*** Serlex has quit IRC | 00:05 | |
*** alisonh has left #kolla | 00:06 | |
*** SiRiuS_ has quit IRC | 00:08 | |
*** salv-orlando has joined #kolla | 00:18 | |
*** daneyon has joined #kolla | 00:21 | |
*** daneyon has quit IRC | 00:21 | |
*** daneyon has joined #kolla | 00:22 | |
*** daneyon has quit IRC | 00:22 | |
*** salv-orlando has quit IRC | 00:24 | |
*** daneyon_ has quit IRC | 00:25 | |
*** Marga_ has quit IRC | 00:53 | |
*** iceyao has joined #kolla | 00:57 | |
*** alisonh has joined #kolla | 01:00 | |
*** iceyao_ has joined #kolla | 01:02 | |
*** iceyao has quit IRC | 01:05 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Copy the logs out of the container https://review.openstack.org/288541 | 01:24 |
---|---|---|
*** daneyon has joined #kolla | 01:25 | |
*** dims has quit IRC | 01:26 | |
*** daneyon_ has joined #kolla | 01:26 | |
*** alisonh has left #kolla | 01:29 | |
*** daneyon has quit IRC | 01:30 | |
*** Allen_Gao has joined #kolla | 01:30 | |
*** skape has joined #kolla | 01:34 | |
*** Jeffrey4l has joined #kolla | 01:37 | |
*** sdake has joined #kolla | 01:37 | |
*** Jeffrey4l has quit IRC | 01:40 | |
*** Jeffrey4l has joined #kolla | 01:40 | |
*** salv-orlando has joined #kolla | 01:43 | |
*** daneyon_ has quit IRC | 01:44 | |
Jeffrey4l | this two bp have been merge. could have approve these two bp? https://blueprints.launchpad.net/kolla/+spec/add-ut-codebase https://blueprints.launchpad.net/kolla/+spec/add-hacking | 01:44 |
Jeffrey4l | sdake, around? see above ^^ | 01:44 |
*** sdake_ has joined #kolla | 01:44 | |
*** tfukushima has joined #kolla | 01:44 | |
*** sdake has quit IRC | 01:44 | |
*** kjelly has joined #kolla | 01:46 | |
*** salv-orlando has quit IRC | 01:50 | |
*** sdake has joined #kolla | 01:52 | |
*** sdake_ has quit IRC | 01:55 | |
*** sdake has quit IRC | 01:57 | |
*** iceyao_ has quit IRC | 02:01 | |
*** iceyao has joined #kolla | 02:01 | |
*** phuongnh has joined #kolla | 02:08 | |
*** iceyao_ has joined #kolla | 02:10 | |
*** iceyao has quit IRC | 02:11 | |
*** stvnoyes has quit IRC | 02:26 | |
*** stvnoyes has joined #kolla | 02:27 | |
*** iceyao has joined #kolla | 02:28 | |
*** vhosakot has joined #kolla | 02:30 | |
*** iceyao_ has quit IRC | 02:31 | |
*** sdake has joined #kolla | 02:33 | |
*** sdake has quit IRC | 02:40 | |
*** tfukushima has quit IRC | 02:55 | |
*** Jeffrey4l has quit IRC | 03:09 | |
*** Jeffrey4l has joined #kolla | 03:15 | |
*** yuanying has quit IRC | 03:22 | |
openstackgerrit | MD NADEEM proposed openstack/kolla: Reconfigure for haproxy https://review.openstack.org/288248 | 03:24 |
SamYaple | Jeffrey4l: you should have perms to approve | 03:24 |
*** coolsvap has joined #kolla | 03:27 | |
Jeffrey4l | SamYaple, no idea about the process. It is me who draft it. Then can I approve it? I thought it should be different user. | 03:30 |
SamYaple | Jeffrey4l: you can approve it unless its a major direction change | 03:33 |
SamYaple | but its best judgement right now | 03:34 |
SamYaple | its not a tool to push through a single persons agenda | 03:34 |
Jeffrey4l | SamYaple, sorry. what's means by 'its not a tool to push through a single persons agenda'. could u explain more. | 03:37 |
*** vhosakot has quit IRC | 03:38 | |
SamYaple | Jeffrey4l: i just mean use your best judgement. If the blueprint changes something fundementally about Kolla then dont approve it yourself. if its simply adding the hacking module or something simliar its fine | 03:38 |
Jeffrey4l | Roger, thanks a lot. | 03:39 |
*** yuanying has joined #kolla | 03:40 | |
*** yuanying has quit IRC | 03:44 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Reconfigure horizon service https://review.openstack.org/289126 | 03:45 |
*** britthouser has joined #kolla | 03:46 | |
*** klint has joined #kolla | 03:49 | |
*** sdake has joined #kolla | 03:51 | |
*** vhosakot has joined #kolla | 03:51 | |
*** sdake has quit IRC | 03:52 | |
*** sdake has joined #kolla | 03:54 | |
*** ayoung has quit IRC | 03:54 | |
*** tfukushima has joined #kolla | 04:07 | |
*** yuanying has joined #kolla | 04:08 | |
*** sdake has quit IRC | 04:11 | |
*** sdake has joined #kolla | 04:12 | |
*** Marga_ has joined #kolla | 04:17 | |
*** sdake has quit IRC | 04:21 | |
openstackgerrit | MD NADEEM proposed openstack/kolla: Update README for Trove image https://review.openstack.org/289133 | 04:24 |
*** salv-orlando has joined #kolla | 04:26 | |
*** salv-orlando has quit IRC | 04:30 | |
*** tfukushima has quit IRC | 04:30 | |
*** Allen_Gao has quit IRC | 04:56 | |
*** vhosakot has quit IRC | 05:07 | |
*** Allen_Gao has joined #kolla | 05:11 | |
*** tfukushima has joined #kolla | 05:13 | |
*** pbourke has quit IRC | 05:19 | |
*** pbourke has joined #kolla | 05:19 | |
openstackgerrit | Merged openstack/kolla: Update README for Trove image https://review.openstack.org/289133 | 05:30 |
*** Jeffrey4l has quit IRC | 05:54 | |
asalkeld | this could go in, if anyone is about: https://review.openstack.org/286799 | 05:57 |
openstackgerrit | Merged openstack/kolla-mesos: Add setup_gate.sh for new infra gates https://review.openstack.org/289041 | 06:01 |
*** Jeffrey4l has joined #kolla | 06:08 | |
openstackgerrit | Merged openstack/kolla-mesos: New per-service CLI access https://review.openstack.org/286799 | 06:10 |
nihilifer | also, thx for commit about py3 job | 06:23 |
nihilifer | uuuum... wat | 06:24 |
nihilifer | looks like i pressed some magic buttons for irssi, sorry | 06:24 |
*** dave-mccowan has quit IRC | 06:31 | |
openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/kolla: [WIP] Drop root for swift https://review.openstack.org/259374 | 06:43 |
openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/kolla: [WIP] Drop root for swift https://review.openstack.org/259374 | 06:45 |
Jeffrey4l | elemoine, around? | 06:55 |
*** chandankumar has joined #kolla | 06:56 | |
*** salv-orlando has joined #kolla | 07:01 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: [WIP] Copy the logs out of the container https://review.openstack.org/288541 | 07:03 |
*** salv-orlando has quit IRC | 07:12 | |
*** Marga_ has quit IRC | 07:17 | |
-openstackstatus- NOTICE: gerrit is going to be restarted due to bad performance | 07:20 | |
*** ChanServ changes topic to "gerrit is going to be restarted due to bad performance" | 07:20 | |
*** 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/VKpPzA" | 07:26 | |
*** tfukushima has quit IRC | 07:28 | |
*** tfukushima has joined #kolla | 07:29 | |
*** sdake has joined #kolla | 07:33 | |
*** macsz has joined #kolla | 07:34 | |
*** coolsvap has quit IRC | 07:36 | |
*** tfukushima has quit IRC | 07:50 | |
*** tfukushima has joined #kolla | 07:52 | |
openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/kolla: Handle kolla-build exception https://review.openstack.org/289211 | 08:03 |
*** fgimenez has joined #kolla | 08:03 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Reconfigure mongodb service https://review.openstack.org/289213 | 08:06 |
*** Allen_Gao has quit IRC | 08:06 | |
*** mikelk has joined #kolla | 08:07 | |
*** salv-orlando has joined #kolla | 08:11 | |
*** akwasnie has joined #kolla | 08:12 | |
*** athomas has joined #kolla | 08:16 | |
*** salv-orlando has quit IRC | 08:19 | |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Support dependencies inside one host https://review.openstack.org/284737 | 08:19 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Dependency between nova-libvirt and nova-compute inside host https://review.openstack.org/284815 | 08:19 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Use constraints only on multinode deployment https://review.openstack.org/285970 | 08:19 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Allow to override resources in globals.yml https://review.openstack.org/284836 | 08:19 |
*** sdake has quit IRC | 08:21 | |
*** Allen_Gao has joined #kolla | 08:22 | |
nihilifer | asalkeld: when you'll have time, can you review this sequence of patches? https://review.openstack.org/#/c/285970 | 08:25 |
nihilifer | except the patch about heka | 08:25 |
openstackgerrit | Artur Zarzycki proposed openstack/kolla-mesos: Fix typo in command https://review.openstack.org/288385 | 08:39 |
openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/kolla: Handle kolla-build exception https://review.openstack.org/289211 | 08:50 |
*** gfidente has joined #kolla | 09:01 | |
openstackgerrit | Merged openstack/kolla: Make memcached reconfigure not block https://review.openstack.org/288963 | 09:03 |
openstackgerrit | Merged openstack/kolla: Upgrade fails at Magnum https://review.openstack.org/288968 | 09:04 |
*** salv-orlando has joined #kolla | 09:04 | |
*** mbound has joined #kolla | 09:05 | |
openstackgerrit | Merged openstack/kolla: Reconfigure horizon service https://review.openstack.org/289126 | 09:11 |
*** shardy has joined #kolla | 09:11 | |
openstackgerrit | Merged openstack/kolla: Updated from global requirements https://review.openstack.org/288890 | 09:12 |
nihilifer | folks, general question | 09:12 |
nihilifer | do we have feature freeze now? | 09:13 |
nihilifer | i assume we have | 09:13 |
nihilifer | if yes - can we create feature branches which will be not merged to master before beginning of N development? | 09:14 |
nihilifer | the thing is that we'd like to do the work on mesos-dns in parallel with bugfixing for mitaka. and propose a bp about that in kolla (because we need a container for that) | 09:17 |
*** coolsvap has joined #kolla | 09:19 | |
openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack/kolla: Handle kolla-build exception https://review.openstack.org/289211 | 09:24 |
*** Allen_Gao has quit IRC | 09:32 | |
*** Jeffrey4l has quit IRC | 09:37 | |
*** Allen_Gao has joined #kolla | 09:47 | |
*** akwasnie has quit IRC | 09:50 | |
*** akwasnie has joined #kolla | 09:51 | |
*** coolsvap has quit IRC | 10:03 | |
*** Serlex has joined #kolla | 10:05 | |
openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/kolla: Fix retrieving api_interface https://review.openstack.org/289241 | 10:06 |
openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/kolla: Fix retrieving api_interface https://review.openstack.org/289241 | 10:31 |
openstackgerrit | Merged openstack/kolla-mesos: Fix typo in command https://review.openstack.org/288385 | 10:40 |
*** dims_ has joined #kolla | 10:41 | |
openstackgerrit | Merged openstack/kolla-mesos: Restart daemons if their config changes https://review.openstack.org/284664 | 10:41 |
*** athomas has quit IRC | 10:46 | |
*** prithiv has joined #kolla | 10:56 | |
openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/kolla: Added documentation for proxy buildards picked from the environment. https://review.openstack.org/289264 | 11:02 |
*** tfukushima has quit IRC | 11:03 | |
*** dims_ has quit IRC | 11:09 | |
*** prithiv has quit IRC | 11:10 | |
*** prithiv has joined #kolla | 11:12 | |
openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/kolla: Add documentation for proxy buildards picked from the environment. https://review.openstack.org/289264 | 11:18 |
*** dims_ has joined #kolla | 11:22 | |
openstackgerrit | Eric Lemoine proposed openstack/kolla: Fix "wait for log socket" in start.sh https://review.openstack.org/289275 | 11:22 |
openstackgerrit | Merged openstack/kolla-mesos: Autodetect number of instances https://review.openstack.org/284643 | 11:24 |
*** britthouser has quit IRC | 11:26 | |
*** britthouser has joined #kolla | 11:26 | |
openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/kolla: Add documentation for proxy buildargs picked from the environment. https://review.openstack.org/289264 | 11:26 |
*** britthouser has quit IRC | 11:27 | |
*** britthouser has joined #kolla | 11:27 | |
asalkeld | nihilifer: done | 11:29 |
nihilifer | i'm resolving merge conflict right now ;) | 11:32 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Support dependencies inside one host https://review.openstack.org/284737 | 11:33 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Dependency between nova-libvirt and nova-compute inside host https://review.openstack.org/284815 | 11:33 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Use constraints only on multinode deployment https://review.openstack.org/285970 | 11:33 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Allow to override resources in globals.yml https://review.openstack.org/284836 | 11:33 |
nihilifer | this one ;) | 11:33 |
nihilifer | but thx for merging the 1st patch | 11:34 |
openstackgerrit | MD NADEEM proposed openstack/kolla: Reconfigure for haproxy https://review.openstack.org/288248 | 11:40 |
*** phuongnh has quit IRC | 11:44 | |
*** jhyang_ has joined #kolla | 11:45 | |
*** jhyang_ has quit IRC | 11:45 | |
*** jhyang_ has joined #kolla | 11:46 | |
*** salv-orlando has quit IRC | 11:49 | |
*** salv-orlando has joined #kolla | 11:49 | |
*** salv-orlando has quit IRC | 11:51 | |
*** salv-orlando has joined #kolla | 11:51 | |
*** salv-orlando has quit IRC | 11:51 | |
*** salv-orlando has joined #kolla | 11:53 | |
*** dave-mccowan has joined #kolla | 12:01 | |
*** prithiv has quit IRC | 12:02 | |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Support dependencies inside one host https://review.openstack.org/284737 | 12:02 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Dependency between nova-libvirt and nova-compute inside host https://review.openstack.org/284815 | 12:02 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Use constraints only on multinode deployment https://review.openstack.org/285970 | 12:02 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Allow to override resources in globals.yml https://review.openstack.org/284836 | 12:02 |
*** mlima has joined #kolla | 12:02 | |
openstackgerrit | Angus Salkeld proposed openstack/kolla-mesos: CLI: Implement service definition https://review.openstack.org/286992 | 12:02 |
openstackgerrit | Angus Salkeld proposed openstack/kolla-mesos: Move common config functions into a new file https://review.openstack.org/286995 | 12:02 |
openstackgerrit | Angus Salkeld proposed openstack/kolla-mesos: Move Runner classes to service.py https://review.openstack.org/286994 | 12:02 |
openstackgerrit | Angus Salkeld proposed openstack/kolla-mesos: CLI: implement service commands https://review.openstack.org/286996 | 12:02 |
openstackgerrit | Angus Salkeld proposed openstack/kolla-mesos: CLI: implement config commands https://review.openstack.org/289295 | 12:02 |
*** JoseMello has joined #kolla | 12:11 | |
asalkeld | nihilifer: i am headding to bed, i'll review any remaining in the morning | 12:15 |
*** chandankumar has quit IRC | 12:15 | |
nihilifer | asalkeld: ok, no problem, good night | 12:16 |
*** jtriley has joined #kolla | 12:23 | |
openstackgerrit | Eric Lemoine proposed openstack/kolla: Fix "wait for log socket" in start.sh https://review.openstack.org/289275 | 12:25 |
openstackgerrit | Mauricio Lima proposed openstack/kolla: Reconfigure mariadb service https://review.openstack.org/289303 | 12:26 |
*** salv-orlando has quit IRC | 12:37 | |
*** prithiv has joined #kolla | 12:46 | |
*** rhallisey has joined #kolla | 12:51 | |
*** dwalsh has joined #kolla | 12:53 | |
*** Allen_Gao has quit IRC | 12:55 | |
mlima | TASK: [ceph | Looking up hostname] failed: [localhost] => {"changed": false, "cmd": ["getent", "hosts", "10.102.22.89"], "delta": "0:00:00.003176", "end": "2016-02-24 11:17:29.999575", "rc": 2, "start": "2016-02-24 11:17:29.996399", "stdout_lines": [], "warnings": []} | 12:55 |
mlima | someone knows why it happens? | 12:56 |
*** salv-orlando has joined #kolla | 12:56 | |
*** thumpba has joined #kolla | 13:03 | |
*** thumpba has quit IRC | 13:07 | |
*** Serlex has quit IRC | 13:08 | |
*** Jeffrey4l has joined #kolla | 13:16 | |
*** Serlex has joined #kolla | 13:17 | |
*** Allen_Gao_ has joined #kolla | 13:20 | |
*** macsz has left #kolla | 13:22 | |
*** dims_ has quit IRC | 13:25 | |
mlima | TASK: [ceph | Looking up hostname] failed: [localhost] => {"changed": false, "cmd": ["getent", "hosts", "10.102.22.89"], "delta": "0:00:00.003176", "end": "2016-02-24 11:17:29.999575", "rc": 2, "start": "2016-02-24 11:17:29.996399", "stdout_lines": [], "warnings": []} | 13:27 |
mlima | someone knows why it happens? | 13:27 |
*** chandankumar has joined #kolla | 13:31 | |
*** Allen_Gao_ has quit IRC | 13:35 | |
*** Allen_Gao has joined #kolla | 13:42 | |
openstackgerrit | Mauricio Lima proposed openstack/kolla: Reconfigure mariadb service https://review.openstack.org/289303 | 13:53 |
*** sdake has joined #kolla | 13:57 | |
sdake | morning | 13:58 |
*** inc0 has joined #kolla | 14:00 | |
mlima | TASK: [ceph | Looking up hostname] failed: [localhost] => {"changed": false, "cmd": ["getent", "hosts", "10.102.22.89"], "delta": "0:00:00.003176", "end": "2016-02-24 11:17:29.999575", "rc": 2, "start": "2016-02-24 11:17:29.996399", "stdout_lines": [], "warnings": []} | 14:00 |
mlima | someone knows why it happens? | 14:00 |
inc0 | hey | 14:00 |
inc0 | mlima, try to call "getent hosts << ip address of node >>" | 14:02 |
mlima | ok | 14:02 |
inc0 | in this situation it's getent hosts 10.102.22.89 | 14:03 |
openstackgerrit | Thiago Gomes proposed openstack/kolla: Deleting only images from kolla build https://review.openstack.org/286823 | 14:03 |
mlima | inc0 returns nothing | 14:05 |
mlima | :/ | 14:05 |
*** fthiagogv has joined #kolla | 14:05 | |
sdake | hey folks just wanted to announce we have a 90 minute maximum on our gating now | 14:05 |
sdake | to solve the fact that centos times out duringbuild sometiems | 14:05 |
inc0 | mlima, then you need to fix your hostnames | 14:06 |
inc0 | it should return hostname | 14:06 |
inc0 | yay sdake | 14:06 |
sdake | mlima /etc/hosts has to be filled in | 14:06 |
inc0 | sdake, when do we vote for backport of upgrades? | 14:06 |
inc0 | and it will be over ml? | 14:06 |
*** alyson_ has joined #kolla | 14:07 | |
sdake | inc0 yes i'll post that now | 14:07 |
inc0 | thanks | 14:07 |
sdake | been busy with release | 14:07 |
inc0 | this topic kinda hurts me | 14:07 |
sdake | and sunday i was busy catching up on diablo 3 | 14:07 |
mlima | sdake, i'm deploying aio | 14:07 |
sdake | mlima is /etc/hosts fliled out? | 14:08 |
inc0 | I lost my char since I moved to US:( I need start over | 14:08 |
inc0 | mlima, aio or not, you still need to have hostname | 14:08 |
sdake | inc0 that sucks | 14:08 |
mlima | oh | 14:08 |
mlima | okay | 14:08 |
sdake | i hope they itnroduce some new pla yto win stuff | 14:08 |
sdake | I would like to buy stash tabs for eample | 14:08 |
sdake | without having do the season 5 | 14:08 |
mlima | i thought it was necessary only multi node | 14:08 |
sdake | inc0 my closeet https://drive.google.com/file/d/0B8q6xDPETSkHZWxhczhEQjFaVmcwa1pJVWR6ZWdXaURDZFpF/view | 14:09 |
inc0 | nope, and ceph is not the only one which deals with hostnames | 14:09 |
mlima | works haha | 14:09 |
mlima | Thank sdake, inc0 | 14:09 |
mlima | :) | 14:09 |
sdake | mlima inc0 figured it out bro | 14:09 |
inc0 | any time | 14:09 |
sdake | i would have never guessed the problem | 14:09 |
sdake | a commit to the docs would help here | 14:09 |
inc0 | yeah, actually, could you add this note to quickstart guide? | 14:10 |
inc0 | mlima, | 14:10 |
sdake | inc0 what paragon was your toon | 14:10 |
sdake | you can get power leveld by people 1 to 70 in 6-7 minutes | 14:10 |
inc0 | 300-something, nothing too fancy | 14:10 |
sdake | so essentially, scrubby :) | 14:10 |
inc0 | but you still need to have story finished | 14:11 |
sdake | my seasonal character is already paragaon 60 | 14:11 |
inc0 | so I need to go through main story myself again | 14:11 |
*** akwasnie has quit IRC | 14:11 | |
inc0 | anyway, I'm not going back to diablo 3...htc vive is incomming | 14:12 |
inc0 | and I'll have all the VR-friendly games to play:> | 14:12 |
mlima | sdake, you assigned a bug for me, but i can not reproduce it | 14:12 |
sdake | mlima what was the bug | 14:13 |
mlima | i will leave it there until someone confirms it | 14:13 |
sdake | i do alot of things :) | 14:13 |
mlima | oh | 14:13 |
mlima | glance endpoints fails | 14:13 |
mlima | or something | 14:13 |
sdake | lin kbug plz | 14:13 |
sdake | i dont rnadmoly asssign bugs to people | 14:14 |
inc0 | mlima, logs plz | 14:14 |
sdake | if i do it usually for a very good reson | 14:14 |
mlima | okay, I just do not know what to do | 14:14 |
sdake | it may be in invalid state, which mens you solved it | 14:14 |
sdake | so you should get credit for it | 14:15 |
sdake | link the bug | 14:15 |
sbezverk | Good morning | 14:15 |
sbezverk | sdake any guidance on extending the list of tags? | 14:15 |
sdake | sbezverk to do what? | 14:15 |
sbezverk | to add a new one | 14:15 |
sdake | for which purpose | 14:15 |
sdake | iscsi? | 14:16 |
sbezverk | I used mariadb as an example for new iscsi container | 14:16 |
sdake | seems reasonable | 14:16 |
sdake | i'm sure the ansible jockies will have a very specific answer in the review | 14:16 |
sbezverk | I have changed config and manage to build iscso container using build.py | 14:16 |
sdake | is it a thin container? | 14:17 |
sbezverk | now I want to deploy just it using --tags iscsi | 14:17 |
sdake | seems reasonble, like I said :) | 14:17 |
inc0 | sbezverk, you made iscsi work? does it work with cinder? | 14:17 |
*** klint has quit IRC | 14:17 | |
*** Marga_ has joined #kolla | 14:17 | |
sdake | inc0 I think thatis what he is workign on next | 14:17 |
inc0 | also yeah, add tags if you feel it's valid | 14:18 |
sbezverk | sdake unfortunately to run iscsi in the container, I must add systemd binaries. not sure if it makes NOT thin.. | 14:18 |
inc0 | cool, I'd love to have working cinder without ceph | 14:18 |
sdake | zomg | 14:18 |
sdake | what systemd binarnies | 14:18 |
sdake | do you eman you run systemd to run iscsi? | 14:18 |
inc0 | sbezverk, it does...we shouldn't run systemd in container | 14:18 |
sdake | if so you wlil need to rework systemd out of the equation | 14:19 |
sdake | we definately are never running systemd in a container :) | 14:19 |
sdake | sbezverk one of hte challenges of iscsi is it has tgtd | 14:19 |
sbezverk | sdake inc0 ok, I will try but so far the only way to run iscsi in a container was with systemd.. | 14:19 |
sdake | and some other daeon that runs | 14:19 |
sdake | they need two containers | 14:20 |
*** Allen_Gao has quit IRC | 14:20 | |
sdake | and they communicte via a ocket via named volumes | 14:20 |
sdake | just deconstruct how systemd does the job | 14:20 |
sdake | and do the same thing in a container | 14:20 |
*** Allen_Gao has joined #kolla | 14:20 | |
sdake | the problem is systemd runs a million commands | 14:20 |
sdake | sbezverk but did it make lvm work? | 14:20 |
sdake | rather cinder | 14:20 |
sbezverk | sdake I disabled ALL other services systemd would usually start | 14:21 |
openstackgerrit | Michal Rostecki proposed openstack/kolla-mesos: Make neutron-openvswitch-agent dependent on openvswitch-vswitchd https://review.openstack.org/289349 | 14:21 |
*** absubram has joined #kolla | 14:21 | |
sdake | sbezverk i get it, but we just got rid of our last fat containers (neutron) | 14:21 |
sdake | i dont want to add any more back in | 14:21 |
sbezverk | sdake btw lvm also requires systemd as far as I remember | 14:21 |
inc0 | duh | 14:21 |
sdake | but i would probably be a good wy to test out your cinder changes | 14:21 |
sdake | since theoretically they got iscsi right | 14:22 |
*** jmccarthy has quit IRC | 14:22 | |
sdake | i think that all is needed is to run tgtd is a container with the correct options | 14:22 |
inc0 | for the sake of pragmatism I'm for practicality over purity, but please confirm that there is no way you can make this work without systemd | 14:22 |
sdake | systemd wont be merged, simle as that | 14:23 |
sdake | i'm also for prgamatism :) | 14:23 |
*** Allen_Gao has quit IRC | 14:23 | |
inc0 | ok, I won't argue about that one for a change | 14:23 |
*** jmccarthy has joined #kolla | 14:23 | |
*** absubram_ has joined #kolla | 14:23 | |
openstackgerrit | Prithiv proposed openstack/kolla: Installs Curl, required OpenStack Packages https://review.openstack.org/289351 | 14:23 |
inc0 | afk, food | 14:23 |
inc0 | and then meetings \o/ | 14:24 |
sbezverk | sdake inc0 theoretical question, if LVM or iscsi will not work without systemd, would you decide to drop these features? | 14:25 |
*** jpeeler has joined #kolla | 14:27 | |
*** absubram has quit IRC | 14:27 | |
*** absubram_ is now known as absubram | 14:27 | |
*** chandankumar has quit IRC | 14:27 | |
*** Allen_Gao has joined #kolla | 14:31 | |
*** Allen_Gao has quit IRC | 14:31 | |
*** absubram has quit IRC | 14:31 | |
*** Allen_Gao has joined #kolla | 14:32 | |
*** Allen_Gao has quit IRC | 14:32 | |
*** Allen_Gao has joined #kolla | 14:33 | |
inc0 | sbezverk, I would say yes, that's why we suggest to use cinder with ceph | 14:35 |
inc0 | we have this gap filled with something that is way more practical | 14:36 |
*** JoseMello has quit IRC | 14:38 | |
ajafo | sdake: hi, I've one question, we need in kolla-mesos mesos-dns container, can I just create bp and try to push it to kolla or we need some feature branch? | 14:42 |
sdake | bueprint is fine | 14:43 |
sdake | ajafo we dont work with feature branches that I know of? | 14:43 |
ajafo | sdake: I just wasn't sure how to solve it because I'm not sure does kolla is in freeze state at the moment or we can still develop in master futures like this | 14:44 |
sbezverk | inc0 I am afraid droping these would if no just limit, but could potentially kill interest of a large part of operations who uses lvm and iscsi for ages ;-) | 14:44 |
openstackgerrit | Dave McCowan proposed openstack/kolla: Use internalURL for internal communication https://review.openstack.org/289367 | 14:45 |
*** prithiv has quit IRC | 14:45 | |
openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/kolla: Fix retrieving api_interface https://review.openstack.org/289241 | 14:46 |
*** spisarski has joined #kolla | 14:49 | |
inc0 | sbezverk, well, true, but same goes to any technology when you think about it | 14:49 |
inc0 | kolla is opinionated, which means we're not saving everyone | 14:49 |
*** prithiv has joined #kolla | 14:50 | |
inc0 | also if you just keep your running infra, with iscsi and tgt outside kolla, it works just fine | 14:50 |
*** jhyang_ has quit IRC | 14:51 | |
sdake | http://cdn.meme.am/instances/500x/67291220.jpg | 14:51 |
sdake | inc0 thats what I think of your kolla on kolla :) | 14:52 |
inc0 | kollaine, so much kollaine | 14:52 |
sdake | inc0 or how about this one http://cdn.meme.am/instances2/500x/4675154.jpg | 14:53 |
inc0 | you know my take on it | 14:54 |
inc0 | glue sniffing koala | 14:54 |
sdake | i dont know a glue sniffing meme | 14:54 |
sdake | cocaine bear was the best i could come up with :) | 14:54 |
*** dims has joined #kolla | 14:54 | |
inc0 | we talked about our pet animal in Tokio | 14:54 |
sdake | i remember | 14:55 |
sdake | but i dont know a meme for one | 14:55 |
inc0 | koala is obvious, glue is kolla | 14:55 |
inc0 | yeah, we need to create one | 14:55 |
sdake | i think sniffing glue is not ideal for you and would highly recommend against it :) | 14:56 |
inc0 | I would say same thing about cigars | 14:56 |
inc0 | but it does match koalas lifestyle | 14:57 |
sdake | sniffing glue way wosrse for you then smoking cigars ;) | 14:57 |
inc0 | I agree, it is | 14:58 |
inc0 | but it does match koalas lifestyle | 14:58 |
inc0 | and question...how bad is it for koalas? | 14:58 |
inc0 | don't forget we're taling about junkie animal | 14:58 |
*** salv-orlando has quit IRC | 14:59 | |
*** salv-orlando has joined #kolla | 15:00 | |
*** daneyon has joined #kolla | 15:02 | |
*** daneyon_ has joined #kolla | 15:03 | |
*** ayoung has joined #kolla | 15:03 | |
*** inc0 has quit IRC | 15:05 | |
*** daneyon has quit IRC | 15:06 | |
*** inc0 has joined #kolla | 15:10 | |
*** jtriley has quit IRC | 15:12 | |
sbezverk | inc0 sdake if kolla targets to be "OpenStack deployment tool" at least that was my understanding of Kolla's mission, then it has to support ALL options offered by OpenStack, cannot be a leader if you do not cover all.. IMHO.. | 15:13 |
*** chandankumar has joined #kolla | 15:17 | |
*** inc0 has quit IRC | 15:19 | |
*** akwasnie has joined #kolla | 15:19 | |
*** jtriley has joined #kolla | 15:20 | |
sdake | sbezverk nobody said iscsi was out | 15:22 |
sdake | sbezverk we want iscsi | 15:22 |
sdake | sbezverk the issue is we want it in a thin container | 15:22 |
sdake | here is the course of action I suggest you take | 15:23 |
sdake | 1) get cinder or whatever else to work properly with the iscsi container you have mde | 15:23 |
sdake | once that is done | 15:23 |
sdake | submit the cinder changes | 15:23 |
sdake | 2) decompose how systemd launches iscsi and duplicate it, but container-style | 15:23 |
sdake | 2 will involve named volumes for storing sockets | 15:24 |
sdake | 1 may as well | 15:24 |
sdake | you may have to share all of /run and /dev, I'm just not sure | 15:24 |
sdake | brb | 15:24 |
pbourke | "Warning [1] was a pretty heated argument and there may have been some swearing :) " - lol | 15:24 |
sdake | ;) | 15:24 |
sdake | brb | 15:24 |
*** sdake_ has joined #kolla | 15:25 | |
sdake_ | bak | 15:25 |
sbezverk | sdake agreed, at this moment I need help from ansible folks to help me to include new iscsi container into available tags list, so I could try to deploy it using --tags in kolla-ansible deploy | 15:27 |
*** blahRus has joined #kolla | 15:28 | |
*** sdake has quit IRC | 15:29 | |
sdake_ | iscsi container shuould pssibly go in common? | 15:31 |
sdake_ | pbourke what is your thought, common or iscsi? | 15:31 |
*** akwasnie has quit IRC | 15:32 | |
*** haplo37 has joined #kolla | 15:32 | |
pbourke | sdake_: sbezverk: I would say have a separate container for it, but dont make a tag, you can have it as dependency of the cinder role | 15:34 |
pbourke | see memcached for swift | 15:34 |
sdake_ | its not just cinnder that will use iscsi i think | 15:35 |
*** akwasnie has joined #kolla | 15:36 | |
pbourke | yeah | 15:36 |
*** inc0 has joined #kolla | 15:36 | |
nihilifer | https://www.quantifiedcode.com/app/project/gh:nhlfr:kolla https://www.quantifiedcode.com/app/project/gh:nhlfr:kolla-mesos ;) | 15:39 |
sbezverk | pbourke at this point I would like to have it independent of anything, it is just easier to make it working, later when everythign is clean, I will do as you say. so still my question, how I can make it deployed as a separate container using tags?? | 15:40 |
pbourke | sbezverk: that sounds fine, I think I did the same with memcached. You just need to edit site.yml to add the tags | 15:40 |
openstackgerrit | Eric Lemoine proposed openstack/kolla: [WIP] Add cron image and playbook https://review.openstack.org/288432 | 15:41 |
pbourke | nihilifer: cool | 15:41 |
*** inc0 has quit IRC | 15:42 | |
*** akwasnie has quit IRC | 15:44 | |
sbezverk | pbourke, that is exactly what I have done, but still it barks back :-( | 15:44 |
pbourke | paste error? | 15:44 |
sbezverk | pbourke here you go: http://paste.openstack.org/show/489555/ | 15:48 |
pbourke | sdake_: do you have a link to the review they are arguing about in that log | 15:48 |
sdake_ | pbourke no | 15:48 |
sdake_ | pbourke it has a -2 on it tho and it sin the queue | 15:49 |
sdake_ | so it shouldn't be hard to find :) | 15:49 |
pbourke | ok | 15:49 |
sdake_ | pbourke https://review.openstack.org/#/c/287969/ | 15:51 |
sdake_ | Sam Yaple | 15:51 |
sdake_ | Mar 4 11:04 AM | 15:51 |
sdake_ | ↩ | 15:51 |
sdake_ | Patch Set 2: Code-Review-1 | 15:51 |
sdake_ | first, you dont need an action plugin to do what you are doing here and simple module is fine. | 15:51 |
sdake_ | second, im -2 on any new action plugin at all, please remove it | 15:51 |
pbourke | ah, I know the review | 15:51 |
sdake_ | third, this will not properly show changed tasks at all. if we are to adapt this tool to any other place we wont show failed/changed messages appropriately so please remove the string library entirely. | 15:51 |
sdake_ | Please use the method we use for all the oterhservices, a script inside the container (preferably one that spits out json like the keystone script). | 15:51 |
pbourke | just missed the latest ps | 15:51 |
SamYaple | morning | 15:53 |
*** inc0 has joined #kolla | 15:55 | |
rhallisey | so having read the upgrade log.. seems like we don't have a choice but to backport it. Otherwise were expecting people start with kolla 1.1.0, or whichever release that has named-volumes | 15:55 |
rhallisey | historically though, I think upgrades were pushed to mitaka because we never anticipated to issues in liberty that required a backport | 15:57 |
rhallisey | never thought they were intended anywhere in liberty | 15:57 |
*** inc0 has quit IRC | 15:59 | |
sdake_ | rhallisey therein lies the crux of the vote :) | 15:59 |
SamYaple | rhallisey: that is wrong, "we never anticipated to issues in liberty that required a backport" | 16:00 |
SamYaple | we discussed in tokyo pinning and bumping the pins for the stable branch | 16:01 |
SamYaple | we were always planning z upgrades | 16:01 |
rhallisey | I'm talking about named volumes | 16:01 |
SamYaple | that specifically yes | 16:01 |
SamYaple | i agree | 16:01 |
rhallisey | major issues | 16:01 |
rhallisey | that docker helped us fix | 16:01 |
SamYaple | but upgrades aren't part of that discussions. upgrades always were, they arent up for a vote (if we want a stable branch) | 16:01 |
*** inc0 has joined #kolla | 16:01 | |
SamYaple | what we are voting on now is _not_ doing it, because it was always planned | 16:01 |
*** dwalsh has quit IRC | 16:02 | |
rhallisey | we may just have to admit to ourselves here that docker features might be given and *exception* when dealing with backports | 16:04 |
rhallisey | that kinda of stuff is out of our control :/ | 16:05 |
sdake_ | master deploy time for me with compute kit + heat: real3m32.559s | 16:05 |
sdake_ | from registry | 16:05 |
sdake_ | so much faster then previously | 16:05 |
rhallisey | v nice | 16:06 |
SamYaple | rhallisey: i dont think anyone has a problem with that type of backport | 16:06 |
SamYaple | sdake_: multinode? | 16:06 |
sdake_ | single node | 16:06 |
sdake_ | say to run upgrade | 16:06 |
sdake_ | I just change release_version: "2.0.1" | 16:06 |
SamYaple | sdake_: the onyl way you got that time was if you had prestaged the images | 16:06 |
sdake_ | and pull the trigger on upgrades? | 16:06 |
SamYaple | sdake_: yea that should work | 16:07 |
sdake_ | SamYaple my registry is on the same node | 16:07 |
SamYaple | yea i know | 16:07 |
rhallisey | SamYaple, well I don't want people to think these emergency are a thing in kolla, but rather a reaction to docker | 16:07 |
sdake_ | but the images were not prestaged | 16:07 |
SamYaple | but you arent pulling from regitsry that way | 16:07 |
sdake_ | docker images had nothing | 16:07 |
SamYaple | i would test that again if I were you :) | 16:07 |
sdake_ | i will | 16:07 |
SamYaple | 3m30 is accurate for a prestaged image run | 16:07 |
*** dwalsh has joined #kolla | 16:07 | |
openstackgerrit | Dave McCowan proposed openstack/kolla: Sort out internal only and external only services https://review.openstack.org/289421 | 16:08 |
SamYaple | pulling from local registry still wont quite get you there | 16:08 |
sdake_ | after upgrade everything is still running 2.0.0 | 16:09 |
SamYaple | sdake_: does your globals.yml have the value set? | 16:09 |
SamYaple | where did you update 2.0.1? | 16:09 |
*** ssurana has joined #kolla | 16:09 | |
sdake_ | globals.yml | 16:09 |
SamYaple | you shouldnt be updating it there | 16:09 |
SamYaple | it shouldnt be in globals.yml at all, even as doc | 16:10 |
sdake_ | http://paste.fedoraproject.org/335191/66995145/ -> http://paste.fedoraproject.org/335191/66995145 | 16:10 |
sdake_ | how do I udpate it then? | 16:10 |
sdake_ | for the upgrade case? | 16:10 |
SamYaple | well before we tag we should bump the var in group_vars/all.yml | 16:10 |
pbourke | sbezverk: sorry yeah I see your problem | 16:10 |
pbourke | sbezverk: you need to update kolla-ansible with your tag(s), it doesn't take --tags param | 16:11 |
prithiv | Hi Sam | 16:11 |
pbourke | sbezverk: or just use ansible-playbook which is better for dev | 16:11 |
sdake_ | i think operators are going to want custom control over what theycall versions | 16:11 |
sdake_ | SamYaple i pushed containers to the registry with a 2.0.1 tag lready in my local machine | 16:11 |
sdake_ | oh its openstack_release | 16:12 |
SamYaple | yea openstack_release sounds right | 16:12 |
SamYaple | hello prithiv | 16:12 |
prithiv | would you suggest installing curl and clients through ansible precheck role or just break init-runonce in case curl and client not installed.. | 16:13 |
SamYaple | hello can I get another review on https://review.openstack.org/#/c/287580/ | 16:13 |
SamYaple | its breaking my usage of --tags | 16:14 |
prithiv | i preferred to have it installed through init | 16:14 |
sbezverk | pbourke: here is the command which get generated: ansible-playbook -i /root/kolla/ansible/inventory/all-in-one -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml --tags iscsi -e action=deploy /root/kolla/ansible/site.yml | 16:14 |
sbezverk | pbourke: please explain whoch you mean by just using ansible-playbook?? | 16:15 |
openstackgerrit | Merged openstack/kolla: Fix "wait for log socket" in start.sh https://review.openstack.org/289275 | 16:15 |
sdake_ | TASK: [nova | Checking if conductor container needs upgrading] **************** | 16:16 |
sdake_ | failed: [localhost] => {"changed": true, "failed": true} | 16:16 |
sdake_ | msg: TypeError("'NoneType' object has no attribute '__getitem__'",) | 16:16 |
sdake_ | inc0!!!2Help | 16:16 |
SamYaple | prithiv: im not suggesting a precheck at all | 16:16 |
sdake_ | inc0 ^^ | 16:16 |
SamYaple | prithiv: im suggesting a check in init-runonce | 16:16 |
SamYaple | prithiv: but im saying init-runonce should not be installing packages on the system | 16:16 |
prithiv | okay… will modify it. | 16:16 |
inc0 | sdake_, looking | 16:16 |
sdake_ | well some of the containers upgrade -> docker ps http://paste.fedoraproject.org/335194/67413145/ -> http://paste.fedoraproject.org/335194/67413145 | 16:17 |
SamYaple | inc0: have you tried an upgrade from one version to another? | 16:17 |
SamYaple | (i have not) | 16:17 |
inc0 | SamYaple, I did, it worked | 16:17 |
sdake_ | this is what i' mdoing, and it does nto work | 16:17 |
sdake_ | did you change the tag release number? | 16:17 |
openstackgerrit | Thiago Gomes proposed openstack/kolla: Deleting only images from kolla build https://review.openstack.org/286823 | 16:17 |
sdake_ | inc0 ^^ | 16:17 |
inc0 | hmm...that might be issue | 16:17 |
inc0 | let me double check | 16:17 |
openstackgerrit | Hui Kang proposed openstack/kolla: Fix type - python-olso-db for ceilometer https://review.openstack.org/289426 | 16:18 |
*** vhosakot has joined #kolla | 16:20 | |
SamYaple | inc0: yea i meant changing the release | 16:24 |
inc0 | I haven't, and probably it breaks | 16:25 |
sdake_ | inc0 i changed openstack_release: "2.0.1" | 16:25 |
inc0 | so I did that | 16:25 |
SamYaple | ok inc0 | 16:25 |
*** macsz has joined #kolla | 16:25 | |
SamYaple | let me know if you need any help | 16:25 |
pbourke | sbezverk: kolla-ansible calls ansible-playbook under the hood | 16:26 |
pbourke | sbezverk: have a look at the source of that script | 16:26 |
sdake_ | if you can call sehll script a hood | 16:26 |
sdake_ | or a car ;) | 16:26 |
*** d_code has quit IRC | 16:27 | |
inc0 | so logic that checks it | 16:28 |
inc0 | is based on different release | 16:28 |
inc0 | but tell me, do you have build images for 2.0.1? | 16:28 |
SamYaple | I have been in contact with the salt guys. Saltstack now has all the features I need. kolla-salt, here I come | 16:28 |
inc0 | kolla-salt, I want that to happen | 16:29 |
SamYaple | inc0: you will love it. all python "plybooks" | 16:29 |
SamYaple | or they can be shell scripts | 16:29 |
SamYaple | up to us | 16:29 |
SamYaple | or still yaml even | 16:29 |
inc0 | I know I will | 16:29 |
inc0 | I already do | 16:29 |
SamYaple | join me on the dark side will you? | 16:29 |
inc0 | and I kinda grow small amount of hate towards ansible | 16:29 |
SamYaple | i have that as well | 16:30 |
inc0 | the hate flows through me | 16:30 |
*** macsz has quit IRC | 16:30 | |
SamYaple | i started out Yaodu as salt for a reason | 16:30 |
SamYaple | i switched to ansible because it had what i needed to work | 16:30 |
*** d_code has joined #kolla | 16:30 | |
SamYaple | but now..... | 16:30 |
sdake_ | dave-mccowan is it posisble to make tls on the client side not need a certificate | 16:30 |
sdake_ | via keystone configuration options | 16:30 |
SamYaple | sdake_: use real cert | 16:30 |
inc0 | sdake_, do you have both images for 2.0.0 and 2.0.1? | 16:30 |
sdake_ | inc0 yes built in my registry | 16:30 |
SamYaple | the client just needs to trust the cert sdake_ . if its selfsigned then it cant implictyl trust it out of the gate | 16:31 |
inc0 | that might be issue | 16:31 |
inc0 | try to pull and rerun upgrade please | 16:31 |
dave-mccowan | there is a client side option, that i would encourage you not to use. other than using a trusted cert, there is nothing you can change on the server side. | 16:32 |
sdake_ | dave-mccowan OSAD has something they change to use self-signed ceerts without having a cert file | 16:32 |
*** jasonsb_ has quit IRC | 16:32 | |
sdake_ | but with a signed cert, if I didn't have to use the OS_CERT thing, that would be fine | 16:32 |
sbezverk | pbourke I checked the script and it is just takes arguments including --tags, wihtout any verification and adds it to the new commands line now with ansible-playbook. That error comes not from the script, but from ansible-playbook which by some reason do no recognize new tag. | 16:32 |
sdake_ | here is what I want to avoid | 16:32 |
*** absubram has joined #kolla | 16:33 | |
sdake_ | distributing haproxy_crt.ca to everyone | 16:33 |
sdake_ | or whatever the ifile is called | 16:33 |
dave-mccowan | sdake_ authentication is kind of the point of using TLS. no cert == no authentication | 16:33 |
sdake_ | no encryption is the point of tls | 16:34 |
sdake_ | auth is provided by keystone | 16:34 |
dave-mccowan | the cert lets the client authenticate the server. it prevents man in the middle. tls give authentication and privacy. encryption is only half the picture. | 16:34 |
sdake_ | i see | 16:35 |
*** sdake_ is now known as sdake | 16:35 | |
SamYaple | sdake_: dave-mccowan you can do noverify | 16:35 |
sdake | well no pubic cloud distributes a crt file :) | 16:35 |
SamYaple | sdake: all those are signed | 16:35 |
SamYaple | just use a signed cert | 16:35 |
dave-mccowan | sdake this is a dev issue, not an operator issue | 16:36 |
sdake | dave-mccowan just to clarify, do you agree with SamYaple that if the cloud is signed with a proper ca, there is no need to distribute a crt file? | 16:36 |
sdake | cloud/deployment | 16:36 |
dave-mccowan | sdake +1 assuming the client has the standard set of trusted root certificates. | 16:37 |
sdake | ok that wfm then | 16:37 |
sdake | what happens if someone compromises a root certificate? | 16:39 |
sdake | does the intenets break | 16:39 |
dave-mccowan | sdake yep. reference heartbleed. | 16:39 |
sdake | i comprehend its a web of trust | 16:39 |
sdake | so it suffers from a single POF instead of a web of trust? | 16:40 |
sdake | a web structure (in nature) can handle the loss of part of its web without problems | 16:41 |
sdake | where things become a problem is if the web bindings fail :) | 16:41 |
SamYaple | dave-mccowan: heartbleed was _not_ compromising a root certificate | 16:42 |
SamYaple | dave-mccowan: heart-bleed was potentially getting the private key for _that_ server and being able to decrypt that data | 16:42 |
SamYaple | jackpot if it was a wildcard cert | 16:43 |
SamYaple | you couldnt sign and reissue certs though | 16:43 |
sdake | inc0 comapre_image is busted imo | 16:46 |
sdake | it returns none in some circumstances | 16:46 |
SamYaple | sdake: None is by design | 16:47 |
SamYaple | its either True or None | 16:47 |
SamYaple | it matches docker-py return style | 16:47 |
sdake | well ansible blows up on none type | 16:47 |
SamYaple | all functions in kolla-docker module do that | 16:47 |
sdake | yes i noticed that | 16:48 |
SamYaple | inc0: i did comment on how you should normalize that before returning to ansible..... | 16:48 |
sdake | how do you normalize it | 16:48 |
sdake | kolla_docker looks pretty good | 16:49 |
SamYaple | nvm he did normalize that | 16:49 |
SamYaple | i dont know what yure seeing then sdake | 16:49 |
SamYaple | its gauranteed to not return None | 16:49 |
sdake | TASK: [nova | Checking if conductor container needs upgrading] **************** | 16:49 |
sdake | failed: [localhost] => {"changed": true, "failed": true} | 16:49 |
sdake | msg: TypeError("'NoneType' object has no attribute '__getitem__'",) | 16:49 |
*** macsz has joined #kolla | 16:50 | |
sdake | i tink the problem is in the function itself | 16:50 |
SamYaple | sdake: that means something is failing _inside_ the module, not the return value | 16:50 |
sdake | yup | 16:51 |
sdake | i think it is referencing something that doesn't exist | 16:51 |
SamYaple | well thats not ansible though | 16:51 |
*** mikelk has quit IRC | 16:51 | |
SamYaple | the new_image value doesnt exist | 16:51 |
SamYaple | thats what it is | 16:51 |
SamYaple | ill patch, moment | 16:51 |
*** inc0 has quit IRC | 16:52 | |
*** dwalsh has quit IRC | 16:52 | |
*** inc0 has joined #kolla | 16:52 | |
sdake | SamYaple how does this look http://paste.fedoraproject.org/335209/45736956/ -> http://paste.fedoraproject.org/335209/45736956 | 16:53 |
*** Marga_ has quit IRC | 16:53 | |
SamYaple | only need it for new_image | 16:53 |
sdake | i'll try that | 16:54 |
SamYaple | container gauranteed to exist | 16:54 |
sdake | it does work though with both :) | 16:54 |
sdake | yay easy fix | 16:54 |
SamYaple | of course, but thats dead code that will never get run | 16:54 |
sdake | you mean the 2nd part | 16:54 |
SamYaple | yes | 16:54 |
sdake | ya i will remove obviously after i test it | 16:54 |
SamYaple | we already check that info | 16:54 |
SamYaple | you going to submit patch? | 16:54 |
sdake | yup | 16:55 |
SamYaple | k | 16:55 |
sdake | upgrade time real1m25.528s | 16:55 |
sdake | upgraded cloud: http://paste.fedoraproject.org/335210/45736973/ -> http://paste.fedoraproject.org/335210/45736973 | 16:55 |
*** dwalsh has joined #kolla | 16:56 | |
sdake | and i can run init-runonce after an upgrade and it doesn't implode | 16:56 |
sdake | inc0 ^^ fix is above | 16:56 |
*** macsz has quit IRC | 16:57 | |
*** salv-orlando has quit IRC | 17:01 | |
*** macsz has joined #kolla | 17:04 | |
*** absubram has quit IRC | 17:06 | |
sbezverk | SamYaple I extended site.yml to include new container I created for iscsi exercises, but when I try to deploy it with --tags iscsi, I get ERROR: tag(s) not found in playbook: iscsi. possible values: ceph,cinder,elasticsearch,glance,haproxy,heat,horizon,ironic,keystone,kibana,magnum,manila,mariadb,memcache,memcached,mistral,mongodb,murano,neutron,nova,rabbitmq,swift | 17:07 |
sbezverk | Command failed ansible-playbook -i /root/kolla/ansible/inventory/all-in-one -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml --tags iscsi -e action=deploy /root/kolla/ansible/site.yml --verbose | 17:07 |
*** bmace has quit IRC | 17:07 | |
inc0 | sdake, back | 17:08 |
*** bmace has joined #kolla | 17:08 | |
inc0 | so it's fixed? let me review | 17:08 |
sdake | i havent submitted a patch yet moment | 17:08 |
sdake | doing a redeploy - then test with sam's suggested change | 17:08 |
SamYaple | inc0: its just a check to make sure new_image exists | 17:08 |
sdake | ya your indexing new_image | 17:09 |
sdake | and new_image no beuno | 17:09 |
inc0 | yeah, error reporting and input validation | 17:09 |
*** mbound has quit IRC | 17:09 | |
sdake | glad that was easy rather then all the playbooks ;) | 17:09 |
SamYaple | sbezverk: iscsi should be in cinder role and not have its own tag | 17:09 |
inc0 | yeah, I see, it's one if statement really...like most of bugs | 17:10 |
sbezverk | SamYaple agreed but for now for easier debugging and testing it would be much easier for me to have it separate, when it is ready, then by all means | 17:10 |
sdake | docker cache was empty minus registry binary: real3m32.139s | 17:10 |
sdake | for a deploy | 17:10 |
sdake | upgrade with no pull real0m29.270s | 17:11 |
*** macsz has quit IRC | 17:12 | |
SamYaple | inc0: if bug: no_bug=True | 17:13 |
ajafo | guys, can you look in free time on https://blueprints.launchpad.net/kolla/+spec/mesos-dns-container I would like to add this container to unblock work on multinode in kolla-mesos | 17:14 |
sdake | upgrade with pull real2m10.769s | 17:14 |
sdake | overlay fs is smoking fast | 17:14 |
sdake | registery 23. smoking fast | 17:15 |
SamYaple | ohh youre using overlayfs | 17:15 |
SamYaple | yea in that case that sounds right | 17:15 |
SamYaple | with btrfs there are some built-in delays that cannot be avoided | 17:15 |
openstackgerrit | Sebastien Fuchs proposed openstack/kolla: Modify the conditional strategy for cinder and glance regarding ceph: use cinder_volume_driver and glance_volume_driver instead of enable_ceph. https://review.openstack.org/289462 | 17:17 |
*** macsz has joined #kolla | 17:21 | |
pbourke | anyone ever get this traceback in keystone.log when trying to boostrap? http://paste.openstack.org/show/489571/ | 17:22 |
pbourke | *bootstrap | 17:23 |
openstackgerrit | Steven Dake proposed openstack/kolla: Upgrade fails from 2.0.0->2.0.1 https://review.openstack.org/289467 | 17:23 |
sdake | inc0 ^^ | 17:23 |
*** Marga_ has joined #kolla | 17:24 | |
pbourke | inc0: mispelled your name in my reply sorry about that | 17:25 |
inc0 | and I thought we were friends... | 17:25 |
inc0 | :P | 17:25 |
pbourke | I would say I gave you the english spelling but even thats wrong!! | 17:25 |
*** salv-orlando has joined #kolla | 17:27 | |
*** absubram has joined #kolla | 17:29 | |
sdake | inc0 i butcher everyones names | 17:33 |
sdake | so just keep that in mind | 17:33 |
sdake | i worked closely with daneyon for 18 months and I still dont know how to say his naem correctly ;) | 17:34 |
sdake | brb relogging | 17:34 |
openstackgerrit | Daniel Gonzalez Nothnagel proposed openstack/kolla: Unify vagrant bootstrap.sh scripts https://review.openstack.org/288821 | 17:34 |
*** Marga_ has quit IRC | 17:35 | |
sbezverk | SamYaple any chance you could answer my question? | 17:35 |
*** sdake_ has joined #kolla | 17:35 | |
SamYaple | sbezverk: im sorry, i may have missed it | 17:36 |
SamYaple | can you repeat? | 17:36 |
sbezverk | How I can extend (temporarely) list of available tags so iscsi container could be deployed (for testing and debugging) purposes on its own, wihtout any dependecies.. | 17:38 |
*** jasonsb has joined #kolla | 17:38 | |
*** sdake has quit IRC | 17:38 | |
*** salv-orlando has quit IRC | 17:40 | |
*** salv-orlando has joined #kolla | 17:41 | |
SamYaple | sbezverk: tags option should do that. look at how site.yml is setup | 17:42 |
sbezverk | SamYaple that is what I thought, I extended this file with entries, but still not luck.. | 17:43 |
SamYaple | not sure | 17:43 |
sbezverk | http://paste.openstack.org/show/489573/ | 17:44 |
*** prithiv has quit IRC | 17:44 | |
openstackgerrit | Eric Lemoine proposed openstack/kolla: [WIP] Add cron image and playbook https://review.openstack.org/288432 | 17:47 |
*** fgimenez has quit IRC | 17:48 | |
*** Marga_ has joined #kolla | 17:48 | |
*** The_Ball has quit IRC | 17:50 | |
*** Marga_ has quit IRC | 17:51 | |
*** salv-orlando has quit IRC | 17:51 | |
sdake_ | sbezverk it needs to go in your inventory file as well | 17:51 |
*** Marga_ has joined #kolla | 17:51 | |
*** salv-orlando has joined #kolla | 17:51 | |
dave-mccowan | are two servers enough to do multi node? one for (deploy + node) and the second for "node only"? | 17:52 |
*** salv-orl_ has joined #kolla | 17:52 | |
sdake_ | sbezverk https://github.com/openstack/kolla/blob/master/ansible/inventory/all-in-one#L18 | 17:52 |
sdake_ | dave-mccowan that works yes | 17:53 |
sdake_ | however, for HA ou need 3 nodes | 17:53 |
*** macsz has quit IRC | 17:54 | |
*** prithiv has joined #kolla | 17:55 | |
sbezverk | sdake I have that configured as well [iscsi:children] | 17:55 |
sbezverk | control | 17:55 |
*** salv-orlando has quit IRC | 17:56 | |
*** prithiv has quit IRC | 18:03 | |
openstackgerrit | Merged openstack/kolla: Upgrade fails from 2.0.0->2.0.1 https://review.openstack.org/289467 | 18:04 |
*** dave-mccowan has quit IRC | 18:05 | |
*** ayoung has quit IRC | 18:06 | |
*** jasonsb has quit IRC | 18:10 | |
*** salv-orl_ has quit IRC | 18:10 | |
*** salv-orlando has joined #kolla | 18:10 | |
*** salv-orlando has quit IRC | 18:14 | |
*** sdake_ has quit IRC | 18:17 | |
*** salv-orlando has joined #kolla | 18:18 | |
*** salv-orl_ has joined #kolla | 18:19 | |
*** salv-orl_ has quit IRC | 18:19 | |
*** salv-orlando has quit IRC | 18:19 | |
*** salv-orlando has joined #kolla | 18:20 | |
*** skape has quit IRC | 18:20 | |
*** jmccarthy has quit IRC | 18:21 | |
*** salv-orlando has quit IRC | 18:22 | |
*** salv-orlando has joined #kolla | 18:22 | |
*** chandankumar has quit IRC | 18:22 | |
*** Marga_ has quit IRC | 18:23 | |
*** The_Ball has joined #kolla | 18:24 | |
*** inc0 has quit IRC | 18:43 | |
*** Serlex has left #kolla | 18:44 | |
*** Marga_ has joined #kolla | 18:50 | |
*** salv-orlando has quit IRC | 18:54 | |
*** salv-orlando has joined #kolla | 18:54 | |
*** salv-orl_ has joined #kolla | 18:56 | |
*** salv-orl_ has quit IRC | 18:57 | |
*** salv-orlando has quit IRC | 18:57 | |
*** salv-orlando has joined #kolla | 18:58 | |
ccesario | hello guys... any idea about this error ? (trunk version) http://pastebin.com/qHKESap1 | 19:03 |
vhosakot | ccesario: is kolla_keystone_bootstrap up ? "docker ps -a | grep keystone" | 19:07 |
*** bmace has quit IRC | 19:16 | |
*** shardy_ has joined #kolla | 19:18 | |
*** shardy has quit IRC | 19:18 | |
SamYaple | I again ask for an ack on this patch https://review.openstack.org/#/c/287580/ | 19:19 |
SamYaple | need it for using tags when ceph is enabled | 19:20 |
ccesario | vhosakot, yes... it is up! | 19:20 |
ccesario | http://pastebin.com/HdM8qxyr | 19:21 |
vhosakot | ccesario: did "kolla-ansible precehcks" pass ? | 19:21 |
vhosakot | prechecks* | 19:21 |
ccesario | well..... let me check again... because it was working in last week... but after code upgrade (trunk) I ḿ getting this error ... :) | 19:22 |
vhosakot | ccesario: I suggest you uninstall kolla and reinstall... always works for me with the latest code | 19:23 |
*** inc0 has joined #kolla | 19:23 | |
ccesario | vhosakot, I already did it... I destroyed my env test and recreate it..... | 19:24 |
rhallisey | SamYaple, all set | 19:25 |
SamYaple | rhallisey: thanks buddy | 19:27 |
rhallisey | no problamo | 19:27 |
openstackgerrit | Merged openstack/kolla: Move hostname registration to common https://review.openstack.org/287580 | 19:28 |
*** gfidente has quit IRC | 19:29 | |
*** inc0 has quit IRC | 19:30 | |
*** inc0 has joined #kolla | 19:30 | |
*** sdake has joined #kolla | 19:31 | |
ccesario | vhosakot, http://pastebin.com/HCZnjqXj | 19:31 |
vhosakot | ccesario: cool, prechecks passed.. now deply | 19:33 |
vhosakot | deploy* | 19:33 |
*** inc0 has quit IRC | 19:35 | |
*** daneyon_ has quit IRC | 19:36 | |
ccesario | same error... http://pastebin.com/Jibievsv | 19:40 |
ccesario | on the target machine when I try run the cocker command manually .... http://pastebin.com/uxyWBhNC | 19:44 |
ccesario | maybe the return data cannot be interpreted as valid json code.... | 19:45 |
sdake | pbourke re https://review.openstack.org/#/c/246437/ | 19:45 |
sdake | pbourke do you need help with project-config for these gates? | 19:45 |
sdake | or a pointer or anything? | 19:45 |
SamYaple | sdake: hes got it | 19:45 |
SamYaple | patches already up | 19:46 |
sdake | cool thx | 19:46 |
sdake | have a link - infra likes acks from ptls for gate changes | 19:46 |
sdake | addition of gates/removal of gates | 19:46 |
sbezverk | sdake would you have some time tonight to check tags problem? | 19:47 |
*** salv-orlando has quit IRC | 19:47 | |
sdake | sbezverk i have meeting in 12 minutes | 19:47 |
sdake | after that i'm free i think | 19:47 |
sbezverk | sdake great, please ping me when you are ready.. | 19:48 |
vhosakot | ccesario: your keystone is down... try adding debugs in ansible/roles/keystone/tasks/register.yml | 19:54 |
*** inc0 has joined #kolla | 19:55 | |
vhosakot | ccesario: also, is the IP address up and pingable 192.168.201.100 | 19:55 |
ccesario | vhosakot, http://pastebin.com/PCGG74zb | 19:57 |
ccesario | what kind debug!? keystone_bootstrap.stdout by example!? | 19:58 |
vhosakot | ccesario: ok, then I'd add debug messages in ansible/roles/keystone/tasks/register.yml and see why keystone is not able to register | 19:58 |
vhosakot | ansible debug | 19:58 |
vhosakot | -debug: "msg=<blah>" | 19:58 |
sdake | sbezverk just recalled i have something to do f or the foundation, not sure when i can dig in today | 19:59 |
sdake | i need to finish the job on that first | 19:59 |
openstackgerrit | Sam Yaple proposed openstack/kolla: Remove openstack client from keystone bootstrap https://review.openstack.org/289542 | 20:02 |
*** Marga_ has quit IRC | 20:04 | |
*** ssurana has quit IRC | 20:08 | |
*** JoseMello has joined #kolla | 20:11 | |
*** Marga_ has joined #kolla | 20:15 | |
ccesario | vhosakot, this is the returned error :/ http://pastebin.com/bg7RhPBw | 20:17 |
vhosakot | hmm.. works for me fine | 20:18 |
ccesario | vhosakot, multihost ? | 20:20 |
vhosakot | ccesario: nope, AIO | 20:20 |
ccesario | AIO it works ..... the problem is multihost :) | 20:21 |
vhosakot | ccesario: I'd print {{ openstack_auth.username }} {{ openstack_auth.password }} {{ openstack_auth.project_name }} admin {{ keystone_admin_url }} {{ keystone_internal_url }} {{ keystone_public_url }} {{ openstack_region_name }} in debug and see theor values | 20:22 |
*** dave-mccowan has joined #kolla | 20:22 | |
ccesario | let me test it | 20:22 |
ccesario | but I believe thaat this vars its ok.... becuse the run command it shown as "<192.168.201.2> REMOTE_MODULE command docker exec -t keystone kolla_keystone_bootstrap admin password admin admin http://192.168.201.100:35357/v3 http://192.168.201.100:5000/v3 http://192.168.201.100:5000/v3 RegionOne" | 20:24 |
ccesario | all vars are printed :) | 20:24 |
ccesario | and the service it seems ok ... lokk the simple test | 20:27 |
ccesario | http://pastebin.com/f47CEHgA | 20:27 |
ccesario | the token it is generated... | 20:28 |
*** jtriley has quit IRC | 20:31 | |
sdake | ccesario could you go into more detail about what is wrong | 20:34 |
sdake | ccesario hopefully your using master, because if not, there were bootstrap changes recently to keystone | 20:35 |
SamYaple | wow 10m12s run time on ubuntu gate! | 20:37 |
SamYaple | https://review.openstack.org/#/c/289542/ | 20:37 |
SamYaple | 10m12s to booted vm! | 20:37 |
SamYaple | building images and all | 20:37 |
*** ssurana has joined #kolla | 20:38 | |
vhosakot | SamYaple: the centOS binary tests failed with error "You must provide a username or user ID via --os-username, --os-user-id, env[OS_USERNAME] or env[OS_USER_ID]"... why does it happen ? I have seen it too occasionally.... are the environment variables sourced from admin-openrc.sh unset/lost in gate ? | 20:39 |
SamYaple | vhosakot: what failed here? | 20:40 |
vhosakot | SamYaple: I was looking at the same review... looks at the logs in gate-kolla-dsvm-deploy-centos-binary that failed | 20:40 |
SamYaple | oh thats because rdo is slow | 20:41 |
vhosakot | ah ok | 20:41 |
vhosakot | I have seen in ocassioanlly | 20:41 |
vhosakot | openrc.sh | 20:41 |
ccesario | vhosakot, maybe the error make sense..... look the glance register and keystone register.... the glance is json formated ... and the keystone it seem no | 20:42 |
ccesario | sdake, I 'm running master branch :) | 20:42 |
vhosakot | glance | 20:43 |
vhosakot | ccesario: I don't think you can curl the keystone tokens without supplying the OS_username and OS_password variabesl | 20:45 |
ccesario | vhosakot, I only tested with curl to check if keystone service is running :) | 20:46 |
vhosakot | ccesario: ok cool | 20:46 |
ccesario | is make sense json format ? http://pastebin.com/SLK1e0YV | 20:49 |
ccesario | sdake, ^^ | 20:49 |
SamYaple | ccesario: can you post `docker images` `docker ps -a` please | 20:50 |
*** salv-orlando has joined #kolla | 20:54 | |
ccesario | SamYaple, http://pastebin.com/cQd5t4tx | 20:56 |
vhosakot | keys | 20:56 |
vhosakot | sorry typo* | 20:56 |
sbezverk | SamYaple what is the best way to add extra parameter to docker container start.yml? | 20:57 |
sbezverk | SamYaple it is more like a run option: --cap-add=ALL | 20:58 |
ccesario | SamYaple, http://paste.openstack.org/show/489594/ | 20:58 |
vhosakot | May be, keystone_bootstrap that is registered must be first checked if it is in JSON format or not.. | 21:00 |
SamYaple | ccesario: im confused as to what the exact breakage is | 21:01 |
SamYaple | can you post the output from the failed ansible task with verbosity all the way up | 21:02 |
vhosakot | SamYaple: keystone is up in ccesario's env, but "docker exec -t keystone kolla_keystone_bootstrap ..." in ansible/roles/keystone/tasks/register.yml is failing to respond with a JSON | 21:03 |
ccesario | SamYaple, http://paste.openstack.org/show/489595/ | 21:03 |
SamYaple | and whats teh output of "docker exec -t keystone kolla_keystone_bootstrap admin password admin admin http://192.168.201.100:35357/v3 http://192.168.201.100:5000/v3 http://192.168.201.100:5000/v3 RegionOne" | 21:05 |
ccesario | http://paste.openstack.org/show/489596/ | 21:06 |
SamYaple | looks like an improper or partial bootstrap | 21:07 |
*** dims has quit IRC | 21:07 | |
SamYaple | i would recommend dropping the database and removing the keystone containers as well | 21:07 |
*** sdake has quit IRC | 21:07 | |
ccesario | I already did it | 21:07 |
ccesario | docker rm -f $(docker ps | awk '{print $1}' ) | 21:08 |
SamYaple | thats not enough ccesario | 21:08 |
SamYaple | you did not clean the ovlumes if you did that | 21:08 |
ccesario | did you mean "images" built ? | 21:09 |
SamYaple | no | 21:10 |
SamYaple | `docker volume ls` | 21:10 |
ccesario | http://paste.openstack.org/show/489600/ | 21:14 |
ccesario | right!? | 21:14 |
ccesario | it is clean... | 21:14 |
SamYaple | cool. now checkout the latest master of the playbooks and rebuild your imags | 21:15 |
ccesario | sure... !! | 21:16 |
ccesario | thanks .... I'll do it,... and keep you informed ... :) | 21:16 |
sbezverk | SamYaple how I can add --cap-add=ALL to start of docker container with ansible 1.9? It seems in 2.0 there is a parameter for this, but cannot find anything for 1.9, any recommendations? | 21:18 |
SamYaple | sbezverk: we dont use the ansible docker module and our docker module doesnt have the ability to specify that | 21:19 |
SamYaple | _HOWEVER_ that is done by just setting the container mode to privieged | 21:19 |
*** ayoung has joined #kolla | 21:20 | |
sbezverk | SamYaple when I launch manually a container with just priviledge and together with --cap-add=ALL I see different results.. | 21:21 |
sbezverk | it make sense since in 2.0 ansible offers these two parameter. | 21:23 |
SamYaple | sbezverk: that sounds morel ike selinux blocking things than anything else | 21:26 |
SamYaple | unless its changed in docker 1.10 --privleged is all caps | 21:26 |
*** shardy_ has quit IRC | 21:27 | |
*** rstarmer has joined #kolla | 21:27 | |
g3ek | hello all, I would like to know if their is a way to have kolla-build to dont have GPG error in it. I need to run the kolla-build in a ansible command and the return always return me RC=1 and mark everything as RED. I know I can put ignore_error: yes, but I would like to have it Green and error free. | 21:30 |
SamYaple | g3ek: im not aware of any gpg error. so you have a link to this error? | 21:32 |
sbezverk | SamYapl here is paste for two containers 1 priv + cap all and 2 just priv.. http://paste.openstack.org/show/489603/ | 21:34 |
sbezverk | if there is a way just to squeez this extra parameter in docker run command line, please let me know, it is just PoC.. | 21:35 |
SamYaple | there is no command line thats run | 21:35 |
SamYaple | all launched through pythonlibs | 21:35 |
sbezverk | :-( | 21:35 |
SamYaple | but that doesnt mean there is a difference fyi | 21:35 |
SamYaple | not passing cap, but passing privlege means you should have the same caps | 21:36 |
sbezverk | well I see difference | 21:36 |
SamYaple | you see a difference from inspect, not what its capable of | 21:36 |
SamYaple | what cap do yo uthink you are missing? | 21:36 |
*** rstarmer has quit IRC | 21:36 | |
sbezverk | SamYaple not sure, but if I do not specify capp all, dbus fails to start | 21:37 |
sbezverk | and my iscsi daemon is not starting.. | 21:37 |
sbezverk | in case where I specify cap all, then it start no prob.. | 21:37 |
SamYaple | running dbus and/or udev in a container is going to be blocked | 21:37 |
SamYaple | hopefully thats not somethign you are counting on | 21:38 |
sbezverk | ok | 21:38 |
g3ek | SamYaple, I just recheck. I dont know really what fail, I just always have a RC=1 and a lots of red in the screen, but in final everything is working. Here's the last line (resume) : 192.168.2.4 : ok=9 changed=8 unreachable=0 failed=1 | 21:39 |
g3ek | I see sometime error like Ignoring dnspython3: markers u"python_version=='3.4'" don't match your environment | 21:40 |
SamYaple | well thats not an error | 21:41 |
SamYaple | thats a warning | 21:41 |
SamYaple | it may be printed to stderr (and im not sure it is), but its a warning | 21:41 |
g3ek | ok, I tried to research in the scrollback and I have no Error and no error in it (no critical too). I have one time the word failed and it's in the resume part LOL. Why is the command return RC=1 then? | 21:43 |
g3ek | The funny thing is that everything is well installed : ({}, {LOTS OF STUFF HERE},{}) | 21:46 |
*** skape has joined #kolla | 21:46 | |
skape | Hi Guys! I've installed and deployed openstack with kolla, thanks for your help and keep doing this great job!!!! | 21:48 |
SamYaple | skape: welcome to the team! | 21:48 |
SamYaple | you have to help us now | 21:48 |
skape | I have a newby question how do I configure and start addicional services in my case ceilometer the images are in the registry but it isn't running | 21:50 |
*** dwalsh has quit IRC | 21:50 | |
g3ek | skape to add new service, you need to create some container in docker folder and create the ansible role for it. | 21:51 |
sbezverk | Hmm, what about just to add in globals.yml enable_ceilometer: "yes" | 21:52 |
g3ek | this is to activate ceilometer, not to create additional services | 21:53 |
g3ek | I just answer one of the question ^^ | 21:53 |
sbezverk | my bad, I thought I saw ceilometer in roles | 21:54 |
g3ek | No problem, I should have answer both questio to make it more clear | 21:54 |
skape | thx guys, after enable_ceilometer ? still need to create the roles ? | 21:56 |
*** dims has joined #kolla | 21:58 | |
*** dims has quit IRC | 22:01 | |
*** rhallisey has quit IRC | 22:06 | |
g3ek | skape, You just have to create a role and a docker (dockerfile) if you create a new service. Normaly, a lots of project is already in kolla. You just need to create your own if you want to have a custom service or if you want to implement a new openstack service into the kolla project. | 22:16 |
*** salv-orlando has quit IRC | 22:17 | |
*** salv-orlando has joined #kolla | 22:18 | |
skape | g3ek : but ceilometer doesn't have any role directory in inventory, it will be created with enable_ceilometer set to yes ? and should I run kolla-ansible deploy or kolla-ansible reconfigure after changing enable_ceilometer in globals.yml | 22:19 |
g3ek | the role will be in ./ansible/roles/ceilometer | 22:20 |
g3ek | skape: check above | 22:21 |
*** slagle has quit IRC | 22:24 | |
g3ek | SamYaple, have you ever try to kolla-build in a ansible command ? Exemple : ansible -i ~/kolla-deploy/kolla/ansible/inventory/multinode localhost -m shell -a "kolla-build --base ubuntu --registry 192.168.2.4:4000" | 22:26 |
g3ek | always have a RC=1 | 22:27 |
g3ek | No matter what build I tried. | 22:27 |
*** JoseMello has quit IRC | 22:38 | |
*** daneyon has joined #kolla | 22:39 | |
*** daneyon_ has joined #kolla | 22:42 | |
skape | g3ek my fault I ment roles directory I run tools/build.py ceilometer but still no ceilometer directory in the ansible\roles. | 22:42 |
g3ek | skape, yeah just saw that. That's strange. It seems like the dockerfile have been done, but the ansible part isn't | 22:44 |
*** daneyon has quit IRC | 22:45 | |
*** v1k0d3n has joined #kolla | 22:53 | |
*** vhosakot has quit IRC | 22:53 | |
*** haplo37 has quit IRC | 23:00 | |
skape | Does anybody know why the ceilometer directory in the ansible/roles isn't being created, the docker images have been created | 23:04 |
SamYaple | skape: not all images have playbooks for deploying them | 23:05 |
*** absubram has quit IRC | 23:06 | |
skape | hum ok, so nothing wrong with my deployment | 23:09 |
*** salv-orlando has quit IRC | 23:13 | |
*** salv-orlando has joined #kolla | 23:14 | |
openstackgerrit | Merged openstack/kolla: Fix type - python-olso-db for ceilometer https://review.openstack.org/289426 | 23:14 |
openstackgerrit | Merged openstack/kolla-mesos: Fix ubuntu config path search https://review.openstack.org/278963 | 23:14 |
openstackgerrit | Merged openstack/kolla: Use internalURL for internal communication https://review.openstack.org/289367 | 23:15 |
asalkeld | o/ | 23:16 |
SamYaple | o/ asalkeld | 23:17 |
*** inc0 has quit IRC | 23:24 | |
*** britthouser has quit IRC | 23:30 | |
*** Jeffrey4l has quit IRC | 23:30 | |
*** daneyon_ has quit IRC | 23:34 | |
*** daneyon has joined #kolla | 23:34 | |
*** spisarski has quit IRC | 23:35 | |
*** daneyon has quit IRC | 23:35 | |
*** blahRus has quit IRC | 23:50 | |
*** salv-orl_ has joined #kolla | 23:52 | |
*** salv-orlando has quit IRC | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!