| *** sdake has quit IRC | 00:01 | |
| *** vhosakot has joined #kolla | 00:05 | |
| *** dims has joined #kolla | 00:08 | |
| *** sdake_ has quit IRC | 00:08 | |
| *** sdake has joined #kolla | 00:11 | |
| *** Jeffrey4l has joined #kolla | 00:12 | |
| *** fthiagogv has joined #kolla | 00:19 | |
| openstackgerrit | Martin André proposed openstack/kolla: Fix host_manager plugin for vagrant virtualbox https://review.openstack.org/293179 | 00:21 |
|---|---|---|
| openstackgerrit | Martin André proposed openstack/kolla: Fix genconfig on vagrant https://review.openstack.org/293180 | 00:22 |
| *** openstack has joined #kolla | 00:24 | |
| *** tummy has joined #kolla | 00:24 | |
| *** vhosakot has quit IRC | 00:25 | |
| openstackgerrit | Thiago Gomes proposed openstack/kolla: Add Manila deployment document https://review.openstack.org/292588 | 00:27 |
| openstackgerrit | Merged openstack/kolla: Add missing tar command for oraclelinux binary https://review.openstack.org/293027 | 00:28 |
| *** tummy has quit IRC | 00:28 | |
| *** ssurana1 has quit IRC | 00:28 | |
| sdake | asalkeld around? | 00:29 |
| asalkeld | hi | 00:29 |
| sdake | can you get this fixed https://bugs.launchpad.net/python-heatclient/+bug/1557810 | 00:30 |
| openstack | Launchpad bug 1557810 in python-heatclient "invalid json blocking kolla liberty gate" [Undecided,New] | 00:30 |
| sdake | i'm not sure we shoud be validating third party project json code but that is besides the point ;) | 00:30 |
| asalkeld | not sure why that made it through the heatclient gate | 00:32 |
| openstackgerrit | Merged openstack/kolla: Move sysctl out of haproxy container https://review.openstack.org/292892 | 00:32 |
| sdake | do you know the problem? | 00:32 |
| asalkeld | no, i'll have a quick look | 00:33 |
| asalkeld | got a clean python-heatclient and running pep8 | 00:33 |
| sdake | you don't run our validate json script in your gate im sure | 00:34 |
| asalkeld | o, why we doing that? | 00:34 |
| sdake | to make sure the syntax is correct? :) | 00:34 |
| openstackgerrit | Merged openstack/kolla: Switch to with items for sysctl https://review.openstack.org/292893 | 00:34 |
| sdake | do you men why are we validating third party json and yaml files? | 00:35 |
| sdake | no good reason i'm sure - I think master removed that | 00:35 |
| asalkeld | this line failed: https://github.com/openstack/python-heatclient/blob/master/heatclient/tests/test_templates/adopt.json#L2 | 00:36 |
| asalkeld | can't see the problem with that | 00:36 |
| asalkeld | adopt.json failed validation: Expecting property name: line 2 column 5 (char 6) | 00:37 |
| sdake | that is master | 00:39 |
| sdake | i dont knwo what is in the gate | 00:39 |
| sdake | looks like sbaker fixed hte invalid json | 00:39 |
| sdake | probably needs a new tag to get into the gate - i doubt the gate gates on master client packages | 00:40 |
| asalkeld | ok | 00:41 |
| SamYaple | asalkeld: dont worry about that bug steven just posted, its already been fixed and merged upstream | 00:42 |
| SamYaple | https://review.openstack.org/#/c/289468/ | 00:42 |
| asalkeld | k | 00:42 |
| sdake | SamYaple ya that is pretty clear looking at the chanelog for the file | 00:42 |
| sdake | what is needed is a tag to introduce the new heatclient to the gate | 00:43 |
| SamYaple | sdake: also the comment in the kolla patchset | 00:43 |
| sdake | and heatclient probably needs a validatejson | 00:43 |
| SamYaple | eh that json is valid for some parsers | 00:44 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: heatclient has invalid json validation picks up https://review.openstack.org/293185 | 00:45 |
| SamYaple | theres the cherrypick from master | 00:45 |
| *** dims has joined #kolla | 00:45 | |
| sdake | thanks SamYaple | 00:46 |
| sdake | ok folks dinner time | 00:46 |
| sdake | after long hard day of physical labor | 00:46 |
| sdake | i think my remodel will be done about the time kolla remodel is done ;) | 00:46 |
| sdake | asalkeld can you ack that cherrypick | 00:47 |
| sdake | then we dont need a new tag | 00:47 |
| SamYaple | sdake: why did you increate teh gate timeout for centos | 00:49 |
| SamYaple | that is not useful | 00:49 |
| asalkeld | done | 00:49 |
| *** phuongnh has joined #kolla | 00:52 | |
| *** Jeffrey4l has quit IRC | 00:53 | |
| asalkeld | /tmp/hudson4318013726486424771.sh: line 2: tools/setup_gate.sh: No such file or directory | 00:53 |
| asalkeld | SamYaple: ^ | 00:53 |
| asalkeld | is that expected on stable? | 00:53 |
| asalkeld | http://logs.openstack.org/85/293185/1/check/gate-kolla-dsvm-build-centos-binary/b102915/console.html#_2016-03-16_00_49_21_342 | 00:53 |
| SamYaple | asalkeld: we made some changes to that. let me look | 00:54 |
| SamYaple | asalkeld: stable is bust | 00:57 |
| *** dave-mccowan has joined #kolla | 00:57 | |
| *** alisonh has left #kolla | 00:57 | |
| asalkeld | ok, i guess approving the above patch is ok tho'? | 00:57 |
| asalkeld | i.e. less bust | 00:57 |
| *** Kennan has quit IRC | 00:58 | |
| SamYaple | asalkeld: yes its fine | 00:58 |
| SamYaple | ive got another patch to fix gate, please merge that one | 00:58 |
| openstackgerrit | Merged openstack/kolla: heatclient has invalid json validation picks up https://review.openstack.org/293185 | 01:03 |
| *** daneyon has quit IRC | 01:04 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Libvirt needs openvswitch to connect br-int https://review.openstack.org/288075 | 01:05 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Fix stable/liberty gate https://review.openstack.org/293189 | 01:05 |
| *** dims has quit IRC | 01:07 | |
| *** Kennan has joined #kolla | 01:10 | |
| *** dims has joined #kolla | 01:12 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Libvirt needs openvswitch to connect br-int https://review.openstack.org/288075 | 01:12 |
| rhallisey | SamYaple oh that's why it was failing | 01:13 |
| rhallisey | thanks for updating it | 01:13 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Fix stable/liberty gate https://review.openstack.org/293189 | 01:16 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Libvirt needs openvswitch to connect br-int https://review.openstack.org/288075 | 01:16 |
| SamYaple | rhallisey: yea im just trying to see it pass the gate | 01:16 |
| *** dims has quit IRC | 01:29 | |
| *** dims has joined #kolla | 01:29 | |
| *** vhosakot has joined #kolla | 01:31 | |
| *** ccesario has quit IRC | 01:44 | |
| *** ccesario has joined #kolla | 01:45 | |
| *** dims has quit IRC | 01:52 | |
| *** mgoddard has quit IRC | 01:54 | |
| *** dims has joined #kolla | 01:56 | |
| *** tfukushima has joined #kolla | 01:57 | |
| *** tfukushima has quit IRC | 01:58 | |
| *** vhosakot has quit IRC | 02:02 | |
| *** dims has quit IRC | 02:03 | |
| *** dims has joined #kolla | 02:05 | |
| *** vhosakot has joined #kolla | 02:13 | |
| openstackgerrit | Merged openstack/kolla: Remove useless dependency on ifconfig https://review.openstack.org/293181 | 02:17 |
| openstackgerrit | Kuo-tung Kao proposed openstack/kolla: add unit tests for build script https://review.openstack.org/290423 | 02:37 |
| *** unicell has quit IRC | 02:37 | |
| *** vhosakot has quit IRC | 02:39 | |
| *** vhosakot has joined #kolla | 02:43 | |
| *** rhallisey has quit IRC | 02:45 | |
| *** Jeffrey4l has joined #kolla | 02:57 | |
| *** tfukushima has joined #kolla | 03:02 | |
| *** tfukushima has quit IRC | 03:04 | |
| *** fthiagogv has quit IRC | 03:05 | |
| *** ssurana has joined #kolla | 03:06 | |
| *** dave-mccowan has quit IRC | 03:17 | |
| *** yuanying has quit IRC | 03:20 | |
| *** dims has quit IRC | 03:37 | |
| *** gbraad has joined #kolla | 03:38 | |
| openstackgerrit | Gerard Braad proposed openstack/kolla: Changed commented default port for docker_registry https://review.openstack.org/293240 | 03:44 |
| *** tfukushima has joined #kolla | 03:50 | |
| *** tfukushima has quit IRC | 03:52 | |
| openstackgerrit | Merged openstack/kolla: Fix genconfig on vagrant https://review.openstack.org/293180 | 04:00 |
| *** ayoung is now known as ayoung_Zzz__ | 04:01 | |
| *** yuanying has joined #kolla | 04:07 | |
| *** tfukushima has joined #kolla | 04:07 | |
| *** ayoung_Zzz__ has quit IRC | 04:12 | |
| openstackgerrit | Gerard Braad proposed openstack/kolla: Changed port of docker_registry for consistency with documentation https://review.openstack.org/293240 | 04:19 |
| *** vhosakot has quit IRC | 04:21 | |
| openstackgerrit | MD NADEEM proposed openstack/kolla: Fix race condition in haproxy https://review.openstack.org/292782 | 04:41 |
| *** tfukushima has quit IRC | 04:43 | |
| *** alisonh has joined #kolla | 04:47 | |
| *** sdake has quit IRC | 04:50 | |
| *** ccesario has quit IRC | 04:59 | |
| *** ccesario has joined #kolla | 05:00 | |
| *** ssurana has quit IRC | 05:08 | |
| openstackgerrit | Gerard Braad proposed openstack/kolla: Changed port of docker_registry in globals.yml https://review.openstack.org/293240 | 05:09 |
| *** ssurana has joined #kolla | 05:13 | |
| *** SiRiuS_ has joined #kolla | 05:21 | |
| *** tfukushima has joined #kolla | 05:23 | |
| *** unicell has joined #kolla | 05:28 | |
| *** Jeffrey4l has quit IRC | 05:32 | |
| openstackgerrit | Kuo-tung Kao proposed openstack/kolla: add unit tests for build script https://review.openstack.org/290423 | 05:37 |
| *** alisonh has quit IRC | 05:42 | |
| *** alisonh has joined #kolla | 05:46 | |
| *** coolsvap has joined #kolla | 06:03 | |
| openstackgerrit | Martin André proposed openstack/kolla: Fix openstack-base for centos-source https://review.openstack.org/293276 | 06:40 |
| *** ssurana has quit IRC | 06:44 | |
| gbraad | init-runonce needs python-neutronclient to be installed. currently not described in the quickstart | 06:57 |
| *** Marga_ has quit IRC | 07:25 | |
| SiRiuS_ | Martin Andre | 07:27 |
| SiRiuS_ | There is a bug in the Vagrantfile | 07:27 |
| SiRiuS_ | `when "virtualbox_ubuntu"` and `when "virtualbox_centos"` .... should be replaced by just one line `when "virtualbox"` | 07:27 |
| coolsvap | SiRiuS_: i think there is a fix proposed for that today | 07:28 |
| SiRiuS_ | Otherwise the host manager plugin won't function work | 07:28 |
| SiRiuS_ | coolsvap: Oh, OK :) | 07:28 |
| coolsvap | SiRiuS_: https://review.openstack.org/#/c/293179/ | 07:28 |
| mandre | hey SiRiuS_ | 07:29 |
| mandre | I believe I submitted a fix for all the issues your reported | 07:30 |
| coolsvap | mandre: ^^ | 07:30 |
| mandre | SiRiuS_: thanks for the reports by the way | 07:30 |
| SiRiuS_ | manure: I saw, thanks :) | 07:31 |
| mandre | SiRiuS_: next time you find an issue you can report them directly in launchpad so that we don't lose track of them | 07:31 |
| SiRiuS_ | mandre: Ok will do | 07:32 |
| SiRiuS_ | mandre: There is also another issue, but it depends on the hypervisor I think | 07:33 |
| SiRiuS_ | mandre: The problem is with VirtualBox and computers on wi-fi | 07:34 |
| mandre | SiRiuS_: it's good practice to come discuss the issues on IRC like you did, please keep doing it :) | 07:35 |
| SiRiuS_ | mandre: The 3rd (neutron public) interface needs to have promiscuous activated | 07:35 |
| mandre | SiRiuS_: that's entirely depending on your environment | 07:36 |
| SiRiuS_ | mandre: But according to the VirtualBox documentation (and tested by me), promiscuous mode on wi-fi does not work fully | 07:36 |
| SiRiuS_ | mandre: I did a fix to make it work, but yes, I do understand it's environment dependent | 07:37 |
| *** Marga_ has joined #kolla | 07:38 | |
| mandre | did you make a change to the provision script? | 07:38 |
| SiRiuS_ | mandre: no | 07:39 |
| SiRiuS_ | mandre: Here is what I did -> http://pastebin.com/7jnixNrj | 07:39 |
| SiRiuS_ | mandre: I also commented out this line : # config.vm.network "public_network", dev: get_default(:bridge_interface), mode: 'bridge', type: 'bridge' | 07:40 |
| SiRiuS_ | mandre: The idea behind the fix is that in the VirtualBox + wi-fi combination the real physical wi-fi router can't be used, because VirtualBox changes the MAC address | 07:41 |
| *** Marga_ has quit IRC | 07:41 | |
| *** Marga_ has joined #kolla | 07:41 | |
| SiRiuS_ | mandre: So we need to create a gateway on the host, and route all the traffic through it | 07:42 |
| *** Marga_ has quit IRC | 07:42 | |
| SiRiuS_ | mandre: I used the VirtualBox NAT Network service | 07:42 |
| mandre | SiRiuS_: I'm afraid I can't be of much help here, I don't use virtualbox :/ | 07:42 |
| *** Marga_ has joined #kolla | 07:43 | |
| SiRiuS_ | mandre: Since vagrant does not provide a mechanism to do that, It has to be done at a low level. That's why I used vm.customize | 07:43 |
| SiRiuS_ | mandre: Also it's mandatory that vm.customize be used for all the other virtual NICs, even if just t change something insignificant | 07:44 |
| mandre | SiRiuS_, you'll need to find a virtualbox user to test your changes | 07:45 |
| mandre | maybe pbourke? | 07:45 |
| SiRiuS_ | mandre: I tested, and it works for me. The only inconvenience is that, now, the neutron public network, is actually a private network , and to connect to the VM's with the floating IPs we have to do a port forward | 07:47 |
| SiRiuS_ | mandre: The host OS and the VM inside OpenStack with floating IPs, are not in the same network anymore, but this is the only way to get it working on wi-fi | 07:49 |
| mandre | SiRiuS_: oh, I'm sure you've tested and it worked for you, we just need to make sure it works for the general case | 07:50 |
| SiRiuS_ | mandre: Oh yeah, sure :) | 07:51 |
| mandre | sadly I'm not a virtualbox user and I can't confirm the bug | 07:51 |
| mandre | SiRiuS_: please log a bug and send a patch | 07:52 |
| SiRiuS_ | mandre: Acording to VirtualBox documentation, promiscuous mode on wi-fi is a problem of the wi-fi interfaces | 07:52 |
| mandre | it will get reviewed | 07:52 |
| SiRiuS_ | mandre: I never worked in launchpad, is there a quick-start documentation somewhere ? | 07:53 |
| openstackgerrit | Martin André proposed openstack/kolla: Remove oslo-log patch https://review.openstack.org/293294 | 07:55 |
| *** stvnoyes has quit IRC | 08:00 | |
| mandre | SiRiuS_: no worries, we'll guide you | 08:01 |
| *** stvnoyes has joined #kolla | 08:01 | |
| mandre | you'll need to create an account on launchpad.net if you haven't done it already | 08:03 |
| SiRiuS_ | mandre: Ok, I'll first look around and get an idea | 08:04 |
| *** jmccarthy has quit IRC | 08:04 | |
| mandre | also, have a look at http://docs.openstack.org/contributor-guide/first-timers-quickstart.html | 08:04 |
| SiRiuS_ | mandre: Doing that right now :) | 08:04 |
| *** jmccarthy has joined #kolla | 08:04 | |
| *** starmer has joined #kolla | 08:13 | |
| *** mikelk has joined #kolla | 08:21 | |
| *** kbyrne has joined #kolla | 08:22 | |
| openstackgerrit | Merged openstack/kolla: Fix host_manager plugin for vagrant virtualbox https://review.openstack.org/293179 | 08:28 |
| *** starmer has quit IRC | 08:29 | |
| openstackgerrit | MD NADEEM proposed openstack/kolla: Manila ubuntu binary container https://review.openstack.org/286379 | 08:44 |
| openstackgerrit | MD NADEEM proposed openstack/kolla: Manila ubuntu binary container https://review.openstack.org/286379 | 08:44 |
| *** Serlex has joined #kolla | 08:46 | |
| openstackgerrit | Gerard Braad proposed openstack/kolla: Trivial: changed port of docker_registry in globals.yml https://review.openstack.org/293240 | 08:48 |
| *** kproskurin has joined #kolla | 08:55 | |
| *** starmer has joined #kolla | 08:55 | |
| openstackgerrit | Merged openstack/kolla-mesos: Add timestamp to logging and remove useless log string https://review.openstack.org/291692 | 09:03 |
| *** kproskurin has quit IRC | 09:04 | |
| *** coolsvap has quit IRC | 09:07 | |
| *** mbound has joined #kolla | 09:10 | |
| *** salv-orlando has joined #kolla | 09:11 | |
| *** kproskurin has joined #kolla | 09:15 | |
| *** jmccarthy1 has joined #kolla | 09:25 | |
| *** tzn has joined #kolla | 09:31 | |
| *** v1k0d3n has quit IRC | 09:35 | |
| *** v1k0d3n has joined #kolla | 09:36 | |
| *** tfukushima has quit IRC | 09:39 | |
| *** ccesario has quit IRC | 09:45 | |
| *** ccesario has joined #kolla | 09:45 | |
| *** Serlex has quit IRC | 09:48 | |
| *** tfukushima has joined #kolla | 09:48 | |
| *** dims has joined #kolla | 09:51 | |
| *** Serlex has joined #kolla | 09:53 | |
| *** openstackgerrit has quit IRC | 09:53 | |
| *** openstackgerrit_ is now known as openstackgerrit | 09:53 | |
| *** openstackgerrit has quit IRC | 09:53 | |
| *** openstackgerrit_ has joined #kolla | 09:53 | |
| *** openstackgerrit_ is now known as openstackgerrit | 09:54 | |
| *** openstackgerrit_ has joined #kolla | 09:55 | |
| *** SiRiuS_ has quit IRC | 09:55 | |
| *** SiRiuS_ has joined #kolla | 09:55 | |
| *** openstackgerrit has quit IRC | 09:55 | |
| *** openstackgerrit has joined #kolla | 09:56 | |
| *** allen_gao has quit IRC | 10:01 | |
| *** kproskurin has quit IRC | 10:04 | |
| *** allen_gao has joined #kolla | 10:10 | |
| *** pbourke has quit IRC | 10:12 | |
| *** pbourke has joined #kolla | 10:13 | |
| *** gfidente has joined #kolla | 10:13 | |
| *** gfidente has joined #kolla | 10:13 | |
| *** starmer has quit IRC | 10:16 | |
| *** starmer has joined #kolla | 10:17 | |
| *** starmer has quit IRC | 10:22 | |
| *** AndChat|241401 has joined #kolla | 10:24 | |
| *** kproskurin has joined #kolla | 10:26 | |
| *** ccesario has quit IRC | 10:26 | |
| *** akwasnie has joined #kolla | 10:27 | |
| akwasnie | morning | 10:28 |
| *** rhallisey has joined #kolla | 10:31 | |
| *** AndChat|241401 has quit IRC | 10:39 | |
| *** vincent_vdk has quit IRC | 10:40 | |
| *** phuongnh has quit IRC | 10:42 | |
| *** kproskurin has quit IRC | 10:48 | |
| *** kproskurin has joined #kolla | 10:48 | |
| *** dims has quit IRC | 10:53 | |
| *** salv-orlando has quit IRC | 10:56 | |
| *** dims has joined #kolla | 10:59 | |
| *** alisonh has quit IRC | 10:59 | |
| openstackgerrit | Paul Bourke proposed openstack/kolla: Increase the docker parition for rht bases in gate https://review.openstack.org/293031 | 11:00 |
| *** Jeffrey4l has joined #kolla | 11:06 | |
| *** iceyao has joined #kolla | 11:11 | |
| *** tfukushima has quit IRC | 11:12 | |
| *** tfukushima has joined #kolla | 11:13 | |
| *** tfukushima has quit IRC | 11:13 | |
| *** openstackgerrit has quit IRC | 11:18 | |
| *** openstackgerrit has joined #kolla | 11:18 | |
| *** alisonh has joined #kolla | 11:18 | |
| *** akwasnie has quit IRC | 11:30 | |
| *** iceyao has quit IRC | 11:32 | |
| *** alisonh has quit IRC | 11:34 | |
| mlima | morning guys | 11:40 |
| *** alisonh has joined #kolla | 11:40 | |
| mlima | this bug can be closed? https://bugs.launchpad.net/kolla/+bug/1550011 | 11:40 |
| openstack | Launchpad bug 1550011 in kolla " Add the default empty reconfigure.yml file to all roles" [Wishlist,Triaged] - Assigned to Jeffrey Zhang (jeffrey4l) | 11:40 |
| *** dwalsh has joined #kolla | 11:47 | |
| *** allen_gao has quit IRC | 11:50 | |
| *** allen_gao has joined #kolla | 11:51 | |
| *** dave-mccowan has joined #kolla | 11:52 | |
| *** salv-orlando has joined #kolla | 11:57 | |
| ccesario_ | morning! | 12:00 |
| openstackgerrit | Mauricio Lima proposed openstack/kolla: Deploy fails with error in stable/liberty https://review.openstack.org/293390 | 12:05 |
| *** asalkeld has quit IRC | 12:12 | |
| *** JoseMello has joined #kolla | 12:27 | |
| *** salv-orlando has quit IRC | 12:28 | |
| *** vincent_vdk has joined #kolla | 12:31 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Set --ipc option to "host" for kolla_docker https://review.openstack.org/293400 | 12:37 |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Connect to mariadb with dnsname. https://review.openstack.org/290176 | 12:40 |
| *** akwasnie has joined #kolla | 12:44 | |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Switch services to use mesos-dns names https://review.openstack.org/290181 | 12:46 |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement commands list command https://review.openstack.org/293081 | 12:50 |
| *** prithiv has joined #kolla | 12:52 | |
| *** inc0 has joined #kolla | 12:54 | |
| *** inc0_ has joined #kolla | 12:58 | |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement deployment commands https://review.openstack.org/292741 | 12:58 |
| *** inc0 has quit IRC | 12:58 | |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Connect to mariadb with dnsname. https://review.openstack.org/290176 | 12:59 |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Switch services to use mesos-dns names https://review.openstack.org/290181 | 12:59 |
| *** Jeffrey4l has quit IRC | 13:02 | |
| openstackgerrit | Thiago Gomes proposed openstack/kolla: Add Manila deployment document https://review.openstack.org/292588 | 13:11 |
| *** fthiagogv has joined #kolla | 13:13 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Set --ipc option to "host" for kolla_docker https://review.openstack.org/293400 | 13:13 |
| *** Jeffrey4l has joined #kolla | 13:14 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: [WIP] iscsi container with lvm2 support https://review.openstack.org/291285 | 13:15 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: [WIP] iscsi container with lvm2 support https://review.openstack.org/291285 | 13:16 |
| *** iceyao has joined #kolla | 13:17 | |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement commands list command https://review.openstack.org/293081 | 13:24 |
| *** Jeffrey4l has quit IRC | 13:27 | |
| *** salv-orlando has joined #kolla | 13:29 | |
| *** sdake has joined #kolla | 13:33 | |
| *** jtriley has joined #kolla | 13:38 | |
| *** salv-orlando has quit IRC | 13:40 | |
| *** shardy has joined #kolla | 13:42 | |
| *** tzn has quit IRC | 13:45 | |
| sdake | hey shardy | 13:45 |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Connect to mariadb with dnsname. https://review.openstack.org/290176 | 13:48 |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Switch services to use mesos-dns names https://review.openstack.org/290181 | 13:48 |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement deployment commands https://review.openstack.org/292741 | 13:50 |
| *** fthiagogv has quit IRC | 13:50 | |
| SamYaple | morning | 13:51 |
| *** fthiagogv has joined #kolla | 13:51 | |
| *** athomas has quit IRC | 13:51 | |
| SamYaple | mlima: yes | 13:51 |
| *** fthiagogv has quit IRC | 13:51 | |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Connect to mariadb with dnsname. https://review.openstack.org/290176 | 13:51 |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Switch services to use mesos-dns names https://review.openstack.org/290181 | 13:51 |
| *** fthiagogv has joined #kolla | 13:51 | |
| *** fthiagogv has quit IRC | 13:52 | |
| *** fthiagogv has joined #kolla | 13:52 | |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Connect to mariadb with dnsname. https://review.openstack.org/290176 | 13:53 |
| openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Switch services to use mesos-dns names https://review.openstack.org/290181 | 13:53 |
| *** fthiagogv has quit IRC | 13:54 | |
| *** fthiagogv has joined #kolla | 13:54 | |
| *** fthiagogv has quit IRC | 13:55 | |
| *** fthiagogv has joined #kolla | 13:55 | |
| *** athomas has joined #kolla | 13:56 | |
| sbezverk | Good morning | 13:56 |
| sbezverk | SamYaple I removed those "sed" commands | 13:57 |
| sbezverk | it looks like --ipc=host did the magic.. | 13:57 |
| mlima | SamYaple, i was looking some bug in launchpad and i think that these bugs can be closed https://bugs.launchpad.net/kolla/+bug/1546798 | 13:57 |
| openstack | Launchpad bug 1546798 in kolla "neutron failed to deploy in ubuntu binary" [Medium,Triaged] - Assigned to Sam Yaple (s8m) | 13:57 |
| mlima | https://bugs.launchpad.net/kolla/+bug/1552433 | 13:57 |
| mlima | https://bugs.launchpad.net/kolla/+bug/1554780 | 13:57 |
| openstack | Launchpad bug 1552433 in kolla "init-runonce does not pull CirrOS image in fresh node deployment" [Low,Confirmed] - Assigned to Prithiv (prithiv) | 13:57 |
| openstack | Launchpad bug 1554780 in kolla "Update the documentation about installation of curl, nova, neutron clients" [Low,Confirmed] - Assigned to Prithiv (prithiv) | 13:57 |
| SamYaple | sbezverk: oh right! i forgot to tell you abbout that one | 13:57 |
| SamYaple | sbezverk: yea when doing a super container there are a few shared namespaces you want | 13:58 |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement deployment list command https://review.openstack.org/293444 | 13:58 |
| SamYaple | mlima: you have permission to close those bugs, but please close them with a link to the proper review that should have closed them (or mark them as dups of other bugs) | 13:58 |
| SamYaple | mlima: and adjust the owner and branch and what not please | 13:59 |
| sbezverk | SamYaple I submitted 1 line patch to make --ipc=host permanent in kolla_docker.py | 13:59 |
| SamYaple | link? | 13:59 |
| sbezverk | SamYaple https://review.openstack.org/#/c/293400/ | 14:00 |
| sbezverk | I tested it and it works | 14:01 |
| SamYaple | sbezverk: i left a review. please address teh comment and we can merge | 14:01 |
| sbezverk | SamYaple ok | 14:03 |
| *** daneyon has joined #kolla | 14:03 | |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement deployment list command https://review.openstack.org/293444 | 14:06 |
| ccesario_ | Hi guys... only to report - even with ansible - 1.9.5rc1 the problem related there https://bugs.launchpad.net/kolla/+bug/1520728 still happen | 14:06 |
| openstack | Launchpad bug 1520728 in kolla "fatal: [openstack002] => One or more undefined variables: 'dict object' has no attribute 'stdout'" [Critical,In progress] | 14:06 |
| SamYaple | ccesario_: yes its not going ot be fixed inthe 1.9.x branch | 14:08 |
| SamYaple | period, ansible has stated as much | 14:08 |
| ccesario_ | :/ | 14:09 |
| *** absubram has joined #kolla | 14:11 | |
| openstackgerrit | Andrey Pavlov proposed openstack/kolla-mesos: CLI: implement commands list command https://review.openstack.org/293081 | 14:11 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Set --ipc option to "host" for kolla_docker https://review.openstack.org/293400 | 14:11 |
| *** absubram_ has joined #kolla | 14:12 | |
| *** banix has joined #kolla | 14:12 | |
| *** tfukushima has joined #kolla | 14:14 | |
| *** daneyon has quit IRC | 14:15 | |
| *** salv-orlando has joined #kolla | 14:15 | |
| *** absubram has quit IRC | 14:15 | |
| *** absubram_ is now known as absubram | 14:15 | |
| sbezverk | SamYaple your comment has been addressed.. could you please re-check? https://review.openstack.org/293400 | 14:26 |
| *** sdake has quit IRC | 14:27 | |
| *** ayoung has joined #kolla | 14:33 | |
| *** iceyao has quit IRC | 14:36 | |
| *** bugroger has joined #kolla | 14:37 | |
| *** absubram has quit IRC | 14:40 | |
| sbezverk | SamYaple do we need to support ipc_mode change? if not then I do not need to compare current mode, just set it. | 14:41 |
| SamYaple | sbezverk: any container change should be registered, yes | 14:41 |
| *** absubram has joined #kolla | 14:43 | |
| *** daneyon has joined #kolla | 14:46 | |
| inc0_ | akwasnie, ping | 14:47 |
| sbezverk | SamYaple, I tested and by default container has None "" ipc mode, which is the same if I do not specify ipc mode parameter, if I specify it it correctly sets ipc mode to host, see http://paste.openstack.org/show/490721/ | 14:56 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Set --ipc option to "host" for kolla_docker https://review.openstack.org/293400 | 14:57 |
| *** mlima has quit IRC | 15:02 | |
| *** mlima has joined #kolla | 15:05 | |
| *** inc0___ has joined #kolla | 15:08 | |
| *** inc0___ has quit IRC | 15:09 | |
| *** inc0 has joined #kolla | 15:09 | |
| akwasnie | inc0_: hi | 15:10 |
| *** inc0_ has quit IRC | 15:12 | |
| *** mbound has quit IRC | 15:22 | |
| *** sdake has joined #kolla | 15:25 | |
| *** vhosakot has joined #kolla | 15:29 | |
| sdake | sbezverk did you manage to make a ipc model? | 15:33 |
| sdake | rather ipcMode in kolla docker module | 15:33 |
| inc0 | akwasnie, do we have meeting today? | 15:36 |
| inc0 | about prezentation? | 15:36 |
| akwasnie | inc0_: yes | 15:36 |
| SamYaple | inc0: yes 2300 UTC | 15:36 |
| inc0 | SamYaple, not this meeting;) | 15:36 |
| inc0 | but thanks;) | 15:36 |
| SamYaple | oh... | 15:37 |
| inc0 | 22.30UTC akwasnie? on skype or hangouts or here? | 15:37 |
| akwasnie | inc0_: elemoine said that he would like to meet at 4:30 UTC, is it ok for you? | 15:37 |
| inc0 | yeah | 15:37 |
| SamYaple | btw i responded to everyones lightouts concern and have no intention of doing another patch with the comments provided, so please rereview and reiterate exactly what yo uare -1'ing on if you still are | 15:37 |
| *** jasonsb has joined #kolla | 15:37 | |
| SamYaple | https://review.openstack.org/#/c/293161/ | 15:37 |
| *** tfukushima has quit IRC | 15:38 | |
| elemoine_ | inc0: thanks 22:30 UTC is too late for me | 15:38 |
| akwasnie | inc0_: we can start on IRC | 15:38 |
| inc0 | elemoine_, thats what I thought, hence my asking | 15:39 |
| inc0 | so, about 1hr from now then | 15:39 |
| inc0 | ? | 15:39 |
| elemoine_ | perfect | 15:39 |
| akwasnie | yes | 15:39 |
| inc0 | cool, thanks guys | 15:39 |
| sbezverk | sdake yes | 15:40 |
| sbezverk | sdake I submitted patch and needs final reviev.. | 15:40 |
| *** blahRus has joined #kolla | 15:47 | |
| *** akwasnie has quit IRC | 15:48 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Mariadb upgrade https://review.openstack.org/293162 | 15:53 |
| openstackgerrit | Sam Yaple proposed openstack/kolla: MariaDB lights out recovery https://review.openstack.org/293161 | 15:53 |
| openstackgerrit | MD NADEEM proposed openstack/kolla: Fix race condition in haproxy https://review.openstack.org/292782 | 15:56 |
| inc0 | SamYaple, why you not do docs in patch? | 15:59 |
| inc0 | that's where docs should be done | 15:59 |
| SamYaple | inc0: i disagree and tahts why docimpact flag exists | 15:59 |
| SamYaple | people like to nitpick docs and i dont want 30 patchsets with 25 being "two spaces after period please" | 16:00 |
| pbourke | ha | 16:01 |
| SamYaple | dont get me wrong, i think docs _should_ be nitpicked | 16:01 |
| SamYaple | but you know, not in a non-docs patchset | 16:02 |
| inc0 | can you make next patch in dependency tree with docs? | 16:02 |
| SamYaple | nope that docimpact flag will trigger a bug _after_ merge | 16:02 |
| SamYaple | inc0: you ready to switch away from jinja2 else/if and move to includes? | 16:03 |
| SamYaple | i think its time! | 16:03 |
| SamYaple | in netwon that is | 16:03 |
| inc0 | ready - yes, but this is lower priority than other stuff for me | 16:04 |
| inc0 | also saltstack uses jinja? | 16:04 |
| SamYaple | it can use whatever you want, but im talking about the Dockerfile stuff | 16:04 |
| pbourke | SamYaple: yes for includes please | 16:04 |
| SamYaple | thats not related to deploy tool | 16:05 |
| pbourke | thought that should've been done from the start tbh | 16:05 |
| SamYaple | i was the opposition there pbourke | 16:05 |
| inc0 | +1 to pbourke | 16:05 |
| inc0 | from the start | 16:05 |
| SamYaple | people didnt understand jinja2 | 16:05 |
| SamYaple | there was like 3 people who understood jinja2 | 16:05 |
| SamYaple | and our dockerfiles were _not_ uniform | 16:05 |
| inc0 | but not just includes, any feature jinja has to offer | 16:05 |
| SamYaple | now they are and people understand jinja2 thanks to ansible | 16:05 |
| inc0 | if it makes sense | 16:05 |
| pbourke | there is something to be said for one dockerfile per service | 16:05 |
| SamYaple | yea i can get on board with that now inc0 | 16:06 |
| pbourke | but ultimately they're too hard to read | 16:06 |
| SamYaple | well pbourke now we have alot of shared code too | 16:06 |
| pbourke | yeah | 16:06 |
| SamYaple | so there will still be the meat of the content in one file | 16:06 |
| inc0 | but before we start this refactoring | 16:06 |
| inc0 | funcional test plz | 16:06 |
| SamYaple | functional test of what? | 16:07 |
| inc0 | in gates | 16:07 |
| SamYaple | you mean voting gates? | 16:07 |
| SamYaple | because we do functional testing now | 16:07 |
| inc0 | 1 vm spawn is kinda functional testing | 16:07 |
| SamYaple | it uploads to glance, creates networks, adds ssh keys | 16:07 |
| inc0 | but maybe something more advanced would be in order | 16:07 |
| SamYaple | its more than just 1 vm boot | 16:07 |
| inc0 | I know | 16:08 |
| SamYaple | but i agree, it needs more | 16:08 |
| SamYaple | it does show core funcationlity is working though | 16:08 |
| inc0 | and refactoring can hurt if you don't have good CI | 16:08 |
| inc0 | if we start it from day one, that wouldn't be the case | 16:08 |
| inc0 | now we make small bug in jinja and it can break stuff | 16:08 |
| SamYaple | well that we have, since the refactor is all about the building | 16:08 |
| SamYaple | i agree with you inc0 | 16:08 |
| inc0 | but if you forget to install one package | 16:08 |
| inc0 | image will build | 16:08 |
| SamYaple | but this is why we should do it _early_ in newton | 16:08 |
| inc0 | app may even run | 16:09 |
| inc0 | but it will break | 16:09 |
| inc0 | I'd say first good CI, then refactoring | 16:09 |
| inc0 | better CI at least | 16:09 |
| SamYaple | honestly though, thats just presure on the reviewer | 16:09 |
| inc0 | multinode and such | 16:09 |
| SamYaple | we need to refactor to get more gates | 16:09 |
| inc0 | my point is, we made a decision and what we have works | 16:09 |
| SamYaple | infra will give us lots fo gates if we can show only a subset ever run together | 16:10 |
| inc0 | if we change it, it may stop working | 16:10 |
| SamYaple | thats what the initial part of the cycle is for | 16:10 |
| SamYaple | merge then stabilize | 16:10 |
| SamYaple | i dont want to refactor later | 16:10 |
| SamYaple | because what you say is still true | 16:10 |
| SamYaple | no matter the testing | 16:10 |
| SamYaple | there is no 100% converage, it just doesnt exist | 16:11 |
| SamYaple | coverage* rather | 16:11 |
| inc0 | well, my point is | 16:11 |
| inc0 | I'm +2 to refactor | 16:11 |
| *** tzn has joined #kolla | 16:11 | |
| inc0 | but I'd calmer person if we have good gates before that | 16:11 |
| SamYaple | i dont disagree | 16:12 |
| SamYaple | i think we are of like mind | 16:12 |
| sdake | just got out of meeting, reading scrollback | 16:12 |
| *** allen_gao has quit IRC | 16:13 | |
| sdake | so folks re refactor of jinja2 dockerfiless, wfm, but we will come out of summit with a list of priorities and work em in the order we think is appropriate | 16:14 |
| *** allen_gao has joined #kolla | 16:15 | |
| sdake | i would prefer to take a holistic approach to it, rather then saying one feature is mandatory for mitaka-1 | 16:15 |
| sdake | for exmaple, I think reno is #1 on the list | 16:15 |
| inc0 | reno? | 16:15 |
| sdake | but others may not, that is what the austin summit is all about - deciding priorities | 16:15 |
| sdake | reno is the release mangement tool | 16:15 |
| sdake | it lets us add release notes on the fly | 16:16 |
| sdake | and creates sdocs automatically | 16:16 |
| inc0 | a yeah we spoke about it | 16:16 |
| sdake | its highly valuable and allows us to obtain the release:managed tag independently of whatever release model we use | 16:16 |
| sdake | so we no longer have to take responsibility for taggin | 16:16 |
| sdake | of course this depends on a proper ci - where everyone knows our gate is a bit leaky :) | 16:17 |
| sdake | taggin/tagging | 16:17 |
| SamYaple | reno is a single commit. its not a problem | 16:18 |
| sdake | agree, but every commit after requires reno releae notes, atleast blueprints do | 16:18 |
| sdake | so it will be an alteration of our workflow that devs will have to familiarlize themselves with | 16:19 |
| sdake | for example docimpact flag with release notes wont fly ;) | 16:20 |
| *** allen_gao has quit IRC | 16:20 | |
| sdake | the tool uses commit dates to sort out the documentation creatoin on releases.openstack.org | 16:20 |
| SamYaple | youre 100% wrong there sdake | 16:21 |
| *** allen_gao has joined #kolla | 16:23 | |
| *** akwasnie has joined #kolla | 16:24 | |
| inc0 | shots fired;) | 16:25 |
| sbezverk | vhosakot Saw you comment, "" and NOne is it not the same in terms of python? | 16:27 |
| sdake | wrong on what | 16:27 |
| vhosakot | sbezverk: well, I see the other values for "default:" are Python datatypes, aren't they ? | 16:27 |
| SamYaple | sdake: how release notes work with docimpact | 16:27 |
| sdake | dhellmann and I had a 1 hour conversation on the topic | 16:27 |
| SamYaple | http://eavesdrop.openstack.org/irclogs/%23openstack-keystone/%23openstack-keystone.2016-03-15.log.html#t2016-03-15T20:35:00 | 16:27 |
| SamYaple | mind you Sam-I-Am is a docs core, and knows a thing or two | 16:28 |
| sdake | SamYaple i am just speaking of usign docimplact with release notes, not other use cases | 16:31 |
| sdake | release otes have to hit with the commit - or the release notes could end up in the wrong release | 16:31 |
| sdake | if for example someoen forgets to finish the job on the release notes | 16:31 |
| SamYaple | docimpact can still be used with releasenotes | 16:31 |
| sdake | anyway all topics for discussion at summit | 16:31 |
| SamYaple | releasenotes just have to be coupled with commit | 16:32 |
| sdake | i'm not sure reno is smart enough to figure out docimpact couples with the commit | 16:32 |
| sdake | but i'll consult dhelllman about it | 16:32 |
| SamYaple | i dont think you understand the difference though | 16:32 |
| SamYaple | releasenotes are not docs like that. they are yaml files | 16:32 |
| SamYaple | docs that go in our doc/* folder are what the docimpact flag is for | 16:33 |
| sdake | right, ok soudns ike we are on same page | 16:33 |
| SamYaple | ok cool | 16:33 |
| sbezverk | vhosakot have to disagree with you comment pid_mode is done the same way and changing ipc_mode would break consistency | 16:33 |
| sdake | the yaml file needs to be part of the commit is all I'm saing :) | 16:33 |
| SamYaple | yes agree | 16:33 |
| vhosakot | sbezverk: ah ok, let me check something... give me 1 sec | 16:33 |
| sdake | SamYaple can you mediate that problem with the None vs "" in ipc_mode | 16:34 |
| sdake | since you are most familiar with the docker kolla module | 16:34 |
| sdake | rather kolla docker module | 16:34 |
| sdake | sbezverk hae a link review handy? | 16:34 |
| SamYaple | vhosakot: does passing None as the ipc mode break it? | 16:35 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Set --ipc option to "host" for kolla_docker https://review.openstack.org/293400 | 16:35 |
| vhosakot | SamYaple: no not about passing.... aren't default: values Python datatypes ? | 16:35 |
| SamYaple | yes | 16:36 |
| sdake | i thought they were yaml datatypes | 16:36 |
| SamYaple | and in this case the default is None | 16:36 |
| sdake | SamYaple can you leave a comment in the review plz :) | 16:36 |
| vhosakot | SamYaple: so, if t he user does not pass ipc_mode, code thinks it is None ? | 16:36 |
| sdake | irc reviews are hard to track ;) | 16:37 |
| sbezverk | SamYaple it does not break anything, but why one parameters is done one way and second similar by nature parameter by another way?!? No consistency! | 16:37 |
| vhosakot | sbezverk: yes, I agree about | 16:37 |
| SamYaple | sbezverk: thats dokcer-pys problem though | 16:37 |
| SamYaple | vhosakot: correct | 16:37 |
| sdake | use gerrit for reviews plz after sorting out the details so there is a record | 16:37 |
| sdake | sbezverk link this irc discussion in the review - check out the irc logs page | 16:38 |
| sdake | sbezverk nice job on triggering my thought processes around ipc_mode ;) | 16:39 |
| sdake | would have never figured that out solo without you pointing out udev was stuck on a semtimedop | 16:39 |
| sdake | udev/lvm | 16:39 |
| vhosakot | sbezverk: gave +1.... Agreed, this is consistent with the default type of pid_mode. | 16:40 |
| sdake | well before we declare victory lets see what the gate has to say :) | 16:40 |
| sbezverk | sdake really glad that there was a solution satisfying sheeps and wolves | 16:40 |
| sdake | bikeshed ftw ;) | 16:40 |
| sdake | the lvm cat was like "that can't be done" | 16:40 |
| sdake | roflcoptersoysoy ;) | 16:41 |
| vhosakot | sdake: sbezverk: :) cool, ATM, the lvm2 ps is my fav :) great storage iscsi/lvm2 backend for kolla containers | 16:41 |
| SamYaple | vhosakot: itll be a great feature for newton. its one of our oldest bugs | 16:44 |
| sbezverk | sdake lvm folks think that container actually owns all lvm related aspects, where in fact it is just a fancy process wrapper and everything else is ownded by host | 16:44 |
| SamYaple | sbezverk: kernel* | 16:44 |
| SamYaple | lol lvm is just device-mapper wrapper | 16:45 |
| vhosakot | SamYaple: cool | 16:45 |
| nihilifer | we have APAC timezone meering today? | 16:45 |
| sdake | yup odd week | 16:45 |
| sdake | date "+%V" | 16:45 |
| sdake | if you ever need to know the even or odd weeks nihilifer | 16:46 |
| sbezverk | vhosakot few small left which needs to be addressed and then you could probably try this patch on top of master just to see if it works in your test bed.. | 16:47 |
| vhosakot | sbezverk: I am planning to test your PS | 16:47 |
| *** athomas has quit IRC | 16:47 | |
| vhosakot | sbezverk: I have iscsi on my box.. I need to setup lvm2 | 16:47 |
| vhosakot | sbezverk: did you merge some docs about this into maater ? how to setup the host for iscsi/lvm2 ? what kernel modules are needed and blah blah ? | 16:48 |
| vhosakot | master* | 16:48 |
| sbezverk | vhosakot not yet, I still need to find out how to deal without "root" access in kolla_toolbox and cinder_volume.. | 16:49 |
| nihilifer | sdake: nice, thx ;) | 16:49 |
| vhosakot | sbezverk: so, is root access needed just to run few iscsi/lvm2 commands inside the container, or, should the container be _created_ with root user in its Dockerfile ? | 16:52 |
| sdake | the lvm commands should be run as sudo | 16:52 |
| sdake | and sudoers file approraitely modified | 16:53 |
| *** athomas has joined #kolla | 16:53 | |
| vhosakot | sdake: similar to neutron_rootwrap and neutron_sudoers ? | 16:53 |
| sdake | thisis the model our other containers follow and is the best we can do | 16:53 |
| sdake | not similar to neutron_rootwrap but similar to our other sudo operations used throughout our containers | 16:54 |
| vhosakot | sdake: cool | 16:54 |
| *** unicell has quit IRC | 16:56 | |
| SamYaple | sbezverk: vhosakot im a bit confused, what is calling kolla-toolbox in the first place? | 16:56 |
| SamYaple | what task is even needing to use kolla_toolbox | 16:57 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: Add hostname of a node as a lock name. https://review.openstack.org/292878 | 16:59 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: run_once commands locking rework https://review.openstack.org/291693 | 16:59 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: Create separate ZK lock nodes https://review.openstack.org/292877 | 16:59 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: set_state now could use exact path as an arg https://review.openstack.org/292876 | 16:59 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: Change mysql user from test to nobody https://review.openstack.org/293571 | 16:59 |
| *** banix has quit IRC | 17:01 | |
| sdake | 94 people in channel | 17:02 |
| sdake | new record ;-) | 17:02 |
| *** banix has joined #kolla | 17:02 | |
| sdake | speaking of the 95 people in the channel, is it possible to get the openstack-kolla thing going? | 17:02 |
| sdake | i'd do the work myself but not certain how it should be done | 17:02 |
| sdake | SamYaple you mentioned you might have time to take this one | 17:03 |
| openstackgerrit | Merged openstack/kolla: Mariadb upgrade https://review.openstack.org/293162 | 17:08 |
| openstackgerrit | Mauricio Lima proposed openstack/kolla: Remove commented line https://review.openstack.org/293583 | 17:12 |
| *** mikelk has quit IRC | 17:12 | |
| *** akwasnie has quit IRC | 17:15 | |
| *** rajathagasthya has joined #kolla | 17:16 | |
| *** inc0 has quit IRC | 17:16 | |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: Add hostname of a node as a lock name. https://review.openstack.org/292878 | 17:16 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: run_once commands locking rework https://review.openstack.org/291693 | 17:16 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: Create separate ZK lock nodes https://review.openstack.org/292877 | 17:16 |
| openstackgerrit | Proskurin Kirill proposed openstack/kolla-mesos: set_state now could use exact path as an arg https://review.openstack.org/292876 | 17:16 |
| *** kproskurin has quit IRC | 17:24 | |
| *** akwasnie has joined #kolla | 17:24 | |
| *** inc0 has joined #kolla | 17:25 | |
| *** unicell has joined #kolla | 17:27 | |
| *** tzn has quit IRC | 17:27 | |
| *** jasonsb has quit IRC | 17:29 | |
| *** akwasnie has quit IRC | 17:32 | |
| rajathagasthya | Hello! Looking to contribute to diagnostics (EHK) in Kolla. sdake mentioned the need for integrating Kibana dashboard plugin with ES and I'd like to take up that task. | 17:33 |
| rajathagasthya | We've done something similar for Kibana 3. How do I go about doing that with Kibana 4? Maybe akwasnie and elemoine can comment? | 17:33 |
| vhosakot | SamYaple: I think sbezverk is planning to use kolla_toolbox for register / check to check disks, services, etc ? | 17:33 |
| sdake | rajathagasthya elemoine and akwasnie are probably asleep at present, but at our 23:00 UTC meeting would be a good time to sync up on this topic | 17:39 |
| sdake | I think an integrated with OpenStack dashboard would be a fantastic improvement to the implementation - may I suggestyou edit the meeting wiki agenda listed in the topic with your topic? | 17:40 |
| sdake | rajathagasthya ^^ | 17:40 |
| sdake | the meeting is today | 17:40 |
| inc0 | sdake, we just talked about that with akwasne and elemoine | 17:41 |
| inc0 | personally I'd love it as post-deploy task | 17:41 |
| sdake | well it sounds like there are more people wanting to join the party :) | 17:41 |
| sdake | less is more more is not less? :) | 17:41 |
| sdake | inc0 when didy ou discuss it - this morning? | 17:42 |
| sdake | in irc logs? | 17:42 |
| inc0 | sdake, we had priv meeting regarding our presentation | 17:42 |
| inc0 | this just came out | 17:42 |
| inc0 | but I'll be happy to dicuss it in proper meeting | 17:42 |
| sdake | inc0 i've been asking elemoine and ak* to look into it for a about a week | 17:42 |
| *** rajathagasthya has quit IRC | 17:43 | |
| sdake | after elemoine told me about it | 17:43 |
| inc0 | hehe | 17:43 |
| sdake | (that it could be done) | 17:43 |
| inc0 | it's a matter of a single curl | 17:43 |
| sdake | yes I know | 17:43 |
| inc0 | pretty long curl, but curl nonetheless | 17:43 |
| sdake | its also a matter of making a rockin dashboard | 17:43 |
| sdake | json from what I gather | 17:43 |
| *** athomas has quit IRC | 17:52 | |
| *** rajathagasthya has joined #kolla | 17:54 | |
| *** salv-orl_ has joined #kolla | 17:54 | |
| *** salv-orlando has quit IRC | 17:56 | |
| *** akwasnie has joined #kolla | 18:01 | |
| rajathagasthya | sdake - Having some trouble editing the meeting agenda. | 18:03 |
| sdake | give me a topic and I'll put in for you | 18:04 |
| rajathagasthya | "Kibana dashboard integration" | 18:04 |
| *** akwasnie has quit IRC | 18:08 | |
| fthiagogv | sdake, could you look if is okay? https://review.openstack.org/#/c/292588 | 18:14 |
| *** Serlex has left #kolla | 18:17 | |
| *** vhosakot has quit IRC | 18:18 | |
| *** vhosakot_ has joined #kolla | 18:18 | |
| sdake | fthiagogv enjoy thanks for the work :) | 18:29 |
| *** rajathagasthya has quit IRC | 18:34 | |
| openstackgerrit | Merged openstack/kolla: Add Manila deployment document https://review.openstack.org/292588 | 18:34 |
| sbezverk | SamYaple sorry for delay with replay.. kolla_toolbox mounts volumes for swift, as a result it needs visibility in "/dev/{vg-group}/{lv}" | 18:34 |
| sbezverk | when it uses ansible as a user, I could not make it work so far.. | 18:35 |
| sdake | rajath* done :) | 18:36 |
| sdake | sbezverk do as i suggested, put sudoers file in the container and use sudo | 18:36 |
| *** jmccarthy1 has quit IRC | 18:37 | |
| sdake | sbezverk https://github.com/openstack/kolla/blob/master/docker/nova/nova-base/nova_sudoers | 18:39 |
| sdake | sbezverk https://github.com/openstack/kolla/blob/master/docker/nova/nova-base/Dockerfile.j2#L55-L57 | 18:39 |
| *** Marga_ has quit IRC | 18:40 | |
| sbezverk | sdake thanks | 18:46 |
| mlima | root@kolla:~/kolla# neutron net-list | 19:01 |
| mlima | -> Unable to establish connection to http://172.24.56.16:9696/v2.0/networks.json | 19:01 |
| vhosakot_ | Many gate failures not related to kolla today | 19:02 |
| vhosakot_ | Error: Partition(s) on /dev/vdb are being used | 19:02 |
| vhosakot_ | Crash dump was written to: erl_crash.dump | 19:02 |
| vhosakot_ | mlima: Is neutron_server container up ? | 19:05 |
| vhosakot_ | mlima: "docker exec neutron_server netstat -pan | grep 9696" | 19:05 |
| mlima | nothing | 19:06 |
| vhosakot_ | mlima: no output! ? | 19:06 |
| mlima | yes | 19:06 |
| vhosakot_ | mlima: then, neutron is down | 19:06 |
| mlima | control:4000/kollaglue/ubuntu-source-neutron-server:2.0.0 "kolla_start" 8 minutes ago Up 8 minutes neutron_server | 19:06 |
| vhosakot_ | it is up | 19:07 |
| vhosakot_ | not listening to TCP port 9696.. that is not good | 19:07 |
| mlima | but, control:4000/kollaglue/ubuntu-source-neutron-openvswitch-agent:2.0.0 "kolla_start" 8 minutes ago Restarting (1) 45 seconds ago neutron_openvswitch_agent | 19:07 |
| mlima | control:4000/kollaglue/ubuntu-source-rabbitmq:2.0.0 "kolla_start" 10 minutes ago Restarting (1) 3 minutes ago rabbitmq | 19:07 |
| vhosakot_ | mlima: restart all failed neutron services | 19:08 |
| mlima | i have three network interfaces | 19:08 |
| mlima | eth0, eth1 and eth2 | 19:08 |
| SamYaple | sbezverk: kolla_toolbox doesnt mount swift volumes | 19:08 |
| SamYaple | but what does that have to do with lvm | 19:08 |
| vhosakot_ | mlima: multinode or AIO ? | 19:09 |
| mlima | aio | 19:10 |
| *** ayoung has quit IRC | 19:10 | |
| vhosakot_ | mlima: can you pastebin "/etc/kolla/globals.yml", "ifconfig -a", "ip a" and "ps aux | grep docker" | 19:11 |
| mlima | yes | 19:11 |
| mlima | w8 | 19:12 |
| mlima | http://pastebin.com/7gKKyCiV vhosakot_ | 19:14 |
| *** salv-orl_ has quit IRC | 19:15 | |
| mlima | http://pastebin.com/XJsxdNwR vhosakot_ | 19:15 |
| vhosakot_ | you can send all output in one pastebin :) | 19:16 |
| mlima | ok | 19:16 |
| openstackgerrit | Merged openstack/kolla: Set --ipc option to "host" for kolla_docker https://review.openstack.org/293400 | 19:16 |
| *** ssurana has joined #kolla | 19:17 | |
| mlima | enjoy http://pastebin.com/6zGh8fRG vhosakot_ | 19:18 |
| sbezverk | SamYaple, you are right it does not mount them bu it discovers and prepares dis file which is used later for mounting.. The reason this topic came up, is my test bed is configured with swift and I always use lvm volumes for swift.. | 19:19 |
| *** ssurana has quit IRC | 19:22 | |
| *** fthiagogv has quit IRC | 19:22 | |
| *** prithiv has quit IRC | 19:22 | |
| SamYaple | sbezverk: I dont think the permissions block viewing /dev/<vg-group> though do there? | 19:22 |
| vhosakot_ | mlima: your kolla_internal_vip_address in globals.yml is not correct. It is the IP of eth0 which is not right.. please change it to an _unused_ IP address in the 172.24.56.16/24 subnet and rerun "kolla-ansible deploy" | 19:24 |
| vhosakot_ | mlima: also, uncomment kolla_external_vip_address and have kolla_external_vip_address: "{{ kolla_internal_vip_address }}" | 19:26 |
| mlima | uncomment? | 19:26 |
| vhosakot_ | mlima: yes... it is commented | 19:27 |
| mlima | i know, but it is necessary? | 19:27 |
| mlima | globals just replace all.yml, right? | 19:28 |
| vhosakot_ | no, not needed in gloabls... it defaults to kolla_internal_vip_address | 19:29 |
| SamYaple | mlima: globlaes doesnt _replace_ all.yml, it overrides it | 19:29 |
| sdake | mlima you dont need to uncomment kolla-extenral-vip | 19:30 |
| mlima | but in all.yml file i have kolla_external_vip_address: "{{ kolla_internal_vip_address }}" and you said to uncomment it and set kolla_external_vip_address: "{{ kolla_internal_vip_address }}" | 19:30 |
| sdake | if you dont hve an external vip, kolla will do the right hing | 19:30 |
| mlima | yes, yes override is the word | 19:30 |
| mlima | i don't know why i wrote replace | 19:31 |
| SamYaple | mlima: cool just checking | 19:31 |
| mlima | Thank you :) | 19:31 |
| vhosakot_ | mlima: yep, have kolla_external_vip_address in globals if you want to override... so, everything looks goodf except that you are not using an unused IP for kolla_internal_vip_address | 19:32 |
| vhosakot_ | SamYaple: sdake: shouldn't kolla-ansible find this (if the user reuses an existing IP on the box for kolla_internal_vip_address) and display and error and exit ? is this worth an enhacement ? | 19:33 |
| sdake | would make a good precheck | 19:35 |
| *** Marga_ has joined #kolla | 19:36 | |
| vhosakot_ | sdake: ok, let me submit something for that | 19:36 |
| vhosakot_ | mlima: did you see any error from kolla-ansible when you used re-used IP ? I think neutron won't be happy if you start with an existing IP | 19:37 |
| sdake | mlima your external interface should hacve no configuration associated with it | 19:38 |
| sdake | rther your neutron interface | 19:38 |
| vhosakot_ | yes, I saw mlima's pastebin.. neutron_external_interface is pure L2 with no ip no bridge | 19:39 |
| mlima | i have it now kolla_external_vip_address: "{{ kolla_internal_vip_address }}" | 19:39 |
| mlima | :) | 19:39 |
| mlima | deploy is running | 19:40 |
| vhosakot_ | mlima: can you send the error from last time ? | 19:40 |
| mlima | root@kolla:~/kolla# neutron net-list | 19:40 |
| mlima | Unable to establish connection to http://172.24.56.16:9696/v2.0/networks.json | 19:40 |
| mlima | this error ^ | 19:40 |
| vhosakot_ | thanks | 19:41 |
| sdake | mlima you dont want kolla_external_vip_address uncommented unless you have an external vip | 19:42 |
| mlima | it is commented sdake | 19:42 |
| sbezverk | SamYaple I am reverting back to use normal users in container where I had to change to test sudoers to see if it does the job I need and I will show the issue in kolla_toolbox with discovering volumes.. | 19:43 |
| mlima | these files are a mess, i need to organize them | 19:43 |
| *** ssurana has joined #kolla | 19:45 | |
| sdake | mlima which files ? | 19:47 |
| mlima | the globals, in my vm | 19:48 |
| sdake | seem pretty straightforward to me ;) | 19:48 |
| mlima | yes sdake :) | 19:48 |
| mlima | sdake, i want use two network interfaces | 19:51 |
| mlima | it is possible? | 19:51 |
| mlima | eth0,eth1 and eth2 | 19:51 |
| mlima | three network interfaces* | 19:51 |
| ccesario_ | mlima, bond? | 19:54 |
| *** vhosakot_ has quit IRC | 19:54 | |
| sbezverk | sdake even with sudoers, lvcreate does not work http://paste.openstack.org/show/490786/ | 19:59 |
| *** vhosakot has joined #kolla | 19:59 | |
| *** mlima has quit IRC | 20:01 | |
| *** salv-orlando has joined #kolla | 20:01 | |
| vhosakot | mlima: why do you want the third interface eth2 ? as a backup api/neutron interface in case eth1 goes down | 20:02 |
| *** inc0 has quit IRC | 20:06 | |
| sbezverk | sdake if I read this bug right, only root not by design can manage lvm https://bugzilla.redhat.com/show_bug.cgi?id=620571 | 20:08 |
| openstack | bugzilla.redhat.com bug 620571 in lvm2 "non root user can no longer view lvm devices" [Low,Closed: errata] - Assigned to agk | 20:08 |
| sbezverk | sdake sorry what I meant only root now, which is by design can manage lvm devices | 20:09 |
| *** ayoung has joined #kolla | 20:11 | |
| *** bugroger has quit IRC | 20:12 | |
| *** prithiv has joined #kolla | 20:18 | |
| *** ssurana has quit IRC | 20:19 | |
| elmiko | hi folks, i've got a question about redeploying kolla. i wanted to enable a few more services, so i modified the globals.yml, ran `tools/cleanup-containers`, then ran `tools/kolla-ansible deploy` | 20:20 |
| elmiko | unfortunately, i ended up with this: http://paste.openstack.org/show/490789/ | 20:20 |
| elmiko | do i need to run `tools/cleanup-host` as well? | 20:20 |
| vhosakot | elmiko: I have been able to enable services in globals.yml and rerun "kolla-ansible deploy" and it worked.. I did not have to cleanup to enable more services in globals.yml | 20:22 |
| *** ssurana has joined #kolla | 20:22 | |
| elmiko | ah, maybe i should have skipped that | 20:22 |
| elmiko | but in general, shouldn't i be able to cleanup and rerun the deploy? | 20:23 |
| elmiko | (for example, when changing passwords) | 20:23 |
| vhosakot | elmiko: yep, no need to cleanup to enable more services.. this decreases ease of use I think... for me, enabling services in globals.yaml without cleanup and rerunning "kolla-ansible deploy" has worked fine | 20:24 |
| vhosakot | elmiko: I mean, running cleanup everytime a new service needs to be enabled, is a pain for the user | 20:24 |
| elmiko | vhosakot: ok, great. what about if i want to change say the keystone conf and restart it, how would i do that? | 20:25 |
| elmiko | agreed | 20:25 |
| vhosakot | elmiko: reconfig* | 20:25 |
| elmiko | excellent, tyvm vhosakot++ | 20:25 |
| vhosakot | elmiko: wait.. searching a link | 20:25 |
| *** sbezverk has quit IRC | 20:26 | |
| vhosakot | elmiko: I suggest you refer this link to reconfigure keystone - https://etherpad.openstack.org/p/kolla-mitaka-testing-reconfigure | 20:27 |
| *** dencaval has quit IRC | 20:27 | |
| elmiko | excellent! | 20:27 |
| vhosakot | elmiko: :) | 20:27 |
| elmiko | thanks again =) time for more hacking! | 20:27 |
| vhosakot | np | 20:28 |
| *** prithiv has quit IRC | 20:29 | |
| *** diogogmt has joined #kolla | 20:33 | |
| elmiko | vhosakot: one more question, i tried enabling swift and reconfiguring, and i get this http://paste.openstack.org/show/490794/ look familiar? | 20:44 |
| *** banix has quit IRC | 20:45 | |
| vhosakot | elmiko: I haven;t this error | 20:47 |
| elmiko | k, i'm reading through the swift-guide now. maybe i just missed something | 20:47 |
| vhosakot | elmiko: yep, setup all disks, partition tables, rings needed for swift properly | 20:48 |
| elmiko | vhosakot: i'm guessing this is just a config error on my part, sorry for the bother | 20:48 |
| vhosakot | np | 20:49 |
| SamYaple | sdake: back here for a moment since this is not infra related | 21:00 |
| SamYaple | I have been looking into automatic docker builds, and we _can_ do that | 21:00 |
| SamYaple | its is possible | 21:00 |
| SamYaple | rhallisey: jpeeler what are your thoughts on that? ^^ | 21:02 |
| jpeeler | automatic docker builds? | 21:03 |
| SamYaple | yes | 21:03 |
| jpeeler | i talked about doing that long ago so that we wouldn't have to worry about it doing it on the gate infrastructure | 21:03 |
| SamYaple | we cant do them for master | 21:03 |
| SamYaple | but we could do them for stable | 21:03 |
| sdake | you mean separate repositories per package for building? | 21:04 |
| SamYaple | yes and no, not a repo we commit to per packge for building | 21:04 |
| jpeeler | SamYaple: why would master be excluded? | 21:04 |
| SamYaple | jpeeler: the way im thinking of setting it up | 21:04 |
| SamYaple | remember automated builds mean we have no infrastrucutre | 21:04 |
| SamYaple | as in no build.py | 21:04 |
| *** achanda has joined #kolla | 21:05 | |
| jpeeler | true | 21:05 |
| SamYaple | it has to be a non-jinja2 dockerfile with all tarballs in repo | 21:05 |
| SamYaple | which we could do in an automated fashion based on tags | 21:05 |
| SamYaple | but it definetely exlcludes master | 21:05 |
| jpeeler | just because of potential breakage? | 21:05 |
| SamYaple | it would never be up to date | 21:06 |
| SamYaple | i can setup a POC for this trigger/autobuilding process | 21:06 |
| sdake | austin summit disucssion imo ;) | 21:06 |
| jpeeler | heh it would be if people stopped committing for sometime | 21:06 |
| SamYaple | no it wouldnt because other proects | 21:06 |
| SamYaple | remember we pull master tarballs | 21:06 |
| jpeeler | yeah... | 21:07 |
| jpeeler | fair enough. | 21:07 |
| jpeeler | but yes, automated builds for stable would be great in my opinion | 21:07 |
| SamYaple | i think we can do it fairly easily, it would put the load on github and dockerhub | 21:08 |
| sdake | to use the docker infrastructure reuqires separate repositories | 21:08 |
| SamYaple | sdake: yes and no | 21:08 |
| SamYaple | yes it would require seperate repos, no we would never submit patches to them directly | 21:08 |
| SamYaple | or with infra resources | 21:08 |
| jpeeler | how would those repos be synced | 21:09 |
| SamYaple | one of us can trigger it | 21:09 |
| sdake | i can see where your headed sam and i'm not one to stop you, but i'd like stable fixed first ;) | 21:09 |
| SamYaple | sdake: you are welcome to jump in at any time and do reviews (and even patchsets!) as i state everytime you say that | 21:11 |
| sdake | i was waiting for you to finish the job ;) | 21:11 |
| sdake | but if you want reviews now I'm happy to provide them | 21:12 |
| sdake | april 1 is what we agreed for an rc | 21:12 |
| SamYaple | as i have stated everytime, yes please review | 21:12 |
| sdake | ok will do | 21:12 |
| sdake | it is hard for more then one person to do the patches, but i'm willing to contribute to the backporting process as well | 21:13 |
| sdake | if you need help all you have to do is ask :) | 21:13 |
| SamYaple | well i ahve been asking for reviews, you seem to keep forgeting that because we have this conversation everytime | 21:14 |
| sdake | ack on reviews | 21:14 |
| sdake | i'll get the CR team on it | 21:14 |
| SamYaple | theres only 3 patches so far so should be no problem | 21:15 |
| sdake | yes but we need sustained reviews for the life of the patchset | 21:15 |
| sdake | i'll escalate this to the core reviewer team | 21:15 |
| *** sbezverk has joined #kolla | 21:15 | |
| sdake | so we can get a nice stable liberty by april 1 | 21:15 |
| SamYaple | just remember, its green gate on final patch, not neccsarrily before | 21:16 |
| SamYaple | ive blocked the initial patch so were safe to +2+1 | 21:16 |
| sdake | yup i know that - that was discussed in the first comment | 21:16 |
| *** ntpttr has joined #kolla | 21:20 | |
| ntpttr | hey everyone, I'm running into an issue deploying kolla on multiple nodes. In the bootstrap.yml play of the neutron setup, the 'creating neutron database' step is getting skipped, causing 'read json from variable' to fail because the database variable doesn't exist | 21:21 |
| SamYaple | ntpttr: can you post your inventory file | 21:21 |
| ntpttr | is there a setting that I might be missing causing this to get skipped? the precheck all works and everything works up to that point | 21:21 |
| ntpttr | sure | 21:21 |
| SamYaple | ccesario_: i think this is your problem as well ^^ | 21:22 |
| *** inc0 has joined #kolla | 21:22 | |
| ntpttr | here's the top of the file, the rest is unchanged from the default http://paste.openstack.org/show/490801/ | 21:22 |
| SamYaple | ntpttr: yea ok im going to point you to a file and i want you to make a change and see if it fixes it for you | 21:23 |
| ntpttr | SamYaple: sounds good | 21:23 |
| SamYaple | ntpttr: https://github.com/openstack/kolla/blob/master/ansible/site.yml#L107 | 21:23 |
| SamYaple | move neutron-server above compute in that list | 21:23 |
| SamYaple | if that fixes this it is an ansible bug you are running into | 21:23 |
| ntpttr | so just put it up at the top of the hosts list? | 21:24 |
| *** achanda has quit IRC | 21:24 | |
| ntpttr | I feel like you're right, because in the create step, it's skipping over a host that I don't think should be a neutron-server host | 21:24 |
| SamYaple | ntpttr: no in the site.yml file | 21:24 |
| SamYaple | that list, just move neutron-server above compute | 21:25 |
| ntpttr | yeah, that's what I meant sorry | 21:25 |
| SamYaple | yup | 21:25 |
| SamYaple | if that fixes it for you this is an ansible bug thats not fixed until 2.0 | 21:25 |
| ntpttr | SamYaple: okay, but kolla doesn't work with ansible 2 does it? | 21:26 |
| *** achanda has joined #kolla | 21:28 | |
| *** shardy has quit IRC | 21:28 | |
| SamYaple | ntpttr: it will not for mitaka, no | 21:28 |
| SamYaple | ntpttr: what ive described for you is the workaround | 21:28 |
| ntpttr | sounds good | 21:28 |
| SamYaple | if your inventory gets even fancier, you may run into a few more issues like that, all fixed in ansible 2.0 only | 21:29 |
| SamYaple | just fyi | 21:29 |
| ntpttr | SamYaple: still failing at that same spot | 21:31 |
| *** dwalsh has quit IRC | 21:31 | |
| ntpttr | skipping: [172.16.2.6] | 21:32 |
| ntpttr | that host should be 172.16.2.3 if it were using the neutron-server group | 21:32 |
| SamYaple | ntpttr: did you pip install kolla? | 21:34 |
| ntpttr | SamYaple: ah do I need to redo that after making changes? | 21:34 |
| SamYaple | it sounds like youre still running with the old site.yml | 21:34 |
| ntpttr | SamYaple: I'll reinstall and try again | 21:34 |
| SamYaple | _if_ you pip install'd kolla then yes, youll need to reintall it after making git changes | 21:34 |
| ntpttr | unrelated, but is there no script to run the cleanup-containers and cleanup-hosts scrips on all the target hosts? sshing to each one and running them manually is a little slow | 21:35 |
| SamYaple | alternatively you could also make th changes directly to /usr/lib/python* | 21:35 |
| inc0 | ntpttr, you have ansible:) | 21:35 |
| SamYaple | ntpttr: there was some work to do that in the repo, but i havent seen it for a while | 21:35 |
| *** achanda has quit IRC | 21:36 | |
| SamYaple | inc0: did you see the conversation above about the automated docker builds | 21:36 |
| inc0 | ansible -i inventory -m shell -a "tools/cleanup-containers" | 21:36 |
| ntpttr | inc0: I know, I could write my own ansible script, but it'd be cool if there was one built in to kolla | 21:36 |
| inc0 | SamYaple, negative, I did not | 21:36 |
| ntpttr | cool thanks | 21:36 |
| SamYaple | ntpttr: there will be, but proably not mitaka | 21:36 |
| SamYaple | ntpttr: i could help you if you wanted to take on that work yourself | 21:36 |
| sdake | /query inc0 | 21:37 |
| ntpttr | SamYaple: I might be interested in that for sure, I would need to talk to managers about work assignments and such if I were going to spend some time on kolla though | 21:37 |
| SamYaple | inc0: he is looking into you! | 21:37 |
| ntpttr | SamYaple: I'm definitely interested, so far it seems like a great tool | 21:37 |
| SamYaple | ntpttr: cool just let me know if you want any guidance and ill point you the right way | 21:38 |
| ntpttr | SamYaple: thanks | 21:38 |
| SamYaple | inc0: i think i can get automated builds going in the dockerhub itself | 21:38 |
| *** jtriley has quit IRC | 21:39 | |
| vhosakot | sdake: read the first comment in https://bugs.launchpad.net/heat/+bug/1471870 | 21:44 |
| openstack | Launchpad bug 1471870 in heat "Kilo Heat Authorization Failed" [Medium,Triaged] | 21:44 |
| vhosakot | sdake: I'm adding all the info I have into the heat DOA bug | 21:45 |
| sdake | are you working on that bug vhosakot ? | 21:46 |
| vhosakot | sdake: no, I dug into a bit for a day and adding info for whoever is working on it | 21:47 |
| *** achanda has joined #kolla | 21:47 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Add generate.py to generate passwords https://review.openstack.org/293728 | 21:47 |
| sdake | ya the heat domain keystone setup script is pretty much what we need but we need it to be broken out into a separate set of steps | 21:47 |
| sdake | i'll take care of that bug | 21:48 |
| vhosakot | sdake: cool | 21:48 |
| *** absubram has quit IRC | 21:49 | |
| *** absubram has joined #kolla | 21:50 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: update openrc file for keystone v3 https://review.openstack.org/293732 | 21:54 |
| *** asalkeld has joined #kolla | 21:55 | |
| SamYaple | so good news everyone, it was a pita, but i successfully changed all of my passwords on a running openstack cluster | 21:59 |
| SamYaple | there was downtime involved, but it is doable | 21:59 |
| vhosakot | meeting ? | 22:01 |
| SamYaple | vhosakot: one hour | 22:01 |
| SamYaple | 2300 UTC | 22:02 |
| SamYaple | (you forgot about DSL probably) | 22:02 |
| vhosakot | ah ok | 22:02 |
| vhosakot | :) | 22:02 |
| ntpttr | SamYaple: success, the workaround worked | 22:03 |
| ntpttr | SamYaple: thanks! | 22:04 |
| SamYaple | ntpttr: cool. sorry for the trouble. it really is an ansible bug | 22:04 |
| ntpttr | SamYaple: yeah that's definitely clear if changing the order of hosts is what fixed it | 22:05 |
| sdake | SamYaple password change on running openstack is sweet | 22:06 |
| sdake | need a blueprint for that | 22:06 |
| SamYaple | ntpttr: yea still its a bit annoying | 22:06 |
| sdake | sarbanes oxley requirees password changes every 3 months | 22:06 |
| SamYaple | hopefully that alone doesnt taint your kolla experince! | 22:06 |
| SamYaple | sdake: for that to happen we have to switch to shade modules entirely | 22:07 |
| openstackgerrit | Merged openstack/kolla: Fix stable/liberty gate https://review.openstack.org/293189 | 22:07 |
| SamYaple | not all shade modules exist for what we need yet | 22:07 |
| sdake | SamYaple ya this is work for newton | 22:07 |
| SamYaple | i have several patches into shade and ansible to address this | 22:07 |
| SamYaple | but mitaka and liberty will remain PITA to update passwords | 22:07 |
| ntpttr | SamYaple: so my deployment just finished, running kolla/tools/init-runonce is telling me " | 22:07 |
| ntpttr | neutron not installed. please install neutron before proceeding". Do I need to run that command on a network node? | 22:07 |
| SamYaple | ntpttr: pip install neutron | 22:08 |
| SamYaple | its just expecting clients on your host | 22:08 |
| *** vhosakot has quit IRC | 22:08 | |
| sdake | no | 22:08 |
| SamYaple | kolla will not touch your host | 22:08 |
| sdake | pip install python-neutronclient | 22:08 |
| ntpttr | SamYaple: oh makes sense! | 22:08 |
| SamYaple | yea thats right what sdake said ntpttr | 22:08 |
| SamYaple | it just needs the openstack clients is all | 22:08 |
| ntpttr | I already did the pip install python-openstackclient | 22:08 |
| SamYaple | client(s0 | 22:08 |
| ntpttr | I need to do the individual service ones as well? | 22:08 |
| sdake | that doesn't get em all | 22:08 |
| sdake | osc doesn't yet have all the clients | 22:09 |
| ntpttr | cool | 22:09 |
| SamYaple | as in python-novaclient, python-neutronclient etc | 22:09 |
| sdake | you need nova and neutronclients to run init-runonce | 22:09 |
| SamYaple | sdake: it _does_ actually do everything our init script does | 22:09 |
| sdake | we coud probably port it to osc, but have not done so yet | 22:09 |
| SamYaple | we just have to update teh init script | 22:09 |
| sdake | right | 22:09 |
| sdake | definately dont install python-keystoneclient | 22:09 |
| sdake | that wont work at all ;) | 22:10 |
| *** ccesario has joined #kolla | 22:10 | |
| SamYaple | correct. we removed all v2 keystone bits in mitaka | 22:10 |
| ntpttr | cool, so now aftetr running it it runs, but after creating the glance image I get | 22:10 |
| ntpttr | Configuring neutron. | 22:10 |
| ntpttr | <html><body><h1>503 Service Unavailable</h1> | 22:10 |
| ntpttr | No server is available to handle this request. | 22:10 |
| ntpttr | </body></html> | 22:10 |
| SamYaple | ntpttr: youll need to do some troubleshooting | 22:10 |
| SamYaple | neutron agents-list is where id start | 22:10 |
| ntpttr | all right, thanks for all the help! | 22:11 |
| SamYaple | ntpttr: most commonly at this stage (right after initial deployment) there may have been a rabbitmq issue | 22:11 |
| SamYaple | so you might have to do some service restarting, but the logs and normal troubleshooting steps apply here | 22:11 |
| ntpttr | yeah even issuing any neutron command is telling me no server is available | 22:11 |
| ntpttr | and nova service-list works, but says all my services are enabled but down | 22:12 |
| inc0 | rabbit | 22:12 |
| inc0 | definetly | 22:12 |
| SamYaple | sounds like rabbit yea | 22:12 |
| SamYaple | ntpttr: check the logs for nova or neutron | 22:12 |
| SamYaple | also the normal rabbitmq commands for debugging | 22:12 |
| SamYaple | docker exec -it rabbitmq rabbitmqctl cluster_status | 22:13 |
| SamYaple | I would recommend making an alias for commands like that, its what i do | 22:13 |
| SamYaple | alias rabbitmqctl="docker exec -it rabbitmq rabbitmqctl" | 22:13 |
| SamYaple | that way you can do `rabbitmqctl cluster_status` | 22:13 |
| inc0 | also docker exec -it rabbitmq bash -> then navigate to /var/log/kolla | 22:13 |
| inc0 | and look there | 22:13 |
| ntpttr | do rabbit failures happen ofter after installation? | 22:13 |
| SamYaple | ntpttr: only when networking on the backend is complicated | 22:14 |
| SamYaple | they never happen for me, but its 2 switch setup | 22:14 |
| *** SiRiuS_ has quit IRC | 22:14 | |
| ntpttr | yeah our networking is a little complicated, the networks are all vlan tagged and there's a lot of them | 22:14 |
| SamYaple | really what we need to do is check rabbitmqs health before proceeding with the role to catch that early | 22:15 |
| *** prithiv has joined #kolla | 22:15 | |
| SamYaple | inc0: could you file a bug for that since you just did the rabbitmq work? | 22:15 |
| SamYaple | or implement that :P | 22:15 |
| ntpttr | yep here's the rabbit cluster status | 22:15 |
| ntpttr | http://paste.openstack.org/show/490808/ | 22:15 |
| ntpttr | connection refused | 22:15 |
| SamYaple | inc0: i was thinking about recommending people look directly at /var/lib/docker/volume/kolla_logs/_data directly for logs | 22:15 |
| inc0 | ntpttr, do you have all hostnames resolvable by hosts? | 22:16 |
| SamYaple | ntpttr: right. so likely this is the hostname issue inc0 is resolving as we speak | 22:16 |
| SamYaple | we are using ips because somepeople dont have resolveable hostnames thorught there cluster | 22:16 |
| ntpttr | inc0: I used IP addresses | 22:16 |
| SamYaple | ntpttr: for now, you can update all your /etc/hosts files with all the hostnames and ip adresses | 22:16 |
| *** dims_ has joined #kolla | 22:16 | |
| ntpttr | and yeah they're all resolvable | 22:16 |
| SamYaple | ntpttr: THIS specifically is a kolla thing | 22:16 |
| *** prithiv has quit IRC | 22:16 | |
| *** prithiv has joined #kolla | 22:16 | |
| SamYaple | ntpttr: you can resolve 'jfz1r02h03' on all boxes? | 22:17 |
| ntpttr | not sure why it tried to connect to that hostname | 22:17 |
| inc0 | ntpttr, it uses hostname | 22:17 |
| SamYaple | ntpttr: yea this is a kolla issue you see, inc0 should have a patch momemtarily | 22:17 |
| inc0 | that's a bug in kolla, we're fixing it now so it will use ips instead | 22:17 |
| ntpttr | there you have it then, I thought that'd be okay because I used IPs in my inventory file | 22:17 |
| SamYaple | yea thats on us ntpttr | 22:18 |
| SamYaple | workaround is add all your hosts to /etc/hosts file | 22:18 |
| ntpttr | haha no problem, there's always work to do | 22:18 |
| *** dims has quit IRC | 22:18 | |
| ntpttr | I'll add all the hosts to the files and reinstall | 22:18 |
| inc0 | remember to run clear before | 22:18 |
| ntpttr | the plus side is that installation is so quick that hardly matters ;) | 22:18 |
| SamYaple | you shouldnt need to rerun anything ntpttr | 22:18 |
| ntpttr | yep | 22:18 |
| SamYaple | it should recover | 22:18 |
| ntpttr | oh nice! | 22:18 |
| SamYaple | you may need to restart the containers but thats it | 22:18 |
| SamYaple | at this point its a 100% operational openstack cloud.... with a rabbitmq issue (so like most clouds) | 22:19 |
| SamYaple | basic troubleshooting applies | 22:19 |
| SamYaple | ntpttr: i am a bit worried your neutron database didnt truly get bootstrapped because of that issue | 22:22 |
| SamYaple | would you mind doing a 'show tables;' on the database to confirm it is populated | 22:22 |
| sdake | bootstrapping isn't as robust as i'd like | 22:23 |
| sdake | I hope upgrade migration is as robust as i'd like | 22:23 |
| sdake | since bootstrap happens once, upgrade happens often | 22:23 |
| SamYaple | in this case the upgrade script is my recommendation to fix bootstrap | 22:24 |
| SamYaple | since it will db_sync | 22:24 |
| SamYaple | thats only if tables are missing | 22:24 |
| ntpttr | sure, one moment almost done filling in /etc/hosts | 22:24 |
| SamYaple | ntpttr: > not using ansible to populate all hosts | 22:24 |
| SamYaple | ntpttr: > 2016 | 22:24 |
| ntpttr | SamYaple: :( you caught me, I'm still an ansible noob | 22:25 |
| ntpttr | I should do it with ansible for more practice for sure | 22:25 |
| SamYaple | haha im just messing around | 22:25 |
| SamYaple | i wouldnt have either, i would have used tmux with a 9-way split | 22:25 |
| ntpttr | hah, it's true tho | 22:25 |
| *** daneyon has quit IRC | 22:25 | |
| ntpttr | SamYaple: so do I need to attach to a container to run that show tables command? how do I connect to the database | 22:27 |
| SamYaple | same as before, docker exec -it mariadb mysql -h<kolla_internal_vip> -p<database_password> | 22:27 |
| SamYaple | you can also use any mysql client from your host system | 22:27 |
| SamYaple | but we dont install those | 22:27 |
| SamYaple | i personally use an alias there to as well | 22:28 |
| SamYaple | alias mysql='docker exec -it mariadb mysql' | 22:28 |
| ntpttr | kk, yeah I definitely should use those aliases | 22:28 |
| SamYaple | ill post my list | 22:28 |
| SamYaple | ntpttr: http://paste.openstack.org/show/490811/ | 22:29 |
| ntpttr | SamYaple: thanks. Is this what you were looking for? http://paste.openstack.org/show/490812/ | 22:30 |
| SamYaple | ntpttr: hmm im not sure | 22:31 |
| SamYaple | i dont think so | 22:31 |
| ntpttr | I think I got the password wrong, what's the default | 22:31 |
| SamYaple | password :P | 22:31 |
| SamYaple | docker exec -it mariadb mysql -h<kolla_internal_vip> -p<database_password> -e 'show tables;' neutron | 22:32 |
| SamYaple | that should either return a list of tables, or nothign | 22:32 |
| SamYaple | if its nothing it hasnt been bootstrapped | 22:32 |
| SamYaple | as you can tell, kolla modifies your workflow a bit. but with aliases its largely unnoticable | 22:32 |
| sdake | two bags of chips = sdakes meal for the day | 22:32 |
| ntpttr | yeah I bet it wouldn't be too much to get used to | 22:33 |
| ntpttr | mysql doesn't seem to like that command though | 22:34 |
| ntpttr | docker exec -it mariadb mysql -h 172.16.2.203 -p password -e 'show tables;' neutron | 22:34 |
| SamYaple | no space between -p and password | 22:34 |
| SamYaple | docker exec -it mariadb mysql -h 172.16.2.203 -ppassword -e 'show tables;' neutron | 22:34 |
| SamYaple | its a mysql client thing, not a docker thing | 22:34 |
| inc0 | no space after -h I think too | 22:35 |
| ntpttr | gotcha. access denied :( I didn't change the defaults | 22:35 |
| SamYaple | check /etc/kolla/passwords.yml on the deploy node | 22:35 |
| sdake | ya bootstrap = fragile | 22:35 |
| inc0 | .203 is your floating | 22:35 |
| inc0 | ? | 22:35 |
| SamYaple | sdake: thats not whats going on | 22:35 |
| ntpttr | yeah I had gotten rid of the space after h | 22:35 |
| ntpttr | yeah that's it inc0 | 22:35 |
| SamYaple | recommend checking passwords.yml to confirm password. if you ran the playbooks to neutron then you connected to the db with the password in passwords.yml | 22:36 |
| ntpttr | the database password is indeed password | 22:36 |
| SamYaple | oh sorry ntpttr | 22:36 |
| SamYaple | docker exec -it mariadb mysql -h 172.16.2.203 -ppassword -u root -e 'show tables;' neutron | 22:36 |
| SamYaple | we dropped the root user out so it was connecting with wrong user | 22:37 |
| ntpttr | that did it! it's filled in | 22:37 |
| SamYaple | ok cool | 22:37 |
| SamYaple | all that work to say "youre good dont worry about it" :P | 22:37 |
| ntpttr | rabbit cluster is still down though | 22:37 |
| SamYaple | did you restart the nodes? | 22:37 |
| SamYaple | rabbitmq containers* | 22:37 |
| SamYaple | docker restart rabbitmq | 22:37 |
| inc0 | you need to redeploy rabbit | 22:37 |
| inc0 | /etc/hosts didnt get copied | 22:38 |
| SamYaple | you shouldnt need to redeploy rabbitmq | 22:38 |
| SamYaple | oh right yes inc0 is right | 22:38 |
| SamYaple | /etc/hosts in the container is wrong | 22:38 |
| ntpttr | to redeploy one service do i manually delete the containers | 22:38 |
| inc0 | yeah | 22:38 |
| SamYaple | ntpttr: `docker rm -f rabbitmq` on all your nodes running rabbitmq | 22:38 |
| *** JoseMello has quit IRC | 22:39 | |
| SamYaple | then jsut rerun playbooks | 22:39 |
| ntpttr | or can I pass something in to the cleanup containers | 22:39 |
| ntpttr | okay got it | 22:39 |
| SamYaple | sorry ntpttr we arent that fancy yet :) | 22:39 |
| inc0 | then kolla-ansible deploy -t rabbitmq | 22:39 |
| inc0 | it will just deploy rabbitmq | 22:39 |
| SamYaple | if you want to limit yes, you can tag just rabbitmq | 22:39 |
| *** v1k0d3n has quit IRC | 22:39 | |
| *** v1k0d3n has joined #kolla | 22:39 | |
| *** spisarski has quit IRC | 22:40 | |
| ntpttr | okay the playbook is off and running | 22:40 |
| ntpttr | how long until the nodes should come up, it still says down after redeploying | 22:41 |
| inc0 | nova nodes? | 22:41 |
| sdake | ntpttr fwiw I had a demo fail because of a wireless drop out - and i ceanup-containers and redeployed in under 4 minutes :) | 22:41 |
| ntpttr | sdake: I'm thinking of doing that haha. The speed is definitely a big plus | 22:42 |
| sdake | ntpttr it is alot faster to use overlayfs backend storage driver too | 22:42 |
| inc0 | aio for me take ~4 min for full deploy | 22:42 |
| sdake | dave-mccowan said 8x faster | 22:42 |
| ntpttr | sdake: where do I configure that? | 22:42 |
| sdake | aio takes 4 minutes for me to deploy on overlay | 22:42 |
| sdake | ntpttr its a docker option -s overlay | 22:43 |
| sdake | ntpttr depends what your distro is | 22:43 |
| inc0 | you'd need to rebuild all of containers | 22:43 |
| sdake | but you would have to reubild your images | 22:43 |
| *** v1k0d3n_ has joined #kolla | 22:43 | |
| ntpttr | sdake: ubuntu 14.04 | 22:43 |
| ntpttr | I don't mind rebuilding | 22:43 |
| SamYaple | ntpttr: if you want to redeploy i wont stop you | 22:43 |
| sdake | unless your containers are in a docker registry | 22:43 |
| inc0 | also for overlay I think you need to create overlay device | 22:43 |
| sdake | inc0 you dont have to do naything for overlay | 22:43 |
| SamYaple | though i assure you you _can_ recover what you have without much work | 22:43 |
| sdake | inc0 it just uses the filesystem in place | 22:43 |
| SamYaple | *only ext4 | 22:43 |
| inc0 | ok | 22:44 |
| sdake | ok didn't know that | 22:44 |
| sdake | i am on ext4 | 22:44 |
| SamYaple | yup | 22:44 |
| SamYaple | its the only one thats supports overlag in docker atm | 22:44 |
| *** v1k0d3n has quit IRC | 22:44 | |
| inc0 | overlag doesn't sound promissing :P | 22:44 |
| SamYaple | WE REQUIRE MORE BANDWIDTH | 22:44 |
| sdake | ext4 works well enough ;-) | 22:44 |
| sdake | 980 down 960gbit up ;) | 22:45 |
| sdake | g/m | 22:45 |
| SamYaple | you said you couldn't break 600mbit on your nodes | 22:45 |
| sdake | 6000mbit | 22:46 |
| sdake | but my switch isn't configured for jumbo frames | 22:46 |
| sdake | or optimized at all | 22:46 |
| sdake | but my internet connection is nearly gige | 22:47 |
| sdake | i only have 8 ports of 10gig | 22:47 |
| SamYaple | so where did you pull the 980 from? | 22:47 |
| sdake | that is what speedtest shows | 22:47 |
| sdake | that website thing | 22:47 |
| sdake | on my workstation connected over gige | 22:48 |
| sdake | maybe its kbit? | 22:48 |
| sdake | not sure | 22:48 |
| sdake | don't recall at present | 22:48 |
| sdake | 980kbit would be gige right? | 22:48 |
| sdake | if so, its 980kbit ;) | 22:48 |
| SamYaple | 980mbit* | 22:48 |
| inc0 | no...it would be meg;) | 22:48 |
| sdake | right | 22:48 |
| sdake | well thats whati've got | 22:48 |
| sdake | my lab mahcines do 6000mbit between them | 22:49 |
| sdake | not 600mbit | 22:49 |
| inc0 | ntpttr, so, 4 minutes passed | 22:49 |
| inc0 | should be deployed by now: | 22:49 |
| inc0 | :P | 22:49 |
| ntpttr | inc0: but I'm deploying on 10 nodes not one | 22:49 |
| sdake | ntpttr would love o have benchmarks on 10 node deploy | 22:49 |
| sdake | i dont have that much gear | 22:49 |
| inc0 | ntpttr next time you can do kolla-ansible pull | 22:49 |
| *** Jeffrey4l has joined #kolla | 22:49 | |
| inc0 | so it wont transfer images over net any more | 22:49 |
| sdake | Jeffrey4l sup dude | 22:49 |
| SamYaple | did you rebuild the images ntpttr or something? | 22:50 |
| *** sbezverk has quit IRC | 22:50 | |
| ntpttr | SamYaple: no they're in a registry | 22:50 |
| sdake | aio takes 4 minutes - 10 node deploy takes ???? -> profit | 22:50 |
| inc0 | yeah on second thought they should be pulled anyway | 22:50 |
| inc0 | shouldn't be much more | 22:51 |
| Jeffrey4l | whats ntpttr? sdake , do not get your point. | 22:51 |
| sdake | Jeffrey4l its a joke | 22:51 |
| sdake | from where I dont know | 22:51 |
| sdake | something about underpants gnomes | 22:51 |
| sdake | its a US thing | 22:51 |
| *** gfidente has quit IRC | 22:51 | |
| inc0 | right now I'm even afraid to ask | 22:51 |
| Jeffrey4l | :D. just wake up | 22:51 |
| ntpttr | ayy it's done, and it seems to be working | 22:52 |
| sdake | Jeffrey4l https://en.wikipedia.org/wiki/Gnomes_(South_Park) | 22:52 |
| sdake | ntpttr next time you do a 10 node deploy can you run time before the kolla-ansible command? | 22:52 |
| sdake | time sudo kolla-ansible deploy | 22:52 |
| sdake | i'm curious how long it takes | 22:52 |
| ntpttr | sdake: for sure. I would say about 7-10 minutes | 22:52 |
| sdake | super super curios | 22:52 |
| sdake | cool thats fantastic info | 22:52 |
| sdake | thanks! | 22:52 |
| SamYaple | ntpttr: awesomeness | 22:53 |
| Jeffrey4l | that's cool ntpttr | 22:53 |
| sdake | hey guys - t-7 minute seto meeting, to not kill time can folks please update their blueprints status | 22:53 |
| ntpttr | tbh this is super refreshing, I literally strggled for a month getting OSA to work with all the networking preconfig stuff you have to do, this was up and running in one day with no previous kolla experience | 22:54 |
| ntpttr | thanks for the help guys | 22:54 |
| ntpttr | very cool | 22:54 |
| * sdake prints out a screenshot for the office cubicle | 22:54 | |
| ntpttr | i'm definitely interested in getting involved if i can | 22:55 |
| SamYaple | ntpttr: I used to work at RAX. Kolla is the culmination of all my bad OSA experinces | 22:56 |
| *** pbourke-mobile has joined #kolla | 22:56 | |
| sdake | ntpttr are you using master? | 22:57 |
| ntpttr | sdake: yes | 22:57 |
| *** pbourke-pub has joined #kolla | 22:57 | |
| *** vhosakot has joined #kolla | 22:57 | |
| *** pbourke-mobile has quit IRC | 22:58 | |
| SamYaple | meeting time | 22:59 |
| SamYaple | join us in #openstack-meeting-4 ntpttr | 23:00 |
| ntpttr | SamYaple: thanks will do | 23:00 |
| SamYaple | i will make a contributor out of you yet ntpttr! | 23:01 |
| SamYaple | feel free to ping me here to while the meetings going on if youre still working on your cluster | 23:01 |
| *** akwasnie has joined #kolla | 23:01 | |
| *** v1k0d3n_ has quit IRC | 23:04 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: [WIP] iscsi container with lvm2 support https://review.openstack.org/291285 | 23:05 |
| *** sbezverk has joined #kolla | 23:06 | |
| *** blahRus has quit IRC | 23:07 | |
| sbezverk | sdake SamYaple I cleaned up lvm-iscsi patch, there is no dependecy on "root" and other small things are addressed. | 23:07 |
| sbezverk | at least I hope | 23:07 |
| sbezverk | I was wondering if I could ask somebody to test it? | 23:08 |
| SamYaple | sbezverk: now youll have to move all of the sockets to named volumes and stop all the host mounting stuff | 23:08 |
| sbezverk | SamYaple, I think it is done, I might have missed some | 23:09 |
| sbezverk | hopefully review will tell ;-) | 23:09 |
| SamYaple | sbezverk: ok i havent reviewed it yet :) | 23:09 |
| SamYaple | we are in #openstack-meeting-4 sbezverk | 23:09 |
| SamYaple | please join us | 23:10 |
| vhosakot | sbezverk: /join #openstack-meeting-4 | 23:13 |
| sbezverk | already there, thank you and sorry for being late | 23:14 |
| vhosakot | cool, np | 23:16 |
| SamYaple | sbezverk: no problem at all! | 23:17 |
| *** v1k0d3n has joined #kolla | 23:18 | |
| *** rajathagasthya has joined #kolla | 23:24 | |
| *** AndChat|241401 has joined #kolla | 23:25 | |
| *** ccesario has quit IRC | 23:25 | |
| *** nkorabli has joined #kolla | 23:27 | |
| *** pbourke-pub- has joined #kolla | 23:35 | |
| *** achanda has quit IRC | 23:36 | |
| *** pbourke-pub- has quit IRC | 23:36 | |
| *** pbourke-pub- has joined #kolla | 23:36 | |
| *** diogogmt has quit IRC | 23:38 | |
| *** pbourke-pub has quit IRC | 23:39 | |
| *** pbourke-pub- has quit IRC | 23:44 | |
| *** jtriley has joined #kolla | 23:45 | |
| *** jtriley has quit IRC | 23:52 | |
| openstackgerrit | Sam Yaple proposed openstack/kolla: Add generate.py to generate passwords https://review.openstack.org/293728 | 23:52 |
| *** salv-orl_ has joined #kolla | 23:54 | |
| *** salv-orlando has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!