*** tonanhngo has joined #openstack-kolla | 00:07 | |
*** clsacramento has quit IRC | 00:12 | |
*** dims_ has quit IRC | 00:20 | |
*** gfidente|afk has quit IRC | 00:21 | |
*** jrist has quit IRC | 00:21 | |
*** gfidente|afk has joined #openstack-kolla | 00:22 | |
*** jrist has joined #openstack-kolla | 00:22 | |
*** dims has joined #openstack-kolla | 00:31 | |
*** yingjun has joined #openstack-kolla | 00:35 | |
*** yingjun has joined #openstack-kolla | 00:35 | |
*** srwilkers has quit IRC | 00:35 | |
*** srwilkers has joined #openstack-kolla | 00:38 | |
*** g3ek has quit IRC | 00:44 | |
*** haplo37 has quit IRC | 00:44 | |
*** vinkman1 has joined #openstack-kolla | 00:47 | |
*** duonghq has joined #openstack-kolla | 00:47 | |
*** vinkman has quit IRC | 00:50 | |
*** vinkman1 has quit IRC | 00:51 | |
*** haplo37 has joined #openstack-kolla | 00:53 | |
*** g3ek has joined #openstack-kolla | 00:54 | |
*** tonanhngo has quit IRC | 00:58 | |
*** phuongnh has joined #openstack-kolla | 01:06 | |
*** dave-mccowan has joined #openstack-kolla | 01:09 | |
*** srwilkers has quit IRC | 01:09 | |
*** vinkman has joined #openstack-kolla | 01:25 | |
*** zhurong has joined #openstack-kolla | 01:31 | |
*** duonghq_ has joined #openstack-kolla | 01:34 | |
*** srwilkers has joined #openstack-kolla | 01:36 | |
*** tovin07 has joined #openstack-kolla | 01:44 | |
*** zhurong has quit IRC | 01:51 | |
*** ssurana has quit IRC | 02:05 | |
*** srwilkers has quit IRC | 02:05 | |
*** zhurong has joined #openstack-kolla | 02:11 | |
openstackgerrit | prameswar proposed openstack/kolla: added kibana reconfigure https://review.openstack.org/390824 | 02:15 |
---|---|---|
*** vinkman has quit IRC | 02:45 | |
*** schwicht has quit IRC | 02:47 | |
*** yuanying has quit IRC | 02:47 | |
*** yuanying has joined #openstack-kolla | 02:48 | |
*** haplo37_ has quit IRC | 02:50 | |
*** yuanying has quit IRC | 02:52 | |
*** haplo37_ has joined #openstack-kolla | 02:52 | |
*** dave-mccowan has quit IRC | 02:53 | |
openstackgerrit | t.goto proposed openstack/kolla: Previous version of keystone reconfigure fails at adding fernet components to fact variable "keystone_item". Ansible can not create fact variable using newly created variable in a same task. https://review.openstack.org/391920 | 02:55 |
*** CBR09 has joined #openstack-kolla | 03:05 | |
*** tonanhngo has joined #openstack-kolla | 03:10 | |
*** tonanhngo has quit IRC | 03:15 | |
*** Jeffrey4l has joined #openstack-kolla | 03:21 | |
*** v1k0d3n has quit IRC | 03:43 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Bump version for gnocchi mistral nova and rally https://review.openstack.org/392028 | 03:50 |
*** pron has joined #openstack-kolla | 04:27 | |
*** yuanying has joined #openstack-kolla | 04:42 | |
openstackgerrit | caoyuan proposed openstack/kolla: Update the word choice for CONTRIBUTING.rst https://review.openstack.org/392032 | 04:53 |
*** yuanying has quit IRC | 04:57 | |
*** yuanying has joined #openstack-kolla | 04:58 | |
*** yuanying has quit IRC | 05:08 | |
*** yuanying has joined #openstack-kolla | 05:12 | |
*** haplo37 has quit IRC | 05:15 | |
*** g3ek has quit IRC | 05:15 | |
*** haplo37 has joined #openstack-kolla | 05:17 | |
*** g3ek has joined #openstack-kolla | 05:18 | |
*** yuanying has quit IRC | 05:28 | |
*** yuanying has joined #openstack-kolla | 05:32 | |
*** yuanying has quit IRC | 05:36 | |
*** yuanying has joined #openstack-kolla | 05:36 | |
*** tonanhngo has joined #openstack-kolla | 06:11 | |
*** yuanying has quit IRC | 06:11 | |
*** tonanhngo has quit IRC | 06:15 | |
*** CBR09 has quit IRC | 06:18 | |
*** yuanying has joined #openstack-kolla | 06:19 | |
*** zhurong has quit IRC | 06:28 | |
*** zhurong has joined #openstack-kolla | 06:30 | |
*** CBR09 has joined #openstack-kolla | 06:38 | |
*** yuanying has quit IRC | 06:40 | |
openstackgerrit | caoyuan proposed openstack/kolla: Update the word choice for rst file https://review.openstack.org/392032 | 06:56 |
*** yuanying has joined #openstack-kolla | 06:57 | |
*** yuanying has quit IRC | 07:02 | |
*** neilus has joined #openstack-kolla | 07:04 | |
*** ankit has joined #openstack-kolla | 07:05 | |
ankit | when executing kolla-ansible deploy, in task: Starting heka container getting | 07:08 |
*** neilus has quit IRC | 07:08 | |
ankit | error:APIError: 500 Server Error: Internal Server Error (\"{\"message\":\"Get https://192.168.56.101:4000/v1/_ping: dial tcp 192.168.56.101:4000: getsockopt: connection refused\"}\")\\n'"} | 07:08 |
ankit | can anyone help me out please! | 07:09 |
openstackgerrit | Sajauddin proposed openstack/kolla: Added code to assign labels to newly created kolla volumes. https://review.openstack.org/392049 | 07:09 |
*** haplo37_ has quit IRC | 07:12 | |
*** haplo37_ has joined #openstack-kolla | 07:14 | |
*** CBR09 has quit IRC | 07:20 | |
*** yuanying has joined #openstack-kolla | 07:29 | |
*** tonanhngo has joined #openstack-kolla | 07:30 | |
*** tonanhngo has quit IRC | 07:35 | |
ankit | when executing kolla-ansible deploy, in task: Starting heka container getting | 07:39 |
ankit | error:APIError: 500 Server Error: Internal Server Error (\"{\"message\":\"Get https://192.168.56.101:4000/v1/_ping: dial tcp 192.168.56.101:4000: getsockopt: connection refused\"}\")\\n'"} | 07:39 |
*** jmccarthy has quit IRC | 07:47 | |
*** jmccarthy has joined #openstack-kolla | 07:48 | |
*** yuanying has quit IRC | 07:48 | |
*** yuanying has joined #openstack-kolla | 07:50 | |
bjolo | morning | 07:52 |
bjolo | ankit, in what log do you get that? | 07:52 |
*** yuanying has quit IRC | 07:57 | |
*** Pavo has quit IRC | 08:00 | |
*** yuanying has joined #openstack-kolla | 08:00 | |
*** yuanying_ has joined #openstack-kolla | 08:07 | |
*** yuanying has quit IRC | 08:08 | |
*** f13o has joined #openstack-kolla | 08:39 | |
*** Serlex has joined #openstack-kolla | 08:50 | |
openstackgerrit | Sajauddin proposed openstack/kolla: Added code to assign labels to newly created kolla volumes. https://review.openstack.org/392049 | 08:51 |
*** yingjun_ has joined #openstack-kolla | 08:53 | |
*** yingjun has quit IRC | 08:53 | |
*** f13o has quit IRC | 08:58 | |
*** athomas has joined #openstack-kolla | 08:59 | |
ankit | bjolo, when i ran kolla-ansible deploy | 09:06 |
*** mgoddard has joined #openstack-kolla | 09:14 | |
*** portdirect has quit IRC | 09:31 | |
*** portdirect_ has joined #openstack-kolla | 09:31 | |
*** portdirect_ is now known as portdirect | 09:32 | |
*** haplo37 has quit IRC | 09:40 | |
*** duonghq has quit IRC | 09:40 | |
*** g3ek has quit IRC | 09:41 | |
*** haplo37 has joined #openstack-kolla | 09:42 | |
*** g3ek has joined #openstack-kolla | 09:43 | |
*** yuanying_ has quit IRC | 09:44 | |
*** yuanying has joined #openstack-kolla | 09:44 | |
*** duonghq_ has quit IRC | 09:45 | |
*** portdirect has quit IRC | 09:45 | |
bjolo | ankit, please try kolla-ansible -i <inventory file> pull | 09:48 |
bjolo | and kolla-ansible -i <inventory file> prechecks | 09:49 |
*** yingjun_ has quit IRC | 09:53 | |
*** portdirect has joined #openstack-kolla | 09:58 | |
*** zhurong has quit IRC | 10:01 | |
*** tovin07 has quit IRC | 10:06 | |
openstackgerrit | Zhijiang Hu proposed openstack/kolla: Uses tunnel_interface as ovs tunnel instead of api_interface https://review.openstack.org/392094 | 10:13 |
*** shardy has joined #openstack-kolla | 10:18 | |
*** asalkeld has quit IRC | 10:20 | |
*** pbourke has joined #openstack-kolla | 10:21 | |
*** tonanhngo has joined #openstack-kolla | 10:22 | |
*** bachp has quit IRC | 10:23 | |
openstackgerrit | Zhijiang Hu proposed openstack/kolla: Uses tunnel_interface as ovs tunnel instead of api_interface https://review.openstack.org/392094 | 10:23 |
*** tonanhngo has quit IRC | 10:27 | |
*** ntpttr has quit IRC | 10:30 | |
*** phuongnh has quit IRC | 10:31 | |
*** ntpttr has joined #openstack-kolla | 10:31 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO_NOT_MERGE: TEST NEWTON BRANCH https://review.openstack.org/386981 | 10:37 |
ankit | bjolo,thanks for the input but i am trying in all in one node | 10:40 |
ankit | is registry is mandatory for all in one node | 10:40 |
ankit | ? | 10:40 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MASTER BRANCH https://review.openstack.org/326307 | 10:40 |
*** bachp has joined #openstack-kolla | 10:50 | |
*** ppalacios has joined #openstack-kolla | 10:51 | |
*** yuanying has quit IRC | 10:56 | |
*** harmw_ has quit IRC | 11:04 | |
*** harmw has joined #openstack-kolla | 11:05 | |
*** bachp has quit IRC | 11:08 | |
pbourke | ankit: no | 11:10 |
*** bachp has joined #openstack-kolla | 11:19 | |
openstackgerrit | Vladislav Belogrudov proposed openstack/kolla: Allow flexibility in mysql backend for ceilometer https://review.openstack.org/392111 | 11:26 |
openstackgerrit | Merged openstack/kolla: Update the format of bifrost deploy.yml https://review.openstack.org/390861 | 11:28 |
openstackgerrit | Vladislav Belogrudov proposed openstack/kolla: Allow flexibility in mysql backend for ceilometer https://review.openstack.org/392111 | 11:28 |
*** haplo37_ has quit IRC | 11:28 | |
openstackgerrit | Merged openstack/kolla: added kibana reconfigure https://review.openstack.org/390824 | 11:29 |
*** haplo37_ has joined #openstack-kolla | 11:31 | |
*** srwilkers has joined #openstack-kolla | 11:35 | |
mliima | morning guys | 11:37 |
srwilkers | good morning :) | 11:38 |
openstackgerrit | Merged openstack/kolla: Update the way of loading the admin-openrc.sh file https://review.openstack.org/391612 | 11:38 |
openstackgerrit | Merged openstack/kolla: Remove indentation from l3_agent.ini.j2 https://review.openstack.org/391485 | 11:41 |
*** dwalsh has joined #openstack-kolla | 11:44 | |
openstackgerrit | Pete Birley proposed openstack/kolla: DO NOT MERGE: Test squashing docker image delta layers https://review.openstack.org/392115 | 11:47 |
pbourke | o/ | 11:49 |
openstackgerrit | Merged openstack/kolla: change "openstack" to "OpenStack" https://review.openstack.org/390758 | 11:51 |
*** Serlex has quit IRC | 11:58 | |
*** schwicht has joined #openstack-kolla | 12:12 | |
openstackgerrit | Merged openstack/kolla: deprecate external_network_bridge option in neutron https://review.openstack.org/391642 | 12:13 |
openstackgerrit | Merged openstack/kolla: Configure log path for kibana https://review.openstack.org/386980 | 12:14 |
*** portdirect has quit IRC | 12:15 | |
*** Pavo has joined #openstack-kolla | 12:15 | |
*** portdirect has joined #openstack-kolla | 12:18 | |
*** schwicht has quit IRC | 12:22 | |
*** schwicht has joined #openstack-kolla | 12:25 | |
*** JoseMello has joined #openstack-kolla | 12:30 | |
*** f13o has joined #openstack-kolla | 12:39 | |
*** lrensing has joined #openstack-kolla | 12:41 | |
*** bhjis has joined #openstack-kolla | 12:45 | |
*** dave-mccowan has joined #openstack-kolla | 12:46 | |
*** eaguilar has joined #openstack-kolla | 12:49 | |
*** srwilkers has quit IRC | 12:52 | |
*** lrensing has quit IRC | 12:52 | |
*** zhurong has joined #openstack-kolla | 12:53 | |
*** jheroux has joined #openstack-kolla | 12:53 | |
*** schwicht has quit IRC | 12:56 | |
*** ayoung has quit IRC | 12:58 | |
bjolo | howdy all | 12:59 |
bjolo | recovered from barcelona yet :) | 12:59 |
pbourke | no! | 13:04 |
*** Serlex has joined #openstack-kolla | 13:12 | |
*** rhallisey has joined #openstack-kolla | 13:13 | |
*** schwicht has joined #openstack-kolla | 13:14 | |
*** jtriley has joined #openstack-kolla | 13:16 | |
*** lrensing has joined #openstack-kolla | 13:18 | |
*** srwilkers has joined #openstack-kolla | 13:19 | |
*** eaguilar has quit IRC | 13:19 | |
*** eaguilar has joined #openstack-kolla | 13:21 | |
*** neilus has joined #openstack-kolla | 13:21 | |
*** neilus has quit IRC | 13:27 | |
*** fragatina has joined #openstack-kolla | 13:32 | |
*** yuanying has joined #openstack-kolla | 13:32 | |
*** fragatina has quit IRC | 13:32 | |
*** fragatina has joined #openstack-kolla | 13:33 | |
*** absubram has joined #openstack-kolla | 13:36 | |
*** lrensing has quit IRC | 13:41 | |
*** fragatina has quit IRC | 13:48 | |
*** g3ek has quit IRC | 13:51 | |
*** jtriley_ has joined #openstack-kolla | 13:52 | |
*** jtriley has quit IRC | 13:55 | |
*** g3ek has joined #openstack-kolla | 13:56 | |
*** neilus has joined #openstack-kolla | 13:56 | |
*** lrensing has joined #openstack-kolla | 13:59 | |
*** yuanying has quit IRC | 14:01 | |
*** dave-mccowan has quit IRC | 14:02 | |
bjolo | pbourke, not me either :) | 14:06 |
bjolo | but im getting there | 14:06 |
openstackgerrit | Wei Cao proposed openstack/kolla: Add solum container https://review.openstack.org/355408 | 14:06 |
*** caowei has joined #openstack-kolla | 14:09 | |
Pavo | morning | 14:12 |
*** caowei has quit IRC | 14:13 | |
*** neilus has quit IRC | 14:16 | |
*** pomac has joined #openstack-kolla | 14:18 | |
*** zhurong has quit IRC | 14:21 | |
*** ayoung has joined #openstack-kolla | 14:23 | |
*** zhurong has joined #openstack-kolla | 14:25 | |
*** caowei has joined #openstack-kolla | 14:25 | |
*** bmace_ has quit IRC | 14:25 | |
*** awiddersheim has quit IRC | 14:30 | |
*** zhurong has quit IRC | 14:30 | |
*** awiddersheim has joined #openstack-kolla | 14:33 | |
portdirect | hey, can anyone help me work out whats going on with zuul? I'm trying to work out why some tests failed and could do with some pointers on where to start. | 14:36 |
portdirect | When trying out my docker-squash patch I can build the images locally, but I'm running out of disck space while testing I think: http://logs.openstack.org/15/392115/1/check/gate-kolla-dsvm-build-centos-binary-centos-7-nv/47ca162/console.html | 14:38 |
portdirect | is it expected for this to affect other tests? they seem to be failing much earlier than I would expect? https://review.openstack.org/#/c/392115/ | 14:39 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MASTER BRANCH https://review.openstack.org/326307 | 14:40 |
*** MarMat has joined #openstack-kolla | 14:44 | |
*** dave-mccowan has joined #openstack-kolla | 14:48 | |
*** DanyC has joined #openstack-kolla | 14:48 | |
openstackgerrit | Pete Birley proposed openstack/kolla: DO NOT MERGE: Test squashing docker image delta layers https://review.openstack.org/392115 | 14:49 |
*** TxGirlGeek has joined #openstack-kolla | 14:50 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO NOT MERGE: TEST MASTER BRANCH https://review.openstack.org/326307 | 14:50 |
*** DanyC has left #openstack-kolla | 14:54 | |
*** msimonin has joined #openstack-kolla | 14:55 | |
*** msimonin has quit IRC | 14:58 | |
*** inc0 has joined #openstack-kolla | 14:58 | |
inc0 | hello everyone | 14:59 |
inc0 | I finally made it home | 14:59 |
Pavo | welcome home | 15:00 |
rhallisey | inc0, hey :) | 15:00 |
*** TxGirlGeek has quit IRC | 15:00 | |
openstackgerrit | Pete Birley proposed openstack/kolla: DO NOT MERGE: Test squashing docker image delta layers https://review.openstack.org/392115 | 15:00 |
inc0 | today I don't expect to be overly productive | 15:00 |
*** TxGirlGeek has joined #openstack-kolla | 15:00 | |
*** dave-mccowan has quit IRC | 15:00 | |
inc0 | really bad case of jetlag | 15:01 |
Pavo | yeah over seas flights always suck | 15:01 |
inc0 | going from US to EU is ok | 15:01 |
inc0 | but going back sucks | 15:01 |
inc0 | I'm semi-awake since 4am | 15:01 |
pbourke | portdirect: does squashing use a lot of temporary disk space maybe? | 15:01 |
srwilkers | hello inc0 :) | 15:02 |
inc0 | hey Steve, glad to see you all here | 15:03 |
Pavo | longest flight was 31hours from Fort Campbell to Afghanistan | 15:03 |
inc0 | When I initially flown here, to US, total flight (including layover because of missed connection) was over 40 | 15:03 |
Pavo | I am talking a non stop flight, no connections or lay overs | 15:04 |
Pavo | thats the longest flight I have been on | 15:04 |
inc0 | yeah, that's rough | 15:05 |
inc0 | refuelled on the way? | 15:05 |
*** dwalsh has quit IRC | 15:05 | |
portdirect | pbourke: thats probably it... - I'm trying again with a single threaded build, and a bit more cleanup | 15:06 |
*** dave-mccowan has joined #openstack-kolla | 15:12 | |
*** TxGirlGeek has quit IRC | 15:13 | |
*** TxGirlGeek has joined #openstack-kolla | 15:14 | |
openstackgerrit | t.goto proposed openstack/kolla: Fix fernet related components added to fact var properly. https://review.openstack.org/392197 | 15:14 |
Pavo | yeah in air refuel | 15:14 |
inc0 | yeah I figured, I'd love to see this in action one day | 15:15 |
Pavo | on a C5 Galaxy not really noticable | 15:15 |
*** TxGirlGeek has quit IRC | 15:18 | |
*** tgoto has joined #openstack-kolla | 15:19 | |
*** caowei has quit IRC | 15:20 | |
tgoto | Hello, I've just committed a change. It would be great if you review it. https://review.openstack.org/#/c/392197/ | 15:20 |
*** tgoto has quit IRC | 15:21 | |
*** athomas has quit IRC | 15:22 | |
*** vinkman has joined #openstack-kolla | 15:25 | |
*** dwalsh has joined #openstack-kolla | 15:25 | |
*** TxGirlGeek has joined #openstack-kolla | 15:26 | |
*** dwalsh_ has joined #openstack-kolla | 15:28 | |
*** dwalsh has quit IRC | 15:29 | |
*** athomas has joined #openstack-kolla | 15:29 | |
*** absubram has quit IRC | 15:31 | |
inc0 | fyi guys, I'm going to cancel tomorrows meeting | 15:32 |
inc0 | I'll be flying around yet again, and I think everyone need some healing time after summit | 15:33 |
*** dwalsh has joined #openstack-kolla | 15:34 | |
*** dwalsh_ has quit IRC | 15:34 | |
*** absubram has joined #openstack-kolla | 15:36 | |
*** vinkman has quit IRC | 15:36 | |
*** vinkman has joined #openstack-kolla | 15:37 | |
srwilkers | +1 | 15:40 |
*** dwalsh has quit IRC | 15:40 | |
*** TxGirlGeek has quit IRC | 15:45 | |
*** TxGirlGeek has joined #openstack-kolla | 15:49 | |
*** sdake has joined #openstack-kolla | 15:52 | |
*** dwalsh has joined #openstack-kolla | 15:53 | |
*** TxGirlGeek has quit IRC | 15:56 | |
*** DanyC has joined #openstack-kolla | 15:59 | |
*** DanyC has quit IRC | 15:59 | |
*** TxGirlGeek has joined #openstack-kolla | 16:00 | |
*** JoseMello has quit IRC | 16:02 | |
pron | got home and first thing is kolla now :P | 16:08 |
*** classicsnail has quit IRC | 16:10 | |
*** dwalsh has quit IRC | 16:11 | |
*** LamT__ has joined #openstack-kolla | 16:16 | |
*** ankit has quit IRC | 16:23 | |
*** srwilkers has quit IRC | 16:34 | |
*** classicsnail has joined #openstack-kolla | 16:37 | |
*** JoseMello has joined #openstack-kolla | 16:45 | |
*** vinkman has left #openstack-kolla | 16:45 | |
*** vinkman has joined #openstack-kolla | 16:45 | |
sdake | pavo 3.0.1 is out | 16:58 |
Pavo | ok I'll start deployment of that when I get off work | 16:58 |
*** vinkman has quit IRC | 16:58 | |
*** mgoddard_ has joined #openstack-kolla | 16:59 | |
mliima | https://review.openstack.org/#/c/389340/ i need some eyes here | 17:00 |
*** vinkman has joined #openstack-kolla | 17:01 | |
*** Serlex has quit IRC | 17:02 | |
*** sdake_ has joined #openstack-kolla | 17:02 | |
*** mgoddard has quit IRC | 17:02 | |
Pavo | sdake you mind looking over my node settings and globals and inventory files later, because I honestly think its something that is in one of them that is giving me these issues | 17:03 |
*** DanyC has joined #openstack-kolla | 17:03 | |
*** sdake has quit IRC | 17:04 | |
*** DanyC has quit IRC | 17:07 | |
*** eaguilar has quit IRC | 17:08 | |
openstackgerrit | Merged openstack/kolla: Add enable service logic in port check https://review.openstack.org/391971 | 17:09 |
*** harlowja has joined #openstack-kolla | 17:11 | |
inc0 | sdake_, how are we in repo split departament? | 17:11 |
*** eaguilar has joined #openstack-kolla | 17:12 | |
sdake_ | inc0 haven't started | 17:13 |
sdake_ | inc0 i'll get cracking on that when my brain is working right again | 17:13 |
inc0 | kk | 17:13 |
sdake_ | its only an hour or so of work | 17:13 |
inc0 | I'm kinda braindead too | 17:13 |
sdake_ | to do the repo creation | 17:14 |
sdake_ | but its super tedious | 17:14 |
sdake_ | and dont want to foxtrot it up ;) | 17:14 |
*** DanyC has joined #openstack-kolla | 17:14 | |
*** DanyC has quit IRC | 17:15 | |
sdake_ | i'm awlays ready to rock after summit unless there is jetlag involved | 17:15 |
sdake_ | then - not so much ;-) | 17:15 |
inc0 | there is jetlag involved in my case | 17:16 |
inc0 | I slept 4 hrs out of last 32 | 17:17 |
inc0 | so yeah, sleep depravation is real | 17:17 |
Pavo | what you guys mean by split repo? | 17:19 |
portdirect | it's chatlogs like this that make me happy to be in europe, it took me longer to get to and from the airport than the flight - though I wont be so lucky next time... | 17:20 |
portdirect | was great to meet you guys :) | 17:20 |
*** intr1nsic has joined #openstack-kolla | 17:21 | |
portdirect | sdake: if you have time could you run through my commit from yesterday with me - I'd like to get it up to scratch, but am not quite sure how to get there | 17:22 |
sdake_ | portdirect link? | 17:26 |
sdake_ | i did leave comments :) | 17:26 |
portdirect | https://review.openstack.org/#/c/391921/ | 17:26 |
*** nihilifer has quit IRC | 17:27 | |
*** intr1nsic is now known as matt_welch | 17:27 | |
mliima | and here https://review.openstack.org/#/c/386552/ | 17:28 |
*** nihilifer has joined #openstack-kolla | 17:29 | |
sdake_ | portdirect - re binary builds being moot, that doesn't apply to src | 17:30 |
sdake_ | portdirect ther eis a list of pip packages to install somewhere in that file, and it shoudl list pymysqsl was the point of that review | 17:30 |
sdake_ | portdirect we dont rely on transititve dependency inclusion with source builds if the source package is installed in < 10 containers | 17:32 |
*** DanyC has joined #openstack-kolla | 17:33 | |
portdirect | re: pip it is already on line 300; I'll update the commit to make it clearer that this was to remove excess baggage? | 17:34 |
sdake_ | let me look at line 300 | 17:38 |
*** srwilkers has joined #openstack-kolla | 17:38 | |
sdake_ | i see | 17:38 |
sdake_ | gerrit is an imperfect tool | 17:38 |
sdake_ | what about toolbox? | 17:38 |
sdake_ | getting that owrking well has been a real challenge | 17:38 |
sdake_ | mostly centered around the python mysql package | 17:39 |
Pavo | I think you guys should have an example of Advanced Configuration like this https://www.gliffy.com/go/publish/11309187 | 17:39 |
sdake_ | portdirect i left a comment in the review- feel free to reply there | 17:39 |
Pavo | so its clear how to setup the kolla_interfaces | 17:40 |
sdake_ | pavo we sort of have one | 17:40 |
Pavo | on the docs? | 17:40 |
sdake_ | have you seen it? | 17:40 |
Pavo | no | 17:40 |
sdake_ | its on my gliffy | 17:40 |
*** DanyC has left #openstack-kolla | 17:40 | |
sdake_ | moment - i'll link | 17:40 |
Pavo | k | 17:40 |
*** shardy has quit IRC | 17:44 | |
portdirect | all good - though i just realised that we are pulling in the wrong lib (i think) explicity in the kolla toolbox, creating another redundancy, gotta love how theres at least 4 different mysql lib's and about 3 package names for each of them depending on python 2/3 and packager... I double check that, address your comments and amend as appropriate. Cheers :) | 17:44 |
sdake_ | pavo i'm not sure where it went | 17:45 |
sdake_ | pavo it was in fliffy no more in gliffy | 17:45 |
*** neilus has joined #openstack-kolla | 17:46 | |
*** dwalsh has joined #openstack-kolla | 17:46 | |
Pavo | ok sdake_ would love to see that diagram though | 17:46 |
sdake_ | me too - i spent like 6 hours drawing the damn thing | 17:48 |
Pavo | lol | 17:48 |
Pavo | well if you want I can setup a gliffy account and give you guys access | 17:49 |
Pavo | so there can always be something that it can reference | 17:50 |
sdake_ | pavo https://etherpad.openstack.org/p/kolla-newton-summit-threat-analysis | 17:53 |
sdake_ | yet the diagram i am looking for is not there | 17:54 |
mliima | hello sweet guys, i need some reviews here https://review.openstack.org/#/c/389340/ and here https://review.openstack.org/#/c/386552/ | 17:54 |
mliima | :) | 17:54 |
*** lamt has joined #openstack-kolla | 17:57 | |
*** lamt has left #openstack-kolla | 17:57 | |
*** mgoddard_ has quit IRC | 17:57 | |
openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Spec - Deploy Kolla-Kubernetes with Helm https://review.openstack.org/392257 | 18:00 |
rhallisey | kfox1111, inc0, srwilkers, sdake_ ^ | 18:00 |
rhallisey | we'll start the discussion there | 18:01 |
srwilkers | perfect | 18:01 |
rhallisey | srwilkers, if you push a POC of any of the services we can use them as reference | 18:01 |
rhallisey | lrensing, ^ you too :) | 18:01 |
*** athomas has quit IRC | 18:02 | |
Pavo | sdake_ that is some incredible information especially on pki | 18:02 |
*** lrensing has quit IRC | 18:03 | |
*** dougs1 has quit IRC | 18:06 | |
*** f13o has quit IRC | 18:07 | |
*** sdake has joined #openstack-kolla | 18:08 | |
*** Jeffrey4l has quit IRC | 18:08 | |
sdake | pavo we have more notes ;-) | 18:09 |
*** sdake_ has quit IRC | 18:11 | |
Pavo | sdake is this the diagram you was talking about | 18:13 |
* Pavo http://i.imgur.com/k6qo5Tf.png | 18:13 | |
sdake | pavo yup | 18:13 |
Pavo | ok based off that couldn't External Routed Network and all Provider Networks be trunked to the same interface? | 18:14 |
Pavo | and API, Storage and Neutron traffic be on another interface | 18:15 |
*** neilus has quit IRC | 18:16 | |
portdirect | Pavo: I'm working on a PR using FreeIPA's dogtag for PKI - I've been using it with k8s for all services (mysql/rabbit/keystone/k8s api & kubelet) - eventually I'd like to get it to a point where we could introduce untrusted nodes into the cluster but this will require some signifigant work at the FreeIPA side of things to operate smoothly. | 18:19 |
openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Spec - Deploy Kolla-Kubernetes with Helm https://review.openstack.org/392257 | 18:22 |
inc0 | Pavo, ad repo split - ansible dir from kolla gets it's own project;) | 18:49 |
inc0 | we split ansible plays from docker images | 18:49 |
Pavo | ah ok | 18:51 |
*** strigazi is now known as strigazi_AFK | 19:09 | |
*** lrensing has joined #openstack-kolla | 19:10 | |
*** dave-mccowan has quit IRC | 19:11 | |
*** Administrator__ has joined #openstack-kolla | 19:11 | |
*** zhugaoxiao has quit IRC | 19:15 | |
*** inc0 has quit IRC | 19:20 | |
harlowja | the split happened? | 19:21 |
*** eaguilar has quit IRC | 19:24 | |
*** portdirect is now known as harbor_log | 19:24 | |
*** clsacramento has joined #openstack-kolla | 19:26 | |
*** mliima has quit IRC | 19:30 | |
*** dave-mccowan has joined #openstack-kolla | 19:32 | |
*** inc0 has joined #openstack-kolla | 19:33 | |
*** eaguilar has joined #openstack-kolla | 19:34 | |
harlowja | sooooo given something like https://gist.github.com/harlowja/227ea1337cda59b3a5d6b47a91a64717 (jenkins job builder format) | 19:36 |
harlowja | i'm about to setup that jenkins groovy script to checkout kolla, and then i guess formulate a run of kolla to just generate glance images | 19:37 |
harlowja | seem doable to folks? | 19:37 |
*** matt_welch has quit IRC | 19:37 | |
*** neilus has joined #openstack-kolla | 19:38 | |
dims | rhallisey : following up to the chat at BCN, see SIG-Openstack links here (https://github.com/kubernetes/community/blob/master/sig-openstack/README.md) | 19:40 |
dims | rhallisey : there's a slack channel and see the google docs for meeting logs | 19:40 |
rhallisey | dims, idk if that's the right place though | 19:40 |
* dims just saw the kolla-kubernetes update :) | 19:40 | |
rhallisey | dims, that's kubernetes on openstack | 19:40 |
rhallisey | I'm not sure if a SIG group will help here | 19:41 |
*** lrensing has quit IRC | 19:41 | |
dims | rhallisey : right, just a pointer to find the folks who are doing stuff between the two... | 19:41 |
rhallisey | dims, of course. Thank you sir | 19:42 |
dims | was just reading through (https://github.com/kubernetes/features/issues/88) and remembered you :) | 19:42 |
rhallisey | dims, do you think the tc would be open to a proposal to make projects more cloud native? | 19:43 |
*** neilus has quit IRC | 19:43 | |
rhallisey | kolla-k8s does inter op here, but the idea is larger than just this project | 19:44 |
rhallisey | dims, oh neat link | 19:44 |
*** dougs has joined #openstack-kolla | 19:45 | |
*** lrensing has joined #openstack-kolla | 19:46 | |
dims | rhallisey : i threw that into the board/tc meeting agenda and talked about what's being done today (https://etherpad.openstack.org/p/bcn-bod+tc+uc-topics). there was some discussion to work better at the foundation level | 19:46 |
*** dougs1 has joined #openstack-kolla | 19:47 | |
dims | rhallisey : starting with showcasing work already being done and get folks to come to each others meetings etc. steve gordon is organizing something at kubecon too | 19:47 |
rhallisey | crap do I have to go to kubecon now | 19:48 |
* rhallisey doesn't have a ticket | 19:48 | |
dims | https://groups.google.com/forum/#!topic/kubernetes-sig-openstack/5MLYimcXw5c | 19:48 |
dims | y, i won't make it as well | 19:49 |
*** dougs1 has quit IRC | 19:49 | |
*** dougs has quit IRC | 19:50 | |
*** dougs has joined #openstack-kolla | 19:50 | |
rhallisey | dims, I think its a good step | 19:50 |
*** neilus has joined #openstack-kolla | 19:50 | |
*** dougs1 has joined #openstack-kolla | 19:51 | |
rhallisey | dims, I'm not quite sure the best way just yet since these are two large communities moving at a million mph | 19:51 |
dims | rhallisey : re-reading what you had sent to the -dev list. if you want to do more around helm, then SIG-Apps may be more appropriate. looks like they have an in-person meeting too (https://groups.google.com/forum/#!msg/kubernetes-users/bO0iImAY3m0/Uzp7SExSAAAJ) | 19:51 |
rhallisey | yes that's the plan. Once the community can identify gaps there I think people will start overlapping between communities | 19:52 |
dims | ++ thanks | 19:52 |
rhallisey | it will be a good bring between the two | 19:53 |
rhallisey | bridge* | 19:53 |
*** neilus has quit IRC | 19:55 | |
*** dougs has quit IRC | 19:55 | |
*** haplo37_ has quit IRC | 20:01 | |
*** JoseMello has quit IRC | 20:02 | |
*** haplo37_ has joined #openstack-kolla | 20:04 | |
jascott1 | rhallisey: was there analysis done on this https://github.com/sapcc/openstack-helm | 20:05 |
rhallisey | jascott1, I've looked at it and it has come a lot father since the community last looked at it | 20:05 |
rhallisey | I think we'll be able to deploy with it with maybe a few hiccups | 20:06 |
rhallisey | I think upgrades, updates, and reconfigure will be more challenging | 20:06 |
jascott1 | ah gotcha | 20:06 |
rhallisey | jascott1, so to answer you question, this fills part of the overall lifecycle gap. I don't think it will fill the entire thing today | 20:07 |
rhallisey | the community will need to work within both projects to make that happen | 20:07 |
jascott1 | so sounds like we will build on that project then | 20:08 |
rhallisey | ya that's the plan | 20:09 |
rhallisey | for work items we'll start with deployment and work up as we go | 20:09 |
rhallisey | so if you want to take on a service feel free to | 20:09 |
sdake | dims work better at the foundation level in which way? | 20:14 |
sdake | jascott1 what id' like is a list of our gaps with helm | 20:17 |
sdake | jascott1 if you want to do an eval to determine that so we can figure out where to point new people that would be good ;-) | 20:17 |
sdake | where helm can't be made to work we need kubernetes controllers | 20:18 |
sdake | we also need a list of what these controllers are/should be | 20:18 |
sdake | jascott1 that will come naturally from a helm gap analysis i suspect | 20:18 |
jascott1 | I will start stabbing at it | 20:18 |
rhallisey | sdake, do you have a link to kubernetes controllers | 20:19 |
rhallisey | are you sure they're called that? | 20:19 |
sdake | rhallisey you mean what htey do? | 20:19 |
sdake | i am pretty sure that is what they are called | 20:19 |
rhallisey | I'd like to include it in the spec | 20:20 |
dims | sdake : tickets/travel/events/meetups etc are possibilities | 20:20 |
sdake | an example: https://github.com/Mirantis/k8s-AppController | 20:20 |
rhallisey | sdake, let's start with deployment. I don't think we're going to be able to identify gaps until we start applying things to helm | 20:20 |
sdake | dims you mean cloudnativecon in particular? | 20:21 |
rhallisey | sdake, ok | 20:21 |
sdake | rhallisey right - that appcontroller work that mirantis has done is to support dependencies in deployment | 20:21 |
dims | opnfv, kubecon others that we can identify | 20:21 |
jascott1 | appcontroller is a mirantis construct | 20:21 |
sdake | seems like we can do something similar to solve th edependencies thing, but directly in helm | 20:21 |
sdake | jascott1 the first gap i know of is helm will not handle package dependencies | 20:22 |
jascott1 | right | 20:22 |
rhallisey | what's the diff between app controller and helm.. | 20:22 |
sdake | it is not an orchestration engine | 20:22 |
jascott1 | helm handles deps on other charts | 20:22 |
sdake | jascott1 e.g. if you launch glance and keystone, two mariadb's are started | 20:22 |
jascott1 | ah | 20:23 |
sdake | hence the chart needs to be al all-in-one of the deployment | 20:23 |
sdake | that is one approach | 20:23 |
sdake | (the crappy way) | 20:23 |
*** dougs1 has quit IRC | 20:23 | |
sdake | or helm needs to handle deps in some way | 20:23 |
sdake | solve that - win :) | 20:23 |
rhallisey | ok gap in helm then | 20:23 |
rhallisey | hehe | 20:23 |
sdake | well its a gap in kubernetes | 20:23 |
sdake | but kubernetes doesn't want deps | 20:23 |
*** ppalacios has quit IRC | 20:23 | |
rhallisey | debatable sdake | 20:23 |
sdake | here, helm isn't doing the right thing | 20:23 |
rhallisey | so no gap then :) | 20:23 |
rhallisey | ah ok | 20:24 |
rhallisey | we in agreement then | 20:24 |
sdake | no | 20:24 |
sdake | openshift has dependencies | 20:24 |
sdake | kubernetes does not have dependencies | 20:24 |
rhallisey | that's an upstream decision | 20:24 |
sdake | got it | 20:25 |
sdake | if kubernetes doesn't want deps it needs to be solved in a different way | 20:25 |
sdake | such as what openshift has done | 20:25 |
sdake | or a master helm file | 20:25 |
rhallisey | tons of discusision on that. They dont want it | 20:25 |
sdake | or making helm understand dependencies! | 20:25 |
rhallisey | yes | 20:25 |
rhallisey | I agree | 20:25 |
rhallisey | whatever I know what you mean | 20:25 |
sdake | right i am expalining to jascott1 :) | 20:25 |
openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Spec - Deploy Kolla-Kubernetes with Helm https://review.openstack.org/392257 | 20:33 |
rhallisey | cool thx sdake | 20:33 |
rhallisey | added to spec | 20:33 |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla: WIP: Tacker NFV support https://review.openstack.org/389790 | 20:34 |
*** phileasF has joined #openstack-kolla | 20:34 | |
*** vinkman has left #openstack-kolla | 20:35 | |
*** neilus has joined #openstack-kolla | 20:38 | |
*** rhallisey has quit IRC | 20:40 | |
*** neilus has quit IRC | 20:42 | |
*** lrensing has quit IRC | 20:43 | |
*** tummy has joined #openstack-kolla | 20:48 | |
*** phileasF has quit IRC | 20:52 | |
*** neilus has joined #openstack-kolla | 20:56 | |
harlowja | anyone seen the following http://paste.openstack.org/show/587586/ ? | 20:56 |
*** srwilkers has quit IRC | 20:58 | |
inc0 | harlowja, check syslog | 20:58 |
harlowja | k | 20:58 |
inc0 | seems like docker deamon issue | 20:58 |
harlowja | ya, thats my guess | 20:58 |
*** neilus has quit IRC | 20:59 | |
TxGirlGeek | qwang jascott1 Jira board is live, want a walkthrough? | 21:06 |
qwang | TxGirlGeek: sure | 21:06 |
jascott1 | cool thanks | 21:06 |
*** dwalsh has quit IRC | 21:07 | |
TxGirlGeek | jascott1 qwang hit me on lync | 21:08 |
jascott1 | one sec gotta change networks | 21:08 |
*** fguillot has quit IRC | 21:09 | |
Pavo | sdake ok kolla_internal_vip_address is for API network correct? | 21:10 |
*** matrohon has joined #openstack-kolla | 21:11 | |
Pavo | which should be used on network_interface correct? | 21:11 |
*** absubram has quit IRC | 21:12 | |
*** tummy has quit IRC | 21:21 | |
*** jtriley_ has quit IRC | 21:25 | |
*** sdake has quit IRC | 21:27 | |
*** schwicht has quit IRC | 21:33 | |
*** jheroux has quit IRC | 21:35 | |
*** schwicht has joined #openstack-kolla | 21:48 | |
Pavo | inc0 how does this look as a network setup? https://www.gliffy.com/go/publish/11309187 | 21:50 |
Pavo | with that type of network layout what interfaces in globals would you use? | 21:50 |
inc0 | you have openstack api and mgmt on separate nets right? you might want to add dedicated neutron_external_interface | 21:51 |
inc0 | and dedicated tunnel interface | 21:52 |
inc0 | vlan+qos works | 21:52 |
inc0 | Pavo, with or without ceph? | 21:52 |
Pavo | without | 21:53 |
Pavo | using Cinder | 21:53 |
Pavo | for storage | 21:53 |
inc0 | which backend? | 21:53 |
Pavo | LVM | 21:53 |
inc0 | lvm+iscsi is not ideal tho, be careful | 21:53 |
inc0 | no HA for your data | 21:53 |
Pavo | yeah this is just a PoC | 21:54 |
inc0 | cinder node dies-> data goes out | 21:54 |
inc0 | kk | 21:54 |
Pavo | ideal would use Ceph | 21:54 |
Pavo | but for PoC | 21:54 |
inc0 | it's better in almost every way | 21:54 |
Pavo | just using Cinder | 21:54 |
inc0 | well, if you're using ceph, you also use cinder | 21:54 |
Pavo | right now we only have 1 node that has the storage | 21:54 |
inc0 | cinder will use ceph as backend | 21:54 |
inc0 | ok | 21:54 |
Pavo | as you can see from the diagram | 21:55 |
inc0 | 1 node rules out ceph for any serious thing | 21:55 |
Pavo | Controller, Network and Storage on the same node | 21:55 |
inc0 | thats fine | 21:55 |
inc0 | make sure to have separate interface for neutron_external_interface tho | 21:55 |
Pavo | yeah here is an updated one | 21:56 |
Pavo | https://www.gliffy.com/go/publish/11309187 | 21:56 |
inc0 | you can just make tagged iface on trunked net | 21:56 |
inc0 | still no neutron external:) | 21:56 |
inc0 | this will be iface for flat network | 21:57 |
Pavo | couldn't that be done on the trunk? | 21:57 |
Pavo | on the Network node | 21:57 |
*** schwicht has quit IRC | 21:57 | |
inc0 | it has to be on compute as well | 21:57 |
Pavo | or does flat and VLAN Provider network have to be seperate? | 21:57 |
inc0 | and yeah, trunk works if you do eth0.vlantag | 21:57 |
Pavo | hmmm | 21:58 |
inc0 | you need a bit more config for vlan provider | 21:58 |
Pavo | I thought that the Compute node would use the OpenStack API to do Tenent network traffic | 21:58 |
inc0 | ahh you're right | 21:59 |
inc0 | well, you won't be able to spawn vm in flat network | 21:59 |
inc0 | but openstack api will be used for tunnels | 21:59 |
Pavo | and the Network Node is what needed the Neutron External (ie... Internet) and all other Provider Network traffic | 21:59 |
inc0 | (sorry, sleep depravation | 21:59 |
inc0 | ) | 21:59 |
*** neilus has joined #openstack-kolla | 21:59 | |
inc0 | well if you want to spawn vm in vlan, that's ok too | 22:00 |
inc0 | just need to use provider vlan and add trunked iface on compute node | 22:00 |
Pavo | so to be able to use the Trunk on the network node for both Neutron Ext and all other VLAN Provider nets I would have to manually add those correct? Or would BR-EX take care of that | 22:00 |
*** matrohon has quit IRC | 22:01 | |
Pavo | so I need to add another 10Gb Trunk to Compute? | 22:01 |
awiddersheim | inc0: can you review? https://review.openstack.org/#/c/279913 | 22:01 |
awiddersheim | trying to get that merged | 22:01 |
inc0 | Pavo, if you want to spawn vms in vlan - then yes | 22:02 |
Pavo | hmmm | 22:02 |
inc0 | if you're ok with using tunnels, your config is fine | 22:02 |
Pavo | ok would I also need to add the Network Node to the Compute node? | 22:03 |
Pavo | in Kolla inventory? | 22:03 |
inc0 | awiddersheim, sorry, my brain doesn't work today;) I slept too little to review complex changes | 22:03 |
awiddersheim | inc0: no worries | 22:03 |
awiddersheim | maybe when you get some rest | 22:03 |
inc0 | I promise to review it when I get rid of sleep deprivation aftereffects | 22:04 |
inc0 | Pavo, uhh....I'm lost in these cables;) so your config on gliffy should just work | 22:04 |
inc0 | for tenant networks:vxlan which is default | 22:04 |
Pavo | inc0 without having to add another Trunk to the Compute node? | 22:05 |
inc0 | yeah | 22:06 |
Pavo | ok cool | 22:06 |
inc0 | adding trunk to compute would make sense if you'd want to spawn vms in vlans | 22:06 |
Pavo | well that also would be ideal | 22:06 |
inc0 | but that needs slight configuration, not too much | 22:06 |
Pavo | inc0 so like this https://www.gliffy.com/go/publish/11309187? | 22:07 |
inc0 | yeah | 22:08 |
inc0 | you'll need to do config override of this var tho https://github.com/openstack/kolla/blob/master/ansible/roles/neutron/templates/ml2_conf.ini.j2#L26 | 22:09 |
Pavo | yeah I know I gotta make some custom ini files | 22:11 |
Pavo | does this make more sense? | 22:11 |
Pavo | https://www.gliffy.com/go/publish/11309187 | 22:11 |
*** neilus has quit IRC | 22:11 | |
inc0 | yeah, if you use vlan it's super easy to expose vms to internet | 22:12 |
inc0 | just put gateway there and conf neutron dhcp to use it | 22:12 |
Pavo | inc0 ok cool, now I gotta figure out what is causing me the issue of kolla not creating the br-ex on the network node | 22:13 |
*** schwicht has joined #openstack-kolla | 22:13 | |
inc0 | as much as I'd like to help, I think I'll go take a nap;) I still live in Barcelona timezone | 22:14 |
Pavo | in the globals file I have eth0 as my kolla_internal_vip_address and openstack api interface and eth1 as my kolla_external_vip_interface and kolla_external_vip_address | 22:14 |
Pavo | ok no problem, get some rest man | 22:15 |
inc0 | br_ex lands on neutron_external_iface | 22:15 |
Pavo | which is what interface in globals? | 22:15 |
inc0 | neutron_external_interface :) | 22:15 |
Pavo | ah | 22:15 |
inc0 | this has to be separate | 22:16 |
Pavo | and that has be on both Network and Compute node right | 22:17 |
Pavo | or just the Network node? | 22:18 |
inc0 | correct | 22:18 |
*** haplo37 has quit IRC | 22:18 | |
Pavo | lol which one is correct? | 22:18 |
inc0 | network and compute | 22:18 |
Pavo | which is what I have in my diagram as the 10Gb Trunk | 22:19 |
Pavo | ok cool | 22:19 |
Pavo | now does those interfaces have to be up but no IP assigned when you deploy? | 22:19 |
Pavo | or does kolla bring them up? after configuring br-ex | 22:20 |
inc0 | I think they don't even have to be up when bridge takes over | 22:21 |
Pavo | ok | 22:21 |
inc0 | IP, if it's there, will simply be ignored | 22:21 |
Pavo | thanks gonna get this all configured now | 22:21 |
inc0 | not 100% about "don't have to be up" tho | 22:21 |
Pavo | and do a deployment test | 22:21 |
Pavo | ok | 22:21 |
Pavo | I can test both | 22:21 |
Pavo | one more thing before you go | 22:21 |
inc0 | go ahead | 22:22 |
Pavo | mind taking a quick look at the settings that I am using on my target nodes to make sure I am not messing something up that could stop the br-ex interface from being created? | 22:22 |
inc0 | bring them on!:) | 22:22 |
Pavo | http://pastebin.com/TN0f3YKX | 22:23 |
Pavo | that is all I do to the target host | 22:23 |
dmsimard | Cores, can we merge https://review.openstack.org/#/c/390030/ ? It's an easy one :) | 22:23 |
Pavo | ie... Controller, Compute node | 22:23 |
inc0 | btw Pavo if you'd like to take on making selinux policies we could get rid of this disable requirement | 22:24 |
*** haplo37_ has quit IRC | 22:25 | |
inc0 | rhallisey was meant to do it but he was officially marked as slacker on summit | 22:25 |
Pavo | inc0 true and I am working on those, I do have a few SELINUX policies made for openstack services, dunno how that effect dockers though | 22:25 |
inc0 | also we have bootstrap-servers playbook that will do most of this stuff | 22:25 |
Pavo | inc0 yeah I do not know how to use that at all | 22:26 |
Pavo | no docs on how to use it | 22:26 |
inc0 | https://github.com/openstack/kolla/tree/master/ansible/roles/baremetal/tasks | 22:26 |
Pavo | doesn;t that require bifronst? | 22:26 |
inc0 | https://github.com/openstack/kolla/blob/master/doc/quickstart.rst#automatic-host-bootstrap | 22:26 |
inc0 | no | 22:26 |
inc0 | this will turn clean ubuntu/centos -> ready for install | 22:27 |
Pavo | hmmm does this also make the cinder-volumes? | 22:27 |
inc0 | just make sure to prep deployment node first and conf globals.yml | 22:27 |
*** haplo37 has joined #openstack-kolla | 22:27 | |
inc0 | no, cinder lvm stuff has to be done manually | 22:27 |
inc0 | but it installs docker, sets up hostnames and all that | 22:28 |
Pavo | hmmm nice | 22:28 |
Pavo | here is my deployer node settings | 22:28 |
Pavo | http://pastebin.com/FMh2KgJJ | 22:28 |
*** eaguilar has quit IRC | 22:29 | |
inc0 | show me globals.yml plz | 22:29 |
*** TxGirlGeek has quit IRC | 22:29 | |
Pavo | ok one sec | 22:30 |
*** g3ek has quit IRC | 22:30 | |
*** schwicht has quit IRC | 22:32 | |
Pavo | inc0 globals http://pastebin.com/yxZWfmNK | 22:32 |
*** haplo37 has quit IRC | 22:32 | |
Pavo | inc0 inventory http://pastebin.com/cNnUukNJ | 22:33 |
*** schwicht has joined #openstack-kolla | 22:34 | |
inc0 | you want to specify kolla_external_vip_interface if this is different than internal | 22:34 |
inc0 | which it seems to be the case looking at addresses | 22:35 |
Pavo | yeah | 22:35 |
inc0 | (unles it's 192.168.0.0/16) | 22:35 |
Pavo | nah | 22:35 |
Pavo | internal is for admins which is on NodeMGT vlan | 22:35 |
inc0 | eth1 -> trunk | 22:35 |
inc0 | eth0 -> api traffic | 22:35 |
Pavo | yes | 22:35 |
inc0 | internal will land on eth0 tho | 22:35 |
*** schwicht has quit IRC | 22:36 | |
Pavo | which should be correct since its on api traffic right? | 22:36 |
Pavo | or is that wrong | 22:36 |
inc0 | yeah, it's correct | 22:36 |
Pavo | ok | 22:36 |
Pavo | and kolla_external_vip_interfacewould land on neutron_external interface right | 22:37 |
inc0 | no | 22:37 |
inc0 | kolla_external_vip_interface will default to eth0 | 22:37 |
inc0 | you want to specify dedicated one | 22:37 |
Pavo | or for external access to horizon right | 22:38 |
inc0 | with 192.168.4.0/24 addr | 22:38 |
inc0 | horizon and apis, yes | 22:38 |
Pavo | oh ok | 22:38 |
Pavo | ok so hang on | 22:38 |
*** haplo37 has joined #openstack-kolla | 22:38 | |
inc0 | sorry | 22:38 |
inc0 | 192,168.2.0/24 is external | 22:39 |
Pavo | kolla_external_vip_interface and kolla_internal_vip_interface lands on api traffic? | 22:39 |
inc0 | no, not exactly | 22:39 |
inc0 | so API will be available on both | 22:39 |
inc0 | external and internal | 22:39 |
inc0 | 2 different endpoints | 22:39 |
inc0 | internal -> this is for openstack services to talk to each other (nova to neutron etc) | 22:39 |
inc0 | external -> for users who want to use APIs or CLI clients | 22:40 |
Pavo | I understand that but how it get routed on the actual NICs | 22:40 |
*** g3ek has joined #openstack-kolla | 22:40 | |
inc0 | both will be created in HAProxy | 22:40 |
*** haplo37_ has joined #openstack-kolla | 22:40 | |
inc0 | users never really hit service api itself | 22:40 |
inc0 | always through haproxy | 22:40 |
Pavo | ah | 22:40 |
Pavo | which is routed through api traffic | 22:41 |
inc0 | yeah, haproxy will expose 192.168.2.100:5000 for keystone | 22:41 |
Pavo | since I am using 1 10Gb NIC for all API traffic | 22:41 |
inc0 | and then load balance between 192.168.2.<<node octet>>:5000 to all keystone nodes | 22:42 |
Pavo | and only Neutron services will use the second 10Gb Trunk for Tenant and Provider network traffic correct? | 22:43 |
inc0 | well, neutron services will set them up | 22:43 |
inc0 | VMs will do actual usage | 22:43 |
inc0 | neutron service will setup ovs and connect it to vm and stuff like that | 22:43 |
Pavo | yeah thats what I meant | 22:43 |
Pavo | ok I think I understand it completely now | 22:44 |
inc0 | and if you add native vlan there that's gonna be flat network | 22:44 |
inc0 | yeah all the terminology might be confusing | 22:44 |
inc0 | but there is no magic there really;) | 22:44 |
Pavo | yeah think I am gonna have 192.168.2.0/24 network set as native vlan so it can be used for Internet (External) network | 22:44 |
inc0 | wait until we add pacemaker container tho.. | 22:45 |
Pavo | now I have a cinder-volume setup that is 30Tb on the Controller node | 22:45 |
inc0 | so be careful with that, you still need neutron_extenral_interface separate | 22:45 |
Pavo | but after deployment Horizon is only showing 50Gb as storage | 22:45 |
inc0 | I think that's matter of lvm conf | 22:46 |
Pavo | hmmm | 22:46 |
inc0 | I'm not good at this space tbh, I always run ceph | 22:46 |
Pavo | wished I had 2 more node and I would use Ceph | 22:46 |
Pavo | I have never setup Ceph | 22:46 |
inc0 | you can use aio ceph with a bit of hack | 22:47 |
Pavo | would be a good learning experience | 22:47 |
inc0 | it's totally worth the time | 22:47 |
Pavo | well maybe you can help me set that up after you get some rest | 22:47 |
inc0 | I'm flying tomorrow (yes...plane again) | 22:48 |
inc0 | but I'm sure someone here would help | 22:48 |
inc0 | https://github.com/openstack/kolla/blob/master/doc/ceph-guide.rst | 22:48 |
Pavo | ok cool | 22:48 |
inc0 | it's pretty straightforward tho | 22:48 |
Pavo | thanks for all your help | 22:48 |
Pavo | have a safe flight | 22:48 |
Pavo | and get some rest | 22:49 |
inc0 | thanks:) | 22:49 |
inc0 | good luck! | 22:49 |
*** inc0 has quit IRC | 22:52 | |
*** lamt has joined #openstack-kolla | 22:53 | |
*** harbor has joined #openstack-kolla | 23:18 | |
*** harbor is now known as portdirect_ | 23:18 | |
jascott1 | Pavo: you can also look at setup_gate.sh to see how you can setup a single node ceph for testing | 23:21 |
*** lamt has quit IRC | 23:21 | |
*** asalkeld has joined #openstack-kolla | 23:36 | |
*** yuanying has joined #openstack-kolla | 23:37 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!