| *** lukepatrick has joined #openstack-kolla | 00:02 | |
| *** sean-md has joined #openstack-kolla | 00:05 | |
| *** jtriley has joined #openstack-kolla | 00:06 | |
| *** sean-md has quit IRC | 00:09 | |
| *** ntpttr_laptop__ has quit IRC | 00:10 | |
| *** yangyape_ has quit IRC | 00:13 | |
| *** yangyapeng has joined #openstack-kolla | 00:13 | |
| *** yangyapeng has quit IRC | 00:18 | |
| *** yingjun has joined #openstack-kolla | 00:18 | |
| *** yangyapeng has joined #openstack-kolla | 00:21 | |
| *** oanson has quit IRC | 00:22 | |
| *** sean-md has joined #openstack-kolla | 00:23 | |
| *** oanson has joined #openstack-kolla | 00:24 | |
| *** chas has quit IRC | 00:25 | |
| *** yangyapeng has quit IRC | 00:25 | |
| *** jtriley has quit IRC | 00:25 | |
| *** chas has joined #openstack-kolla | 00:25 | |
| *** sean-md has quit IRC | 00:27 | |
| *** chas has quit IRC | 00:30 | |
| *** tovin07_ has joined #openstack-kolla | 00:36 | |
| *** lukepatrick has quit IRC | 00:37 | |
| *** lukepatrick has joined #openstack-kolla | 00:38 | |
| *** sean-md has joined #openstack-kolla | 00:41 | |
| *** sean-md has quit IRC | 00:45 | |
| *** huanxie has joined #openstack-kolla | 00:46 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla-kubernetes master: Updated from global requirements https://review.openstack.org/509443 | 00:48 |
|---|---|---|
| openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla-kubernetes master: Updated from global requirements https://review.openstack.org/509443 | 00:55 |
| *** zhurong has joined #openstack-kolla | 00:55 | |
| *** lukepatrick has quit IRC | 00:57 | |
| *** sean-md has joined #openstack-kolla | 00:59 | |
| *** sean-md has quit IRC | 01:03 | |
| *** jtriley has joined #openstack-kolla | 01:08 | |
| *** zhangfei has joined #openstack-kolla | 01:10 | |
| *** jtriley has quit IRC | 01:13 | |
| *** sean-md has joined #openstack-kolla | 01:17 | |
| *** tone_z has joined #openstack-kolla | 01:20 | |
| *** sean-md has quit IRC | 01:21 | |
| inc0 | duonghq Jeffrey4l around? | 01:23 |
| openstackgerrit | jinke proposed openstack/kolla-ansible master: Add dvr_no_external l3 dvr agent mode https://review.openstack.org/510327 | 01:27 |
| *** yangyapeng has joined #openstack-kolla | 01:27 | |
| *** yangyapeng has quit IRC | 01:29 | |
| *** yangyapeng has joined #openstack-kolla | 01:29 | |
| *** yangyape_ has joined #openstack-kolla | 01:31 | |
| *** yangyapeng has quit IRC | 01:31 | |
| *** itlinux has joined #openstack-kolla | 01:32 | |
| *** caowei has joined #openstack-kolla | 01:39 | |
| *** danardelean has joined #openstack-kolla | 01:41 | |
| *** danardelean has quit IRC | 01:46 | |
| *** yangyapeng has joined #openstack-kolla | 01:51 | |
| *** oanson has quit IRC | 01:52 | |
| *** zhubingbing_ has quit IRC | 01:52 | |
| *** zhubingbing__ has joined #openstack-kolla | 01:52 | |
| *** yangyape_ has quit IRC | 01:52 | |
| *** oanson has joined #openstack-kolla | 01:52 | |
| *** itlinux has quit IRC | 02:03 | |
| *** unicell has quit IRC | 02:03 | |
| zhubingbing__ | Hi guys | 02:04 |
| inc0 | hey Zhubing | 02:11 |
| duonghq | hi inc0 | 02:17 |
| Jeffrey4l | sup inc0 | 02:19 |
| inc0 | hey guys | 02:19 |
| inc0 | Jeffrey4l: I've made few small edits to slides | 02:20 |
| inc0 | how do you feel before summit? | 02:20 |
| Jeffrey4l | about the onboarding session, do you have any ppt for this? | 02:20 |
| inc0 | anything I can help you with? | 02:20 |
| inc0 | no, I didn't have any | 02:20 |
| Jeffrey4l | thanks. | 02:20 |
| inc0 | I just used whiteboard and explained structure of project | 02:20 |
| Jeffrey4l | nervous ;p | 02:20 |
| inc0 | like, "imagea are in docker directory, they're rendered from jinja2" | 02:20 |
| Jeffrey4l | is projector possible? | 02:21 |
| inc0 | I'm not sure, you should ask Kendall | 02:21 |
| Jeffrey4l | should i show a online demo? | 02:21 |
| inc0 | send her mail | 02:21 |
| Jeffrey4l | ok. | 02:21 |
| inc0 | so goal of onboarding is to teaach new contributors to understand code | 02:22 |
| inc0 | people it's designed for are people who would like to join our community but don't know where to start | 02:22 |
| Jeffrey4l | so mainly developer, not operator, right? | 02:22 |
| inc0 | well, our community is made of operators too | 02:22 |
| inc0 | but yeah, people who will write kolla code | 02:23 |
| inc0 | hopefully;) | 02:23 |
| Jeffrey4l | got | 02:23 |
| inc0 | duonghq you'll be around next week right? | 02:23 |
| inc0 | it's gonna be quiet time in openstack which means it's good time to work on gates - can you help me? | 02:24 |
| duonghq | inc0, sure, I cannot go to Sydney, so I can have more free time | 02:24 |
| inc0 | cool:) | 02:25 |
| inc0 | so about gates | 02:25 |
| inc0 | https://review.openstack.org/#/c/512779/ can you guys take a look at this failure? I'm lost | 02:25 |
| *** salv-orl_ has joined #openstack-kolla | 02:25 | |
| * duonghq opening the linj | 02:26 | |
| inc0 | https://review.openstack.org/#/c/513855/ <- also let's please merge this one asap, so we'll have time to test out pushing mechanism | 02:26 |
| inc0 | https://review.openstack.org/#/c/516817/ I think we'll need this one too (responded to your commend duonghq) | 02:26 |
| *** Jeffrey4l has quit IRC | 02:26 | |
| duonghq | got it | 02:27 |
| *** chas has joined #openstack-kolla | 02:27 | |
| *** salv-orlando has quit IRC | 02:28 | |
| *** Jeffrey4l has joined #openstack-kolla | 02:29 | |
| *** zhurong has quit IRC | 02:30 | |
| *** sean-md has joined #openstack-kolla | 02:30 | |
| *** chas has quit IRC | 02:31 | |
| *** sean-md has quit IRC | 02:34 | |
| *** yangyapeng has quit IRC | 02:37 | |
| *** yangyapeng has joined #openstack-kolla | 02:38 | |
| *** sean-md has joined #openstack-kolla | 02:40 | |
| *** zhurong has joined #openstack-kolla | 02:40 | |
| *** sean-md has quit IRC | 02:44 | |
| *** sean-md has joined #openstack-kolla | 02:57 | |
| *** goldyfruit has quit IRC | 02:59 | |
| *** sean-md has quit IRC | 03:02 | |
| *** wcen has quit IRC | 03:05 | |
| *** yangyapeng has quit IRC | 03:11 | |
| *** yangyapeng has joined #openstack-kolla | 03:11 | |
| *** sean-md has joined #openstack-kolla | 03:15 | |
| *** sean-md has quit IRC | 03:20 | |
| *** mwynne has quit IRC | 03:20 | |
| *** gcb has joined #openstack-kolla | 03:20 | |
| *** wcen has joined #openstack-kolla | 03:24 | |
| *** yingjun has quit IRC | 03:26 | |
| *** sean-md has joined #openstack-kolla | 03:33 | |
| *** zhangfei has quit IRC | 03:35 | |
| *** sean-md has quit IRC | 03:38 | |
| *** xinliang has quit IRC | 03:43 | |
| *** yangyapeng has quit IRC | 03:45 | |
| *** yangyapeng has joined #openstack-kolla | 03:46 | |
| *** zhangfei has joined #openstack-kolla | 03:48 | |
| *** sean-md has joined #openstack-kolla | 03:52 | |
| *** xinliang has joined #openstack-kolla | 03:55 | |
| *** sean-md has quit IRC | 03:56 | |
| *** coolsvap has joined #openstack-kolla | 03:58 | |
| *** binbincong has joined #openstack-kolla | 04:01 | |
| *** binbincong has quit IRC | 04:02 | |
| *** bbc_ has joined #openstack-kolla | 04:03 | |
| *** kiennt26 has quit IRC | 04:04 | |
| spsurya | morning guys | 04:04 |
| *** sean-md has joined #openstack-kolla | 04:10 | |
| *** sean-md has quit IRC | 04:15 | |
| SamYaple | /w/win 20 | 04:20 |
| *** huanxie has quit IRC | 04:25 | |
| *** chas has joined #openstack-kolla | 04:28 | |
| *** sean-md has joined #openstack-kolla | 04:28 | |
| *** chas has quit IRC | 04:32 | |
| *** sean-md has quit IRC | 04:32 | |
| japestinho | which docker-py package should I use for pike? from pip or centos repo one? | 04:38 |
| japestinho | I got this error | 04:38 |
| japestinho | https://www.irccloud.com/pastebin/SJQmk690/ | 04:38 |
| *** skramaja has joined #openstack-kolla | 04:39 | |
| *** huanxie has joined #openstack-kolla | 04:42 | |
| *** zhurong has quit IRC | 04:43 | |
| *** sean-md has joined #openstack-kolla | 04:46 | |
| openstackgerrit | Merged openstack/kolla master: update kafka version. https://review.openstack.org/517180 | 04:49 |
| *** unicell has joined #openstack-kolla | 04:50 | |
| *** sean-md has quit IRC | 04:50 | |
| *** lukepatrick has joined #openstack-kolla | 04:50 | |
| spsurya | duonghq: IIUC need to merge this one https://review.openstack.org/#/c/516817/ | 04:54 |
| *** sean-md has joined #openstack-kolla | 05:05 | |
| *** caowei has quit IRC | 05:05 | |
| *** janki has joined #openstack-kolla | 05:08 | |
| *** sean-md has quit IRC | 05:09 | |
| duonghq | spsurya, binary gates has not been stable atm. | 05:14 |
| *** unicell1 has joined #openstack-kolla | 05:15 | |
| *** unicell has quit IRC | 05:16 | |
| *** lukepatrick has quit IRC | 05:16 | |
| *** sean-md has joined #openstack-kolla | 05:23 | |
| *** sean-md has quit IRC | 05:27 | |
| *** caowei has joined #openstack-kolla | 05:33 | |
| *** oanson has quit IRC | 05:34 | |
| *** oanson has joined #openstack-kolla | 05:34 | |
| spsurya | duonghq: i think merging this will not create any problem | 05:35 |
| *** sean-md has joined #openstack-kolla | 05:41 | |
| *** sean-md has quit IRC | 05:45 | |
| *** zhurong has joined #openstack-kolla | 05:58 | |
| duonghq | spsurya, of course | 06:02 |
| spsurya | duonghq: done | 06:03 |
| duonghq | spsurya, nice | 06:12 |
| *** chas has joined #openstack-kolla | 06:28 | |
| *** huanxie has quit IRC | 06:29 | |
| *** hachi has joined #openstack-kolla | 06:31 | |
| *** chas has quit IRC | 06:33 | |
| openstackgerrit | zhichao zhu proposed openstack/kolla-ansible master: fix shell error to create ipsec-vpn https://review.openstack.org/517553 | 06:35 |
| openstackgerrit | zhichao zhu proposed openstack/kolla-ansible master: fix shell error about tools/init-vpn https://review.openstack.org/517553 | 06:38 |
| *** jaosorior has joined #openstack-kolla | 06:46 | |
| *** sean-md has joined #openstack-kolla | 06:50 | |
| *** unicell1 has quit IRC | 06:52 | |
| *** sean-md has joined #openstack-kolla | 06:53 | |
| *** chas has joined #openstack-kolla | 06:54 | |
| *** kiennt26 has joined #openstack-kolla | 06:54 | |
| *** cah_link has joined #openstack-kolla | 06:59 | |
| *** cah_link has quit IRC | 07:01 | |
| *** hoaile has quit IRC | 07:02 | |
| *** hoaile has joined #openstack-kolla | 07:04 | |
| *** cah_link has joined #openstack-kolla | 07:05 | |
| *** egonzalez has joined #openstack-kolla | 07:05 | |
| *** hrw has quit IRC | 07:06 | |
| *** hrw has joined #openstack-kolla | 07:08 | |
| *** gcb has quit IRC | 07:12 | |
| *** huanxie has joined #openstack-kolla | 07:15 | |
| openstackgerrit | Merged openstack/kolla master: Make binary gates non-voting https://review.openstack.org/516817 | 07:15 |
| *** hachi__ has joined #openstack-kolla | 07:17 | |
| *** hachi has quit IRC | 07:18 | |
| *** pcaruana has joined #openstack-kolla | 07:24 | |
| *** dardelean has joined #openstack-kolla | 07:25 | |
| *** sean-md has quit IRC | 07:33 | |
| *** zhurong has quit IRC | 07:36 | |
| *** adac has joined #openstack-kolla | 07:38 | |
| *** sean-md has joined #openstack-kolla | 07:38 | |
| *** hachi__ has quit IRC | 07:51 | |
| *** mnasiadka has joined #openstack-kolla | 07:53 | |
| *** danardelean has joined #openstack-kolla | 07:55 | |
| *** danardelean has quit IRC | 07:55 | |
| *** danardelean has joined #openstack-kolla | 07:56 | |
| *** hachi has joined #openstack-kolla | 07:56 | |
| *** mgoddard has joined #openstack-kolla | 08:01 | |
| *** Zophar has quit IRC | 08:04 | |
| *** __sagar has quit IRC | 08:06 | |
| *** egonzalez has quit IRC | 08:10 | |
| *** gkadam has joined #openstack-kolla | 08:17 | |
| *** serlex has joined #openstack-kolla | 08:18 | |
| *** serlex has quit IRC | 08:18 | |
| *** egonzalez has joined #openstack-kolla | 08:26 | |
| *** danardelean has quit IRC | 08:40 | |
| *** dims has quit IRC | 08:46 | |
| *** danardelean has joined #openstack-kolla | 08:47 | |
| *** dims has joined #openstack-kolla | 08:48 | |
| openstackgerrit | Huan Peng proposed openstack/kolla-ansible master: Delete kolla_internal_address variable https://review.openstack.org/517580 | 08:58 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/kolla master: Separate rsyslog image definition https://review.openstack.org/517581 | 08:59 |
| *** sean-md has quit IRC | 09:06 | |
| *** dougsz has joined #openstack-kolla | 09:10 | |
| *** sean-md has joined #openstack-kolla | 09:12 | |
| *** sean-md has quit IRC | 09:17 | |
| *** sean-md has joined #openstack-kolla | 09:31 | |
| *** danardelean has quit IRC | 09:33 | |
| *** sean-md has quit IRC | 09:35 | |
| *** dougsz has quit IRC | 09:36 | |
| *** danardelean has joined #openstack-kolla | 09:43 | |
| *** sean-md has joined #openstack-kolla | 09:49 | |
| *** yangyapeng has quit IRC | 09:52 | |
| *** sean-md has quit IRC | 09:54 | |
| *** matrohon_ has joined #openstack-kolla | 09:54 | |
| *** kiennt26 has quit IRC | 10:03 | |
| *** bbc_ has quit IRC | 10:06 | |
| *** matrohon_ has quit IRC | 10:06 | |
| *** caowei has quit IRC | 10:06 | |
| *** bbc_ has joined #openstack-kolla | 10:06 | |
| *** genek has joined #openstack-kolla | 10:06 | |
| *** sean-md has joined #openstack-kolla | 10:07 | |
| *** daidv has quit IRC | 10:10 | |
| *** sean-md has quit IRC | 10:11 | |
| *** tovin07_ has quit IRC | 10:12 | |
| *** caowei has joined #openstack-kolla | 10:13 | |
| *** caowei has quit IRC | 10:16 | |
| *** hieulq has quit IRC | 10:16 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/kolla master: Separate rsyslog image definition https://review.openstack.org/517581 | 10:18 |
| *** tone_z has quit IRC | 10:19 | |
| *** duonghq has quit IRC | 10:20 | |
| *** jmccarthy has joined #openstack-kolla | 10:31 | |
| *** huanxie has quit IRC | 10:32 | |
| *** Zophar has joined #openstack-kolla | 10:44 | |
| Zophar | https://review.openstack.org/#/c/514676/ any new code review will be appreciated ;) | 10:45 |
| *** hachi has quit IRC | 10:47 | |
| *** hachi has joined #openstack-kolla | 10:48 | |
| *** serlex has joined #openstack-kolla | 10:51 | |
| *** serlex has left #openstack-kolla | 10:51 | |
| *** serlex has quit IRC | 10:51 | |
| *** hachi has quit IRC | 10:57 | |
| *** pbourke has quit IRC | 10:59 | |
| *** pbourke has joined #openstack-kolla | 11:00 | |
| *** hachi has joined #openstack-kolla | 11:07 | |
| *** huanxie has joined #openstack-kolla | 11:12 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/kolla master: Separate rsyslog image definition https://review.openstack.org/517581 | 11:13 |
| *** dougsz has joined #openstack-kolla | 11:13 | |
| *** ansmith has quit IRC | 11:17 | |
| openstackgerrit | Doug Szumski proposed openstack/kolla-ansible master: Add configurable rabbitmq monitoring user https://review.openstack.org/517386 | 11:19 |
| *** sean-md has joined #openstack-kolla | 11:24 | |
| *** sean-md has quit IRC | 11:28 | |
| *** hachi has quit IRC | 11:37 | |
| *** sean-md has joined #openstack-kolla | 11:42 | |
| *** janki has quit IRC | 11:45 | |
| *** sean-md has quit IRC | 11:46 | |
| *** mgoddard has quit IRC | 11:49 | |
| *** sean-md has joined #openstack-kolla | 12:00 | |
| *** rmart04 has joined #openstack-kolla | 12:01 | |
| *** lpetrut has joined #openstack-kolla | 12:02 | |
| *** sean-md has quit IRC | 12:04 | |
| *** zhenguo has quit IRC | 12:05 | |
| *** rhallisey has joined #openstack-kolla | 12:08 | |
| openstackgerrit | Doug Szumski proposed openstack/kolla-ansible master: Add configurable rabbitmq monitoring user https://review.openstack.org/517386 | 12:08 |
| openstackgerrit | Taeha Kim proposed openstack/kolla-ansible master: Add missing hostname to nodename in RabbitMQ for multinode deployment https://review.openstack.org/516847 | 12:12 |
| *** goldyfruit has joined #openstack-kolla | 12:13 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Allow object versioning by default https://review.openstack.org/517281 | 12:17 |
| *** sean-md has joined #openstack-kolla | 12:18 | |
| *** janki has joined #openstack-kolla | 12:18 | |
| *** goldyfruit has quit IRC | 12:18 | |
| *** zhangfei has quit IRC | 12:23 | |
| *** sean-md has quit IRC | 12:24 | |
| *** sean-md has joined #openstack-kolla | 12:37 | |
| *** genek has quit IRC | 12:38 | |
| *** genek has joined #openstack-kolla | 12:38 | |
| *** zhangfei has joined #openstack-kolla | 12:40 | |
| *** sean-md has quit IRC | 12:42 | |
| *** jmccarthy has left #openstack-kolla | 12:45 | |
| *** gkadam has quit IRC | 12:47 | |
| *** ansmith has joined #openstack-kolla | 12:48 | |
| openstackgerrit | Taeha Kim proposed openstack/kolla-ansible master: Add missing hostname to nodename in RabbitMQ for multinode deployment https://review.openstack.org/516847 | 12:51 |
| *** sean-md has joined #openstack-kolla | 12:56 | |
| *** sean-md has quit IRC | 13:00 | |
| *** hrw has quit IRC | 13:07 | |
| *** hrw has joined #openstack-kolla | 13:08 | |
| *** huanxie has quit IRC | 13:11 | |
| *** coolsvap has quit IRC | 13:11 | |
| *** goldyfruit has joined #openstack-kolla | 13:12 | |
| *** hrw has quit IRC | 13:13 | |
| jaosorior | mandre: could you check this out https://review.openstack.org/#/c/517581 ? | 13:13 |
| *** sean-md has joined #openstack-kolla | 13:14 | |
| jaosorior | mandre: thanks | 13:16 |
| *** hrw has joined #openstack-kolla | 13:18 | |
| *** sean-md has quit IRC | 13:18 | |
| *** bastafidli has joined #openstack-kolla | 13:27 | |
| *** mgoddard has joined #openstack-kolla | 13:27 | |
| *** bastafidli has quit IRC | 13:27 | |
| *** bastafidli has joined #openstack-kolla | 13:28 | |
| *** skramaja has quit IRC | 13:28 | |
| openstackgerrit | zhichao zhu proposed openstack/kolla-ansible master: fix shell error about tools/init-vpn https://review.openstack.org/517553 | 13:30 |
| *** jtriley has joined #openstack-kolla | 13:31 | |
| *** sean-md has joined #openstack-kolla | 13:32 | |
| *** hachi has joined #openstack-kolla | 13:36 | |
| *** sean-md has quit IRC | 13:37 | |
| *** bbc_ has quit IRC | 13:39 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla-kubernetes master: Updated from global requirements https://review.openstack.org/509443 | 13:42 |
| *** jtriley has quit IRC | 13:47 | |
| *** egonzalez has quit IRC | 13:59 | |
| *** bastafidli__ has joined #openstack-kolla | 14:01 | |
| *** ansmith has quit IRC | 14:04 | |
| *** bastafidli has quit IRC | 14:05 | |
| *** jtriley has joined #openstack-kolla | 14:05 | |
| *** jbrooks has quit IRC | 14:11 | |
| *** serlex has joined #openstack-kolla | 14:12 | |
| *** serlex has left #openstack-kolla | 14:13 | |
| *** serlex has quit IRC | 14:13 | |
| *** rmart04 has quit IRC | 14:13 | |
| *** genek has quit IRC | 14:13 | |
| *** genek has joined #openstack-kolla | 14:18 | |
| openstackgerrit | Merged openstack/kolla master: Separate rsyslog image definition https://review.openstack.org/517581 | 14:19 |
| *** sean-md has joined #openstack-kolla | 14:23 | |
| *** sbezverk has quit IRC | 14:23 | |
| *** jbrooks has joined #openstack-kolla | 14:26 | |
| *** sean-md has quit IRC | 14:27 | |
| *** genek has quit IRC | 14:28 | |
| *** hachi has quit IRC | 14:28 | |
| *** ansmith has joined #openstack-kolla | 14:28 | |
| *** hachi has joined #openstack-kolla | 14:29 | |
| *** Pavo has joined #openstack-kolla | 14:30 | |
| *** sean-md has joined #openstack-kolla | 14:32 | |
| *** jgriffith has quit IRC | 14:34 | |
| *** hrw has quit IRC | 14:34 | |
| *** hachi has quit IRC | 14:35 | |
| *** sean-md has quit IRC | 14:36 | |
| *** mnasiadka has quit IRC | 14:37 | |
| *** jgriffith has joined #openstack-kolla | 14:38 | |
| *** yangyapeng has joined #openstack-kolla | 14:40 | |
| *** sean-md has joined #openstack-kolla | 14:41 | |
| *** sean-md has quit IRC | 14:46 | |
| *** sambetts|afk has quit IRC | 14:51 | |
| *** itlinux has joined #openstack-kolla | 14:52 | |
| *** sean-md has joined #openstack-kolla | 14:59 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Fix Ansible WARNING logs on reserved name: action https://review.openstack.org/511473 | 15:00 |
| *** zhangfei has quit IRC | 15:01 | |
| *** hrw has joined #openstack-kolla | 15:01 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Fix iscsi and tgt containers deployment case https://review.openstack.org/501729 | 15:01 |
| *** sean-md has quit IRC | 15:03 | |
| *** sean-md has joined #openstack-kolla | 15:17 | |
| *** itlinux has quit IRC | 15:20 | |
| *** janki has quit IRC | 15:20 | |
| *** janki has joined #openstack-kolla | 15:21 | |
| *** sean-md has quit IRC | 15:21 | |
| *** janki has quit IRC | 15:22 | |
| *** lukepatrick has joined #openstack-kolla | 15:25 | |
| *** jaosorior has quit IRC | 15:27 | |
| *** itlinux has joined #openstack-kolla | 15:28 | |
| *** Zophar has quit IRC | 15:32 | |
| *** sean-md has joined #openstack-kolla | 15:35 | |
| kfox1111 | morning | 15:36 |
| *** Pavo has quit IRC | 15:37 | |
| *** sean-md has quit IRC | 15:39 | |
| *** lukepatrick has quit IRC | 15:44 | |
| *** sean-md has joined #openstack-kolla | 15:45 | |
| *** Pavo has joined #openstack-kolla | 15:47 | |
| *** lukepatrick has joined #openstack-kolla | 15:47 | |
| *** sean-md has quit IRC | 15:50 | |
| *** cah_link has quit IRC | 15:54 | |
| inc0 | good morning | 15:56 |
| inc0 | kfox1111 jlvillal got time today?;) | 15:56 |
| *** Pavo has quit IRC | 15:57 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla master: Dockerhub publisher job https://review.openstack.org/513855 | 15:58 |
| *** sean-md has joined #openstack-kolla | 16:03 | |
| *** Pavo has joined #openstack-kolla | 16:04 | |
| *** sean-md has quit IRC | 16:07 | |
| *** livelace-link has joined #openstack-kolla | 16:12 | |
| *** genek has joined #openstack-kolla | 16:14 | |
| *** ktibi has joined #openstack-kolla | 16:15 | |
| ktibi | Hi, anyone can check my bug => https://bugs.launchpad.net/kolla-ansible/+bug/1729566 | 16:15 |
| openstack | Launchpad bug 1729566 in kolla-ansible "No update for /etc/hosts after adding a new compute" [Undecided,New] | 16:15 |
| inc0 | ktibi: only playbook that would add it is bootstrap-servers | 16:16 |
| inc0 | but frankly, I don't advise running bootstrap-servers on working cluster | 16:17 |
| inc0 | it's not something we tested or support | 16:17 |
| *** livelace-link has quit IRC | 16:19 | |
| *** unicell has joined #openstack-kolla | 16:20 | |
| *** unicell has quit IRC | 16:20 | |
| *** sean-md has joined #openstack-kolla | 16:21 | |
| *** livelace-link has joined #openstack-kolla | 16:21 | |
| ktibi | inc0 ok so maybe need to add a check for update /etc/hosts every time the cluster is deploy | 16:23 |
| ktibi | or update | 16:24 |
| inc0 | or setup nameserver | 16:24 |
| inc0 | intersting experiment would be to see if libvirt inside container would share same etc/hosts as host without restart | 16:24 |
| ktibi | I guess compute cluster need to be autonomous | 16:25 |
| *** sean-md has quit IRC | 16:25 | |
| ktibi | If I lose my NS, cluster need to be up | 16:25 |
| inc0 | well you can quickly make ansible playbook to make it work | 16:26 |
| ktibi | Yes, I did this | 16:26 |
| inc0 | https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/baremetal/tasks/pre-install.yml#L19-L35 | 16:26 |
| ktibi | but it's better if in kolla :) | 16:27 |
| inc0 | well yes and no | 16:27 |
| inc0 | we have kinda standard that we don't touch host after boostrap servers | 16:27 |
| inc0 | which is meant to be run on clean OS | 16:27 |
| *** danardel_ has joined #openstack-kolla | 16:35 | |
| *** danardelean has quit IRC | 16:38 | |
| *** bmace has quit IRC | 16:39 | |
| *** sean-md has joined #openstack-kolla | 16:39 | |
| *** bmace has joined #openstack-kolla | 16:39 | |
| *** itlinux has quit IRC | 16:39 | |
| *** salv-orl_ has quit IRC | 16:41 | |
| *** sean-md has quit IRC | 16:43 | |
| *** itlinux has joined #openstack-kolla | 16:43 | |
| *** sean-md has joined #openstack-kolla | 16:57 | |
| *** sean-md has quit IRC | 17:02 | |
| *** genek has quit IRC | 17:16 | |
| *** chas has quit IRC | 17:17 | |
| *** chas has joined #openstack-kolla | 17:17 | |
| *** genek has joined #openstack-kolla | 17:18 | |
| *** chas has quit IRC | 17:22 | |
| *** sean-md has joined #openstack-kolla | 17:25 | |
| *** genek has quit IRC | 17:25 | |
| kfox1111 | inc0: maybe this afternoon sometime? | 17:27 |
| *** sean-md has quit IRC | 17:29 | |
| inc0 | jlvillal: ^ works for you? | 17:29 |
| inc0 | I have time till ~2.45 | 17:29 |
| jlvillal | inc0: Sure | 17:29 |
| inc0 | pm | 17:29 |
| jlvillal | inc0: I should be back from lunch by 12:30pm | 17:30 |
| inc0 | kfox1111: that works for you? | 17:30 |
| kfox1111 | let me double checck my calendar.... | 17:30 |
| *** lpetrut has quit IRC | 17:31 | |
| kfox1111 | i've got a 2:30-3:30 but otherwise free this after noon. | 17:31 |
| inc0 | ok, so anytime after 12.30 let's have discussion about healthchecks | 17:31 |
| inc0 | thanks | 17:31 |
| kfox1111 | k | 17:33 |
| *** unicell has joined #openstack-kolla | 17:34 | |
| *** klindgren_ has quit IRC | 17:39 | |
| *** genek has joined #openstack-kolla | 17:41 | |
| *** salv-orlando has joined #openstack-kolla | 17:41 | |
| *** sean-md has joined #openstack-kolla | 17:43 | |
| *** danardel_ has quit IRC | 17:46 | |
| *** salv-orlando has quit IRC | 17:47 | |
| *** sean-md has quit IRC | 17:47 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla-kubernetes master: Updated from global requirements https://review.openstack.org/509443 | 17:48 |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla-ansible master: Enable multinode scenarios in zuulv3 https://review.openstack.org/512779 | 17:48 |
| *** itlinux has quit IRC | 17:51 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/kolla-kubernetes master: Updated from global requirements https://review.openstack.org/509443 | 17:51 |
| *** Pavo has quit IRC | 17:55 | |
| *** itlinux has joined #openstack-kolla | 18:00 | |
| *** sean-md has joined #openstack-kolla | 18:01 | |
| *** dougsz has quit IRC | 18:04 | |
| *** sean-md has quit IRC | 18:05 | |
| *** mgoddard has quit IRC | 18:09 | |
| *** danardelean has joined #openstack-kolla | 18:15 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla-ansible master: Enable multinode scenarios in zuulv3 https://review.openstack.org/512779 | 18:18 |
| *** sean-md has joined #openstack-kolla | 18:19 | |
| *** danardelean has quit IRC | 18:20 | |
| *** sean-md has quit IRC | 18:23 | |
| *** jascott1 has joined #openstack-kolla | 18:26 | |
| jascott1 | o/ | 18:26 |
| jascott1 | no messages in the slack linkage since 8:12am PST today so wanted to come in and make sure there wasnt a nuclear event in here or somethin' | 18:27 |
| inc0 | I think our slack bot died | 18:27 |
| inc0 | sdake ^ | 18:27 |
| sdake | inc0 working it moment | 18:27 |
| inc0 | thanks Steve | 18:28 |
| sdake | the slackbot is on auto-reboot | 18:28 |
| sdake | so the fact it didn't come back online is worrisome ;) | 18:28 |
| jascott1 | me too | 18:28 |
| jascott1 | every morn | 18:28 |
| sdake | jascott1 :) | 18:28 |
| sdake | morning btw (a5am here0 | 18:28 |
| inc0 | yeah I remember this type of jetlag | 18:29 |
| inc0 | when I came from Europe to US | 18:29 |
| inc0 | you wake up at 4am | 18:29 |
| jascott1 | ride a kolla for me! | 18:29 |
| jascott1 | koala | 18:29 |
| jascott1 | haha | 18:29 |
| sdake | inc0 do you want ssh access to the machine? | 18:30 |
| sdake | jascott1 sadly i haven'gt seen any | 18:30 |
| sdake | I guess I need to go to the oo | 18:31 |
| jascott1 | i figured they would be complimentary at the hotel | 18:32 |
| inc0 | sdake: shoot, I can try to fix it if you can't | 18:33 |
| sdake | i can fix it I'm sure | 18:33 |
| sdake | just offering in case you want access to the machine - ptl should probably have access (as I travel _ETOOMUCH!) | 18:34 |
| inc0 | well you've set it up;) we should move it to less bus-factory place tho | 18:34 |
| sdake | inc0 it is on a cloud provider | 18:36 |
| sdake | i seem to recall they were booting their VMs enmass tho | 18:36 |
| sdake | to fix bus factor = second admin | 18:37 |
| *** sean-md has joined #openstack-kolla | 18:37 | |
| sdake | i had offered to portdirect to add his bot in my vm as well | 18:38 |
| sdake | although he has yet to taek me up on the offer :) | 18:38 |
| inc0 | https://review.openstack.org/#/c/513855/ <- when this merges, we will have (hopefully) images on dockerhub next day:) | 18:39 |
| *** lpetrut has joined #openstack-kolla | 18:40 | |
| jascott1 | nice | 18:40 |
| sdake | inc0 could you epand on images? | 18:41 |
| sdake | inc0 a dail ybuild or a stable build? | 18:41 |
| inc0 | daily top of master | 18:41 |
| *** sean-md has quit IRC | 18:41 | |
| inc0 | but when we'll backport it to stable later | 18:41 |
| sdake | well that is a start | 18:41 |
| sdake | inc0++ :) | 18:42 |
| inc0 | just cherrypick | 18:42 |
| *** salv-orlando has joined #openstack-kolla | 18:42 | |
| sdake | sorry for spam: | 18:44 |
| sdake | START: 2017-11-02 07:00 UTC | 18:44 |
| sdake | END: 2017-11-02 15:00 UTC | 18:44 |
| sdake | Hi there, | 18:44 |
| sdake | Our engineers have identified a problem with your droplet's hypervisor which may cause random unplanned outages. In order for you to have the most optimal experience, we are scheduling a controlled and scheduled reboot of the hypervisor to implement our fixes which should prevent future outages from this issue. | 18:44 |
| sdake | We're scheduling an eight hour maintenance window to perform this reboot starting at 2017-11-02 07:00 UTC, which is 2017-11-02 00:00 in San Francisco where your droplet is located. | 18:44 |
| *** itlinux has quit IRC | 18:44 | |
| inc0 | well it's 3rd Nov;) | 18:45 |
| jascott1 | 8 hrs? jeebus | 18:46 |
| sdake | There were 5019 failed login attempts since the last successful login. | 18:46 |
| inc0 | we should talk to them about time it takes to reboot machine;) | 18:46 |
| sdake | can't expect great SLAs for 5$/mo | 18:46 |
| jascott1 | true | 18:47 |
| *** salv-orlando has quit IRC | 18:47 | |
| jascott1 | idk might get a tiny at AMZN/GOOD/M$ for that | 18:47 |
| *** kolla-slack has joined #openstack-kolla | 18:48 | |
| sdake | cheers :) | 18:48 |
| inc0 | welcome back kolla-slack | 18:49 |
| sdake | somehow the docker run operation disappeared | 18:49 |
| sdake | going to reboot the machine make sure it comes up properly | 18:49 |
| sdake | inc0 pls send me your public key in a PM | 18:50 |
| *** kolla-slack has quit IRC | 18:51 | |
| *** lpetrut has quit IRC | 18:53 | |
| *** kolla-slack has joined #openstack-kolla | 18:54 | |
| *** kolla-slack has quit IRC | 18:55 | |
| *** kolla-slack has joined #openstack-kolla | 18:55 | |
| kolla-slack | <sdake> test of lack | 18:59 |
| *** kevzha01_ has joined #openstack-kolla | 19:02 | |
| *** kevinz has quit IRC | 19:05 | |
| *** lukepatrick has quit IRC | 19:11 | |
| *** itlinux has joined #openstack-kolla | 19:14 | |
| *** rwsu has quit IRC | 19:18 | |
| *** chas has joined #openstack-kolla | 19:18 | |
| *** chas has quit IRC | 19:22 | |
| *** jrist has quit IRC | 19:25 | |
| portdirect | sdake: you coming to Sydney? | 19:31 |
| sdake | yo bro | 19:31 |
| sdake | i'm here | 19:31 |
| portdirect | Oh nice, see ya in 20 hours :) | 19:32 |
| kfox1111 | hi guys. | 19:32 |
| *** pcaruana has quit IRC | 19:32 | |
| kfox1111 | i'm not making it. :( | 19:32 |
| *** lukepatrick has joined #openstack-kolla | 19:32 | |
| portdirect | Will mark the point I've made it round this lump o rock | 19:32 |
| portdirect | kfox1111: that sux dude | 19:32 |
| portdirect | Will put a shrimp on the bbq for ya. | 19:33 |
| * inc0 ordering card deck for blackjack to make our own summit | 19:33 | |
| sdake | portdirect ZUG - that flight sucks :) | 19:33 |
| sdake | kfox1111 were you coming prior? | 19:33 |
| kfox1111 | portdirect: thanks. :) | 19:34 |
| kfox1111 | sdake: no. | 19:34 |
| sdake | kfox1111 roger | 19:34 |
| kfox1111 | inc0: whats the plan? | 19:34 |
| sdake | anyone know which kolla cores will be here? | 19:34 |
| sdake | seems pretty slim :) | 19:34 |
| *** jrist has joined #openstack-kolla | 19:36 | |
| bkero | I'd expect attendance overall to be pretty slim | 19:36 |
| inc0 | jlvillal: whenever you're ready;) | 19:36 |
| sdake | bkero ya - I think I heard half as many people as last summit | 19:36 |
| sdake | (just because of cost to get here) | 19:37 |
| jlvillal | inc0: I'm listening | 19:37 |
| inc0 | cool;) | 19:37 |
| inc0 | portdirect: in fact this will interest you as well | 19:37 |
| inc0 | sooo...healthchecks | 19:37 |
| sdake | portdirect that shrimp on the barbie joke was terribad :) | 19:37 |
| inc0 | let me do intros | 19:37 |
| inc0 | jlvillal - core in Ironic, sits close to myself in Intel;) | 19:38 |
| sdake | jlvillal o/ ironic ftw :) | 19:38 |
| inc0 | kfox1111 - operator and core in kolla-kubernetes, if something can break, he had seen that breaking | 19:38 |
| *** lpetrut has joined #openstack-kolla | 19:38 | |
| inc0 | portdirect - Scottish | 19:38 |
| inc0 | ;) | 19:38 |
| inc0 | also author of probably 90ish % of openstack-helm | 19:39 |
| inc0 | sooo let's talk healthchecks | 19:39 |
| kfox1111 | healthchecks | 19:39 |
| * jlvillal hears Scottish and thinks of "Eleven" and two guys in an elevator :) | 19:40 | |
| inc0 | on PTG we started this discussion and I think Ironic would be great prototype for proper healthchecks for others projects to mimic | 19:40 |
| *** sean-md has joined #openstack-kolla | 19:41 | |
| inc0 | it has both REST api services and conductor thingy that's harder to monitor as it doesn't have any tangable endpoings | 19:41 |
| inc0 | soo, kfox1111 why won't you start with your prometheus idea and explain what's this is about? | 19:42 |
| kfox1111 | hmm.. | 19:42 |
| kfox1111 | well, prometheus is a monitoring service that uses metric gathering to let you do alerting | 19:43 |
| *** salv-orlando has joined #openstack-kolla | 19:43 | |
| kfox1111 | its format is super simple and being standardized for adoption for other monitoring systems too. | 19:43 |
| kfox1111 | its protocol is basically a text file sitting on an http server. | 19:43 |
| kfox1111 | so its super easy to integrate into services that already provide http, and its still pretty easy to add a simple web service to services that dont. | 19:44 |
| *** sean-md has quit IRC | 19:46 | |
| kfox1111 | inc0: anything else you can think of? | 19:47 |
| inc0 | one question I'm asking myself is how to monitor ironic-conductor | 19:48 |
| inc0 | or any other RPC-driven service | 19:48 |
| *** salv-orlando has quit IRC | 19:48 | |
| *** jrist_ has joined #openstack-kolla | 19:49 | |
| inc0 | so for example | 19:49 |
| inc0 | an ironic conductor slowed down to 5% of msg/hour consumed | 19:49 |
| *** jrist_ has quit IRC | 19:49 | |
| *** jrist has quit IRC | 19:49 | |
| kfox1111 | if the conductor exposed prometheus metrics on how many messages per hour it processed, | 19:50 |
| inc0 | how do we expose this info? we could have ironic conductor calling some arbitrary code (plugin based) whenever it cosumes a msg | 19:50 |
| kfox1111 | you can alert on that. | 19:50 |
| *** jrist has joined #openstack-kolla | 19:50 | |
| *** jrist has quit IRC | 19:50 | |
| *** jrist has joined #openstack-kolla | 19:50 | |
| inc0 | and plugin would hit prometheus endpoint | 19:51 |
| kfox1111 | prometheus works the other way around. | 19:51 |
| sdake | did anyone fill out the cncf survey | 19:51 |
| kfox1111 | ironic-conductor would have an http based service that exposes something like /metrics | 19:51 |
| portdirect | inc0: I'm actually a lost Aussie, I'm going home ;) | 19:52 |
| jlvillal | I doubt we want to add an http based service to the conductor. My initial gut feeling | 19:52 |
| kfox1111 | where curling it outputs stuff like rabbit_messages_processed{q=foo} 10032 | 19:52 |
| inc0 | right but without adding storage to conductor, which itself is quite an arch change, it's not feasable to have this "buffery" type metric | 19:52 |
| jlvillal | Can we ask RabbitMQ instead? | 19:52 |
| inc0 | and that | 19:52 |
| inc0 | I don't think rabbitmq exposes this kind of data | 19:53 |
| kfox1111 | you could get metrics from rabbit. there is an expoerter for it too. | 19:53 |
| kfox1111 | it is potentially harder to tell some info from there though, as the relationship between queue's and clients is nontrivial. | 19:53 |
| inc0 | alternatively we could have some log-parsing mechansim which will act as middleware, get info from conductor via logs and expose it in metric-based way | 19:53 |
| kfox1111 | hacky. | 19:53 |
| kfox1111 | the states in question are not very big and its not really a problem if they get reset on restart of the service. | 19:54 |
| inc0 | api will suffer from similar issue too - how to measure processed requests per hour? | 19:54 |
| jlvillal | Now, could possibly add getting metrics about the conductor via the API. | 19:54 |
| kfox1111 | the counters are assumed to be incrementing, and if it desrements, is assumed to be a reset. | 19:54 |
| inc0 | jlvillal: that's gonna add huge amount of complexity | 19:54 |
| kfox1111 | you don't measure requests per hour, you output number of requests handled. | 19:54 |
| jlvillal | inc0: How is that? I guess my gut feeling is no. | 19:55 |
| kfox1111 | deriving requests per hour or minute or (insert time unit here) can b e done after the fact. | 19:55 |
| *** klindgren has joined #openstack-kolla | 19:55 | |
| inc0 | and will mean that monitoring of X depends of well being of Y, which makes failure detection harder | 19:55 |
| kfox1111 | ie, each request handled just requests++ | 19:55 |
| kfox1111 | expoerter just 'printf "requests_handled $requests"' | 19:56 |
| inc0 | right, but exposing http interface just for this purpose in service that normally doesn't have http is an issue | 19:56 |
| kfox1111 | true, but I think the effort is worh the cost. you can gather many many metrics yo ucoudln't normally gather. | 19:56 |
| kfox1111 | number of times that rabbit disconnects? | 19:56 |
| kfox1111 | number of times you can't acess the database | 19:56 |
| kfox1111 | internal count of a failure of the function foo | 19:57 |
| kfox1111 | whatever is appropriate to the system. | 19:57 |
| inc0 | right, but all that could be derived from sidecart container with http service parsing logs | 19:58 |
| inc0 | MSG123 - rabbitmq message parsed | 19:58 |
| kfox1111 | you can do higher level constructs... count of times the conductor failed to schedule something (if it schedules things) | 19:58 |
| inc0 | and then service will count MSG123 per hour and expose it as metric | 19:58 |
| kfox1111 | log parsing is fragile as hell. | 19:58 |
| kfox1111 | and has a lot of overhead not needed. | 19:58 |
| inc0 | to me issue is communication interface to conductor | 19:59 |
| *** sean-md has joined #openstack-kolla | 19:59 | |
| kfox1111 | there is also an issue of activeness. | 19:59 |
| kfox1111 | when you curl a webserver, you know when it doesn't respond. | 19:59 |
| kfox1111 | if you call a sidecar webserver, it can't know if the logger broke. | 19:59 |
| kfox1111 | yeah. and http is a communication interface. | 19:59 |
| inc0 | which is significant overhead in non-http service | 20:00 |
| kfox1111 | I think we're arguing on the definition of small. ;) | 20:00 |
| kfox1111 | to me a webserver is trivial these days. 10 lines of code and a few k of ram. | 20:01 |
| inc0 | well what I'd like to achieve is something that ideally will be portable to any other service with minimal effort | 20:01 |
| kfox1111 | lets just call it 2mb to be generious. | 20:01 |
| inc0 | opening another port is not that | 20:01 |
| kfox1111 | drop in the bucket. :) | 20:01 |
| inc0 | resources are not what I'm worried about | 20:01 |
| inc0 | it's effort it takes to implement it and integrate it closely with code | 20:01 |
| kfox1111 | 99.9999% of the effort is gathering the metrics, not how to export it. | 20:02 |
| inc0 | especially if you'd want to have for example monitoring endpoint listening on different network than regular communication | 20:02 |
| kfox1111 | the integrate part. | 20:02 |
| jlvillal | Thanks that percentage may be hyperbolic... | 20:02 |
| kfox1111 | still think thats a pretty trivial amount of code. | 20:02 |
| jlvillal | s/Thanks/Thinks/ | 20:03 |
| *** sean-md has quit IRC | 20:03 | |
| kfox1111 | I'm thinking adding a prometheus exporter web service to oslo somehow? | 20:03 |
| kfox1111 | the effort to integrate it to rpc services would be minimal then. you load the extra library and hand it a config chunk from oslo.conf? | 20:03 |
| inc0 | or even create new oslo library | 20:04 |
| kfox1111 | that would work too. | 20:04 |
| inc0 | but http would require external service | 20:04 |
| kfox1111 | what do you mean? | 20:04 |
| inc0 | how about socket communication? | 20:04 |
| inc0 | oslo doesn't have "running services" | 20:05 |
| kfox1111 | still don't follow | 20:05 |
| kfox1111 | how is a unix socket really any different then a tcp one? | 20:05 |
| inc0 | yeah I guess it doesn't that much | 20:06 |
| inc0 | so, oslo library that would listen on tcp port X for prometheus? | 20:06 |
| kfox1111 | I'd be just fine if it coudl speek prometheus format over http over tcp or unix socket. | 20:06 |
| kfox1111 | yeah. | 20:06 |
| *** hachi has joined #openstack-kolla | 20:07 | |
| kfox1111 | listen=0.0.0.0:9294 or listen=unix:/var/run/mystats.sock kind of thing | 20:07 |
| inc0 | dims: around? (or any other oslo person?:)) | 20:08 |
| inc0 | harlowja: ^ | 20:08 |
| inc0 | ok, let's assume for a second that we'll have oslo.metrics library | 20:10 |
| inc0 | however oslo.metrics communicates with external world, tbd | 20:10 |
| kfox1111 | ok. | 20:11 |
| inc0 | we could do something like oslo.metrics.add_metric(name="requests_counter", value=1) whenever ironic api parses a http | 20:11 |
| harlowja | hola | 20:12 |
| inc0 | hey harlowja, we're debating common monitoring mechanism for openstack services | 20:12 |
| kfox1111 | inc0: maybe some osloification of https://github.com/prometheus/client_python ? | 20:12 |
| inc0 | based of prometheus | 20:12 |
| inc0 | little background: in Denver in deployment tools wg meeting we found out that we, operators and deployment people, all strugle with similar issue | 20:13 |
| inc0 | how to say that for example ironic conductor is alive and thriving | 20:13 |
| harlowja | hmmmm | 20:14 |
| harlowja | so at godaddy we are all sensu | 20:14 |
| inc0 | ok, but what sensu measuer | 20:14 |
| inc0 | what we're thinking about is to have shim in oslo or services | 20:14 |
| *** rwsu has joined #openstack-kolla | 20:14 | |
| inc0 | which will enable service to expose events or metrics to sensu, prometheus or whatever is listening | 20:15 |
| inc0 | np, conductor on node X processed 100 messages during last 5 hrs | 20:15 |
| inc0 | not ideal | 20:15 |
| harlowja | https://imgur.com/a/ctglx ... | 20:15 |
| harlowja | i can get u all the sensu stuff | 20:15 |
| harlowja | maybe | 20:15 |
| harlowja | there's a lot of them | 20:15 |
| harlowja | basic metrics would be good :-P | 20:16 |
| harlowja | klindgren ^ | 20:16 |
| klindgren | um what | 20:16 |
| harlowja | prometheus sure, but i'd rather get clint in here (he's been working on some k8s + prometheus stuff and prob has some ideas) | 20:17 |
| harlowja | so i'd like to hear his thoughts on prometheus first | 20:17 |
| *** sean-md has joined #openstack-kolla | 20:17 | |
| inc0 | ok, point of this discussion is not sensu vs prometheus | 20:18 |
| harlowja | klindgren had some stuff he did to solve ` ironic conductor is alive and thriving` | 20:18 |
| harlowja | inc0 ok | 20:18 |
| inc0 | it's how to make metrics we care about exposed from services | 20:18 |
| klindgren | at least at deployment time. | 20:18 |
| inc0 | in a way - what code services would need to write to make monitoring easier | 20:18 |
| harlowja | have each service expose a status.json file | 20:18 |
| kfox1111 | the problem i have with status kinds of things, | 20:19 |
| harlowja | http://zuulv3.openstack.org/status.json (or other) | 20:19 |
| kfox1111 | is often you may want to redefine what kinds of things can be a problem after you discover what they are. | 20:19 |
| harlowja | ^ just one example | 20:19 |
| *** ansmith has quit IRC | 20:19 | |
| harlowja | ok | 20:19 |
| kfox1111 | so getting the raw metrics let you add new status things once you desciver what those things are. | 20:19 |
| kfox1111 | like, is handling 5 requests per second good or bad? | 20:20 |
| harlowja | i'd rather have any status.json at this point, lol | 20:20 |
| harlowja | not ideal status, ha | 20:20 |
| kfox1111 | if its a busy projduction server, that may be super low, and "broken" | 20:20 |
| harlowja | ya, i get it | 20:20 |
| *** SpamapS has joined #openstack-kolla | 20:21 | |
| SpamapS | o/ | 20:21 |
| harlowja | kfox1111 inc0 so alive, vs dead, that'd be good | 20:21 |
| *** sean-md has quit IRC | 20:21 | |
| inc0 | right, and what does alive means | 20:22 |
| inc0 | if pid is on, but service hangs, that's dead | 20:22 |
| inc0 | but most of monitoring services won't notice it's death | 20:22 |
| SpamapS | hola kolla | 20:22 |
| kfox1111 | if you can curl the endpoint, its alive enough to respond. | 20:22 |
| inc0 | howdy SpamapS | 20:22 |
| kfox1111 | past that, I think its what meterics are exported? | 20:22 |
| SpamapS | its death ;-) | 20:22 |
| harlowja | ya, klindgren connected into the systemd event to determine alive vs not | 20:22 |
| SpamapS | it's what | 20:23 |
| * SpamapS will stop now | 20:23 | |
| inc0 | but if you don't have endpoint like in ironic-conductor, you won't notice anythign | 20:23 |
| harlowja | have a endpoint? | 20:23 |
| SpamapS | can somebody catch me up to speed. I heard there was talk of Sensu vs. Prometheus for something. I have thoughts. :) | 20:23 |
| harlowja | there was sorta :-P | 20:23 |
| inc0 | SpamapS: it's more about our discussion in PTG regarding healthchecs across the board for openstack services | 20:23 |
| inc0 | I think that as it is today we don't have means to monitor services very well | 20:24 |
| inc0 | if it's API, sure, just make a request | 20:24 |
| harlowja | i mean why can't ironic-conductor expose some tiny REST endpoint (along-side the rest of the crap it does) | 20:24 |
| harlowja | same with nova-conductor ... | 20:24 |
| kfox1111 | harlowja: yeah. thats what I'm thinking. | 20:24 |
| klindgren | or the lack of the ability to interrogate services to see their status. | 20:24 |
| kfox1111 | then expose metrics from there. | 20:24 |
| klindgren | apis are stupid simple | 20:24 |
| inc0 | harlowja: or even have oslo piece that will do it | 20:24 |
| klindgren | but anything thats rpc - is pretty much shit | 20:24 |
| SpamapS | The industry standard right now is, even if you aren't "an API service" that you have an HTTP endpoint that can be queried by health checks. | 20:25 |
| inc0 | right and we want to have means to do it | 20:25 |
| klindgren | the best I was able to figure out is to check to see if its connected to rabbitmq | 20:25 |
| SpamapS | So the plumbing pieces in the middle would do well to have something like that. But then we get into "eventlet vs. something else" | 20:25 |
| klindgren | only some of the processess will emmit a started event on the systemd socket | 20:25 |
| harlowja | inc0 https://docs.python.org/2/library/wsgiref.html#wsgiref.simple_server.WSGIServer exists; so meh, not sure oslo needed | 20:25 |
| klindgren | so for those that do that, I also create the socket and wait some time for the process to emit the event. | 20:25 |
| inc0 | harlowja: well, what I was thinking is to have oslo wrapper around sth like that | 20:26 |
| kfox1111 | klindgren: and that doesn't work when in docker. no systemd. | 20:26 |
| klindgren | if it does I assume that it started succesfully (combined with rmq connectivity check) | 20:26 |
| SpamapS | I thought oslo.service implemented the systemd thing so basically everything works with that now. | 20:26 |
| harlowja | ah, but i think klindgren did get it to work in docker kfox1111 :) | 20:26 |
| klindgren | kfox1111, yes it does | 20:26 |
| inc0 | so in services you'll be able to put method calls in correct places (or even connect it to logger) | 20:26 |
| klindgren | we have python code that creates the socket and listens in on it inside the container | 20:26 |
| harlowja | ^ not super great python code, ha | 20:26 |
| inc0 | oslo.healthcheck.send_message("connection_lost") | 20:26 |
| SpamapS | Also is this about startup, or ongoing health? | 20:27 |
| harlowja | yes | 20:27 |
| klindgren | startup only | 20:27 |
| harlowja | lol | 20:27 |
| klindgren | it only emmits once | 20:27 |
| kfox1111 | yeah, I'm interested in ongoing health. | 20:27 |
| inc0 | and then prometheus pings oslo-managed endpoint "how many connection_lost was during last 1hr"? | 20:27 |
| klindgren | which is :-/ | 20:27 |
| harlowja | perhaps split 2 discussions? :-P | 20:27 |
| harlowja | 2+ | 20:27 |
| harlowja | lol | 20:27 |
| kfox1111 | inc0: number of losses is much better then having the thing try and determine frequency itself. | 20:27 |
| klindgren | but if we could send a singal to the process and have it emit something on a socket, that would work as well | 20:27 |
| kfox1111 | ie, connection_losses += 4 or whatever | 20:28 |
| inc0 | yeah agree kfox1111 but what I'm saying is to have simple service that will parse messages and make counter | 20:28 |
| inc0 | hold counter | 20:28 |
| klindgren | but yes an http endpoint that did some local health checking and gave a :thumbsup: or :thumbsdown: would be awesome. | 20:28 |
| kfox1111 | parsing messages is painful. | 20:28 |
| kfox1111 | and error prone. | 20:28 |
| inc0 | kfox1111: parsing messages dedicated for it | 20:28 |
| SpamapS | So there are two theories of startup operation and they are basically 100% counter to eachother. I thought OpenStack had subscribed to one, but who knows. | 20:28 |
| inc0 | not log text scraping | 20:29 |
| harlowja | klindgren where is that python script :-/ | 20:29 |
| SpamapS | #1 is you startup and keep running if you get no irrecoverable errors. You keep trying to do things that error but might work later, forever. | 20:29 |
| *** klindgren has quit IRC | 20:29 | |
| SpamapS | #2 is you startup and try to verify everything is perfect, and if it's not, exit(some_number). | 20:29 |
| inc0 | right, and things breaks after startup too | 20:30 |
| SpamapS | As I said, I thought OpenStack services/daemons/etc. more or less do #1 | 20:30 |
| kfox1111 | yeah. startup works pretty well today. not too concerned with it. | 20:30 |
| inc0 | I'm working well for last 3 months, but hey, why not break at Friday 9pm | 20:30 |
| kfox1111 | its what happens after that concerns me. | 20:30 |
| kfox1111 | exactly. | 20:30 |
| inc0 | and break in non-trivial way | 20:31 |
| inc0 | like, you have 3 working conductors | 20:31 |
| SpamapS | If you do #1 for startup, then you can't assume a running process is a working process. | 20:31 |
| harlowja | so one option, and we have something running here is that we have rally probing our clouds all the time | 20:31 |
| inc0 | one of them will drop to 5% of regular performance | 20:31 |
| *** itlinux has quit IRC | 20:31 | |
| SpamapS | In #2 world, you fail fast. You exit() whenever things are bad. | 20:31 |
| inc0 | stuff keeps working, but cluster is far from healthy | 20:31 |
| SpamapS | And let process supervisors handle the mitigation. | 20:31 |
| inc0 | if it keeps doing this, at some point you'll end up in world of pain | 20:31 |
| kfox1111 | SpamapS: that does assume stoping actualy can fix it. | 20:31 |
| SpamapS | (that, btw, is how the mysql cluster backend "ndb" works) | 20:31 |
| kfox1111 | doesn't handle things like db entries being "funny" | 20:32 |
| SpamapS | kfox1111: well no, it doesn't really assume that. It just hands off mitigation to "not this process" | 20:32 |
| kfox1111 | otherwise, you just get into crash loops. | 20:32 |
| SpamapS | In fact it reduces assumptions. | 20:32 |
| SpamapS | Because you never go forward with an undefined state. | 20:32 |
| kfox1111 | yeah. but one user can potentially dos other users toopo. | 20:33 |
| SpamapS | It's a much harder process model to follow, but it is good for things that are critical and complex, like, say, MySQL cluster. ;) | 20:33 |
| kfox1111 | too | 20:33 |
| kfox1111 | true. | 20:33 |
| SpamapS | Well the assumption you're making is that the process supervisor is systemd, and not mysqld_safe or mysqld itself, which can possibly use logic and system awareness to handle the failed process. :) | 20:33 |
| SpamapS | Anyway, we do #1 | 20:34 |
| SpamapS | so a running process needs a way to tell mitigators that they need to mitigate. | 20:34 |
| *** hachi has quit IRC | 20:34 | |
| *** jtriley has quit IRC | 20:34 | |
| SpamapS | A socket that one can query on is just as hard to support as an HTTP listener. | 20:34 |
| kfox1111 | true | 20:35 |
| *** sean-md has joined #openstack-kolla | 20:35 | |
| harlowja | ya, i've made one, i can share | 20:35 |
| harlowja | made one yesterday, lol | 20:35 |
| inc0 | jlvillal: would you be open to have this tiny http listener in conductor? | 20:35 |
| harlowja | there is also the eventlet backdoor btw | 20:36 |
| jlvillal | inc0: Uh, I don't know. Gut feeling is no... | 20:36 |
| inc0 | and have some methods all around the code to add events? | 20:36 |
| jlvillal | inc0: But can see what other Ironic people think. | 20:36 |
| harlowja | https://gist.github.com/harlowja/43453ca1fce77b68655cc165e666a220 (shitty telenet server like thing0 | 20:37 |
| kfox1111 | inc0: events != metrics | 20:37 |
| jlvillal | We already have some metric gathering in Ironic | 20:37 |
| sdake | sup harlowja | 20:37 |
| *** itlinux has joined #openstack-kolla | 20:37 | |
| inc0 | kfox1111: but number of events over period of time is a metric | 20:37 |
| harlowja | https://gist.github.com/harlowja/43453ca1fce77b68655cc165e666a220#file-gistfile1-txt-L56-L57 ... | 20:37 |
| sdake | inc0 the standard way to distribute metrics is via amqp publishing | 20:37 |
| harlowja | ^ other code is missing from that, but ya... | 20:37 |
| sdake | not rest api calls | 20:37 |
| inc0 | sdake: key here is to have per-node metrics | 20:37 |
| inc0 | for health monitoring purposes | 20:38 |
| *** hachi has joined #openstack-kolla | 20:38 | |
| *** chas has joined #openstack-kolla | 20:38 | |
| inc0 | and ultimately things like autohealing by k8s | 20:38 |
| kfox1111 | inc0: that could be true, but incurs potentially a lot more overhead | 20:38 |
| sdake | i've always been a fan of adding healthchecking to openstack - I suggested the idea 5 years ago - but it was probably too early :) | 20:38 |
| inc0 | kfox1111: # of messages processed is derived from event | 20:39 |
| inc0 | s | 20:39 |
| inc0 | event being a message was processed | 20:39 |
| kfox1111 | inc0: I'm trying to say, the way you are describing events seems to be a push model. | 20:39 |
| inc0 | you publish this event somehow, even by calling oslo.health which will just do "messages_processed += 1" | 20:39 |
| inc0 | no | 20:39 |
| kfox1111 | I'm uggesting a pull model of return the current variables values. | 20:39 |
| harlowja | sdake sup with u | 20:40 |
| harlowja | lol | 20:40 |
| *** sean-md has quit IRC | 20:40 | |
| inc0 | kfox1111: no, I'm talking internally | 20:40 |
| sdake | harlowja chillin in australia | 20:40 |
| harlowja | nice nice | 20:40 |
| sdake | iterally - its freezing here :) | 20:40 |
| kfox1111 | ie, oslo.metric.register('foo', &value) | 20:40 |
| kfox1111 | value++ | 20:40 |
| kfox1111 | value+=7 | 20:40 |
| kfox1111 | etc | 20:40 |
| inc0 | yeah | 20:40 |
| inc0 | or oslo.metric.append('foo', 2) | 20:40 |
| inc0 | call it 6 times | 20:40 |
| inc0 | and prometheus calls some API terminated by oslo.metrics | 20:41 |
| inc0 | GET foo/since_last_call -> 12 | 20:41 |
| harlowja | sounds like what i have seen from yahoo, where that would go to a local rrd file and then be picked up by some daemon and shove it somewhere | 20:41 |
| inc0 | I think it can even be done in memory | 20:41 |
| inc0 | because we're incrementing integer which will be zeroed every second or so | 20:41 |
| harlowja | ya | 20:41 |
| harlowja | either/or idk | 20:42 |
| inc0 | well, that's a start | 20:42 |
| inc0 | but one big difference over today is | 20:42 |
| inc0 | oslo.health would listen on TCP port | 20:42 |
| inc0 | I don't think this happens at all today right? | 20:43 |
| harlowja | https://gist.github.com/harlowja/43453ca1fce77b68655cc165e666a220#file-gistfile1-txt-L187-L214 (here u go, ha) | 20:43 |
| harlowja | i made it for u, lol | 20:43 |
| harlowja | inc0 the evenlet backdoor is the closest to that right now | 20:43 |
| kfox1111 | inc0: close, but since_last_call is still wrong. | 20:43 |
| kfox1111 | just return the direct values. | 20:43 |
| harlowja | inc0 but idk if people even know the evenlet backdoor is a thing :-P | 20:43 |
| harlowja | http://eventlet.net/doc/modules/backdoor.html | 20:43 |
| kfox1111 | monitoring keeps last value and subtracits it from new value to get per time metrics. | 20:43 |
| inc0 | every service in openstack (outside of wsgi-terminated APIs) uses eventlet today? | 20:44 |
| harlowja | not sure if its 100% anymore | 20:44 |
| harlowja | or 90% or less | 20:44 |
| harlowja | (i hope its less) | 20:44 |
| harlowja | the eventlet backdoor is way to powerful though :-/ | 20:44 |
| *** sean-md has joined #openstack-kolla | 20:44 | |
| *** salv-orlando has joined #openstack-kolla | 20:45 | |
| harlowja | simpler micro telnet server that responds to small things is better (like in https://gist.github.com/harlowja/43453ca1fce77b68655cc165e666a220#file-gistfile1-txt-L54 ) | 20:45 |
| kfox1111 | yeah, an http get only read only way to dump text files doesn't need to be very complicated. | 20:45 |
| harlowja | ya either/or | 20:45 |
| inc0 | ok, back to my question, harlowja do you think it's feasable to have oslo lib doing this? | 20:46 |
| kfox1111 | the advantage is you can use standard http tools, and can extend it easily in the futre if you need some more stuff exported. | 20:46 |
| harlowja | kfox1111 ya, either/or | 20:46 |
| harlowja | inc0 i don't see a problem with that, if u think oslo in the name will help adoption (it may or may not...) | 20:47 |
| harlowja | i still think people will need to be convinced (though i wish they wouldn't need to be...) | 20:47 |
| harlowja | ^ cause people... | 20:47 |
| kfox1111 | could it be added somehow as a plugin? | 20:48 |
| kfox1111 | so that if the projects don't by in, but the operators do, they can plug it in and at least some oslo level metrics could be gathered? | 20:48 |
| kfox1111 | like rabbit statistics, etc? | 20:48 |
| *** salv-orlando has quit IRC | 20:49 | |
| kfox1111 | then as confidence is gained, devs can add new metrics as needed? | 20:49 |
| inc0 | kfox1111: I still think actual event calling would need to be done in code | 20:49 |
| *** sean-md has quit IRC | 20:49 | |
| inc0 | but it would be super easy to implement if done in oslo | 20:49 |
| harlowja | i've seen how osprofiler tried these things :-P | 20:49 |
| kfox1111 | inc0: (still don't like events) but oslo itself has a bunch of metrics it can gather too. | 20:49 |
| harlowja | people were concerned of 'if osprofiler.enabled' taking up to much CPU cycles.. | 20:50 |
| harlowja | so i imagine how metrics coming out will be done, ha | 20:50 |
| inc0 | kfox1111: it's going to be events internally to code | 20:50 |
| inc0 | not push mechanism | 20:50 |
| kfox1111 | heh. well, that is kind of unavoidable I guess. | 20:50 |
| harlowja | i honestly would like to see some kind of TC goal or something | 20:51 |
| kfox1111 | if they are afraid of any code changes at all will take up cycles, there isn't much hope other then proviing with timing runs. | 20:51 |
| harlowja | cause i just imagine the resistance to modifying code... | 20:51 |
| kfox1111 | +1 to TC | 20:51 |
| harlowja | cause i've seen it happen with osprofiler... | 20:51 |
| inc0 | I mentioned it to TC and they said we'd need a prototype first | 20:51 |
| inc0 | hence ironic | 20:51 |
| kfox1111 | but TC says they don't want to ever decide anything technical.... | 20:51 |
| kfox1111 | we really need an archetectural group. :( | 20:51 |
| harlowja | osprofiler is doing lots of trace points, there is POC of what happened, lol | 20:51 |
| inc0 | also TC goal is moslty to "convince slackers to move to standard" rather than "envorce new design upon projects" | 20:51 |
| harlowja | osprofiler still not really fully integrated and .... | 20:51 |
| harlowja | so i imagine actual metrics in various points in code, being ummm, ya, hard... | 20:52 |
| harlowja | the other question, is osprofiler trace also a metric? | 20:52 |
| inc0 | but it's one of few people every operator I know and talked about agrees is needed | 20:52 |
| harlowja | those traces do have timing info and ... | 20:52 |
| harlowja | http://boris-42.github.io/ngk.html (old but an example) | 20:53 |
| inc0 | well, it's gonna to be night impossible to translate profiler trace to metric | 20:53 |
| inc0 | not impossible, hard | 20:53 |
| inc0 | and hacky | 20:53 |
| kfox1111 | is osprofiler tracing intended to do what https://github.com/jaegertracing/jaeger does? | 20:53 |
| inc0 | and heavy;) | 20:53 |
| harlowja | ya, i was just thinking out loud that if osprofiler has taken years, metrics will also take years :-/ | 20:53 |
| harlowja | kfox1111 osprofiler is older, but yes, similar | 20:53 |
| inc0 | well hopefully not as it's lighter | 20:54 |
| kfox1111 | harlowja: yeah. was hoping for that functionality too. | 20:54 |
| kfox1111 | but yeah, metrics and traces are seperate things. | 20:54 |
| kfox1111 | traces are much heavier. | 20:54 |
| kfox1111 | as the state is per request. | 20:54 |
| kfox1111 | metrics are grouped up. | 20:54 |
| harlowja | kfox1111 osprofiler started off as a yahoo thing (made by a prior co-worker) but boris started osprofiler and started to try to get it plugged in | 20:54 |
| *** lpetrut has quit IRC | 20:54 | |
| inc0 | jlvillal: still around? | 20:55 |
| jlvillal | inc0: I am | 20:55 |
| *** lpetrut has joined #openstack-kolla | 20:55 | |
| harlowja | kfox1111 and ... ya, then tracing got more popular (not sure if that's helped osprofiler or hurt osprofiler, due to competition and nobody picking one...) | 20:55 |
| inc0 | how about we do this: hack together oslo lib with http endpoint that will expose incremental event numbers | 20:55 |
| inc0 | or any other metric we can think of | 20:55 |
| kfox1111 | might be an opertunity to collaborate more with the cncf. they seem to be working on stuff that standardizes it. | 20:55 |
| inc0 | and add this to iroinc-conductor as PoC | 20:56 |
| inc0 | and see what sticks | 20:56 |
| harlowja | kfox1111 i'm not sure the CNCF is standardizing much :-P | 20:56 |
| kfox1111 | inc0: +1 | 20:56 |
| harlowja | but fair point | 20:56 |
| harlowja | kfox1111 i think CNCF mostly just accepted anything and all the things | 20:56 |
| kfox1111 | harlowja: true. | 20:56 |
| inc0 | harlowja: would you be interested in helping with this rather lofty goal? | 20:56 |
| kfox1111 | harlowja: kind of. narrowly focused mostly. | 20:56 |
| harlowja | kfox1111 https://github.com/cncf/toc/pull/47/files is pretty broad, lol | 20:57 |
| harlowja | ^ broad and very fuzzy | 20:57 |
| kfox1111 | harlowja: heh. fair enough. :) | 20:57 |
| harlowja | no kingmakers, blah blah, 'fundamentally sound design' ... | 20:57 |
| harlowja | inc0 i can help, but i'd like to get some support from other openstack elders :-P | 20:58 |
| kfox1111 | harlowja: does kind of sound like openstack'scharter too. :( | 20:58 |
| harlowja | fuzzy and warm | 20:58 |
| kfox1111 | "do all the cloud things", "we don't want to pick any winners" | 20:58 |
| * harlowja thought this morning that a lawyer wrote that PR | 20:58 | |
| harlowja | or ex-lawyer | 20:58 |
| harlowja | lol | 20:58 |
| kfox1111 | big pile of stuff that doesn't work together..... | 20:58 |
| harlowja | ` There is no “one true stack”` ... `Has a committed and excited team` (wuttttt) | 20:59 |
| harlowja | `Has a fundamentally sound design` (wut) | 20:59 |
| harlowja | all very fuzzy statements, ha | 20:59 |
| * kfox1111 nods | 20:59 | |
| harlowja | such lawyer, ha | 20:59 |
| inc0 | harlowja: like who?;) PTLs of every deployment tool are on board;) | 21:00 |
| harlowja | the PTLs of the code that will need to be tweaked ;) | 21:01 |
| harlowja | metrics injection is similar to osprofiler (imho) in that it needs metrics 'points' (like 'trace' points) | 21:01 |
| harlowja | and people freaked at trace points :-/ | 21:01 |
| *** bswartz has quit IRC | 21:02 | |
| harlowja | (alot of what osprofiler does is connect under the covers, like in sqlachemy, so peoples code doesn't need modification) | 21:02 |
| kfox1111 | yeah. | 21:02 |
| harlowja | ^ which though nice and all, isn't really ideal :-/ | 21:02 |
| *** sean-md has joined #openstack-kolla | 21:02 | |
| *** bastafidli__ has quit IRC | 21:03 | |
| inc0 | well ironic is our guinea pig and I think community there is agreeable;) reasonably | 21:03 |
| kfox1111 | its kind of a bootstrapping issue. | 21:03 |
| inc0 | talked with Dmitry about it and he wasn't against;) | 21:03 |
| kfox1111 | need to prove its valuable before they add value to it. | 21:03 |
| inc0 | but goal now is PoC | 21:03 |
| inc0 | if we do PoC and come to Dublin with it and stand with all the PTLs saying "yes, this is effing important" | 21:03 |
| inc0 | I hope that'll make people listen;) | 21:03 |
| kfox1111 | or just a pile of operators telling the ptl's "yes, this is effing important" | 21:04 |
| inc0 | that too | 21:04 |
| inc0 | we'll hats with "Make health visible again" | 21:05 |
| *** sean-md has quit IRC | 21:07 | |
| inc0 | anyway, I'll handle advocacy of this across communities | 21:11 |
| inc0 | what we need is good PoC | 21:11 |
| inc0 | I'm hereby enslisting you harlowja to our healthcheck team;) anything you can contribute would be awesome | 21:11 |
| harlowja | lol | 21:12 |
| inc0 | thank you all, great conversation | 21:13 |
| inc0 | https://etherpad.openstack.org/p/oslo-health-design <- let's use this etherpad for discussion | 21:14 |
| *** ansmith has joined #openstack-kolla | 21:15 | |
| *** hachi has quit IRC | 21:19 | |
| *** sean-md has joined #openstack-kolla | 21:21 | |
| *** sean-md has quit IRC | 21:25 | |
| *** sean-md has joined #openstack-kolla | 21:39 | |
| *** sean-md has quit IRC | 21:44 | |
| *** salv-orlando has joined #openstack-kolla | 21:45 | |
| *** itlinux has quit IRC | 21:49 | |
| *** salv-orlando has quit IRC | 21:50 | |
| *** salv-orlando has joined #openstack-kolla | 21:52 | |
| *** lpetrut has quit IRC | 21:56 | |
| *** sean-md has joined #openstack-kolla | 21:57 | |
| *** ktibi has quit IRC | 21:58 | |
| *** sean-md has quit IRC | 22:01 | |
| *** jrist has quit IRC | 22:11 | |
| *** harlowja has quit IRC | 22:12 | |
| *** Pavo has joined #openstack-kolla | 22:13 | |
| *** jrist has joined #openstack-kolla | 22:15 | |
| *** jrist has quit IRC | 22:15 | |
| *** jrist has joined #openstack-kolla | 22:15 | |
| *** sean-md has joined #openstack-kolla | 22:15 | |
| *** sean-md has quit IRC | 22:19 | |
| *** Pavo has quit IRC | 22:27 | |
| *** chas has quit IRC | 22:31 | |
| *** sean-md has joined #openstack-kolla | 22:33 | |
| *** sean-md has quit IRC | 22:38 | |
| *** itlinux has joined #openstack-kolla | 22:50 | |
| *** sean-md has joined #openstack-kolla | 22:51 | |
| *** itlinux has quit IRC | 22:53 | |
| *** lukepatrick has quit IRC | 22:54 | |
| *** sean-md has quit IRC | 22:55 | |
| *** adac has quit IRC | 23:00 | |
| *** itlinux has joined #openstack-kolla | 23:01 | |
| *** harlowja has joined #openstack-kolla | 23:04 | |
| *** itlinux has quit IRC | 23:04 | |
| *** sean-md has joined #openstack-kolla | 23:06 | |
| *** mwynne has joined #openstack-kolla | 23:07 | |
| *** sean-md has quit IRC | 23:10 | |
| *** genek has quit IRC | 23:21 | |
| *** genek has joined #openstack-kolla | 23:21 | |
| *** sean-md has joined #openstack-kolla | 23:24 | |
| *** manjeets has quit IRC | 23:27 | |
| *** sean-md has quit IRC | 23:28 | |
| *** sean-md has joined #openstack-kolla | 23:42 | |
| *** sean-md has quit IRC | 23:47 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!