| *** sacharya has quit IRC | 00:09 | |
| *** serverascode has quit IRC | 00:17 | |
| *** DuncanT has quit IRC | 00:17 | |
| *** NobodyCam has quit IRC | 00:17 | |
| *** elemoine_ has quit IRC | 00:18 | |
| *** DuncanT has joined #openstack-kolla | 00:19 | |
| *** serverascode has joined #openstack-kolla | 00:19 | |
| *** elemoine_ has joined #openstack-kolla | 00:20 | |
| *** NobodyCam has joined #openstack-kolla | 00:20 | |
| *** diogogmt has quit IRC | 00:22 | |
| *** daneyon_ has joined #openstack-kolla | 00:22 | |
| *** daneyon_ has quit IRC | 00:27 | |
| *** ssurana has quit IRC | 00:36 | |
| *** ravig has joined #openstack-kolla | 00:36 | |
| *** zigo has quit IRC | 00:37 | |
| *** Jeffrey4l_ has quit IRC | 00:39 | |
| *** Jeffrey4l_ has joined #openstack-kolla | 00:43 | |
| *** sacharya has joined #openstack-kolla | 00:44 | |
| murphyslawbbs | The kolla-ansible precheck first reads the ipv4 address of the network_interface, but then requires that the address isn't pingable... Isn't that a chicken and egg problem? | 00:45 |
|---|---|---|
| murphyslawbbs | it fails on the first if the ipv4 isn't there, then fails on the second if it is | 00:46 |
| openstackgerrit | Joshua Harlow proposed openstack/kolla: Stop using a global logger for all the things https://review.openstack.org/321884 | 00:52 |
| *** ravig has quit IRC | 00:52 | |
| *** ravig has joined #openstack-kolla | 00:55 | |
| *** ravig has quit IRC | 00:57 | |
| *** ravig has joined #openstack-kolla | 00:58 | |
| *** ravig has quit IRC | 00:58 | |
| *** JoseMello has quit IRC | 00:59 | |
| *** zigo has joined #openstack-kolla | 01:03 | |
| *** murphyslawbbs has quit IRC | 01:07 | |
| *** ayoung has joined #openstack-kolla | 01:07 | |
| openstackgerrit | Hui Kang proposed openstack/kolla: Implement ceilometer ansible role https://review.openstack.org/300574 | 01:24 |
| *** sdake has joined #openstack-kolla | 01:25 | |
| *** jtriley has quit IRC | 01:27 | |
| *** mummer has quit IRC | 01:29 | |
| *** mummer has joined #openstack-kolla | 01:30 | |
| *** tfukushima has joined #openstack-kolla | 01:58 | |
| *** diogogmt has joined #openstack-kolla | 02:06 | |
| *** murphyslawbbs has joined #openstack-kolla | 02:06 | |
| *** ravig has joined #openstack-kolla | 02:08 | |
| *** ravig has quit IRC | 02:16 | |
| *** ravig has joined #openstack-kolla | 02:16 | |
| *** dave-mccowan has quit IRC | 02:34 | |
| *** coolsvap has joined #openstack-kolla | 02:49 | |
| *** yuanying has quit IRC | 02:51 | |
| *** daneyon_ has joined #openstack-kolla | 03:05 | |
| *** daneyon_ has quit IRC | 03:09 | |
| *** neophy has joined #openstack-kolla | 03:13 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added unit test for start container https://review.openstack.org/327247 | 03:13 |
| openstackgerrit | Merged openstack/kolla: Add non-intrusive ctrl-c handling https://review.openstack.org/321913 | 03:14 |
| *** jtriley has joined #openstack-kolla | 03:22 | |
| *** jtriley has quit IRC | 03:27 | |
| mdnadeem_ | coolsvap, Hi could you please let me know , what you mean by "Add the documentation to index" in commit message of PS >> https://review.openstack.org/#/c/285138/15//COMMIT_MSG | 03:28 |
| patchbot | mdnadeem_: patch 285138 - kolla - Added unit test for kolla_docker | 03:28 |
| coolsvap | mdnadeem_, you need to add the documentation link to https://github.com/openstack/kolla/blob/master/doc/index.rst | 03:29 |
| coolsvap | its the first page of kolla documentation | 03:29 |
| coolsvap | from there one can navigate to the other pages | 03:29 |
| mdnadeem_ | coolsvap, Thanks , one more query for test document, shoud i create a new <kolla Test> Section or it comes under <kolla overview> ? | 03:32 |
| coolsvap | i think developer docs would be the right place for unit test docs | 03:35 |
| *** iceyao has joined #openstack-kolla | 03:36 | |
| mdnadeem_ | coolsvap, Ok, will add there :) | 03:37 |
| *** sdake has quit IRC | 03:41 | |
| *** iceyao has quit IRC | 03:42 | |
| *** murphyslawbbs has quit IRC | 03:46 | |
| *** yuanying has joined #openstack-kolla | 03:48 | |
| *** murphyslawbbs has joined #openstack-kolla | 03:49 | |
| *** ravig has quit IRC | 03:50 | |
| *** ravig has joined #openstack-kolla | 03:53 | |
| *** ravig has quit IRC | 03:54 | |
| *** ravig has joined #openstack-kolla | 03:59 | |
| *** coolsvap_ has joined #openstack-kolla | 04:31 | |
| *** coolsvap has quit IRC | 04:35 | |
| *** coolsvap_ is now known as coolsvap | 04:36 | |
| *** tfukushima has quit IRC | 04:41 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added test document https://review.openstack.org/328069 | 04:46 |
| *** sdake has joined #openstack-kolla | 04:46 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added test document https://review.openstack.org/328069 | 04:47 |
| openstackgerrit | David Wang proposed openstack/kolla: Enable kolla k8s to override bind api bind address in genconfig https://review.openstack.org/327925 | 04:47 |
| *** muhammadshahzeb has joined #openstack-kolla | 05:04 | |
| *** muhammadshahzeb has quit IRC | 05:08 | |
| *** muhammadshahzeb has joined #openstack-kolla | 05:09 | |
| *** muhammadshahzeb has quit IRC | 05:09 | |
| Mech422 | pbourke: Ping ? | 05:12 |
| *** jtriley has joined #openstack-kolla | 05:23 | |
| sdake | Mech422 in the eu, he is fsat asleep im sure | 05:26 |
| *** salv-orlando has joined #openstack-kolla | 05:28 | |
| *** jtriley has quit IRC | 05:30 | |
| *** salv-orlando has quit IRC | 05:31 | |
| *** salv-orl_ has quit IRC | 05:31 | |
| *** salv-orlando has joined #openstack-kolla | 05:31 | |
| *** murphyslawbbs has quit IRC | 05:37 | |
| *** murphyslawbbs has joined #openstack-kolla | 05:41 | |
| Mech422 | sdake: its like 6 am there ....he's sleeping late :-P | 05:42 |
| *** tfukushima has joined #openstack-kolla | 05:58 | |
| *** salv-orlando has quit IRC | 06:06 | |
| *** salv-orlando has joined #openstack-kolla | 06:07 | |
| *** belmoreira has joined #openstack-kolla | 06:12 | |
| *** zhiwei has joined #openstack-kolla | 06:12 | |
| *** fragatina has quit IRC | 06:13 | |
| *** ravig has quit IRC | 06:15 | |
| *** ravig has joined #openstack-kolla | 06:19 | |
| mdnadeem_ | sdake, Have you face the Mariadb issue while upgrade openstack liberty to mitaka < [ERROR] Found 1 prepared transactions! It means that mysqld was not shut down properly last time and critical > | 06:32 |
| mdnadeem_ | >> https://bugs.launchpad.net/kolla/+bug/1590639 | 06:32 |
| openstack | Launchpad bug 1590639 in kolla "Kolla upgrade fail Mariadb keeps on restarting" [Undecided,New] | 06:32 |
| *** openstackgerrit has quit IRC | 06:32 | |
| *** openstackgerrit has joined #openstack-kolla | 06:33 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added test document https://review.openstack.org/328069 | 06:33 |
| *** fragatina has joined #openstack-kolla | 06:33 | |
| *** fragatina has quit IRC | 06:36 | |
| *** fragatina has joined #openstack-kolla | 06:37 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added test document https://review.openstack.org/328069 | 06:38 |
| *** ravig has quit IRC | 06:38 | |
| *** murphyslawbbs has quit IRC | 06:40 | |
| *** murphyslawbbs has joined #openstack-kolla | 06:41 | |
| *** daneyon_ has joined #openstack-kolla | 06:41 | |
| *** Mr_Broken has joined #openstack-kolla | 06:43 | |
| *** murphyslawbbs has quit IRC | 06:45 | |
| *** daneyon_ has quit IRC | 06:46 | |
| *** salv-orlando has quit IRC | 06:52 | |
| *** salv-orlando has joined #openstack-kolla | 06:53 | |
| *** murphyslawbbs has joined #openstack-kolla | 06:54 | |
| *** zhiwei has quit IRC | 07:01 | |
| mdnadeem_ | Jeffrey4l_, ping | 07:12 |
| mdnadeem_ | Jeffrey4l_, regarding your 1st comments on PS https://review.openstack.org/#/c/285138/15/tests/test_kolla_docker.py | 07:12 |
| patchbot | mdnadeem_: patch 285138 - kolla - Added unit test for kolla_docker | 07:12 |
| mdnadeem_ | Jeffrey4l_, We can update the <params> as per test case requirement by updating <self.dw.params> , like have a look at line no: 132 of PS https://review.openstack.org/#/c/327247/3/tests/test_kolla_docker.py | 07:15 |
| patchbot | mdnadeem_: patch 327247 - kolla - Added unit test for start container | 07:15 |
| *** sacharya has quit IRC | 07:22 | |
| *** jtriley has joined #openstack-kolla | 07:25 | |
| *** tfukushi_ has joined #openstack-kolla | 07:26 | |
| *** tfukushima has quit IRC | 07:29 | |
| *** jtriley has quit IRC | 07:30 | |
| *** openstackgerrit has quit IRC | 07:33 | |
| *** openstackgerrit has joined #openstack-kolla | 07:33 | |
| *** coolsvap has quit IRC | 07:59 | |
| *** Serlex has joined #openstack-kolla | 07:59 | |
| *** shardy has joined #openstack-kolla | 08:03 | |
| *** reddydodda has joined #openstack-kolla | 08:10 | |
| *** athomas has joined #openstack-kolla | 08:11 | |
| *** coolsvap has joined #openstack-kolla | 08:11 | |
| Mech422 | pbourke: Ping ? | 08:14 |
| *** sdake has quit IRC | 08:21 | |
| *** sacharya has joined #openstack-kolla | 08:22 | |
| *** sacharya has quit IRC | 08:27 | |
| *** daneyon_ has joined #openstack-kolla | 08:30 | |
| *** daneyon_ has quit IRC | 08:34 | |
| *** coolsvap has quit IRC | 08:47 | |
| *** tzn has joined #openstack-kolla | 08:49 | |
| *** coolsvap has joined #openstack-kolla | 08:52 | |
| *** coolsvap has quit IRC | 08:52 | |
| *** coolsvap has joined #openstack-kolla | 08:52 | |
| *** coolsvap has quit IRC | 09:03 | |
| *** SiRiuS__ has joined #openstack-kolla | 09:09 | |
| *** coolsvap has joined #openstack-kolla | 09:10 | |
| *** coolsvap has quit IRC | 09:15 | |
| *** coolsvap has joined #openstack-kolla | 09:22 | |
| * Mech422 yawns | 09:26 | |
| *** jtriley has joined #openstack-kolla | 09:26 | |
| *** jtriley has quit IRC | 09:31 | |
| *** tzn has quit IRC | 09:44 | |
| *** murphyslawbbs has quit IRC | 09:46 | |
| *** coolsvap is now known as coolsvap|afk | 09:46 | |
| *** mbound has joined #openstack-kolla | 09:48 | |
| *** athomas has quit IRC | 09:57 | |
| *** athomas has joined #openstack-kolla | 10:03 | |
| *** iceyao has joined #openstack-kolla | 10:04 | |
| *** sacharya has joined #openstack-kolla | 10:05 | |
| Mech422 | pbourke: Ping ? | 10:06 |
| *** hps_ has joined #openstack-kolla | 10:08 | |
| *** tfukushi_ has quit IRC | 10:09 | |
| *** iceyao has quit IRC | 10:09 | |
| *** sacharya has quit IRC | 10:10 | |
| *** hps_ has quit IRC | 10:10 | |
| *** pbourke has quit IRC | 10:13 | |
| *** pbourke has joined #openstack-kolla | 10:13 | |
| *** daneyon_ has joined #openstack-kolla | 10:18 | |
| *** daneyon_ has quit IRC | 10:23 | |
| *** neophy has quit IRC | 10:23 | |
| *** coolsvap|afk is now known as coolsvap | 10:29 | |
| *** reddydodda has quit IRC | 10:31 | |
| *** zhiwei has joined #openstack-kolla | 10:38 | |
| *** jmccarthy has joined #openstack-kolla | 10:39 | |
| *** zhiwei has quit IRC | 10:42 | |
| *** jtriley has joined #openstack-kolla | 10:58 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added unit test for kolla_docker https://review.openstack.org/285138 | 10:59 |
| *** Mech422 has quit IRC | 11:02 | |
| *** mliima has joined #openstack-kolla | 11:03 | |
| *** mliima_ has joined #openstack-kolla | 11:03 | |
| mliima | Morning all | 11:04 |
| *** coolsvap has quit IRC | 11:08 | |
| sean-k-mooney | mliima: o/ | 11:09 |
| *** dmk0202 has joined #openstack-kolla | 11:10 | |
| *** salv-orlando has quit IRC | 11:13 | |
| *** salv-orlando has joined #openstack-kolla | 11:14 | |
| *** dmk0202 has quit IRC | 11:23 | |
| *** jtriley has quit IRC | 11:24 | |
| mliima | sean-k-mooney, :) | 11:25 |
| *** salv-orl_ has joined #openstack-kolla | 11:28 | |
| *** rhallisey has joined #openstack-kolla | 11:29 | |
| *** salv-orlando has quit IRC | 11:31 | |
| *** dmk0202 has joined #openstack-kolla | 11:36 | |
| *** jmccarthy has left #openstack-kolla | 11:41 | |
| *** jmccarthy has quit IRC | 11:41 | |
| *** jmccarthy has joined #openstack-kolla | 11:44 | |
| *** jtriley has joined #openstack-kolla | 11:45 | |
| *** Mr_Broken has quit IRC | 11:47 | |
| *** Mr_Broken has joined #openstack-kolla | 11:52 | |
| openstackgerrit | Steve Hindle proposed openstack/kolla: Read partition tables directly, bypassing stale kernel data for ceph. https://review.openstack.org/326609 | 11:57 |
| *** Mech422 has joined #openstack-kolla | 12:00 | |
| Mech422 | pbourke: just resent those Ceph/partion patches - hopefully this should be the last iteration | 12:00 |
| *** jtriley has quit IRC | 12:05 | |
| *** zhiwei has joined #openstack-kolla | 12:07 | |
| *** SiRiuS__ has quit IRC | 12:09 | |
| *** zhiwei has quit IRC | 12:11 | |
| openstackgerrit | Steve Hindle proposed openstack/kolla: Read partition tables directly, bypassing stale kernel data for ceph. https://review.openstack.org/326609 | 12:14 |
| *** ayoung has quit IRC | 12:15 | |
| *** dwalsh has joined #openstack-kolla | 12:23 | |
| *** dmk0202 has quit IRC | 12:25 | |
| *** dmsimard is now known as dmsimard|afk | 12:37 | |
| *** dmk0202 has joined #openstack-kolla | 12:39 | |
| *** tzn has joined #openstack-kolla | 12:45 | |
| *** SiRiuS__ has joined #openstack-kolla | 12:46 | |
| *** kbyrne has quit IRC | 12:48 | |
| *** tzn has quit IRC | 12:49 | |
| *** kbyrne has joined #openstack-kolla | 12:49 | |
| *** jtriley has joined #openstack-kolla | 12:54 | |
| *** stvnoyes has quit IRC | 12:57 | |
| *** stvnoyes has joined #openstack-kolla | 12:58 | |
| *** dmk0202 has quit IRC | 12:58 | |
| *** dmk0202 has joined #openstack-kolla | 12:59 | |
| *** daneyon_ has joined #openstack-kolla | 13:00 | |
| *** tzn has joined #openstack-kolla | 13:02 | |
| *** daneyon_ has quit IRC | 13:05 | |
| *** jmccarthy has left #openstack-kolla | 13:16 | |
| *** ayoung has joined #openstack-kolla | 13:17 | |
| *** jmccarthy has joined #openstack-kolla | 13:23 | |
| *** ppowell has joined #openstack-kolla | 13:23 | |
| *** dmk0202 has quit IRC | 13:29 | |
| *** dave-mccowan has joined #openstack-kolla | 13:33 | |
| *** esharao has joined #openstack-kolla | 13:34 | |
| *** salv-orl_ has quit IRC | 13:39 | |
| *** salv-orlando has joined #openstack-kolla | 13:39 | |
| *** ravig has joined #openstack-kolla | 13:45 | |
| *** diogogmt has quit IRC | 13:54 | |
| *** diogogmt has joined #openstack-kolla | 13:55 | |
| *** alyson_ has quit IRC | 13:58 | |
| *** dwalsh has quit IRC | 13:58 | |
| *** diogogmt has quit IRC | 14:03 | |
| *** SiRiuS__ has quit IRC | 14:03 | |
| *** inc0 has joined #openstack-kolla | 14:05 | |
| inc0 | good morning | 14:05 |
| *** sacharya has joined #openstack-kolla | 14:06 | |
| inc0 | guys, can we move either way with these patchsets? https://review.openstack.org/#/c/324665/ | 14:06 |
| patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 14:06 |
| Mech422 | morning inc0 | 14:08 |
| *** sacharya has quit IRC | 14:11 | |
| Mech422 | inc0: hmm - so far, my only thought is it would be nice to have an override directory rather then single file... | 14:12 |
| inc0 | Mech422, you can add imports there | 14:12 |
| Mech422 | inc0: thats much easier to script, sorta like its easier to drop files in sudoers.d rather then modify sudoers | 14:12 |
| inc0 | I'd like to start with one file, test it out and see | 14:13 |
| Mech422 | inc0: hmm - good point | 14:13 |
| inc0 | we can change it before release | 14:13 |
| inc0 | anytime | 14:14 |
| inc0 | and it will be easy | 14:14 |
| Mech422 | inc0: still think directory would be 'cleaner' - but if you can add includes, its effectively the same | 14:14 |
| Mech422 | sounds good to me :-) | 14:14 |
| inc0 | yeah, that's what I'm saying | 14:14 |
| inc0 | depends how people use it really | 14:14 |
| inc0 | I mean I doubt there will be lots of use cases of changes more than 1-2 containers | 14:14 |
| Mech422 | inc0: I think we're gonna see more people 'pre-scripting' kolla... | 14:15 |
| Mech422 | inc0: like for the calico stuff, etc | 14:15 |
| inc0 | yeah, and I love that | 14:15 |
| inc0 | I mean we'll have calico in tree probably | 14:15 |
| Mech422 | so its nice to be able to have a 'script friendly' way to manage the customizations | 14:15 |
| inc0 | but there will be vendors selling "drivers" | 14:15 |
| Mech422 | inc0: yep | 14:15 |
| Mech422 | 5.2 earthquake in San Diego ? | 14:16 |
| *** Mr_Broken has quit IRC | 14:16 | |
| *** dwalsh has joined #openstack-kolla | 14:16 | |
| Mech422 | actually, 5.2 isn't so bad by cali. standards | 14:16 |
| inc0 | so writing this stuff will not be super easy, will need knowledge and I can easily see it being small project on it's own | 14:16 |
| sean-k-mooney | Mech422: personally i i was doing this i would have my single file be in a seperate git repo and just use includes | 14:16 |
| *** Mr_Broken has joined #openstack-kolla | 14:16 | |
| Mech422 | sean-k-mooney: yeah - thing is, what if you have 2 'custom' products ? | 14:17 |
| sean-k-mooney | Mech422: feature branches | 14:17 |
| inc0 | Mech422, still includes | 14:17 |
| Mech422 | sean-k-mooney: instead of each product jsut being able to drop a config, you have to edit the includes | 14:17 |
| inc0 | if you want them together | 14:17 |
| Mech422 | inc0: yeah - I know I know - it performs the same function :-P | 14:17 |
| Mech422 | inc0: I'm not arguing - just saying why I tend to prefer dir. based stuff :-) | 14:18 |
| inc0 | also we can try to do either | 14:18 |
| inc0 | but that will be harder | 14:18 |
| inc0 | anyway, that's a valuable feedback, thanks | 14:18 |
| Mech422 | inc0: and I agree 100% that something is better then nothing... | 14:18 |
| Mech422 | working code trumps everything :-) | 14:19 |
| Mech422 | inc0: shhh - I don't want people to expect me to be useful :-P | 14:19 |
| inc0 | so thing is, I doubt people will customize full stack whole bunch | 14:20 |
| inc0 | but now when I'm thinking about it | 14:20 |
| inc0 | what if you want to get custom network driver AND storage driver | 14:20 |
| inc0 | with single file that would require merging 2 of them manually | 14:21 |
| *** Mr_Broken has quit IRC | 14:21 | |
| Mech422 | inc0: yeah - and whatever comes out in the future.. | 14:24 |
| Mech422 | inc0: say custom keystone token driver thingies | 14:24 |
| inc0 | yeah | 14:24 |
| inc0 | so one of ways would be to provide multiple files | 14:25 |
| inc0 | like --override=foo --override=bar | 14:25 |
| inc0 | will require a bit of jinja2 magics but should be doable | 14:25 |
| inc0 | that would be best of 2 worlds imho | 14:25 |
| *** Jeffrey4l_ has quit IRC | 14:26 | |
| Mech422 | inc0: cool :-) | 14:26 |
| inc0 | ok I'll make that work | 14:27 |
| Mech422 | inc0: its easy enough to 'script' a bunch of comman line args | 14:27 |
| inc0 | Mech422, easier even to put that into kolla-build.conf | 14:27 |
| inc0 | that's not something that's gonna change per build | 14:27 |
| inc0 | that will require some jinja2 magic tho, will do that in separate patch | 14:28 |
| Mech422 | inc0: either way - it would be easier for the end user then maintaining 1 big file :-) | 14:28 |
| inc0 | true | 14:29 |
| Mech422 | inc0: I could also see people wanting to swap out infrastructure - like say you have an in-house mysql cluster up already, or rabbitmq or use a hardware loadbalancer instead of HAproxy ? | 14:30 |
| inc0 | so it's possible now | 14:30 |
| Mech422 | inc0: oh awesome | 14:30 |
| inc0 | enable_haproxy: false, vip_address: your own | 14:30 |
| inc0 | enable_mysql: false, write config overrides | 14:31 |
| Mech422 | inc0: ahh - yeah, I wasn't sure if that was all there was to it, or if there was more 'baked into' stuff | 14:31 |
| inc0 | that should be it | 14:32 |
| inc0 | with mysql it will be a bit bitchy as you need to write config override for every service | 14:32 |
| inc0 | but well, still not horrible | 14:32 |
| *** thumpba has joined #openstack-kolla | 14:33 | |
| Mech422 | inc0: btw - I got an email from pburke about https://bugs.launchpad.net/kolla/+bug/1585185 | 14:34 |
| openstack | Launchpad bug 1585185 in kolla "Ceph/Swift find_disk.py fails to find GPT disklabels on some systems" [Undecided,New] | 14:34 |
| Mech422 | inc0: he was wondering if it might be related to the stale partition table stuff | 14:34 |
| Mech422 | inc0: but I haven't managed to catch back up with him | 14:35 |
| inc0 | hmm, no idea | 14:35 |
| Mech422 | inc0: yeah - I'll catch up with him sooner or later | 14:36 |
| inc0 | but I think find_disks needs changing anyway | 14:36 |
| Mech422 | inc0: dang it - I just got it past jenkins :-P what needs changing now ? | 14:36 |
| inc0 | well, enable osd on partitins | 14:37 |
| inc0 | or file | 14:37 |
| inc0 | it seems pyudev has issues | 14:37 |
| Mech422 | inc0: oh - I fixed that | 14:38 |
| Mech422 | inc0: https://review.openstack.org/#/c/326609/4 | 14:38 |
| patchbot | Mech422: patch 326609 - kolla - Read partition tables directly, bypassing stale ke... | 14:38 |
| *** SiRiuS__ has joined #openstack-kolla | 14:38 | |
| Mech422 | inc0: yeah - pyudev was getting stale partition name data from the kernel | 14:39 |
| Mech422 | inc0: so I changed it to shell out sgdisk and scrape the partition name | 14:39 |
| *** iceyao has joined #openstack-kolla | 14:39 | |
| inc0 | so we started with parted -l and screenscrappin | 14:41 |
| inc0 | g | 14:41 |
| inc0 | it was horrible | 14:41 |
| inc0 | but maybe you'll have better luck | 14:41 |
| *** ravig has quit IRC | 14:43 | |
| Mech422 | inc0: its just a grep on partition name and a split... | 14:43 |
| Mech422 | inc0: I was worried it might be 'twitchy' but its working pretty good here | 14:43 |
| *** iceyao has quit IRC | 14:44 | |
| Mech422 | inc0: the 'twitchy' one for me seems to be mariadb - if I run cleanup-*, and try to re-run the kolla deploy - it appears mariadb gets stuck trying to recover itself | 14:45 |
| Mech422 | inc0: but I haven't really dug into it too much yet | 14:45 |
| *** daneyon_ has joined #openstack-kolla | 14:49 | |
| *** daneyon_ has quit IRC | 14:54 | |
| *** _tzn has joined #openstack-kolla | 14:56 | |
| *** tzn has quit IRC | 14:59 | |
| wirehead_ | okay, the large pot of tea has been made. Thinking may commence. | 15:06 |
| Mech422 | wirehead_: Nevah! | 15:06 |
| inc0 | I'm getting done with my first cup of green tea today | 15:07 |
| inc0 | thinking is a no-no yet | 15:07 |
| *** jtriley_ has joined #openstack-kolla | 15:10 | |
| *** jtriley has quit IRC | 15:12 | |
| *** salv-orlando has quit IRC | 15:22 | |
| *** salv-orlando has joined #openstack-kolla | 15:23 | |
| *** sacharya has joined #openstack-kolla | 15:25 | |
| *** jmccarthy has left #openstack-kolla | 15:30 | |
| *** mliima_ has joined #openstack-kolla | 15:30 | |
| inc0 | mag009_, soo...did you talk about etcd with wirehead_ and rhallisey ? | 15:30 |
| inc0 | you both implement etcd..\ | 15:31 |
| Lyncos | i would love to poke mag but he's not at the office | 15:32 |
| wirehead_ | I commented yesterday, but I'm not sure where we are with that. | 15:32 |
| *** mliima has quit IRC | 15:32 | |
| *** jmccarthy has joined #openstack-kolla | 15:33 | |
| *** cloudnau_ has joined #openstack-kolla | 15:33 | |
| *** salv-orlando has quit IRC | 15:34 | |
| *** salv-orlando has joined #openstack-kolla | 15:34 | |
| mag009_ | well mine is ready (worst case they can adjust it to their need by modifying mine) | 15:35 |
| mag009_ | or we go seperate ways | 15:35 |
| wirehead_ | So, we're talking about the etcd that Kubernetes uses that Kolla-Kubernetes runs atop, versus the etcd that some newer stuff that runs atop Kolla is using? | 15:36 |
| mag009_ | yes | 15:36 |
| mag009_ | i can rename my role to calic_etcd or something.. | 15:36 |
| mag009_ | calico* | 15:37 |
| wirehead_ | So, I think the best use of the Kolla-Kubernetes time is to not write another artisinal handmade installer. | 15:37 |
| wirehead_ | Ergo, there's a perfectly good Ansible playbook that installs a HA-Kubernetes config, including the etcd. | 15:37 |
| mag009_ | are you guys ready... | 15:37 |
| mag009_ | can i use kubernates stuff | 15:38 |
| wirehead_ | And the Kubernetes folks take a fairly dim view of the orchestrated services using Kubernetes's etcd. | 15:38 |
| mag009_ | sure i get that | 15:39 |
| mag009_ | my question is at the moment I see nothing in the master repo about k8s | 15:39 |
| mag009_ | what am i suppose to do | 15:39 |
| wirehead_ | Kolla-kubernetes repo | 15:40 |
| Lyncos | We are not using kubernete and we do not want to use it | 15:40 |
| mag009_ | when will it get merge to master? | 15:40 |
| Lyncos | we need to go prod asap | 15:40 |
| *** belmoreira has quit IRC | 15:41 | |
| wirehead_ | So, you should be able to clone the kolla-kubernetes repo and follow the quickstart (only doc patch not merged: https://review.openstack.org/#/c/324116/ ) and be able to get it kinda-sorta installing. | 15:41 |
| patchbot | wirehead_: patch 324116 - kolla-kubernetes - Added documentation for building Kolla containers. | 15:41 |
| *** matrohon has quit IRC | 15:42 | |
| wirehead_ | With the patches out there, we're just about to have a demonstratable config. A few more days likely. | 15:42 |
| *** mliima__ has joined #openstack-kolla | 15:42 | |
| *** SiRiuS__ has quit IRC | 15:42 | |
| mag009_ | look, i've done the work i'm not going to redo my stuff until it's merge in master repo | 15:43 |
| mag009_ | until than I'll just fork my stuff locally | 15:43 |
| wirehead_ | *shrug* Kolla-kubernetes still uses Kolla's infra. | 15:43 |
| *** mliima__ is now known as mliima | 15:44 | |
| wirehead_ | So it's probably just fine for you to continue your callico and etcd work in Kolla. | 15:44 |
| *** mliima_ has quit IRC | 15:44 | |
| mag009_ | i'd like my stuff to get merge if there are no problem then ? | 15:45 |
| mag009_ | unless other stuff get in conflict? | 15:46 |
| *** ssurana has joined #openstack-kolla | 15:47 | |
| mag009_ | inc0: you tell me... this k8s stuff I have no clue how its going to work it'd be nice to have some sort of flow | 15:49 |
| wirehead_ | I don't anticipate your patch being in conflict. | 15:49 |
| mag009_ | wirehead_: alright then | 15:50 |
| wirehead_ | If Kolla-Kubernetes installs Calico as part of the Kubernetes undercloud, we'd do it at the same time as we install Kubernetes. | 15:50 |
| inc0 | wirehead_, what I mean is, I don't want 2 sets of etcd containers | 15:51 |
| inc0 | write one etcd dockerfile that both of you will use | 15:51 |
| wirehead_ | Yeah, I'm not aware of anyone in Kolla-Kubernetes who's writing etcd containers. | 15:51 |
| inc0 | to my eyes, mag009_ wrote pretty solid one | 15:51 |
| inc0 | ahh so this is not a conflict, fair enough | 15:52 |
| mag009_ | but in review it says it is | 15:52 |
| mag009_ | https://review.openstack.org/#/c/298451/16 | 15:52 |
| patchbot | mag009_: patch 298451 - kolla - Add etcd Docker container | 15:52 |
| mag009_ | this one | 15:52 |
| *** ssurana1 has joined #openstack-kolla | 15:53 | |
| wirehead_ | Okay, that's for the kuryr plugin. | 15:53 |
| inc0 | hmm...hui should be here rather evening-ish | 15:55 |
| inc0 | mag009_, so can you review this etcd patch hui produced and see if it works for you? | 15:55 |
| wirehead_ | They look very similar. | 15:55 |
| wirehead_ | for the docker part. | 15:55 |
| *** SiRiuS__ has joined #openstack-kolla | 15:55 | |
| inc0 | I'd expect so | 15:55 |
| mag009_ | doesn't i need client as well | 15:56 |
| mag009_ | mine install etcd on each compute where calico is running | 15:56 |
| inc0 | not sure, check it out please, I don't know etcd nearly enough | 15:56 |
| mag009_ | acting as client | 15:56 |
| wirehead_ | etcd has a very simple REST API. | 15:56 |
| *** ssurana has quit IRC | 15:57 | |
| inc0 | mag009_, if we could use raw REST api, less containers running is good | 15:57 |
| inc0 | not sure how calico deals with it | 15:57 |
| inc0 | but if it doesn't deal with it well, I'm ok with etcd container | 15:58 |
| inc0 | (however if it's not running service, jsut a client, maybe put it into calico container?) | 15:58 |
| Lyncos | when you talking about client it is the etcd proxy ? | 15:58 |
| mag009_ | yes | 15:58 |
| Lyncos | proxy is needed .. you don't want 1 millions tcp connection to your etcd servers | 15:59 |
| *** dwalsh has quit IRC | 15:59 | |
| Lyncos | it is in the etcd best practice to use a local proxy on every node connecting to the cluster | 15:59 |
| inc0 | ok, I understand | 15:59 |
| inc0 | so how about that | 16:00 |
| *** ssurana has joined #openstack-kolla | 16:00 | |
| inc0 | you rebase your patch to huis | 16:00 |
| inc0 | and change only stuff to run it on compute | 16:00 |
| mag009_ | how about I do it from my stuff -_- i'll just make a flag to disable the client | 16:01 |
| mag009_ | when enable_calico | 16:01 |
| mag009_ | so if you just need the cluster on controller it will only deploy the etcd | 16:01 |
| inc0 | no, if it's needed on compute, it's needed there | 16:01 |
| *** mark-casey has joined #openstack-kolla | 16:01 | |
| *** Serlex has left #openstack-kolla | 16:01 | |
| mag009_ | for my calico yes | 16:01 |
| *** SiRiuS__ has quit IRC | 16:01 | |
| inc0 | issue is | 16:02 |
| inc0 | if both of these patches merge | 16:02 |
| inc0 | we get 2 etcd dockerfiles | 16:02 |
| inc0 | and I don't want that | 16:02 |
| mag009_ | just abandon hui stuff ? | 16:02 |
| mag009_ | my merge does more than what he need | 16:03 |
| mag009_ | i just make a flag to disable the extra that he doesn't need | 16:03 |
| inc0 | so let's do that | 16:03 |
| inc0 | -1 hui's patch now, point to your review | 16:03 |
| inc0 | and discuss how to combine efforts ok? | 16:04 |
| inc0 | he should be on later today | 16:04 |
| *** ssurana1 has quit IRC | 16:04 | |
| mag009_ | i'm leaving at 3pm | 16:04 |
| mag009_ | est | 16:04 |
| inc0 | duh | 16:04 |
| mag009_ | :) | 16:04 |
| inc0 | then talk over review comments;) | 16:04 |
| mag009_ | i will submit a patch | 16:04 |
| mag009_ | he can then review | 16:05 |
| inc0 | mag009_, sooo...etcd compute | 16:06 |
| inc0 | LISTEN_CLIENT_URLS binds to IP of host | 16:06 |
| inc0 | while INITIAL_ADVERTISE_PEER binds to localhost | 16:07 |
| inc0 | is that correct? | 16:07 |
| inc0 | and why? | 16:07 |
| inc0 | ok, mag009_ few comments, we need to change your inventory;) | 16:10 |
| wirehead_ | Oh, for a second I thought this conversation had devolved to yelling. Then I realized it was just uppercase config variable names. | 16:13 |
| *** ravig has joined #openstack-kolla | 16:13 | |
| inc0 | well, I guess ETCD devs are jumpy folk | 16:14 |
| wirehead_ | I used to work with some of them. | 16:14 |
| inc0 | since all configs are uppercase | 16:14 |
| inc0 | are they jumpy? | 16:14 |
| *** mummer has joined #openstack-kolla | 16:18 | |
| *** mummer has quit IRC | 16:18 | |
| wirehead_ | Not as jumpy as all-uppercase variable names would suggest | 16:19 |
| mag009_ | inc0: what is it ? | 16:19 |
| inc0 | soo, mag009_ I posted a review | 16:20 |
| inc0 | for etcd | 16:20 |
| inc0 | will continue reviewing felix now | 16:20 |
| inc0 | take look at comments and see if you know what I mean | 16:21 |
| *** ssurana has quit IRC | 16:21 | |
| *** diogogmt has joined #openstack-kolla | 16:22 | |
| *** ssurana has joined #openstack-kolla | 16:22 | |
| inc0 | mag009_, Lyncos btw, how tenant networks works in calico? | 16:23 |
| Lyncos | it is ranges | 16:24 |
| inc0 | is it only for giant flat networking? | 16:24 |
| Lyncos | you chose a specific range | 16:24 |
| inc0 | IP ranges? | 16:24 |
| Lyncos | and ip from that range will be asigned | 16:24 |
| Lyncos | yeah | 16:24 |
| inc0 | ok, so it's a flat network with multiple ranges | 16:24 |
| *** athomas has quit IRC | 16:24 | |
| inc0 | can you define ACL with it? | 16:24 |
| Lyncos | yeah | 16:24 |
| Lyncos | the ACL is made at the security groups level in openstack | 16:25 |
| inc0 | ok I see | 16:25 |
| Lyncos | the only isolation you have is via firwall rules (automated by OS) | 16:25 |
| *** ssurana has quit IRC | 16:29 | |
| *** ayoung has quit IRC | 16:33 | |
| *** ssurana has joined #openstack-kolla | 16:34 | |
| mag009_ | there's more than one line different from compute vs control | 16:38 |
| mag009_ | for my etcd config | 16:38 |
| mag009_ | can i still use one template with an if.. | 16:38 |
| mag009_ | I had to change it at your first request | 16:38 |
| inc0 | I know | 16:38 |
| inc0 | but you had 2 huge blocks | 16:38 |
| mag009_ | i get what you mean | 16:38 |
| inc0 | with an if | 16:38 |
| inc0 | I didn't notice that changes are so slim | 16:39 |
| mag009_ | well i can do the if for the diff only | 16:39 |
| inc0 | that'd be better way | 16:39 |
| inc0 | also, this localhost is required? | 16:39 |
| mag009_ | yes for the proxy | 16:39 |
| inc0 | can it be an IP? | 16:39 |
| inc0 | localhost can resolve to any ip on host | 16:39 |
| inc0 | and while it probably keep working, we tend to bind to IP | 16:40 |
| inc0 | because of security | 16:40 |
| *** alyson_ has joined #openstack-kolla | 16:40 | |
| inc0 | also if localhost resolves to 127.0.0.1 and service would use 192.168.0.100 (ip of api iface) it might break | 16:40 |
| harlowja | btw, mdorman put a bunch of good stuff on https://etherpad.openstack.org/p/kolla-gd-thoughts-concerns-remarks if people get some time to look at | 16:40 |
| inc0 | so unless there is reason to use localhost, our standard is to use interface itsefl | 16:41 |
| mag009_ | i can use the ip | 16:42 |
| inc0 | yeah, do that then plz | 16:42 |
| inc0 | also won't that make it closer to controller config? | 16:42 |
| *** ravig has quit IRC | 16:44 | |
| mag009_ | i have to compare maybe | 16:44 |
| *** ravig has joined #openstack-kolla | 16:45 | |
| *** ravig has quit IRC | 16:48 | |
| *** sdake has joined #openstack-kolla | 16:50 | |
| *** ravig has joined #openstack-kolla | 16:51 | |
| *** vhosakot has joined #openstack-kolla | 16:52 | |
| inc0 | harlowja, answered remakrs | 16:53 |
| esharao | hi.. should the tags for the lokolla/centos-binary-heka be 2.0.0 or 3.0.0? | 16:53 |
| esharao | i see that in the registry its 3.0.0 while the ansible playbook is searching for 2.0.0 | 16:54 |
| inc0 | so main thing I see is that you really want your own puppet modules right?;) | 16:54 |
| inc0 | esharao, hold on | 16:54 |
| inc0 | https://github.com/openstack/kolla/blob/master/ansible/group_vars/all.yml#L163 that's the reason | 16:55 |
| inc0 | wanna patch it up or I should do it?;) | 16:55 |
| *** ravig has quit IRC | 16:55 | |
| inc0 | you found that, you should take the credit;) | 16:55 |
| *** ravig has joined #openstack-kolla | 16:56 | |
| harlowja | inc0 thx | 16:58 |
| sdake | morning | 17:00 |
| harlowja | inc0 i think mdorman would like it to be a staged thing, which i think is becoming ok/not impossible | 17:06 |
| inc0 | yeah, well, you are first one asking for puppet really, but we figured that will be the case | 17:07 |
| inc0 | so our dockerfiles can be easily consumed without ansible | 17:07 |
| harlowja | right, think we all learned that :) | 17:07 |
| harlowja | which is cool | 17:07 |
| inc0 | so your case will drop a lot of complexity right now | 17:07 |
| harlowja | kolla-build, kolla-ansible... | 17:07 |
| inc0 | as most of complexity comes from bootstrapping/orchiestration | 17:08 |
| inc0 | starting containers is easiest part;) | 17:08 |
| inc0 | just be careful to mimic config which ansible uses during startup | 17:08 |
| inc0 | like mounts and such | 17:08 |
| harlowja | right | 17:08 |
| inc0 | and you'll going to be ok | 17:08 |
| harlowja | do u swear to that | 17:09 |
| harlowja | lol | 17:09 |
| harlowja | ____ sign | 17:09 |
| harlowja | lol | 17:09 |
| inc0 | ok, I take that back, you're using puppet, you're in the world of pain | 17:09 |
| inc0 | but I guess you should be used to it by now | 17:09 |
| inc0 | maybe even like it | 17:10 |
| *** ayoung has joined #openstack-kolla | 17:14 | |
| *** ravig has quit IRC | 17:16 | |
| cloudnau_ | I have a networking question, I apologize if its basic, I have two VMs on my laptop each with a bridge and a veth pair. One end on the bridge the other given to neutron with OVS. | 17:17 |
| harlowja | inc0 :-P | 17:17 |
| *** ravig has joined #openstack-kolla | 17:17 | |
| harlowja | lol | 17:17 |
| harlowja | to the pain! | 17:17 |
| cloudnau_ | How do I configure the host so that I can ping the default 10.0.2.x ips | 17:17 |
| cloudnau_ | Its madness with the vswitch, and bridges :) | 17:18 |
| harlowja | https://www.youtube.com/watch?v=I_keWS1i3RA (to the pain!) | 17:18 |
| inc0 | cloudnau_, sooo... | 17:18 |
| cloudnau_ | Or is there a good article that can point me in the right direction | 17:18 |
| inc0 | when vm is created | 17:18 |
| inc0 | it creates veth, as you said, you add this to ovs | 17:18 |
| inc0 | however that's what neutron should be oding... | 17:18 |
| cloudnau_ | It did | 17:19 |
| cloudnau_ | I think | 17:19 |
| cloudnau_ | I give Kolla the vbr1 iface for neutron | 17:19 |
| cloudnau_ | it adds it to ovs-system? | 17:19 |
| cloudnau_ | it looksl ike | 17:19 |
| cloudnau_ | which then has it tied to br-ex | 17:19 |
| inc0 | do you have to give bridge to kolla? | 17:20 |
| inc0 | well not kolla, neutron adds it to ovs | 17:20 |
| cloudnau_ | I have a single interface on the box | 17:20 |
| cloudnau_ | I could add a second interface to the boxes... but I find myself in the same situation... | 17:21 |
| cloudnau_ | no idea how to reach the VMs :) | 17:21 |
| inc0 | btw, did you check namespaces?;) | 17:21 |
| inc0 | vms are up and running? | 17:21 |
| inc0 | if so, check ip netns | 17:21 |
| *** sdake has quit IRC | 17:22 | |
| sean-k-mooney | cloudnau_: if you are using neutron and you want to reach vms on a private network you will either have to assign a private ip or change in to the network name space of the neutron networks dhcp server | 17:23 |
| cloudnau_ | I have console in horizon to the VM | 17:23 |
| cloudnau_ | I have router and dhcp on the controller node | 17:23 |
| inc0 | cloudnau_, paste output of ip netns plz | 17:23 |
| sean-k-mooney | s/private ip/floating ip/ | 17:23 |
| cloudnau_ | [root@rancher-controller ~]# ip netns | 17:24 |
| cloudnau_ | RTNETLINK answers: Invalid argument | 17:24 |
| cloudnau_ | RTNETLINK answers: Invalid argument | 17:24 |
| cloudnau_ | RTNETLINK answers: Invalid argument | 17:24 |
| cloudnau_ | qrouter-13b735a3-89f7-4af7-9747-becdb9e1b42d | 17:24 |
| cloudnau_ | RTNETLINK answers: Invalid argument | 17:24 |
| cloudnau_ | qdhcp-41a3345b-c91a-41b0-b7f8-8249f99dd8e4 | 17:24 |
| inc0 | and ip of vm is 10.0.2.1? | 17:24 |
| inc0 | well in any case | 17:24 |
| cloudnau_ | thats the gateway | 17:24 |
| cloudnau_ | the VM got 10.0.2.151 | 17:24 |
| inc0 | try ip netns exec qdhcp-41a3345b-c91a-41b0-b7f8-8249f99dd8e4 ping 10.0.2.151 | 17:25 |
| cloudnau_ | I'm getting an invalid argument :-\ | 17:25 |
| cloudnau_ | error | 17:25 |
| cloudnau_ | sean-k-mooney, where would I point the floating IP? | 17:26 |
| sean-k-mooney | cloudnau_: you will need to run that with sodo if you are not root | 17:26 |
| *** ravig has quit IRC | 17:26 | |
| cloudnau_ | I'm root. | 17:26 |
| inc0 | hmm...I've seen it somewhere, can't remember where tho, your network namespaces seems dirty | 17:27 |
| sean-k-mooney | cloudnau_: you would point the floating ip to the vms private ip 10.0.2.151 in this case | 17:27 |
| cloudnau_ | its possible. I have been running through this 100+ times :) | 17:27 |
| sean-k-mooney | cloudnau_: if you cant ping form the dhcp network namespace i would guess floating ips wont work either | 17:28 |
| cloudnau_ | Its been deployed using Rancher | 17:28 |
| sean-k-mooney | cloudnau_: are you using rancherOS? | 17:28 |
| *** salv-orl_ has joined #openstack-kolla | 17:28 | |
| cloudnau_ | Indeed | 17:28 |
| cloudnau_ | I work at rancher | 17:28 |
| *** ravig has joined #openstack-kolla | 17:28 | |
| sean-k-mooney | cloudnau_: ah cool i tried that once and didnt get it to work with kolla. | 17:29 |
| cloudnau_ | I'm using our latest builds... we need to add 'shared' mount support, but for the most part it seems good | 17:29 |
| sean-k-mooney | cloudnau_: rancherOS is an interesting os to use to host a kolla cloud | 17:29 |
| inc0 | cloudnau_, any chances you could try this out on fresh OS? | 17:30 |
| *** ravig has quit IRC | 17:30 | |
| cloudnau_ | Yeah, its nicely suited for container workloads. I'm just deploying a wrapper container that sets up enough for ansible to run | 17:30 |
| cloudnau_ | and letting Kolla do the rest | 17:30 |
| cloudnau_ | Yeah inc0 | 17:30 |
| cloudnau_ | I'll give that a shot | 17:30 |
| *** mdnadeem_home has joined #openstack-kolla | 17:31 | |
| inc0 | please do | 17:31 |
| inc0 | with docker 1.10 docker changed way it deals with network namespaces | 17:31 |
| inc0 | it user to do bad things before | 17:31 |
| *** salv-orlando has quit IRC | 17:32 | |
| sean-k-mooney | inc0: before 1.10 you could not see them correctly with ip netns if i remember correctly right | 17:32 |
| inc0 | yeah, and that invalid argument was raised | 17:33 |
| inc0 | so that seems familiar;) | 17:33 |
| sean-k-mooney | cloudnau_: what version of docker is the user docker in your build of rancherOS? | 17:33 |
| inc0 | keep in mind kolla requires 1.10 | 17:34 |
| inc0 | or later | 17:34 |
| *** salv-orl_ has quit IRC | 17:35 | |
| *** salv-orlando has joined #openstack-kolla | 17:36 | |
| mdnadeem_home | inc0, have you face any similar issue while upgrade from liberty to mitaka : https://bugs.launchpad.net/kolla/+bug/1590639 | 17:37 |
| openstack | Launchpad bug 1590639 in kolla "Kolla upgrade fail Mariadb keeps on restarting" [Undecided,New] | 17:37 |
| inc0 | hmm | 17:38 |
| inc0 | no, but that might be because we didnt test this case | 17:38 |
| inc0 | it's aio? | 17:38 |
| mdnadeem_home | inc0, yeh, centOs | 17:39 |
| inc0 | so with multinode it might be better | 17:39 |
| inc0 | what happend is that you performed upgrade during transaction | 17:39 |
| inc0 | I'm not sure if multinode will handle that gracefully too | 17:39 |
| mdnadeem_home | inc0, some container gets upgrade to mitaka, then mariadb container get fail and script exit | 17:41 |
| cloudnau_ | docker 1.11.2 | 17:41 |
| mdnadeem_home | inc0, have a look my container status : http://pastebin.com/ny4tz5D0 | 17:42 |
| inc0 | mdnadeem_home, did you try mariadb_recovery? | 17:43 |
| *** ravig has joined #openstack-kolla | 17:43 | |
| mdnadeem_home | inc0, not yet, could you please let me know how to do it? | 17:43 |
| inc0 | kolla-ansible mariadb_recovery | 17:45 |
| cinerama | sean-k-mooney, awesome work, thanks for updating your change. there is one tiny thing i spotted which i think snuck in during your rebase but so far everything else looks good | 17:45 |
| mdnadeem_home | inc0, ok, I am trying it | 17:45 |
| *** klindgren has quit IRC | 17:46 | |
| mdnadeem_home | inc0, it fails by saying msg: There are running MariaDB nodes, please stop them first. | 17:48 |
| inc0 | well....there you go | 17:48 |
| inc0 | you need to stop them first... | 17:48 |
| mdnadeem_home | inc0, currently mariadb container is in restart state | 17:48 |
| inc0 | so issue is, if you do full restart of all galera nodes | 17:48 |
| inc0 | which is single one in aio | 17:49 |
| inc0 | you can't just start them up | 17:49 |
| *** _tzn has quit IRC | 17:49 | |
| inc0 | you can mess up your data | 17:49 |
| inc0 | which is probably what you did:P | 17:49 |
| mdnadeem_home | inc0, so currently, I just use docker stop mariadb and again execute mariadb recover right? | 17:51 |
| inc0 | try that | 17:51 |
| mdnadeem_home | hmmm | 17:52 |
| *** mbound has quit IRC | 17:55 | |
| *** shardy has quit IRC | 17:56 | |
| *** tzn has joined #openstack-kolla | 17:58 | |
| *** salv-orlando has quit IRC | 17:58 | |
| mdnadeem_home | inc0, again mariadb recovery fail :(>> msg: Timeout when waiting for search string MariaDB in 192.168.121.165:3306 | 17:58 |
| *** salv-orlando has joined #openstack-kolla | 17:59 | |
| mdnadeem_home | inc0, its getting too late here, feeling sleepy, will let you know if get any solution for it, Thanks :) | 18:05 |
| inc0 | thanks mdnadeem_ | 18:06 |
| inc0 | thanks mdnadeem_home | 18:06 |
| *** mdnadeem_home has quit IRC | 18:09 | |
| *** sacharya has quit IRC | 18:13 | |
| *** dwalsh has joined #openstack-kolla | 18:22 | |
| *** vhosakot has quit IRC | 18:25 | |
| *** daneyon_ has joined #openstack-kolla | 18:26 | |
| *** tzn has quit IRC | 18:27 | |
| *** ravig has quit IRC | 18:31 | |
| *** sacharya has joined #openstack-kolla | 18:31 | |
| *** daneyon_ has quit IRC | 18:31 | |
| *** tzn has joined #openstack-kolla | 18:31 | |
| *** ssurana has quit IRC | 18:42 | |
| *** ssurana has joined #openstack-kolla | 18:42 | |
| *** ssurana has quit IRC | 18:43 | |
| *** dwalsh has quit IRC | 18:45 | |
| *** ssurana has joined #openstack-kolla | 18:46 | |
| *** daneyon_ has joined #openstack-kolla | 18:48 | |
| *** tzn has quit IRC | 18:52 | |
| *** dwalsh has joined #openstack-kolla | 18:53 | |
| *** ssurana has quit IRC | 18:53 | |
| *** vhosakot has joined #openstack-kolla | 18:54 | |
| *** ravig has joined #openstack-kolla | 18:55 | |
| *** ssurana has joined #openstack-kolla | 18:55 | |
| *** daneyon_ has quit IRC | 18:55 | |
| *** mbound has joined #openstack-kolla | 18:56 | |
| *** Serlex has joined #openstack-kolla | 18:57 | |
| *** dwalsh has quit IRC | 18:58 | |
| *** fragatina has quit IRC | 18:59 | |
| *** fragatina has joined #openstack-kolla | 18:59 | |
| *** mbound has quit IRC | 19:01 | |
| *** ppowell has quit IRC | 19:04 | |
| *** fragatina has quit IRC | 19:04 | |
| *** athomas has joined #openstack-kolla | 19:06 | |
| esharao | hi.. another question.. i keep on failing at kolla_start when i try to run kolla-ansible deploy for the very first task - starting heka container.. | 19:22 |
| esharao | i just tried doing a docker run and it fails with the "docker: Error response from daemon: Container command 'kolla_start' not found or does not exist.." | 19:22 |
| esharao | what did i miss? | 19:22 |
| mdorman | cloudnau_: o/ ! saw your tweet about kolla, we are looking at it too :) | 19:23 |
| mdorman | inc0: / harlowja, sorry i was out all morning so just now getting updated on irc. thanks for the comments on the etherpad. again, it’s not so much puppet vs. ansible as it is having the freedom/ability to choose your own orchestration tool. so the kolla-ansible/kolla-build idea IMO is a good direction. We can take a look at the advanced configuration stuff more, and yeah i think you’re probably right, we should be | 19:26 |
| mdorman | able to leverage all this. I am relatively green on both kolla and docker, so there are probably some assumptions i have in my head that are wrong. | 19:26 |
| mdorman | now just to find some time to work on this :-/ | 19:26 |
| harlowja | ya, lab mdorman seems like a good place to do this | 19:26 |
| harlowja | (when lab is enabled/arrivfed/ready) | 19:26 |
| mdorman | yah, as soon as they get all those cables plugged in! | 19:27 |
| *** dwalsh has joined #openstack-kolla | 19:28 | |
| inc0 | mdorman, let me know if you run into hurdles, I'll be happy to help | 19:28 |
| harlowja | inc0 u have been enlisted | 19:29 |
| harlowja | lol | 19:29 |
| mdorman | thanks i appreciate it. | 19:29 |
| inc0 | harlowja, sure, will send you invoice later | 19:29 |
| mark-casey | could someone remind me the difference between upgrade and reconfigure? | 19:30 |
| mark-casey | (in the action taken, I obviously get the end goals for each :D ) | 19:30 |
| inc0 | mark-casey, upgrade == release upgrade | 19:31 |
| inc0 | reconfigure won't redeploy configs | 19:31 |
| inc0 | (at least shouldnt) | 19:31 |
| inc0 | also upgrade has some more mechanics build in | 19:32 |
| harlowja | inc0 beer invoice is ok | 19:32 |
| harlowja | free as in beer right | 19:32 |
| harlowja | lol | 19:32 |
| mark-casey | reconfigure as in... the configs that are on the nodes do not change? even if changed on deploying host? | 19:32 |
| inc0 | configs will change | 19:33 |
| inc0 | but now when I look at it | 19:33 |
| inc0 | it doesn't look super nice | 19:33 |
| inc0 | at least for nova case | 19:33 |
| *** mbound has joined #openstack-kolla | 19:34 | |
| inc0 | so with COPY_ALWAYS it will just restart containers | 19:35 |
| inc0 | with COPY_ONCE it really messes stuff up | 19:35 |
| inc0 | I'll submit bug about that | 19:35 |
| inc0 | also...question to operators... anyone cares about copy_once config strategy? | 19:35 |
| *** mliima has quit IRC | 19:36 | |
| *** tzn has joined #openstack-kolla | 19:36 | |
| *** tzn has quit IRC | 19:36 | |
| inc0 | mark-casey, another thing, for reconfigure you don't really call syncdb | 19:41 |
| *** mark-casey has quit IRC | 19:43 | |
| *** mark-casey has joined #openstack-kolla | 19:47 | |
| *** cloudnau_ has quit IRC | 19:49 | |
| *** dwalsh has quit IRC | 19:51 | |
| *** sdake has joined #openstack-kolla | 19:52 | |
| *** esharao has quit IRC | 20:04 | |
| inc0 | soo Mech422 I'm thinking about implementation of this multi-file | 20:05 |
| *** rhallisey has quit IRC | 20:05 | |
| inc0 | and best idea I have so far is to have jinja tpl that renders jinja tpl :D | 20:06 |
| inc0 | meta-templating | 20:06 |
| sdake | inc0 loves jinja! | 20:06 |
| inc0 | sdake, speaking of which, review a patch stream | 20:07 |
| inc0 | you wanted it most | 20:08 |
| sdake | jusst woke up | 20:08 |
| sdake | havinga bad day today it appears | 20:08 |
| inc0 | I need this through so I can start working on a) refactoring and b) additional features | 20:09 |
| inc0 | like multiple override files | 20:09 |
| sdake | link | 20:10 |
| inc0 | thats first https://review.openstack.org/#/c/324665/ | 20:10 |
| patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 20:10 |
| *** jtriley_ has quit IRC | 20:14 | |
| sdake | inc0 i don't understand jeffrey4l's comment | 20:17 |
| sdake | could you explain it pls | 20:17 |
| *** vhosakot has quit IRC | 20:18 | |
| inc0 | instead of having one file he suggested having a lot | 20:18 |
| sdake | zug | 20:18 |
| inc0 | like you provide heat/heat-api/heat-api-customizatin | 20:18 |
| inc0 | you mirror our structure | 20:18 |
| inc0 | but I think that's not good idea | 20:18 |
| sdake | next link | 20:25 |
| *** Mr_Broken has joined #openstack-kolla | 20:25 | |
| inc0 | sdake, you have it in related changes | 20:26 |
| inc0 | https://review.openstack.org/#/c/326688/ | 20:27 |
| patchbot | inc0: patch 326688 - kolla - Add customizations for apt keys and packages | 20:27 |
| sdake | read a review for the 1080gtx | 20:29 |
| sdake | its through the roof faster | 20:29 |
| inc0 | I just bought new radeon few months ago, I'm good for few years | 20:30 |
| *** cloudnau_ has joined #openstack-kolla | 20:32 | |
| cloudnau_ | mdorman cool :) | 20:33 |
| *** d_code has quit IRC | 20:35 | |
| cloudnau_ | inc0 what would be the ideal network config if I have two VMWare Fusion VMs on a mac? two nics bridged to the phys net? or 1 physnet one local net? | 20:37 |
| inc0 | doesn't really matter | 20:39 |
| *** fragatina has joined #openstack-kolla | 20:39 | |
| inc0 | both will allow you to spawn and ping vms | 20:40 |
| *** Serlex has quit IRC | 20:40 | |
| *** Mr_Broken has quit IRC | 20:40 | |
| cloudnau_ | K. Thanks | 20:40 |
| *** ravig has quit IRC | 20:46 | |
| *** ravig has joined #openstack-kolla | 20:49 | |
| *** ayoung has quit IRC | 20:49 | |
| *** ravig has quit IRC | 20:54 | |
| *** ravig has joined #openstack-kolla | 20:57 | |
| *** d_code has joined #openstack-kolla | 21:03 | |
| kklimonda | is there a reason why role dependencies are only listing common, and not all the dependencies? | 21:24 |
| inc0 | kklimonda, well, technically only role dependency is common | 21:26 |
| *** d_code has quit IRC | 21:31 | |
| *** inc0 has quit IRC | 21:31 | |
| *** sdake has quit IRC | 21:36 | |
| *** sdake has joined #openstack-kolla | 21:41 | |
| *** sdake_ has joined #openstack-kolla | 21:43 | |
| *** sdake has quit IRC | 21:45 | |
| *** iceyao has joined #openstack-kolla | 21:54 | |
| *** d_code has joined #openstack-kolla | 21:58 | |
| *** iceyao has quit IRC | 21:59 | |
| *** dave-mccowan has quit IRC | 22:02 | |
| *** fragatin_ has joined #openstack-kolla | 22:03 | |
| *** fragatina has quit IRC | 22:07 | |
| *** alyson_ has quit IRC | 22:08 | |
| *** vhosakot has joined #openstack-kolla | 22:15 | |
| *** ravig has quit IRC | 22:35 | |
| *** vhosakot has quit IRC | 22:36 | |
| *** ravig has joined #openstack-kolla | 22:44 | |
| kklimonda | ha, I got a ~working contrail+openstack deployment with kolla | 22:46 |
| *** Jeffrey4l_ has joined #openstack-kolla | 23:01 | |
| *** fragatin_ has quit IRC | 23:22 | |
| *** fragatina has joined #openstack-kolla | 23:23 | |
| *** salv-orl_ has joined #openstack-kolla | 23:28 | |
| *** salv-orlando has quit IRC | 23:31 | |
| *** sacharya has quit IRC | 23:34 | |
| *** mbound has quit IRC | 23:36 | |
| *** mark-casey has quit IRC | 23:39 | |
| openstackgerrit | Ken Wronkiewicz proposed openstack/kolla-kubernetes: Convert Keystone to HostNetwork=False, Fix 2/3rds of bootstrap https://review.openstack.org/328567 | 23:45 |
| *** ravig has quit IRC | 23:48 | |
| openstackgerrit | Joshua Harlow proposed openstack/kolla: Make kolla-ansible a top level python command https://review.openstack.org/324982 | 23:49 |
| *** ravig has joined #openstack-kolla | 23:55 | |
| openstackgerrit | Ken Wronkiewicz proposed openstack/kolla: Keystone interface address and memcached override https://review.openstack.org/328569 | 23:55 |
| *** ravig has quit IRC | 23:55 | |
| *** cloudnau_ has quit IRC | 23:56 | |
| openstackgerrit | Joshua Harlow proposed openstack/kolla: Add a cloud-init RHEL(ish) bootstrap + run kolla script https://review.openstack.org/321802 | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!