input received: Connector doesn't have required information: initiator\n"] | 14:27 |
| *** sdake_ has quit IRC | 14:28 | |
| sbezverk | Mr_Broken quick test. Can you run this command: tgtd tgtadm --lld iscsi --mode target --op show | 14:28 |
| Mr_Broken | sbezverk: yes but bare in mind that this Targer I created myself | 14:29 |
| Mr_Broken | root@compute4:/var/lib/cinder/volumes# tgtadm --lld iscsi --mode target --op show | 14:29 |
| Mr_Broken | Target 1: iqn.2001-04.example.com:target.lvm.iscsi | 14:29 |
| Mr_Broken | System information: | 14:29 |
| Mr_Broken | Driver: iscsi | 14:29 |
| Mr_Broken | State: ready | 14:29 |
| Mr_Broken | I_T nexus information: | 14:29 |
| *** sdake has joined #openstack-kolla | 14:29 | |
| Mr_Broken | LUN information: | 14:29 |
| Mr_Broken | LUN: 0 | 14:29 |
| Mr_Broken | Type: controller | 14:29 |
| Mr_Broken | SCSI ID: IET 00010000 | 14:29 |
| Mr_Broken | SCSI SN: beaf10 | 14:29 |
| Mr_Broken | Size: 0 MB, Block size: 1 | 14:29 |
| Mr_Broken | Online: Yes | 14:29 |
| Mr_Broken | Removable media: No | 14:29 |
| Mr_Broken | Prevent removal: No | 14:29 |
| Mr_Broken | Readonly: No | 14:29 |
| Mr_Broken | SWP: No | 14:29 |
| Mr_Broken | Thin-provisioning: No | 14:29 |
| Mr_Broken | Backing store type: null | 14:29 |
| Mr_Broken | Backing store path: None | 14:29 |
| Mr_Broken | Backing store flags: | 14:29 |
| Mr_Broken | Account information: | 14:29 |
| Mr_Broken | ACL information: | 14:30 |
| Mr_Broken | ALL | 14:30 |
| Mr_Broken | sbezverk: but i used this myself : tgtadm --lld iscsi --op new --mode target --tid 1 -T iqn.2001-04.example.com:target.lvm.iscsi | 14:30 |
| Mr_Broken | and its the only target created | 14:30 |
| Mr_Broken | sbzeverk: after restarting tgt and cinder-volume : | 14:31 |
| Mr_Broken | root@compute4:/var/lib/cinder/volumes# tgtadm --lld iscsi --mode target --op show | 14:31 |
| Mr_Broken | root@compute4:/var/lib/cinder/volumes# | 14:31 |
| sbezverk | Mr_Broken please use paste.opensatck.org in future.. in this case tgtd looks good | 14:31 |
| Mr_Broken | sbezverk: sorry for the long paste, will paste it next time | 14:31 |
| sbezverk | Mr_Broken in this case I suggest to enable debug for cinder and nova and see if something can be spotted which explains this issue.. | 14:32 |
| Mr_Broken | I have enabled debug for all services for now | 14:32 |
| Mr_Broken | cant find anything userfull in cinder logs | 14:33 |
| mag009_ | ok i think I have an idea | 14:33 |
| Mr_Broken | let me check nova | 14:33 |
| mag009_ | :) | 14:33 |
| Mr_Broken | sbezverk: you mean nova_compute ? | 14:33 |
| *** sdake has quit IRC | 14:35 | |
| sbezverk | Mr_Broken yes | 14:36 |
| *** berendt has joined #openstack-kolla | 14:36 | |
| sbezverk | Mr_Broken try this command: grep -r 6993e02a-8786-4370-a07a-ba05663063f1 /var/log | 14:37 |
| *** Liuqing_ has joined #openstack-kolla | 14:37 | |
| Mr_Broken | sbezverk: ClientException: The server has either erred or is incapable of performing the requested operation. (HTTP 500) (Request-ID: req-6993e02a-8786-4370-a07a-ba05663063f1) | 14:38 |
| Mr_Broken | interesting ;] | 14:38 |
| *** Liuqing has quit IRC | 14:38 | |
| *** Liuqing_ has quit IRC | 14:40 | |
| sbezverk | Mr_Broken Hm, just one line?? should be way more than that.. | 14:40 |
| *** Liuqing has joined #openstack-kolla | 14:41 | |
| Mr_Broken | sbezverk: i think only this is relevant, but here you go : http://paste.openstack.org/show/496743/ | 14:41 |
| sbezverk | Mr_Broken how about this one: grep -r b77f780a-600f-4f99-894a-eb4f4e97f10a /var/log | 14:44 |
| Mr_Broken | sbezverk: and this is from nova_libvirt (was out of permissions on last grep) http://paste.openstack.org/show/496744/ | 14:44 |
| sbezverk | Mr_Broken, one second, I see that your nova compute is container | 14:45 |
| Mr_Broken | sbezverk: http://paste.openstack.org/show/496745/ - systool not installed whats cought my attention | 14:45 |
| *** Liuqing has quit IRC | 14:45 | |
| Mr_Broken | sbezverk: yes i only removed cinder-volume and cinder-backup containers from compute node | 14:46 |
| sbezverk | Mr_Broken with iscsi patch from master nova_compute does not know how to deal with iscsi targets | 14:46 |
| sbezverk | Mr_Broken, I meant WIHTOUT that patch it will not work | 14:46 |
| Mr_Broken | should I try to install compute service outside of container , and just remove nova_compute and nova_libvirt containers ? | 14:47 |
| Mr_Broken | could you paste a link to this patch ? | 14:47 |
| *** Liuqing has joined #openstack-kolla | 14:49 | |
| sbezverk | Mr_Broken as you can see from your last paste, nova compute cannot even access initiatorname.iscsi file and bunch of other things are missing | 14:49 |
| sbezverk | Mr_Broken Let me find it, but it has not been tested with mitaka, so no guarantee it will work.. | 14:50 |
| *** dwalsh has joined #openstack-kolla | 14:50 | |
| Mr_Broken | sbezverk: :D I guess i will be the first one to test it then :] | 14:50 |
| Mr_Broken | sbezverk: if it will not work, I guess i will move more compute services outside of containers for now :/ | 14:50 |
| sbezverk | Mr_Broken You need these three patches: http://paste.openstack.org/show/496747/ | 14:55 |
| sbezverk | then you need to rebuild images and re-deploy | 14:55 |
| *** prithiv has joined #openstack-kolla | 14:56 | |
| Mr_Broken | sbezverk: oh rebuild and redeploy sounds bad :D need to have some show-off environment by tommorow afternoon ;] | 14:56 |
| Mr_Broken | sbezverk: i guess i will check this patch during the weekend earlies | 14:57 |
| Mr_Broken | sbezverk: thanks for the great support, sorry for bad pastes and some typos :> | 14:57 |
| Mr_Broken | gotta switch locations, be back in around 45 mins | 14:58 |
| Mr_Broken | thanks again guys! | 14:58 |
| *** Mr_Broken has quit IRC | 14:58 | |
| *** mlima has quit IRC | 14:58 | |
| *** Mr_Broken has joined #openstack-kolla | 14:58 | |
| sbezverk | Mr_Broken np and good luck!! | 14:59 |
| *** reddydodda has joined #openstack-kolla | 15:00 | |
| *** absubram has joined #openstack-kolla | 15:03 | |
| *** sdake has joined #openstack-kolla | 15:03 | |
| *** Mr_Broken has quit IRC | 15:03 | |
| sdake | morning cats | 15:05 |
| *** haplo37 has joined #openstack-kolla | 15:05 | |
| *** ravig has quit IRC | 15:08 | |
| *** dwalsh has quit IRC | 15:10 | |
| *** Liuqing has quit IRC | 15:12 | |
| *** diogogmt has joined #openstack-kolla | 15:14 | |
| *** salv-orl_ has quit IRC | 15:17 | |
| *** mlima has joined #openstack-kolla | 15:18 | |
| sdake | rhallisey ping me when your avalable, you need to finish the job on the release | 15:20 |
| nihilifer | do we have a meeting today? | 15:20 |
| sdake | nihilifer ack meeting today | 15:20 |
| sdake | at 23:00 UTC iirc | 15:21 |
| *** salv-orlando has joined #openstack-kolla | 15:21 | |
| nihilifer | ok | 15:21 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: WIP - Add support for third party plugins for Keystone https://review.openstack.org/315068 | 15:23 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: [WIP] Add support for third party plugins https://review.openstack.org/315027 | 15:23 |
| rhallisey | sdake, hey | 15:23 |
| *** vhosakot has joined #openstack-kolla | 15:24 | |
| *** mark-casey has joined #openstack-kolla | 15:25 | |
| sdake | rhallisey any bug that wasn't fixed in 1.1.0 needs to be moved to 1.1.0 | 15:26 |
| sdake | 1.1.1 | 15:26 |
| sdake | moved to 1.1.1 | 15:26 |
| rhallisey | looking.. | 15:26 |
| sdake | after you finish that ping me | 15:26 |
| sdake | and then we will create the release in lauchpad | 15:26 |
| sdake | we being you, so you can learn how its done | 15:26 |
| *** prithiv has quit IRC | 15:32 | |
| *** sacharya has joined #openstack-kolla | 15:33 | |
| *** diogogmt has quit IRC | 15:33 | |
| sdake | rhallisey first step, create a 1.1.1 milestone | 15:36 |
| sdake | click this link: https://launchpad.net/kolla/+milestone/1.1.0 | 15:37 |
| rhallisey | ya I"m just making sure I use the correct formate | 15:37 |
| *** mikelk has quit IRC | 15:37 | |
| rhallisey | I got it one sec | 15:37 |
| rhallisey | what's the target data? | 15:38 |
| *** sacharya has quit IRC | 15:38 | |
| rhallisey | date | 15:38 |
| rhallisey | sdake, ^ | 15:38 |
| sdake | no dont release | 15:38 |
| sdake | we need to move stuff to 1.1.1 first | 15:38 |
| rhallisey | no I'm creating a new milestone | 15:38 |
| sdake | oh | 15:38 |
| sdake | 30 days out | 15:38 |
| rhallisey | ok | 15:38 |
| *** berendt has quit IRC | 15:39 | |
| sdake | actually 45 from beginning of summit | 15:39 |
| sdake | try june 27th | 15:39 |
| sdake | wait thats wrong | 15:40 |
| sdake | june 6th is the date | 15:40 |
| *** ccesario___ has quit IRC | 15:40 | |
| *** ccesario___ has joined #openstack-kolla | 15:40 | |
| *** huikang has joined #openstack-kolla | 15:40 | |
| *** sacharya has joined #openstack-kolla | 15:41 | |
| *** huikang has quit IRC | 15:41 | |
| sdake | tuesday release days | 15:41 |
| *** huikang has joined #openstack-kolla | 15:42 | |
| rhallisey | sdake, so anything that is merged into master that just needs a backport kick? | 15:43 |
| *** aNupoisc has joined #openstack-kolla | 15:43 | |
| sdake | huh? | 15:43 |
| rhallisey | since already merged the hash for the release | 15:43 |
| rhallisey | https://bugs.launchpad.net/kolla/+bug/1577498 | 15:43 |
| openstack | Launchpad bug 1577498 in kolla mitaka "Bandit #nosec tags should be applied in a few instances" [Critical,Confirmed] - Assigned to Travis McPeak (travis-mcpeak) | 15:43 |
| rhallisey | just needs a backport to be fixed | 15:44 |
| rhallisey | nevermind I'll just kick to 1.1.0 and backport | 15:44 |
| *** ccesario___ has quit IRC | 15:44 | |
| rhallisey | 1.1.1* | 15:44 |
| *** ccesario___ has joined #openstack-kolla | 15:44 | |
| sdake | ya anything that isn't fix commiteed should be kicked | 15:45 |
| sdake | inprogress should be moved too | 15:49 |
| sdake | since they haven't hit the repo | 15:49 |
| sdake | rhallisey now you understand why we need bug ids for every patch that needs a backport | 15:49 |
| sdake | isnt' launchpad handy here? | 15:49 |
| rhallisey | ya launchpad is handy | 15:49 |
| rhallisey | sdake, ok all set | 15:50 |
| mlima | hello everyone | 15:50 |
| sdake | hey mlima | 15:50 |
| sdake | rhallisey in 1.1.0 click "create release" | 15:51 |
| sdake | you shouold click the button that disablees new bugs in the release | 15:51 |
| sdake | i forget what it is | 15:51 |
| rhallisey | yup | 15:51 |
| sdake | it shoudl be obvious - feel free to ask | 15:51 |
| rhallisey | keel this milestone active | 15:51 |
| sdake | dont keep it active | 15:52 |
| rhallisey | any release notes | 15:52 |
| sdake | no release notes | 15:52 |
| rhallisey | changelog? | 15:52 |
| sdake | we have nother tools to do that work | 15:52 |
| sdake | we have other tools to do that | 15:52 |
| rhallisey | k | 15:52 |
| sdake | i leae rest empty | 15:52 |
| rhallisey | releasing... | 15:52 |
| rhallisey | ok released | 15:52 |
| sdake | tectnically its already been relased ;) | 15:52 |
| rhallisey | well in launchpad :) | 15:52 |
| sdake | ok thats it- nice job | 15:53 |
| *** ccesario___ has quit IRC | 15:53 | |
| rhallisey | sweet | 15:53 |
| *** ccesario___ has joined #openstack-kolla | 15:53 | |
| *** huikang has quit IRC | 15:55 | |
| *** diogogmt has joined #openstack-kolla | 15:59 | |
| openstackgerrit | Steven Dake proposed openstack/kolla: Update liberty deployment warning https://review.openstack.org/315144 | 16:01 |
| *** Avi82 has joined #openstack-kolla | 16:02 | |
| sdake | need acks on that review above plz ^^ | 16:02 |
| *** Serlex has quit IRC | 16:03 | |
| *** dmk0202 has quit IRC | 16:07 | |
| *** huikang has joined #openstack-kolla | 16:13 | |
| *** athomas has quit IRC | 16:15 | |
| vhosakot | sdake: I reviewed it... should the recommended Ansible version (1.9.4) for Liberty also be mentioned in the Liberty deployment warning ? | 16:17 |
| sdake | people can read the docs ;) | 16:17 |
| vhosakot | cool | 16:17 |
| *** SiRiuS__ has quit IRC | 16:21 | |
| *** athomas has joined #openstack-kolla | 16:22 | |
| vhosakot | sdake: ack'ed.. looks good... glad liberty rocks now :) | 16:27 |
| *** mummer has joined #openstack-kolla | 16:29 | |
| *** daneyon has joined #openstack-kolla | 16:29 | |
| *** daneyon has quit IRC | 16:34 | |
| *** aNupoisc has quit IRC | 16:35 | |
| *** Jeffrey4l has quit IRC | 16:43 | |
| *** arafferty_ has joined #openstack-kolla | 16:47 | |
| *** huikang has quit IRC | 16:49 | |
| *** mummer has quit IRC | 16:49 | |
| *** gfidente has quit IRC | 16:53 | |
| Lyncos | Guys, are you running any ansible-lint or docker-lint in Travis | 16:55 |
| sdake | no | 17:03 |
| *** unicell has quit IRC | 17:04 | |
| Lyncos | ok | 17:07 |
| Lyncos | is there any way for me to see what exactly is being done ? | 17:08 |
| vhosakot | Lyncos: see what exactly is being done where ? at the gate ? | 17:10 |
| Lyncos | all the 'tests' that are run | 17:10 |
| *** aNupoisc has joined #openstack-kolla | 17:11 | |
| Lyncos | the purpose is to reflect these tests in our own ci | 17:11 |
| Lyncos | so when we push patches we can be sure it will pass | 17:12 |
| *** ccesario___ has quit IRC | 17:13 | |
| vhosakot | Lyncos: tox.ini has the basic pep8, py27, py34 tests --> https://github.com/openstack/kolla/blob/master/tox.ini we boot a nova VM at the gate in https://github.com/openstack/kolla/blob/master/tools/deploy_aio.sh | 17:13 |
| *** ccesario___ has joined #openstack-kolla | 17:13 | |
| vhosakot | Lyncos: in Newton, more kolla gates will be added --> https://etherpad.openstack.org/p/kolla-newton-summit-kolla-functional-gating | 17:13 |
| Lyncos | thanks | 17:14 |
| *** huikang has joined #openstack-kolla | 17:19 | |
| *** berendt has joined #openstack-kolla | 17:20 | |
| *** ravig has joined #openstack-kolla | 17:29 | |
| vhosakot | cool | 17:31 |
| *** unicell has joined #openstack-kolla | 17:31 | |
| *** huikang has quit IRC | 17:32 | |
| *** openstackgerrit has quit IRC | 17:33 | |
| *** openstackgerrit has joined #openstack-kolla | 17:33 | |
| *** ccesario___ has quit IRC | 17:34 | |
| *** ravig has quit IRC | 17:35 | |
| *** unicell has quit IRC | 17:35 | |
| *** unicell has joined #openstack-kolla | 17:35 | |
| *** ccesario___ has joined #openstack-kolla | 17:42 | |
| *** ravig has joined #openstack-kolla | 17:45 | |
| sdake | Lyncos apologies was usy with somthing, so we dont use travis, we use openstack's ci system | 17:48 |
| sdake | all our gate scripts are in tools directory | 17:48 |
| sdake | the stuff that sets up the gate jobs is in github.com/openstack-infra/project-config | 17:48 |
| Lyncos | anyway .. we want to implement something similar in jenkins | 17:48 |
| Lyncos | ok I'll take a look | 17:49 |
| *** ccesario___ has quit IRC | 17:53 | |
| *** thumpba has quit IRC | 17:54 | |
| *** berendt has quit IRC | 18:03 | |
| *** unicell has quit IRC | 18:04 | |
| *** unicell has joined #openstack-kolla | 18:04 | |
| *** thumpba has joined #openstack-kolla | 18:05 | |
| *** signed8bit is now known as signed8bit_Zzz | 18:06 | |
| *** signed8bit_Zzz is now known as signed8bit | 18:09 | |
| *** dwalsh has joined #openstack-kolla | 18:09 | |
| *** ravig has quit IRC | 18:11 | |
| *** ravig has joined #openstack-kolla | 18:11 | |
| *** ravig has quit IRC | 18:18 | |
| *** ravig has joined #openstack-kolla | 18:18 | |
| *** ravig has quit IRC | 18:23 | |
| *** ravig has joined #openstack-kolla | 18:29 | |
| *** ravig has quit IRC | 18:30 | |
| *** ccesario___ has joined #openstack-kolla | 18:30 | |
| *** huikang has joined #openstack-kolla | 18:33 | |
| *** huikang has quit IRC | 18:34 | |
| *** huikang has joined #openstack-kolla | 18:34 | |
| *** dave-mccowan has quit IRC | 18:35 | |
| dims | sbezverk : i checked, i have options in the CLI for these | 18:44 |
| dims | kolla_kubernetes options: | 18:44 |
| dims | --kolla_kubernetes-host KOLLA_KUBERNETES_HOST | 18:44 |
| dims | --kolla_kubernetes-kubectl_path KOLLA_KUBERNETES_KUBECTL_PATH | 18:44 |
| dims | sbezverk : should be enough right? | 18:45 |
| dims | sbezverk : if u run with "-d" you can see the command(s) we run. example | 18:46 |
| dims | 2016-05-11 14:43:21.309 96056 INFO root [-] Command : ['kubectl', '--server=http://localhost:8080', 'delete', '-f', '/var/folders/wr/8hznwwy91lqbbt8b2k99hxtw0000gn/T/kolla-2016-05-11_14-43-21_0ZkusJ/kubernetes'] | 18:46 |
| *** sdake has quit IRC | 18:48 | |
| *** dave-mccowan has joined #openstack-kolla | 18:53 | |
| sbezverk | dims right I can specify when I use cli, but if I run te script which is using kubectl it needs to get it from somewhere, right? either from env. variable or from command line parameter | 18:53 |
| dims | sbezverk : see --server in the line above | 18:54 |
| sbezverk | dims yes, that is what I currently using | 18:54 |
| sbezverk | dims --kolla_kubernetes-host can I specify port as well? | 18:55 |
| sbezverk | example --kolla_kubernetes-host=A.A.A.A:8080 | 18:55 |
| dims | --kolla_kubernetes-host=http://localhost:8080 | 18:55 |
| dims | yes i think so | 18:55 |
| sbezverk | dims ok, then it should be good enough | 18:56 |
| *** fragatina has quit IRC | 19:01 | |
| *** godleon has joined #openstack-kolla | 19:10 | |
| dims | rhallisey : any tips on what i may be missing with mariadb? https://review.openstack.org/#/c/314990/1/services/mariadb/mariadb-pod.yml.j2 | 19:11 |
| patchbot | dims: patch 314990 - kolla-kubernetes - [WIP] Add Mariadb service | 19:11 |
| dims | rhallisey : here's the log - http://paste.openstack.org/show/496803/ | 19:12 |
| *** daneyon has joined #openstack-kolla | 19:12 | |
| *** daneyon has quit IRC | 19:17 | |
| rhallisey | dims, you need to bootstrap the db | 19:17 |
| rhallisey | which mean running with the ENV var KOLLA_BOOTSTRAP | 19:18 |
| dims | rhallisey : the container is run twice? | 19:19 |
| rhallisey | dims, yes, we bootstrap all the services before they start | 19:19 |
| rhallisey | I outlined it in the spec under 'bootstrapping' | 19:20 |
| dims | ack. looking now. thanks | 19:20 |
| rhallisey | I've heard some cases where we can pair db init/db sync with the service | 19:20 |
| rhallisey | in the pod | 19:20 |
| rhallisey | but kolla has tried this in the past and it didn't go well | 19:21 |
| *** fragatina has joined #openstack-kolla | 19:26 | |
| *** aNupoisc has quit IRC | 19:28 | |
| sbezverk | rhallisey what about keeping two containers in the pod 1 - bootstrap with run once 2 - actual mariadb, all containers in pod can share volumes so lesss headache? | 19:34 |
| *** dcwangmit01 has quit IRC | 19:35 | |
| rhallisey | sbezverk, so that's the other popular solution I've heard. It's more kubernetes like. | 19:35 |
| *** aNupoisc has joined #openstack-kolla | 19:36 | |
| rhallisey | sbezverk, the problem with it though is that you need to make sure you don't db sync every time | 19:36 |
| rhallisey | you only do that once | 19:36 |
| *** shardy has quit IRC | 19:36 | |
| rhallisey | sbezverk, if we decouple bootstrapping and deployment it will provide more flexibility | 19:36 |
| sbezverk | rhallisey got it, thanks | 19:37 |
| rhallisey | sbezverk, check out some of the comments about this task. kfox has some good points as to why we should decouple | 19:38 |
| *** dcwangmit01 has joined #openstack-kolla | 19:38 | |
| *** ravig has joined #openstack-kolla | 19:38 | |
| *** thumpba has quit IRC | 19:38 | |
| rhallisey | for a few of the iterations on the spec I had it as an alternative | 19:38 |
| rhallisey | but removed it | 19:38 |
| Lyncos | is there any reason why we have to put shared instead of slave in systemd file for docker ? | 19:38 |
| *** dmk0202 has joined #openstack-kolla | 19:38 | |
| rhallisey | sbezverk, SAP folks do it that way | 19:38 |
| rhallisey | Lyncos, neutron | 19:39 |
| *** dcwangmit01 has quit IRC | 19:39 | |
| *** dcwangmit01 has joined #openstack-kolla | 19:39 | |
| rhallisey | Lyncos, in order to run neutron in thin containers we need mounts to be shared between the neutron containers and the host | 19:39 |
| rhallisey | if the mounts don't propogate, then a container can create a namespace that the host can't see and other containers can't see | 19:40 |
| *** thumpba has joined #openstack-kolla | 19:42 | |
| *** thumpba has quit IRC | 19:43 | |
| Lyncos | which part of neutron | 19:43 |
| Lyncos | the neutron agents ? | 19:43 |
| *** thumpba has joined #openstack-kolla | 19:43 | |
| Lyncos | ok for namespaces | 19:43 |
| vhosakot | Lyncos: yes, the neutron-dhcp-agent creates the namespaces | 19:45 |
| *** aNupoisc has quit IRC | 19:45 | |
| *** thumpba has quit IRC | 19:45 | |
| *** thumpba has joined #openstack-kolla | 19:45 | |
| *** ravig has quit IRC | 19:46 | |
| Lyncos | maybe you guys knows | 19:46 |
| *** ravig has joined #openstack-kolla | 19:46 | |
| Lyncos | but with shared enabled... our docker become really slow and unusable | 19:46 |
| Lyncos | we just put it back to slave and it fixed everything | 19:46 |
| vhosakot | Lyncos: MountFlags=shared is docker.service makes docker slow for you ? slow during what ? when building the images ? | 19:48 |
| *** matrohon has joined #openstack-kolla | 19:49 | |
| Lyncos | it was working fine | 19:49 |
| Lyncos | ans all of sudden... all command 'docker ps' or docker rm become really slow... | 19:49 |
| Lyncos | even systemctl stop docker is very slow | 19:49 |
| Lyncos | and when we do 'mount' it takes forever to show all the millions mounts (I did not count them but there was a lot of them) | 19:50 |
| vhosakot | Lyncos: and changing to slave fixed it and made the docker daemon fast ? | 19:50 |
| Lyncos | yeah | 19:50 |
| vhosakot | Lyncos: yes, lots of mounts will show commands like "docker volume ls" | 19:50 |
| Lyncos | not only docker is slow | 19:50 |
| vhosakot | slow* | 19:50 |
| Lyncos | df -h was slow | 19:50 |
| Lyncos | ssh in was slow | 19:50 |
| vhosakot | those are not related to docker/kolla :) | 19:50 |
| Lyncos | docker start used 100% CPU for like 10 minutes | 19:50 |
| *** mlima has quit IRC | 19:51 | |
| vhosakot | Lyncos: may your docker daemon is overloaded | 19:51 |
| Lyncos | man | 19:51 |
| Lyncos | we have only 4-5 containers | 19:51 |
| vhosakot | Lyncos: can you check its CPU usage ? | 19:51 |
| Lyncos | right now we reverted back to Slave | 19:51 |
| Lyncos | CPU is 100% idle | 19:52 |
| vhosakot | Lyncos: can you paste the output of "grep "/run" /proc/self/mountinfo" form the node on which the docker daemon was slow | 19:53 |
| Lyncos | To add to the wierdness .. 2hosts started to do the problem at same time | 19:53 |
| Lyncos | http://paste.openstack.org/show/496807/ | 19:54 |
| Lyncos | the servers is a dual 10 core Xeon with 384 gig of ram... running on Raid 1 of enterpise SSD | 19:55 |
| vhosakot | Lyncos: nice servers... your mounts look good... | 19:57 |
| Lyncos | let me put this back to shared | 19:57 |
| Lyncos | now the grep takes forever | 19:58 |
| Lyncos | and return a Huge list | 19:58 |
| Lyncos | will paste | 19:58 |
| Lyncos | Ah man | 19:58 |
| Lyncos | maybe becuase we configure the hugepages | 19:58 |
| *** athomas has quit IRC | 19:59 | |
| *** ravig has quit IRC | 20:00 | |
| Lyncos | http://paste.openstack.org/show/496811/ | 20:00 |
| Lyncos | I did not paste everything | 20:01 |
| *** SiRiuS__ has joined #openstack-kolla | 20:01 | |
| Lyncos | There is 2109458 lines..... | 20:01 |
| vhosakot | Lyncos: why do you use hugepages ? :) wont the default work ? | 20:01 |
| Lyncos | That host was configured for performance | 20:02 |
| Lyncos | but we had the same problem on another host without the huge page | 20:02 |
| Lyncos | I can remove the hugepage config | 20:02 |
| vhosakot | ah ok | 20:02 |
| Lyncos | we'll | 20:02 |
| Lyncos | see | 20:02 |
| *** salv-orl_ has joined #openstack-kolla | 20:03 | |
| dims | rhallisey : this looks handy - https://github.com/kubernetes/kubernetes/pull/23567 | 20:04 |
| Lyncos | 2 millions mount points seems to be a bit excessive | 20:04 |
| vhosakot | Lyncos: kolla does not create so many ;) | 20:04 |
| Lyncos | it's doing that only when shared | 20:05 |
| *** sdake has joined #openstack-kolla | 20:05 | |
| Lyncos | when not shared we don't have all these | 20:05 |
| Lyncos | ok rebooting without hugepages | 20:05 |
| *** salv-orlando has quit IRC | 20:06 | |
| vhosakot | Lyncos: centos or Ubuntu >= 15 ? | 20:06 |
| Lyncos | our host is an ubuntu 16.04 I think | 20:07 |
| Lyncos | we deploy with ubuntu sources | 20:07 |
| Lyncos | 14.04 | 20:07 |
| *** huikang has quit IRC | 20:09 | |
| vhosakot | cool | 20:11 |
| *** ravig has joined #openstack-kolla | 20:13 | |
| rhallisey | weird. | 20:13 |
| rhallisey | dims, that's interesting | 20:13 |
| * dims peeling layers of onions :) | 20:14 | |
| *** huikang has joined #openstack-kolla | 20:18 | |
| *** ravig has quit IRC | 20:20 | |
| Lyncos | after rebooting I have only 106K mount points | 20:21 |
| *** SiRiuS__ has quit IRC | 20:22 | |
| *** absubram has quit IRC | 20:38 | |
| *** ccesario___ has quit IRC | 20:40 | |
| *** ccesario___ has joined #openstack-kolla | 20:41 | |
| *** SiRiuS__ has joined #openstack-kolla | 20:41 | |
| *** ayoung has quit IRC | 20:41 | |
| matrohon | hi all | 20:42 |
| matrohon | I'm testing kolla, so I have a newbi question | 20:42 |
| matrohon | I want to understand how kolla deals with ovs | 20:43 |
| *** ravig has joined #openstack-kolla | 20:44 | |
| matrohon | so I entered the neutron_ovs_agent container with $ docker exec -t -i neutron_openvswitch_agent bash | 20:45 |
| *** aNupoisc has joined #openstack-kolla | 20:45 | |
| matrohon | and I need a pwd to run ovs-vsctl show | 20:45 |
| matrohon | the same command didn't need a pwd in the ovs_vswitchd container | 20:46 |
| sbezverk | rhallisey in specs I did not find anything with regards to logging. Are you planning to add it later or it will not be a part of specs? | 20:47 |
| *** ravig has quit IRC | 20:47 | |
| Lyncos | sbezverk: what you mean by logging ... it's already there no ? | 20:47 |
| rhallisey | sbezverk, I figured we would use what is already containerized | 20:47 |
| rhallisey | so heka | 20:47 |
| Lyncos | matrohon: I cannot help we ditched ovs | 20:48 |
| openstackgerrit | Hui Kang proposed openstack/kolla: Do not use AUFS for centos/fedora/oraclelinux container images on ubuntu host https://review.openstack.org/314205 | 20:48 |
| matrohon | Lyncos, can you help me figure out what pwd show I use to run a sudo cmd in a container managed by kolla? | 20:49 |
| matrohon | s/show/should | 20:49 |
| vhosakot | matrohon: try logging in an root | 20:49 |
| vhosakot | matrohon: docker exec -t -i -u root neutron_openvswitch_agent bash | 20:49 |
| Lyncos | usually you don't have to sudo inside a container | 20:49 |
| Lyncos | You guys want to see something wierd ? | 20:50 |
| sbezverk | rhallisey yeah, but sharing volumes between pods is not recommended practice | 20:50 |
| matrohon | vhosakot, thanks, it works | 20:50 |
| Lyncos | grep /run /proc/self/mountinfo |wc -l | 20:50 |
| Lyncos | 4206609 | 20:50 |
| sbezverk | so we will not be able to od it as easily as it was with heka shared volume | 20:50 |
| vhosakot | matrohon: cool | 20:50 |
| *** ccesario___ has quit IRC | 20:50 | |
| vhosakot | matrohon: some commands insdie the container needs privileged access... and "ovs-vsctl ..." is one such command | 20:51 |
| matrohon | vhosakot, however, It's strange that this command didn't need root privileges in the ovs_vswitchd container! | 20:51 |
| sbezverk | rhallisey I had an idea about logging, but since there is no section of specs about it I was not sure where I can express it.. | 20:53 |
| rhallisey | sbezverk, add it in as a comment | 20:53 |
| rhallisey | anywhere | 20:53 |
| rhallisey | I'm still thinking | 20:53 |
| vhosakot | matrohon: neutron_openvswitch_agent container runs as neutron user https://github.com/openstack/kolla/blob/master/docker/neutron/neutron-openvswitch-agent/Dockerfile.j2#L25 | 20:54 |
| *** ccesario___ has joined #openstack-kolla | 20:54 | |
| sbezverk | rhallisey ok | 20:54 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/kolla-kubernetes: Add Mariadb service https://review.openstack.org/314990 | 20:54 |
| dims | rhallisey : worked once - so ship it! ^^ :) | 20:54 |
| rhallisey | dims, nice! | 20:54 |
| matrohon | vhosakot, ok, I need to learn all the kolla magic :) | 20:55 |
| matrohon | vhosakot, thanks | 20:55 |
| vhosakot | matrohon: cool, np... you can see in the output of "ps -ef" inside the container and see why kolla runs it as root... | 20:56 |
| *** thumpba has quit IRC | 20:56 | |
| *** ccesario___ has quit IRC | 20:58 | |
| sdake | matrohon welcome to the party ;) | 21:01 |
| matrohon | sdake, thanks :) hopefully kolla will meet my need! | 21:03 |
| matrohon | sdake, by the way with the aio install, is there any default domain/user/pwd I can use in keystone to test a VM deployment? | 21:04 |
| vhosakot | Lyncos: I have not seen so many mounts... is that before or after deploying kolla ? | 21:04 |
| Lyncos | after many many deploy/cleanup | 21:06 |
| *** papacz has quit IRC | 21:06 | |
| *** papacz has joined #openstack-kolla | 21:07 | |
| vhosakot | Lyncos: cleanup deletes the volumes that triggers unmount... | 21:07 |
| Lyncos | when I switch docker to 'slave' they all disapear | 21:08 |
| sbezverk | rhallisey I added comment, curious what you think about this approach.. | 21:08 |
| matrohon | sdake, it seems post-deploy script is here to help me | 21:08 |
| rhallisey | sbezverk, awesome thx | 21:08 |
| vhosakot | Lyncos: would be worth checking is those many mounts are seen after deleting kolla and uninstalling kolla | 21:08 |
| Lyncos | and when in shared mode ... the number of volumes is kinda random | 21:08 |
| Lyncos | Want me to try a cleanup ? | 21:09 |
| vhosakot | Lyncos: the docker daemon must be restarted when MountFlags is changed in docker.service | 21:09 |
| Lyncos | yes I know | 21:09 |
| sdake | matrohon ya - you can use kolla-ansible post-deploy to generate the post deploy admin-openrc.sh | 21:09 |
| vhosakot | Lyncos: yeah, cleanup everything, reboot, and then check the output of "/proc/self/mountinfo"... | 21:10 |
| Lyncos | Ok | 21:13 |
| Lyncos | getting more infio | 21:13 |
| Lyncos | After cleaning up | 21:13 |
| Lyncos | and changing to 'shared' (was at slavewhen cleanup) | 21:13 |
| Lyncos | seems to be better now | 21:13 |
| Lyncos | docker volume ls |wc -l | 21:13 |
| Lyncos | 38 | 21:13 |
| Lyncos | I'll reboot to make sure | 21:13 |
| vhosakot | Lyncos: "docker volume ls | wc -l" should display 2 after cleanup :) | 21:14 |
| Lyncos | I can't check .. Already rebooting :-) will check when it's back.... in 5 min (these dell servers are so slow to boot) | 21:15 |
| *** ayoung has joined #openstack-kolla | 21:15 | |
| *** esharao_ has quit IRC | 21:16 | |
| *** jtriley has quit IRC | 21:16 | |
| *** ccesario___ has joined #openstack-kolla | 21:17 | |
| vhosakot | Lyncos: so, your deploy node is a bare metal node ? | 21:17 |
| Lyncos | yeah | 21:17 |
| vhosakot | cool | 21:18 |
| Lyncos | all our nodes are baremetal | 21:18 |
| Lyncos | :-) | 21:18 |
| Lyncos | Why would we deploy openstack in vm's ? | 21:18 |
| *** matrohon has quit IRC | 21:20 | |
| Lyncos | Still have 38 volumes in docker | 21:21 |
| *** ccesario___ has quit IRC | 21:21 | |
| Lyncos | but after a cleanup everything works | 21:22 |
| Lyncos | I think it was caused by my coworker testing a new container for calico | 21:24 |
| Lyncos | the container was constantly restarting.... maybe something happened with volumes | 21:24 |
| *** rhallisey has quit IRC | 21:24 | |
| vhosakot | Lyncos: cool... reboot all the things! :) | 21:25 |
| Lyncos | what fixed it is the cleanup | 21:25 |
| *** esharao_ has joined #openstack-kolla | 21:25 | |
| vhosakot | Lyncos: yep, cleanup+reboot should take back the cloud to how to was before kolla was installed | 21:26 |
| *** ccesario___ has joined #openstack-kolla | 21:26 | |
| *** haplo37 has quit IRC | 21:30 | |
| Lyncos | we are re-deployingh | 21:31 |
| Lyncos | without the 'h' | 21:31 |
| *** ccesario___ has quit IRC | 21:34 | |
| vhosakot | cool | 21:35 |
| *** ccesario___ has joined #openstack-kolla | 21:36 | |
| *** SiRiuS__ has quit IRC | 21:43 | |
| *** ccesario___ has quit IRC | 21:45 | |
| *** huikang has quit IRC | 21:46 | |
| *** huikang has joined #openstack-kolla | 21:53 | |
| *** daneyon has joined #openstack-kolla | 21:54 | |
| *** daneyon has quit IRC | 21:59 | |
| *** ccesario___ has joined #openstack-kolla | 22:02 | |
| *** gumbyblake2 has joined #openstack-kolla | 22:13 | |
| *** gumbyblake has joined #openstack-kolla | 22:14 | |
| *** gumbyblake2 has quit IRC | 22:14 | |
| huikang | is there any meeting today? | 22:15 |
| *** sean-k-mooney has joined #openstack-kolla | 22:18 | |
| vhosakot | yep... in 41 mins | 22:19 |
| *** sean-k-m1 has quit IRC | 22:19 | |
| huikang | thanks, vhosakot | 22:19 |
| vhosakot | cool | 22:19 |
| *** huikang has quit IRC | 22:20 | |
| *** ozialien10 has joined #openstack-kolla | 22:21 | |
| *** dwalsh has quit IRC | 22:32 | |
| *** mark-casey has left #openstack-kolla | 22:38 | |
| *** mbound has joined #openstack-kolla | 22:41 | |
| *** esharao_ has quit IRC | 22:42 | |
| *** dmk0202 has quit IRC | 22:47 | |
| *** aNupoisc has quit IRC | 22:53 | |
| vhosakot | meeting in 6 minutes... | 22:54 |
| *** Jeffrey4l has joined #openstack-kolla | 22:59 | |
| *** aNupoisc has joined #openstack-kolla | 22:59 | |
| *** asalkeld has joined #openstack-kolla | 23:00 | |
| vhosakot | meeting time in #openstack-meeting-4 | 23:00 |
| sdake | meeting time folks | 23:00 |
| *** huikang has joined #openstack-kolla | 23:01 | |
| d_code | sdake: you mentioned about deploying as non-root user. I set group ownership on all nodes for /etc/kolla to the “kolla” group that I created…my ansible user is in that group and also docker | 23:05 |
| d_code | it fails on “creating log volume" | 23:06 |
| d_code | looks like it’s calling the docker client api, so I would think being in the docker group would suffice | 23:06 |
| sdake | i'd think so too | 23:07 |
| gumbyblake | Hi, I am having an issues starting instance after kolla deploy. Instance fails on Block Device mapping task with this message " Failed to perform requested operation on instance "cirros0", the instance has an error status: Please try again later [Error: Build of instance 1a68341e-69ea-44c5-b2c7-41da69d393ee aborted: Block Device Mapping is Invalid.]." Not sure where I find logs to find out what is wrong | 23:07 |
| d_code | k…I”ll keep digging | 23:08 |
| d_code | error is permission denied | 23:08 |
| *** godleon has quit IRC | 23:09 | |
| sdake | gumbyblake hang tight a minute we hae our team meeting | 23:10 |
| sdake | but everyone is on vacation it appears | 23:10 |
| sdake | so we are ending shortly (early) | 23:10 |
| gumbyblake | sdake no problem, any assistance welcome | 23:10 |
| *** yuanying has joined #openstack-kolla | 23:10 | |
| *** beekhof has quit IRC | 23:13 | |
| sdake | gumbyblake what distro are you using | 23:14 |
| *** jtriley has joined #openstack-kolla | 23:14 | |
| *** zigo has quit IRC | 23:14 | |
| *** zigo has joined #openstack-kolla | 23:15 | |
| gumbyblake | sdake centos 7 | 23:16 |
| *** huikang has quit IRC | 23:16 | |
| sdake | gumbyblake ok - running liberty or mitaka? | 23:16 |
| *** Avi82 has quit IRC | 23:16 | |
| Jeffrey4l | huikang, sdake should we will support all kinda of infra service source install? Or the ovn from source will be replaced by the binary in the future? I think we should define this first. :) | 23:16 |
| gumbyblake | sdake mitaka I think | 23:17 |
| *** huikang has joined #openstack-kolla | 23:17 | |
| sdake | gumbyblake did you git checkout stable/mitaka? | 23:17 |
| sdake | or download the tarball | 23:17 |
| sdake | or pip install | 23:17 |
| sdake | or what? :) | 23:17 |
| gumbyblake | sdake pip install following online docs | 23:17 |
| sdake | ok those docs unfortunately don't have instructions to check out the stable branch | 23:18 |
| sdake | they need such instructions | 23:18 |
| gumbyblake | bummer, does that mean I am running unstable deploy ? | 23:18 |
| sdake | you are running master, not stable/mitaka | 23:18 |
| sdake | master worked last time i tried it (day or two ago) | 23:19 |
| sdake | could start vms | 23:19 |
| *** vhosakot has quit IRC | 23:19 | |
| sdake | did you build your own images? | 23:19 |
| d_code | sdake: ah…think my issue is my docker setup….I’ve been using btrfs driver, but I forgot to setup it up on this host…not that that would cause this issue…but I ran it with become and still didn’t work…so…no kolla problem | 23:19 |
| sdake | iwhy use becomem? | 23:19 |
| gumbyblake | image build as per docs too, not custom image. Should I modify the globals.yaml to install stable releaser set ? | 23:19 |
| sdake | become? | 23:19 |
| sdake | gumbyblake what I'd do is comment out the line in globals.yml that specifies a registry and just use the images we push to the docker hub | 23:20 |
| *** weiyu_ has joined #openstack-kolla | 23:20 | |
| sdake | (commenting that line will use the images from docker hub) | 23:20 |
| gumbyblake | sdake let me check, I thought I had registry commented already | 23:21 |
| *** huikang has quit IRC | 23:21 | |
| d_code | sdake: I manually ran the ansible playbook that kolla-ansible deploy runs, but with the —become flag | 23:21 |
| d_code | b/c I can’t ssh as root | 23:21 |
| *** jtriley has quit IRC | 23:22 | |
| gumbyblake | sdake checked registry lines commented out in globals.yaml | 23:23 |
| *** beekhof has joined #openstack-kolla | 23:24 | |
| *** Liuqing has joined #openstack-kolla | 23:25 | |
| *** weiyu_ has quit IRC | 23:26 | |
| *** jtriley has joined #openstack-kolla | 23:27 | |
| *** ozialien10 has quit IRC | 23:27 | |
| *** ozialien10 has joined #openstack-kolla | 23:28 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MASTER BRANCH https://review.openstack.org/294906 | 23:29 |
| *** Liuqing has quit IRC | 23:30 | |
| *** salv-orl_ has quit IRC | 23:32 | |
| *** arafferty_ has quit IRC | 23:32 | |
| gumbyblake | sdake I would think I need to rebuild images when changing to stable/mitaka | 23:34 |
| *** diogogmt_ has joined #openstack-kolla | 23:34 | |
| *** diogogmt_ has quit IRC | 23:34 | |
| *** diogogmt has quit IRC | 23:37 | |
| *** ozialien10 has quit IRC | 23:37 | |
| *** ozialien10 has joined #openstack-kolla | 23:38 | |
| *** papacz has quit IRC | 23:42 | |
| *** papacz has joined #openstack-kolla | 23:43 | |
| sdake | gumbyblake do cleanup-containers | 23:43 |
| sdake | cleanup-images -f | 23:43 |
| sdake | cleanup-host | 23:43 |
| sdake | reboot | 23:43 |
| sdake | and then deploy from the docker hub | 23:43 |
| *** daneyon has joined #openstack-kolla | 23:43 | |
| sdake | but yes, if you want to build your own images, you will need to rebuild and push to your local registry when changing to stabkle/mitaka | 23:43 |
| *** jtriley_ has joined #openstack-kolla | 23:44 | |
| gumbyblake | sdake excellent will do the rebuild and see how I go | 23:46 |
| *** gumbyblake has left #openstack-kolla | 23:46 | |
| sdake | make sure to clean up first ;) | 23:47 |
| *** jtriley has quit IRC | 23:47 | |
| *** daneyon has quit IRC | 23:47 | |
| *** gumbyblake has joined #openstack-kolla | 23:47 | |
| *** erhudy has quit IRC | 23:51 | |
| *** godleon has joined #openstack-kolla | 23:55 | |
| godleon | d_code: I hit the issue as well before, then I solved it by switching to root | 23:56 |
| godleon | d_code: I suggest you use root first for avoiding permission issues. | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!