*** sdake_ has quit IRC | 00:04 | |
*** jascott1_ has joined #openstack-kolla | 00:08 | |
*** haplo37 has quit IRC | 00:10 | |
*** haplo37 has joined #openstack-kolla | 00:13 | |
awiddersheim | could use a review on this | 00:23 |
---|---|---|
awiddersheim | https://review.openstack.org/#/c/409909/ | 00:23 |
awiddersheim | without it kolla-genpwd doesn't work :( | 00:24 |
*** Pavo has quit IRC | 00:29 | |
*** yingjun has joined #openstack-kolla | 00:30 | |
*** ipsecguy_ has quit IRC | 00:30 | |
*** ipsecguy has joined #openstack-kolla | 00:31 | |
*** sdake has joined #openstack-kolla | 00:34 | |
*** Pavo has joined #openstack-kolla | 00:34 | |
*** ipsecguy has quit IRC | 00:39 | |
*** yingjun has quit IRC | 00:44 | |
*** ipsecguy has joined #openstack-kolla | 00:44 | |
*** duonghq has joined #openstack-kolla | 00:52 | |
duonghq | morning guys | 00:54 |
srwilkers | hey duonghq | 00:57 |
duonghq | nice to see you srwilkers | 00:58 |
srwilkers | you as well duonghq :) | 00:58 |
Jeffrey4l | srwilker-, re kolla-k8s meeting, i think it is possible. but one issue is kolla is a common for kolla-ansible kolla-k8s, some guys may have to attend two meeting a week, and take some issue to sync the decision made by other in the meeting. and more guys may have no idea about the decision made by other meeting. we need find a better solution for handing this issue. | 01:01 |
*** diogogmt has joined #openstack-kolla | 01:01 | |
Jeffrey4l | and we also need check with TC whether we can get another 1 hour for kolla. | 01:02 |
srwilkers | Jeffrey4l, if it's a common meeting, we need to ensure that at least each component gets enough time for discussion with a hard stop, regardless of the agenda for the meeting. i understand the predicament you mention, and i agree | 01:03 |
srwilkers | these one-off discussions we keep having here are very disruptive and introduce a lot of latency | 01:03 |
srwilkers | re: kolla-k8s that is | 01:03 |
srwilkers | i also feel that introducing walls of texts for architectural discussions here drown out others that may have concerns re: kolla/kolla-ansible. just seems very disruptive for all :( | 01:05 |
sdake | sup duonghq | 01:06 |
sdake | sup srwilkers | 01:06 |
sdake | sup Jeffrey4l | 01:06 |
sdake | Jeffrey4l hard stop seems reasonable | 01:06 |
srwilkers | hey sdake | 01:06 |
Jeffrey4l | morning sdake | 01:06 |
sdake | kolla deliverable is on autopilot | 01:06 |
duonghq | morning sdake | 01:07 |
sdake | kolla-ansible is less on autopilot but needs less attention | 01:07 |
sdake | kolla-kubernetes needs 30 minutes and not a second more :) | 01:07 |
srwilkers | sdake, +1 | 01:07 |
duonghq | sdake, +1 | 01:08 |
sdake | i think that solves the problem for everyone | 01:08 |
sdake | worth an experiment atleasat | 01:08 |
Jeffrey4l | is it possible if we have a 1.5 hours meeting? 30 min for kolla, 30 min for kolla-ansible and 30 for kolla-k8s | 01:08 |
sdake | flavio's email in that thread was right on too | 01:08 |
sdake | Jeffrey4l the schedule is locked down on meeting times | 01:09 |
sdake | there is no room to budge | 01:09 |
*** huzhengchuan has joined #openstack-kolla | 01:09 | |
sdake | flavio's thread: http://lists.openstack.org/pipermail/openstack-dev/2015-February/056551.html | 01:09 |
sdake | sometimes ther ewill be no agenda space for any deliverable - only planning for ptgs or summits | 01:10 |
sdake | it is clear from the governance documentation that all meetings must occur on irc and must occur in the open | 01:11 |
*** yuanying has quit IRC | 01:11 | |
*** yuanying has joined #openstack-kolla | 01:11 | |
*** fragatin_ has joined #openstack-kolla | 01:12 | |
srwilkers | i agree with that thread 100% | 01:12 |
sdake | another option is 20 / 20 /20 | 01:13 |
sdake | th downside of this is kolla deliverable really needs no attention in the meetings | 01:14 |
sdake | rollcall and announcements usually burn up 10 minutes | 01:14 |
sdake | so we coulld do 10/10/20/20 | 01:14 |
*** fragatin_ has quit IRC | 01:14 | |
*** fragatina has quit IRC | 01:15 | |
*** fragatina has joined #openstack-kolla | 01:15 | |
*** zhurong has joined #openstack-kolla | 01:15 | |
*** zhubingbing has joined #openstack-kolla | 01:20 | |
Jeffrey4l | sdake, yep. we can try that in the next meeting. and try to see whether it will make team happy. | 01:20 |
*** zhurong has quit IRC | 01:20 | |
sdake | soundsgood | 01:20 |
sdake | we can always overflow if needed | 01:20 |
sdake | not for hours on end | 01:20 |
sdake | but agree the walls of text are distracting ;) | 01:21 |
sdake | need to feed family bbl | 01:21 |
*** zhurong has joined #openstack-kolla | 01:22 | |
*** ayoung has quit IRC | 01:25 | |
*** tovin07_ has joined #openstack-kolla | 01:35 | |
*** tonanhngo_ has quit IRC | 01:36 | |
*** wanghongxu has joined #openstack-kolla | 01:36 | |
*** tonanhngo has joined #openstack-kolla | 01:38 | |
*** tonanhngo has quit IRC | 01:42 | |
*** srwilkers has quit IRC | 01:42 | |
*** tonanhngo has joined #openstack-kolla | 01:42 | |
*** sayantan_ has quit IRC | 01:44 | |
*** sdake has quit IRC | 01:44 | |
*** sayantan_ has joined #openstack-kolla | 01:45 | |
*** tonanhngo has quit IRC | 01:47 | |
*** elynn has joined #openstack-kolla | 01:51 | |
*** adrian_otto has quit IRC | 01:54 | |
*** elynn has quit IRC | 01:56 | |
*** elynn has joined #openstack-kolla | 01:56 | |
*** sdake has joined #openstack-kolla | 01:57 | |
*** yingjun has joined #openstack-kolla | 02:03 | |
*** gomarivera has joined #openstack-kolla | 02:05 | |
*** fragatina has quit IRC | 02:06 | |
*** fragatina has joined #openstack-kolla | 02:07 | |
*** sayantan_ has quit IRC | 02:07 | |
openstackgerrit | wangwei proposed openstack/kolla-ansible: Add remove-garbage function in kolla-ansible https://review.openstack.org/402362 | 02:09 |
*** gomarivera has quit IRC | 02:10 | |
*** jascott1_ has quit IRC | 02:11 | |
*** yuanying has quit IRC | 02:11 | |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helm-ize Heat API and CFN https://review.openstack.org/409959 | 02:12 |
*** huzhengchuan has quit IRC | 02:14 | |
*** jascott1_ has joined #openstack-kolla | 02:14 | |
v1k0d3n | back sdake | 02:15 |
*** sdake has quit IRC | 02:17 | |
*** sdake has joined #openstack-kolla | 02:20 | |
*** huzhengchuan has joined #openstack-kolla | 02:26 | |
*** rhallisey has quit IRC | 02:27 | |
*** Pavo has quit IRC | 02:29 | |
*** Pavo has joined #openstack-kolla | 02:33 | |
*** sdake has quit IRC | 02:40 | |
*** sdake has joined #openstack-kolla | 02:42 | |
sdake | sup peeps | 02:42 |
*** unicell has quit IRC | 02:44 | |
*** sdake_ has joined #openstack-kolla | 02:45 | |
*** Dishwasha has joined #openstack-kolla | 02:49 | |
*** sdake has quit IRC | 02:49 | |
Dishwasha | Howdy folks | 02:49 |
Dishwasha | I've done an all-in-one kolla installation successfully and am now moving on to a multi-node install | 02:50 |
*** sayantan_ has joined #openstack-kolla | 02:51 | |
Dishwasha | When I'm running the prechecks, I get a failure in the port checks complaining that dict object has no attribute u'ansible_<my_kolla_external_vip_interface_name>' | 02:52 |
Dishwasha | Checking free port for Aodh API | 02:52 |
Dishwasha | api interface name is the same | 02:52 |
sdake_ | Dishwasha see subject pls :) | 02:53 |
Dishwasha | Do you think its because my interface name has a hyphen in it? | 02:53 |
sdake_ | if you ask there, it will b eanswered | 02:53 |
sdake_ | and then everyone benefits | 02:53 |
Dishwasha | gotcha | 02:54 |
duonghq | sdake_, seems that srwilker is working on helm glance, so I guess that I should come back to Operator(s) of Ryan, do you have any suggestion on it? | 02:54 |
sdake_ | thanks :) | 02:54 |
sdake_ | duonghq 8pm here dude | 02:54 |
sdake_ | duonghq out of steam | 02:54 |
sdake_ | duonghq give me 5-10 misn - just got home from dinner | 02:55 |
duonghq | sdake_, sorry, don't no you go out for dinner | 02:55 |
duonghq | *know | 02:55 |
sdake_ | its all good | 02:55 |
sdake_ | we pretty much eat out 7 days a week | 02:55 |
sdake_ | optimizes our time :) | 02:55 |
sdake_ | (at the expense of money) | 02:55 |
duonghq | ya | 02:56 |
sdake_ | more money then time atm | 02:56 |
duonghq | restaurant is near your home? | 02:56 |
v1k0d3n | hey sdake_ | 03:01 |
sdake_ | duonghq ya 15 min drive | 03:01 |
sdake_ | sup v1k0d3n | 03:02 |
v1k0d3n | told you when i'd be back | 03:02 |
sdake_ | huh? | 03:02 |
v1k0d3n | nevermind. | 03:02 |
sdake_ | ok :) | 03:03 |
v1k0d3n | so i think in a constructive way, flipping the conversation around....perhaps a document covering the goal of the current workflow would be good? | 03:04 |
v1k0d3n | your thoughts? | 03:04 |
v1k0d3n | this way, it could be reviewable by other possible contributors, and outside resources like helm core developers | 03:04 |
sdake_ | anyone is free to submit any review to gerrit that they like | 03:07 |
sdake_ | as long as they have signed the cla | 03:07 |
sdake_ | I think gerrit enforces that | 03:08 |
v1k0d3n | right | 03:08 |
sdake_ | duonghq bbi10 mins need to flog my children to do their homework | 03:09 |
*** v1k0d3n has quit IRC | 03:09 | |
*** zhubingbing has quit IRC | 03:10 | |
*** zhubingbing has joined #openstack-kolla | 03:11 | |
*** jtriley has joined #openstack-kolla | 03:13 | |
sdake_ | duonghq done with thata | 03:17 |
duonghq | sdake_, ya | 03:17 |
sdake_ | duonghq it looks like in the blueprint your still marked as glance microservices | 03:17 |
sdake_ | duonghq i see srwilker- has heat atm | 03:18 |
duonghq | yup, but srwilker is working on that, w/ at least 1 ps is merged (pv/pvc one) | 03:18 |
sdake_ | ok srwilker- can you change the tracker pls | 03:18 |
duonghq | As I see, srwilker- , portdirect ... nearly has compute kit (at microservice layer) is done or good progress on review queue, so I guess that it's time for me to move to Operator, finding way to make it works w/ Helm layer? | 03:19 |
*** diablo_rojo has joined #openstack-kolla | 03:20 | |
sdake_ | portdirect https://blueprints.launchpad.net/kolla-kubernetes/+spec/kubeadm-quickstart is that done? | 03:20 |
sdake_ | portdirect i think it is | 03:20 |
*** jtriley has quit IRC | 03:20 | |
sdake_ | duonghq so next layer is service layer | 03:21 |
sdake_ | duonghq i think portdirect and sbezverk_ were goin gto work on that | 03:21 |
duonghq | sdake_, you mean the layer 4 in Ryan specs? Executes Helm packages for a service. | 03:22 |
sdake_ | duonghq moment | 03:22 |
*** v1k0d3n has joined #openstack-kolla | 03:23 | |
sdake_ | duonghq we are still on layer 3 | 03:25 |
sdake_ | duonghq layer 3 is broken into 3 parts - openstack, service, microservice | 03:25 |
*** zhubingbing has quit IRC | 03:26 | |
*** zhubingbing has joined #openstack-kolla | 03:26 | |
duonghq | sdake_, okay, so this layer is the last layer in layer 3? | 03:28 |
*** unicell has joined #openstack-kolla | 03:28 | |
sdake_ | this is middle layer in layer 3 | 03:28 |
sdake_ | sbezverk_ you about | 03:28 |
duonghq | sdake_, I'm in some kind of miss direction w/ naming things here | 03:28 |
sdake_ | sbezverk_ this needs an approval - i can't do it myself: https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-services | 03:28 |
sdake_ | oops mispellled | 03:29 |
sdake_ | moment | 03:29 |
sdake_ | nah spelled right | 03:29 |
sdake_ | will need another cor eto approve that - in meantime needs all services filled out in work items | 03:29 |
duonghq | kfox1111, is the person? | 03:30 |
sdake_ | duonghq you can probably cut and paste the microservices layer work items to service work items | 03:30 |
sdake_ | duonghq any core can approve | 03:30 |
sdake_ | duonghq lets assume it will be approved ;) | 03:30 |
duonghq | you mean copy-paste? | 03:30 |
sdake_ | ya copy and paste the microservices layer workitems to services workitems | 03:30 |
duonghq | awful network here, I'm still not go out of all lucks till I can still talk w/ you | 03:31 |
sdake_ | huh? | 03:31 |
duonghq | Lag 1.5s (hexchat tell me that) | 03:31 |
sdake_ | C&P work items from microservices to services blueprint above | 03:31 |
sdake_ | set all to unassigned | 03:31 |
sdake_ | people can self-assign | 03:31 |
duonghq | ah, I mean it's such a long waiting time for the page is loaded | 03:32 |
sdake_ | oh gotcha | 03:32 |
duonghq | moment, hope that it's ok in about 1mins | 03:32 |
sdake_ | ya not in a rush besides need to hit the rack :) | 03:32 |
duonghq | guess that we haven't had any ps merged in this layer? | 03:32 |
*** unicell has quit IRC | 03:32 | |
sdake_ | no work has been done on this layer | 03:32 |
duonghq | okay | 03:33 |
sdake_ | one way to think of the layer 3 is 3.3, 3.2, 3.1 | 03:33 |
sdake_ | where 3.1 = microservice layer | 03:33 |
openstackgerrit | Merged openstack/kolla-ansible: Implement octavia ansible role https://review.openstack.org/404648 | 03:33 |
sdake_ | 3.2 = service alyer | 03:33 |
sdake_ | 3.3 = meta layer | 03:34 |
*** zhurong has quit IRC | 03:34 | |
sdake_ | zhubingbing you about | 03:34 |
zhubingbing | i am hear | 03:34 |
sdake_ | https://review.openstack.org/#/c/409965/ | 03:34 |
sdake_ | can you use your reviewing prowess to take a look at that pls :) | 03:34 |
zhubingbing | doning | 03:35 |
zhubingbing | ok | 03:35 |
duonghq | sdake_, bp workitems is updated | 03:35 |
sdake_ | mliima you about? | 03:35 |
duonghq | sdake_, some days or weeks ago, you told that we move from 1 sublayer to other sublayer in layer 3, | 03:36 |
zhubingbing | i test it now | 03:36 |
duonghq | helm packing to helm microservice? | 03:36 |
sdake_ | duonghq i dont understand the q | 03:37 |
duonghq | last time I talk with you, you told that we was moving from helm packing bp to helm microservice bp | 03:37 |
duonghq | iirc | 03:37 |
*** unicell has joined #openstack-kolla | 03:38 | |
sdake_ | right let me fix that up in teh tracker moment | 03:38 |
duonghq | zhubingbing, can you make it role https://review.openstack.org/#/c/408981/, quite simple ps but hang around for quite a long time | 03:38 |
zhubingbing | this link not Not Found | 03:39 |
zhubingbing | this link not found | 03:39 |
duonghq | 2 more ps: https://review.openstack.org/#/c/399707/ https://review.openstack.org/#/c/407815/ | 03:40 |
sdake_ | Jeffrey4l are you able to approve https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-services | 03:42 |
*** unicell has quit IRC | 03:42 | |
Jeffrey4l | seems no. i am not in the kolla-kubernetes-drivers team. sdake_ | 03:43 |
*** harlowja has quit IRC | 03:43 | |
sdake_ | hmm well anyone is welcome in the drivers group that contributes to the project especially the proxy-ptl :) | 03:43 |
sdake_ | Jeffrey4l try now | 03:44 |
Jeffrey4l | sdake_, done. thanks. | 03:44 |
sdake_ | https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-meta | 03:44 |
sdake_ | another one plz | 03:44 |
duonghq | I'm brb in about 2mins | 03:45 |
sdake_ | Jeffrey4l click approver->pick me plz | 03:45 |
duonghq | I'm back | 03:48 |
Jeffrey4l | done | 03:48 |
duonghq | zhubingbing, I can access all 3 links, | 03:48 |
duonghq | zhubingbing, can you recheck | 03:49 |
zhubingbing | ok | 03:51 |
zhubingbing | doning | 03:51 |
zhubingbing | i testing zookeeper | 03:51 |
duonghq | zhubingbing, thanks, don't want this kind of conflict run into merge conflict | 03:51 |
duonghq | *kind of ps | 03:51 |
zhubingbing | i know | 03:52 |
sdake_ | duonghq https://blueprints.launchpad.net/kolla-kubernetes/+spec/helm-meta | 03:55 |
sdake_ | check out the dependecy tree for more clarification | 03:55 |
*** mliima_ has joined #openstack-kolla | 03:55 | |
duonghq | sdake_, thanks | 03:55 |
sdake_ | how run into a merge conflict duonghq ? | 03:57 |
duonghq | sdake_, many ps hang up in review queue enough long will take a good chance to conflict with something, | 03:57 |
sdake_ | lol | 03:58 |
sdake_ | no dobut bout it | 03:58 |
*** Jeffrey4l has quit IRC | 03:58 | |
*** mliima has quit IRC | 03:58 | |
openstackgerrit | Merged openstack/kolla-ansible: Optimize the code https://review.openstack.org/407815 | 03:58 |
duonghq | and this kind of simple ps make the review queue too long no necessary | 03:58 |
*** diogogmt has quit IRC | 04:00 | |
*** Jeffrey4l has joined #openstack-kolla | 04:03 | |
openstackgerrit | Merged openstack/kolla-ansible: Horizon custom policies https://review.openstack.org/399707 | 04:04 |
sdake_ | does this make more sense now duonghq ? https://launchpad.net/kolla-kubernetes/+milestone/0.4.0 | 04:09 |
duonghq | sdake_, yup, thanks | 04:10 |
sdake_ | check out the 0.5.0 milestone while your about it | 04:10 |
duonghq | sdake_, seen, | 04:11 |
duonghq | sdake_, the service mean OpenStack service like Nova, Neutron...? | 04:12 |
sdake_ | roger | 04:12 |
duonghq | how about the meta layer? | 04:12 |
sdake_ | helm install openstack | 04:13 |
duonghq | the entrypoint for end-user? | 04:13 |
sdake_ | all of them are entrypoints for the operator | 04:13 |
*** yingjun has quit IRC | 04:13 | |
duonghq | ok | 04:13 |
*** yingjun has joined #openstack-kolla | 04:14 | |
*** williamcaban has quit IRC | 04:15 | |
*** williamcaban has joined #openstack-kolla | 04:16 | |
*** yingjun has quit IRC | 04:18 | |
sdake_ | i'll probably be receiving corosync notifications until i'm dead from github | 04:20 |
*** williamcaban has quit IRC | 04:20 | |
*** harlowja has joined #openstack-kolla | 04:21 | |
sdake_ | harlowja sup dude | 04:22 |
harlowja | sup sup | 04:22 |
harlowja | just at home chilling, u | 04:22 |
sdake_ | beating launchpad into submission | 04:22 |
harlowja | fun fun, ha | 04:23 |
sdake_ | better then the irs or the dentist or road rage ;) | 04:23 |
harlowja | i spent a bunch of time today trying to rip apart a coredump file (from uwsgi) | 04:23 |
*** JRobinson__ is now known as JRobinson__afk | 04:23 | |
harlowja | it is really a pain in the ass to figure out what is in its memory :-P | 04:23 |
sdake_ | gdb ftw | 04:23 |
harlowja | it was absorbing 5gb for some reason | 04:23 |
harlowja | ya, i tried gdb | 04:23 |
sdake_ | no luck huh? | 04:24 |
harlowja | still couldn't figure out the memory mapping of what tookup the most | 04:24 |
harlowja | cause is it in uwsgi or is it in python | 04:24 |
harlowja | (cause uwsgi connects into python) | 04:24 |
harlowja | but ya, not much luck | 04:24 |
Dishwasha | FYI, ansible dislikes hyphens for hostvars names | 04:25 |
sdake_ | why uswgi isntead of apache? | 04:25 |
harlowja | ya, i asked that :-P | 04:25 |
harlowja | i forget the answer though | 04:26 |
harlowja | i'm just gonna push for apache (since its default in kolla) | 04:26 |
harlowja | i think klindgren_ is ok with that | 04:26 |
sdake_ | harlowja probably better to forget ;) | 04:26 |
harlowja | but uwsgi was before my time, perhaps it does something good, not really sure, lol | 04:26 |
harlowja | sdake_ ya | 04:26 |
harlowja | lol | 04:26 |
harlowja | we put haproxy infront of it anyway | 04:27 |
sdake_ | ya kolla does that already | 04:27 |
sdake_ | but you may not be doing that in your ver | 04:27 |
harlowja | klindgren_ is figuring out that all, ha (more of the ansible side of things) | 04:27 |
sdake_ | kolla does that mainly to provide ha and tls | 04:28 |
harlowja | ya, sounds famiailr/similar | 04:28 |
sdake_ | klindgren_ can probably just use the ansible haproxy config file | 04:28 |
sdake_ | and be good to go for the most part | 04:28 |
harlowja | ya | 04:28 |
sdake_ | ok i must P T F O | 04:28 |
sdake_ | night all | 04:28 |
harlowja | peace | 04:29 |
*** Pavo has quit IRC | 04:29 | |
*** elynn has quit IRC | 04:29 | |
duonghq | night sdake_ | 04:31 |
*** sdake_ has quit IRC | 04:31 | |
*** yuanying has joined #openstack-kolla | 04:33 | |
*** Pavo has joined #openstack-kolla | 04:33 | |
*** yingjun has joined #openstack-kolla | 04:37 | |
*** unicell has joined #openstack-kolla | 04:38 | |
*** adrian_otto has joined #openstack-kolla | 04:41 | |
*** v1k0d3n has quit IRC | 04:41 | |
*** unicell has quit IRC | 04:43 | |
*** gomarivera has joined #openstack-kolla | 04:44 | |
*** HyperJohnGraham has joined #openstack-kolla | 04:46 | |
*** Dishwasha has quit IRC | 04:48 | |
*** elynn has joined #openstack-kolla | 04:49 | |
*** unicell has joined #openstack-kolla | 04:50 | |
*** JRobinson__afk is now known as JRobinson__ | 04:51 | |
*** elynn has quit IRC | 04:54 | |
*** elynn has joined #openstack-kolla | 04:54 | |
*** sdake has joined #openstack-kolla | 05:00 | |
*** sdake_ has joined #openstack-kolla | 05:03 | |
*** sdake has quit IRC | 05:05 | |
*** sp_ has quit IRC | 05:07 | |
*** yatin_on_leave has quit IRC | 05:07 | |
*** sdake_ has quit IRC | 05:07 | |
*** harlowja has quit IRC | 05:08 | |
*** mdnadeem has joined #openstack-kolla | 05:18 | |
*** krtaylor has quit IRC | 05:19 | |
*** sayantan_ has quit IRC | 05:19 | |
*** sayantan_ has joined #openstack-kolla | 05:19 | |
*** sp_ has joined #openstack-kolla | 05:20 | |
*** yatin_on_leave has joined #openstack-kolla | 05:21 | |
openstackgerrit | zhubingbing proposed openstack/kolla-ansible: Fix aodh upgrade permission issue https://review.openstack.org/410046 | 05:22 |
*** fragatina has quit IRC | 05:26 | |
*** fragatina has joined #openstack-kolla | 05:26 | |
*** sp_ has quit IRC | 05:27 | |
*** yatin_on_leave has quit IRC | 05:28 | |
*** unicell has quit IRC | 05:29 | |
*** unicell has joined #openstack-kolla | 05:29 | |
*** yingjun has quit IRC | 05:32 | |
*** yingjun has joined #openstack-kolla | 05:32 | |
*** yingjun_ has joined #openstack-kolla | 05:33 | |
*** yingjun has quit IRC | 05:33 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm Keystone K8s Services https://review.openstack.org/408963 | 05:38 |
*** adrian_otto has quit IRC | 05:39 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm Keystone K8s Services https://review.openstack.org/408963 | 05:39 |
*** yatin_on_leave has joined #openstack-kolla | 05:40 | |
cliles | anyone know the uid nova/libvirt tries to access volumes as over nfs? | 05:41 |
*** sp_ has joined #openstack-kolla | 05:41 | |
*** adrian_otto has joined #openstack-kolla | 05:41 | |
cliles | I keep getting permission denied, trying to figure out what to add to the nfs acl.... | 05:41 |
jascott1_ | no sorry | 05:45 |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helmize Heat-engine https://review.openstack.org/410048 | 05:46 |
*** huzhengchuan has quit IRC | 05:47 | |
cliles | it was the qem uid I was missing | 05:48 |
cliles | cinder, nova, and qemu | 05:48 |
cliles | :) | 05:48 |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helmize Heat-engine https://review.openstack.org/410048 | 05:48 |
*** yatin_on_leave has quit IRC | 05:50 | |
*** huzhengchuan has joined #openstack-kolla | 05:50 | |
*** harlowja has joined #openstack-kolla | 05:56 | |
*** sp_ has quit IRC | 05:57 | |
portdirect | anyone know what could have caused this? http://logs.openstack.org/63/408963/4/check/gate-kolla-kubernetes-deploy-centos-binary-external-ovs-nv/86e1cc8/console.html | 06:00 |
*** yingjun_ has quit IRC | 06:00 | |
*** hieulq has quit IRC | 06:01 | |
*** tovin07_ has quit IRC | 06:01 | |
*** duonghq has quit IRC | 06:01 | |
*** tovin07_ has joined #openstack-kolla | 06:01 | |
*** hieulq has joined #openstack-kolla | 06:01 | |
*** duonghq has joined #openstack-kolla | 06:01 | |
*** diablo_rojo has quit IRC | 06:02 | |
*** yatin_on_leave has joined #openstack-kolla | 06:03 | |
jascott1_ | portdirect weird | 06:05 |
jascott1_ | did you recheck yet? | 06:05 |
portdirect | kolla-ansible broken? | 06:05 |
portdirect | yeah | 06:05 |
jascott1_ | oh ok | 06:05 |
jascott1_ | hmm | 06:05 |
portdirect | I see you're getting the same :/ | 06:06 |
*** tovin07 has quit IRC | 06:06 | |
*** tovin07__ has joined #openstack-kolla | 06:06 | |
jascott1_ | ansible folder changed 2 hrs aho | 06:06 |
jascott1_ | ago | 06:06 |
portdirect | looks like our gate doesn't like it as its that change thats thowing the error? | 06:07 |
jascott1_ | yup its horizon | 06:07 |
jascott1_ | think so anyway | 06:07 |
jascott1_ | that 2hr change is horizon folder, still looking | 06:07 |
portdirect | i'll modify my ps to use an older checkout and see how that goes | 06:07 |
jascott1_ | maybe this? https://github.com/openstack/kolla-ansible/commit/199015ce5d0986ca2c22fa26afce7dd0ac50e875 | 06:08 |
*** JRobinson__ has quit IRC | 06:09 | |
portdirect | thats the one i think | 06:10 |
portdirect | got it: https://github.com/openstack/kolla-ansible/commit/199015ce5d0986ca2c22fa26afce7dd0ac50e875#diff-dc672b240356486d7af4d616ca5976c8R47 | 06:10 |
portdirect | enabled" | 06:10 |
portdirect | same at line 69 | 06:10 |
jascott1_ | yup | 06:10 |
jascott1_ | good verk! | 06:11 |
jascott1_ | haha i was staring at it | 06:11 |
jascott1_ | how did that make it? | 06:11 |
portdirect | easy done - weird how the ansible gate doesn't care, its like a honey badger or something | 06:11 |
jascott1_ | lol | 06:11 |
*** mdnadeem has quit IRC | 06:11 | |
*** yuanying has quit IRC | 06:12 | |
*** yuanying has joined #openstack-kolla | 06:12 | |
*** yuanying has quit IRC | 06:12 | |
jascott1_ | ansible: "idk wtf happened. do you wanna retry or not? whatevs." | 06:12 |
*** sp_ has joined #openstack-kolla | 06:13 | |
*** yuanying has joined #openstack-kolla | 06:13 | |
jascott1_ | anyone around can fix? | 06:13 |
*** mdnadeem has joined #openstack-kolla | 06:13 | |
jascott1_ | sorry imeant any cores around that can fix this broken kolla-ansible merge? | 06:14 |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm Keystone K8s Services https://review.openstack.org/408963 | 06:15 |
*** yuanying has quit IRC | 06:17 | |
*** gomarivera has quit IRC | 06:17 | |
*** yatin_on_leave has quit IRC | 06:27 | |
jascott1_ | zhubingbing, pbourke ? | 06:28 |
zhubingbing | i am hear | 06:29 |
zhubingbing | here | 06:29 |
*** Pavo has quit IRC | 06:29 | |
zhubingbing | what's matter | 06:29 |
jascott1_ | hi | 06:29 |
jascott1_ | so looks like kolla-ansible latest commit has some wrong chars and is breaking gate | 06:29 |
zhubingbing | hi | 06:29 |
jascott1_ | for kolla-k8s | 06:30 |
openstackgerrit | Pete Birley proposed openstack/kolla-ansible: BugFix: Horizon custom policies https://review.openstack.org/410056 | 06:30 |
jascott1_ | heres the broken gate for kolla-k8s http://logs.openstack.org/63/408963/4/check/gate-kolla-kubernetes-deploy-centos-binary-external-ovs-nv/86e1cc8/console.html | 06:30 |
portdirect | zhubingbing: the ps I just posted should hopefully fix it | 06:31 |
jascott1_ | and here is the problem commit in kolla-ansible | 06:31 |
jascott1_ | https://github.com/openstack/kolla-ansible/commit/199015ce5d0986ca2c22fa26afce7dd0ac50e875#diff-dc672b240356486d7af4d616ca5976c8R47 | 06:31 |
zhubingbing | ok | 06:31 |
zhubingbing | good | 06:31 |
jascott1_ | portdirect awesome! | 06:31 |
*** tovin07_ is now known as tovin0 | 06:32 | |
*** tovin0 is now known as tovin07 | 06:32 | |
Jeffrey4l | we need enable horizon in gate~ | 06:33 |
*** gomarivera has joined #openstack-kolla | 06:33 | |
portdirect | Jeffrey4l, is it not currently for kolla-ansible? | 06:33 |
Jeffrey4l | portdirect, yes. it is remove for some reason long time ago. | 06:34 |
*** Pavo has joined #openstack-kolla | 06:34 | |
*** tovin07__ has quit IRC | 06:34 | |
*** duonghq is now known as duonghq_ | 06:34 | |
*** tovin07__ has joined #openstack-kolla | 06:35 | |
*** tovin07__ has quit IRC | 06:35 | |
*** tovin07__ has joined #openstack-kolla | 06:35 | |
portdirect | ach well - if that ps I've put up goes threw your gate will one of you be able to WF it? can work around it in the mean timeso no worries :) | 06:36 |
zhubingbing | jascott1 i fix it now | 06:37 |
jascott1_ | great thank you | 06:37 |
portdirect | zhubingbing: https://review.openstack.org/#/c/410056/ | 06:37 |
zhubingbing | <portdirect> good commit | 06:38 |
zhubingbing | i review it | 06:38 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Enable horizon deployment in gate https://review.openstack.org/410057 | 06:38 |
portdirect | np | 06:38 |
openstackgerrit | Kuo-tung Kao proposed openstack/kolla: fix typo in extend_start in collectd docker https://review.openstack.org/410058 | 06:38 |
zhubingbing | done | 06:39 |
zhubingbing | done portdirect | 06:39 |
portdirect | zhubingbing: i dont follow? | 06:39 |
Jeffrey4l | portdirect, remove the implement bp or closes bug. | 06:39 |
Jeffrey4l | we do not use pb and bug at the same time. | 06:39 |
portdirect | whoops! i see it :) | 06:39 |
zhubingbing | you miss ":" | 06:39 |
zhubingbing | yes | 06:40 |
zhubingbing | and you should add a bug | 06:40 |
portdirect | shows how easy it is :) | 06:40 |
openstackgerrit | Pete Birley proposed openstack/kolla-ansible: BugFix: Horizon custom policies https://review.openstack.org/410056 | 06:41 |
*** yatin_on_leave has joined #openstack-kolla | 06:41 | |
*** tovin07__ is now known as tovin07_afk | 06:42 | |
*** adrian_otto has quit IRC | 06:42 | |
*** duonghq_ is now known as duonghq__ | 06:43 | |
*** duonghq__ is now known as duonghq_away | 06:44 | |
zhubingbing | portdirect good boy:) thank your commit | 06:44 |
*** duonghq_away is now known as duonghq | 06:44 | |
portdirect | no problem - thanks dude :) | 06:45 |
jascott1_ | thanks team kolla | 06:45 |
portdirect | we should make a film | 06:46 |
*** yingjun has joined #openstack-kolla | 06:46 | |
jascott1_ | slo mo walking away as the gate explodes? | 06:46 |
jascott1_ | now that yo mention it | 06:46 |
jascott1_ | maybe we take for granted all the cooperation | 06:47 |
jascott1_ | non tech people might get something out of seeing how the opensource is made | 06:47 |
jascott1_ | and if not they can go to sleep ;) | 06:47 |
openstackgerrit | Pete Birley proposed openstack/kolla-ansible: Fix syntax error in horizon custom policies file https://review.openstack.org/410056 | 06:48 |
jascott1_ | some of the people i know with more physical jobs would probably say " you chat on the internet all day for your job?" | 06:48 |
jascott1_ | to which I reply "and all night" | 06:49 |
*** tovin07_afk has quit IRC | 06:50 | |
duonghq | Jeffrey4l, sorry but why we need Horizon in gate? | 06:52 |
Jeffrey4l | duonghq, it ensure the horizon deploy role work. | 06:53 |
duonghq | Jeffrey4l, hmm, ok | 06:53 |
Jeffrey4l | tbh, i want enable all service in gate. ( but the gate vm is not that big enough, need some workaround for this ) | 06:54 |
duonghq | sure, I'm thinking on this too | 06:55 |
duonghq | wonder why we have many service but only some enable | 06:55 |
portdirect | kollabot help | 06:55 |
*** kollabot has quit IRC | 06:55 | |
*** kollabot has joined #openstack-kolla | 06:56 | |
Jeffrey4l | the size for gate vm is limited. we can not get any larger vm. | 06:56 |
duonghq | roger | 06:56 |
duonghq | Can we enable some of these on some tests | 06:56 |
portdirect | kollabot remind to jascott1_ in 5s that robots are superior to biologicals | 06:57 |
jascott1_ | that one works but needs a little tweaking iirc | 06:57 |
portdirect | :) | 06:57 |
Jeffrey4l | duonghq, what's that mean? | 06:57 |
jascott1_ | i think if I rejoin it works | 06:57 |
jascott1_ | so kinda dud | 06:57 |
jascott1_ | oh you said 5m | 06:58 |
duonghq | I mean if we need test project A, we enable it and test in this test | 06:58 |
duonghq | only this test | 06:58 |
jascott1_ | portdirect idk if wolfram is listed in help? | 06:58 |
jascott1_ | kollabot wfa how many bytes in a gigabyte? | 06:59 |
kollabot | Input interpretation convert 1 GB (gigabyte) to bytes | 06:59 |
kollabot | Result 1×10^9 bytes | 06:59 |
kollabot | undefined 0.9313 GiB (gibibytes) | 06:59 |
kollabot | undefined ~~ 1.5 × standard compact disc capacity ( 650 to 703 MB ) | 06:59 |
Jeffrey4l | duonghq, possible. no every devs know how it works. | 06:59 |
Jeffrey4l | duonghq, another solution is set up some test scenario. | 06:59 |
duonghq | ya, yes, I mean the scenario | 06:59 |
Jeffrey4l | like scenario 1: test default profiles. | 06:59 |
duonghq | just forgot the word | 07:00 |
Jeffrey4l | scenario 2: test heat + murano + senlin | 07:00 |
Jeffrey4l | etc | 07:00 |
Jeffrey4l | hope this can be done in this cycle ;) | 07:00 |
duonghq | ya | 07:00 |
duonghq | Jeffrey4l, can you draft a bp/ | 07:00 |
duonghq | iirc we have not had any bp like this | 07:01 |
Jeffrey4l | np. | 07:01 |
Jeffrey4l | this is talked in the summit. | 07:01 |
Jeffrey4l | i will register one bp for this. | 07:01 |
duonghq | Jeffrey4l, do you link the etherpad of bp summarize? | 07:01 |
duonghq | *have link | 07:01 |
Jeffrey4l | duonghq, https://blueprints.launchpad.net/kolla-ansible/+spec/functional-gating | 07:02 |
*** saneax-_-|AFK is now known as saneax | 07:02 | |
Jeffrey4l | it is registered already. | 07:02 |
Jeffrey4l | this is the roadmap which is talked in the summit. | 07:02 |
Jeffrey4l | https://etherpad.openstack.org/p/kolla-o-summit-roadmap | 07:03 |
duonghq | Jeffrey4l, maybe we need copy some of these to work items? | 07:04 |
Jeffrey4l | from the roadmap page? | 07:04 |
duonghq | but the etherpad does not have split scenarios? | 07:04 |
duonghq | from functional gating | 07:05 |
*** fragatin_ has joined #openstack-kolla | 07:05 | |
Jeffrey4l | duonghq, yep. this is just talked. but without any further detail how to split the service into different scenarios. | 07:06 |
duonghq | ah, ok | 07:06 |
duonghq | seems that I missed many things :( | 07:06 |
*** fragatin_ has quit IRC | 07:06 | |
*** fragatin_ has joined #openstack-kolla | 07:07 | |
Jeffrey4l | ah, not that much. but you'd better review the all the etherpad wrote in the summit. ;) | 07:07 |
duonghq | do we have meta etherpad? | 07:07 |
Jeffrey4l | i have some idea for the split scenarios. will write something to explain it later. | 07:07 |
Jeffrey4l | yep | 07:08 |
duonghq | can you give a the link? | 07:08 |
*** fragatina has quit IRC | 07:08 | |
duonghq | or it's roadmap? | 07:08 |
Jeffrey4l | duonghq, check this https://wiki.openstack.org/wiki/Design_Summit/Newton/Etherpads#Kolla | 07:08 |
Jeffrey4l | hrm. wrong link. | 07:09 |
Jeffrey4l | this is newton summit. | 07:09 |
Jeffrey4l | duonghq, this one https://wiki.openstack.org/wiki/Design_Summit/Ocata/Etherpads#Kolla | 07:09 |
duonghq | ah, thank you | 07:09 |
*** imcsk8_ has quit IRC | 07:09 | |
*** imcsk8_ has joined #openstack-kolla | 07:09 | |
*** tovin07 has quit IRC | 07:10 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm Keystone K8s Services https://review.openstack.org/408963 | 07:10 |
*** HyperJohnGraham has quit IRC | 07:10 | |
*** tovin07_ has joined #openstack-kolla | 07:10 | |
*** tovin07__ has joined #openstack-kolla | 07:11 | |
*** fragatin_ has quit IRC | 07:11 | |
*** unicell has quit IRC | 07:13 | |
*** unicell1 has joined #openstack-kolla | 07:13 | |
SamYaple | Jeffrey4l: was there any work in progress on keep a set uid/gid for the container user? | 07:14 |
Jeffrey4l | SamYaple, yep. there is a spec WIP. | 07:14 |
SamYaple | do you think it will be implemented in Ocata? | 07:14 |
Jeffrey4l | SamYaple, https://review.openstack.org/405647 | 07:14 |
*** sayanta__ has joined #openstack-kolla | 07:15 | |
Jeffrey4l | SamYaple, yep. this is not that complex. I and klindgren have reached some agreement on how to implement this. | 07:15 |
portdirect | Jeffrey4l: if we could use the same UID/GID that rdo uses by default it would be great from my pov | 07:16 |
Jeffrey4l | btw i reproduced can not reach the vms issue during upgrade. debugging why. | 07:16 |
Jeffrey4l | portdirect, rdo will not re-create the user if it already exist. and how could we handle oraclelinux and ubuntu? | 07:17 |
Jeffrey4l | so answer should be on. i prefer to use 10000+ id, which is never used automatically by OS. | 07:17 |
SamYaple | Jeffrey4l: a very high id like that will make userns mapping difficult/impossible in the future | 07:18 |
*** sayantan_ has quit IRC | 07:18 | |
portdirect | I belive that oracle uses the same .spec as rdo - it would make it esier to intergrate with existing security tooling - i can check to see what ubuntu uses if that would help? | 07:18 |
*** zhubingbing has quit IRC | 07:18 | |
SamYaple | portdirect: rdo uses a static uid/gid? | 07:19 |
*** zhubingbing has joined #openstack-kolla | 07:19 | |
Jeffrey4l | let me check the largest and lowest id the OS use. we should not use that range. | 07:19 |
Jeffrey4l | SamYaple, yep. | 07:19 |
Jeffrey4l | rod use static uid. | 07:19 |
Jeffrey4l | rdo* | 07:19 |
portdirect | SamYaple: https://github.com/rdo-packages/nova-distgit/blob/rpm-master/openstack-nova.spec#L654-L656 | 07:19 |
SamYaple | what happens when the uid/gid is taken? | 07:19 |
Jeffrey4l | if they do not use static uid, they will meet the upgrade issue. | 07:20 |
SamYaple | i understand, but what do they do with conflicts | 07:20 |
Jeffrey4l | SamYaple, sorry? what conflicts? | 07:21 |
SamYaple | when the package is installed on a host that already has that uid taken | 07:21 |
Jeffrey4l | SamYaple, we create the user with specified id before install rdo. | 07:21 |
SamYaple | and what if the system already has a user with that uid? | 07:22 |
portdirect | yeah - SamYaple that link i posted shows exactly what they do | 07:22 |
Jeffrey4l | SamYaple, normally the os create user in a range. | 07:22 |
SamYaple | portdirect: i thought i was just reading it wrong. they coopt the user?! | 07:23 |
Jeffrey4l | in docker image, i do not any user will create with specified id. | 07:23 |
portdirect | ... | 07:23 |
portdirect | yup | 07:23 |
SamYaple | wth | 07:23 |
*** Jeffrey4l has quit IRC | 07:24 | |
SamYaple | Jeffrey4l: we still have to take the host into account of mapping of uids. bindmounts and such | 07:24 |
SamYaple | i think going with rdo mappings is fine | 07:25 |
*** harlowja has quit IRC | 07:25 | |
*** Jeffrey4l has joined #openstack-kolla | 07:25 | |
*** gomarivera has quit IRC | 07:26 | |
Jeffrey4l | sorry, network issue. offlined just now. | 07:26 |
SamYaple | portdirect: im checking now if ubuntu uses static as well | 07:27 |
Jeffrey4l | did i miss anything? SamYaple ? ( offlined just now ) | 07:27 |
SamYaple | Jeffrey4l: because of host to container bind mounts, we still have to consider the host uid/gid | 07:28 |
portdirect | SamYaple: nice, thanks - I'm not familiar with that world really :( | 07:28 |
SamYaple | not all deployment projects will follow kolla-ansible with "don't touch the host" | 07:28 |
Jeffrey4l | SamYaple, why? we do not care the uid right now. it works well expect during upgrade. | 07:28 |
SamYaple | kolla-ansible does not care | 07:28 |
SamYaple | i certainly care | 07:28 |
Jeffrey4l | SamYaple whats your expectation? | 07:29 |
portdirect | +1 | 07:29 |
portdirect | to me able to migrate to kolla with mimial effort regarless of how the conatiners are deployed | 07:29 |
SamYaple | Jeffrey4l: static mappings, low as possible. perhaps matching rdo (and ubuntu whatever they are) | 07:29 |
portdirect | *to be | 07:29 |
SamYaple | portdirect: yea did you see my kolla-salt ether pad? | 07:29 |
Jeffrey4l | in the spec, you can override the user/id mapping by add some configuration to use your own during build stage. | 07:29 |
portdirect | no! | 07:29 |
portdirect | where | 07:29 |
Jeffrey4l | is this work for you? SamYaple | 07:30 |
SamYaple | portdirect: config files from the host, from the default locations _on_ the host, bind into containers | 07:30 |
Jeffrey4l | so you can use whatever mapping you want. | 07:30 |
SamYaple | portdirect: should be able to do a bare-metal to contaienr conversion in place with kolla-salt | 07:30 |
SamYaple | Jeffrey4l: this isn't for me alone. im talking about official kolla images | 07:30 |
Jeffrey4l | but in kolla-ansible's implementation, we use a static pre-configuration user/id mapping. | 07:30 |
SamYaple | Jeffrey4l: at this point, everyone wants static uid/gid, we just need to define _what_ they are | 07:31 |
SamYaple | portdirect: https://etherpad.openstack.org/p/kolla-salt | 07:31 |
Jeffrey4l | so what u expect? SamYaple | 07:31 |
SamYaple | 07:30:56 < SamYaple> Jeffrey4l: static mappings, low as possible. perhaps matching rdo (and ubuntu whatever they are) | 07:31 |
Jeffrey4l | SamYaple, so how about the source installation? | 07:32 |
portdirect | use the same mappings ideally | 07:32 |
SamYaple | what about it? | 07:32 |
Jeffrey4l | and remember, some infra service do not have static user/id mapping. | 07:32 |
SamYaple | yea same mappings | 07:32 |
Jeffrey4l | like rabbitmq . | 07:32 |
SamYaple | right but we are building the container | 07:32 |
SamYaple | no one is asking for random mappings | 07:32 |
SamYaple | where no "standard" exists, a mapping we pick is fine | 07:33 |
Jeffrey4l | so we will use static use/id mapping. then another q is: why we need follow rdo and cloud-archive magic number? | 07:33 |
SamYaple | for consistency? | 07:34 |
Jeffrey4l | some there are some openstack service haven't rdo package. | 07:34 |
SamYaple | and potentially to take a baremetal rdo deploy, and convert to a kolla deploy | 07:34 |
Jeffrey4l | how to handle this? | 07:34 |
SamYaple | 07:34:33 < SamYaple> where no "standard" exists, a mapping we pick is fine | 07:34 |
Jeffrey4l | this will make world confused. | 07:34 |
portdirect | Jeffrey4l: i think creating a new standard would do that? | 07:35 |
SamYaple | your arguement is because some mapping at some point in the future may not work, don't even try | 07:35 |
SamYaple | mine is, work with the community for consistency | 07:35 |
portdirect | +1 | 07:35 |
SamYaple | differences may occur, but we can solve them when they arise | 07:35 |
openstackgerrit | Merged openstack/kolla-ansible: Fix syntax error in horizon custom policies file https://review.openstack.org/410056 | 07:36 |
Jeffrey4l | imaging: rdo may not provide senlin package ( for example ), so we choose 10001 as user id. | 07:36 |
Jeffrey4l | what happend when rdo provide senlin package with 155 as user id? | 07:36 |
Jeffrey4l | SamYaple, ^^ | 07:36 |
SamYaple | as i said, we can deal with it then. if that means a migration of ownership during an upgrade, so be it | 07:37 |
SamYaple | or we can choose 155 and ask rdo to follow | 07:37 |
SamYaple | thats where teh "working with the community" part comes in | 07:37 |
SamYaple | we all communicate | 07:37 |
Jeffrey4l | i will not have upgrade/migrate ownership script after fix uid bp is finished. | 07:37 |
Jeffrey4l | if we have such script, we do not need fix uid at all. | 07:38 |
SamYaple | now thats just silly statement to make | 07:38 |
Jeffrey4l | make file owner migration is hard in some cause. | 07:38 |
Jeffrey4l | for example nova case. | 07:38 |
SamYaple | there is nothing wrong with migration scripts between versions for _whatever_ reason | 07:38 |
Jeffrey4l | not all file in /var/lib/nova is owned by nova:nova. | 07:38 |
SamYaple | no. we abolsultly still need migration scripts | 07:39 |
SamYaple | what if someone deploys kolla images and needs to change the uid/gid | 07:39 |
Jeffrey4l | ok i do not agree on that point ;) | 07:39 |
SamYaple | what if they use native kolla images and have to build thier own | 07:39 |
SamYaple | having migration is no problem | 07:39 |
Jeffrey4l | migration is hard and need many test, which kolla can not cover. | 07:40 |
SamYaple | and Jeffrey4l what if systemd reserves some uid/gid for some reason and we need to move it? | 07:40 |
SamYaple | or they use it to own some /sys files? | 07:40 |
SamYaple | then you have a security risk | 07:40 |
Jeffrey4l | we do not prevent the user/developer use its own uid mapping. but you need know what are u doing. becuase kolla do not provide the migrate script. | 07:41 |
SamYaple | "i will not have upgrade/migrate ownership script" is short sighted | 07:41 |
Jeffrey4l | SamYaple, the system use the uid in a range. ( let me find ) | 07:41 |
SamYaple | well whatever the case is Jeffrey4l, if the need arises to migrate permissions, of course we are going to | 07:42 |
*** sacharya has quit IRC | 07:42 | |
Jeffrey4l | SamYaple, yep. | 07:42 |
SamYaple | so all of this is beside the point | 07:42 |
Jeffrey4l | we can talk this with the team in next meeting. will u be there? | 07:42 |
SamYaple | i dont even know what time the meetings are anymore | 07:43 |
Jeffrey4l | SamYaple, in the wiki page https://wiki.openstack.org/wiki/Meetings/Kolla | 07:43 |
portdirect | 16:00 UTC wed | 07:43 |
Jeffrey4l | MEETING TIME: Wednesdays at 16:00 UTC #openstack-meeting-4. | 07:44 |
Jeffrey4l | anyway, the uid migration script will be there for at least one cycle. even if the fixed uid is done. | 07:44 |
Jeffrey4l | so we can talk this later ;) | 07:44 |
Jeffrey4l | back to the uid choose strategy. following the rdo is not a good idea. | 07:46 |
SamYaple | strongly disagree | 07:46 |
SamYaple | following any current standard is better than making our own | 07:46 |
Jeffrey4l | i meant in my example: imaging xxx service is not provide by rdo. we implement this, so we pick on by random like 10001 | 07:47 |
Jeffrey4l | after 1 year, rdo provide xxx service finally. | 07:47 |
Jeffrey4l | with uid 146, | 07:47 |
SamYaple | well for one, we should talk to rdo first and ask to agree on a uid | 07:47 |
SamYaple | for consistency across communities | 07:47 |
portdirect | exactly | 07:47 |
SamYaple | if we _can't_ agree, then its because someone is being stubborn | 07:48 |
Jeffrey4l | it is hard tbh. imaging after talked with rdo, we choosed 146. | 07:48 |
SamYaple | actually it would be because both peopel are being stubborn | 07:48 |
Jeffrey4l | but in harf year, service yyy choose 146 first. | 07:48 |
Jeffrey4l | because no one will ensure this. | 07:49 |
Jeffrey4l | no gate no cross ci. | 07:49 |
Jeffrey4l | just a agreement is not reliable. | 07:49 |
portdirect | Jeffrey4l: remember that these projects are very well estabilished, centos uses the same UID/GID for example, we would be the odd ones out | 07:49 |
openstackgerrit | Javier Castillo AlcÃbar proposed openstack/kolla-ansible: Fix ceilometer bootstrap https://review.openstack.org/398190 | 07:49 |
SamYaple | youll never win me over with your arguements here Jeffrey4l. Saying something *might* not work so lets not do it is a bad arguement | 07:50 |
SamYaple | we should work with the communities | 07:50 |
SamYaple | if a uid conflict occurs, we will deal with it then | 07:50 |
SamYaple | *IF* | 07:50 |
*** yuanying has joined #openstack-kolla | 07:51 | |
Jeffrey4l | when dealing with it, it makes world more complicate and confused for the operator. | 07:51 |
SamYaple | well the operator should never know about this according to you | 07:51 |
SamYaple | as for "makes world more complicate and confused" welcome to collaboration | 07:52 |
SamYaple | the idea is the complication is outweighed by the benefits | 07:52 |
Jeffrey4l | anyway, this is worth a talk in the meeting . | 07:54 |
*** mkoderer has joined #openstack-kolla | 07:54 | |
Jeffrey4l | could u leave your idea in the spec? SamYaple | 07:54 |
n0isyn0ise | hi all | 07:56 |
n0isyn0ise | I'm experiencing a problem when building keystone-base image with kolla on stable/newton | 07:57 |
n0isyn0ise | I got the following error : ERROR:kolla.image.build.keystone-base:The command '/bin/sh -c yum -y install httpd mod_wsgi python-ldappool && yum clean all' returned a non-zero code: 1 | 07:57 |
n0isyn0ise | can someone give me hint about how can to solve it please ? | 07:57 |
Jeffrey4l | n0isyn0ise, seem like a network issue? | 07:58 |
Jeffrey4l | n0isyn0ise, did u try one more time? | 07:58 |
n0isyn0ise | hmm that's strange , I don't have error with other container images | 07:58 |
*** unicell has joined #openstack-kolla | 07:58 | |
n0isyn0ise | I tried several times before asking here | 07:59 |
n0isyn0ise | but i'm beginning with kolla | 07:59 |
Jeffrey4l | try run `install httpd mod_wsgi python-ldappool` manually on openstack-base conainer. | 07:59 |
Jeffrey4l | yum -y install httpd mod_wsgi python-ldappool | 07:59 |
*** msimonin has joined #openstack-kolla | 08:00 | |
*** unicell1 has quit IRC | 08:00 | |
*** sacharya has joined #openstack-kolla | 08:00 | |
n0isyn0ise | ok thx Jeffrey, i will try that | 08:00 |
openstackgerrit | zhubingbing proposed openstack/kolla-ansible: WIP add fluentd role https://review.openstack.org/407392 | 08:00 |
*** dupengfei has joined #openstack-kolla | 08:02 | |
*** sacharya has quit IRC | 08:05 | |
Jeffrey4l | SamYaple, still around? found the root cause that vms is lost during upgrade. the vm's nic detached from br-int ( but no idea why). after add the port to the br-int, it works. | 08:05 |
openstackgerrit | dupengfei proposed openstack/kolla: Not to pull the base image when building. https://review.openstack.org/408940 | 08:05 |
SamYaple | Jeffrey4l: are you changing versions of openvswitch? | 08:06 |
Jeffrey4l | SamYaple, absolutely not. | 08:06 |
Jeffrey4l | just upgrade from 2 weeks ago image to the latest. | 08:06 |
SamYaple | does this occur when the openvswitch container upgrades, or when neutron-agent upgrades? | 08:07 |
SamYaple | when does the detach happen | 08:07 |
Jeffrey4l | seem when upgrade dhcp-agent. i am not sure. will try again. | 08:07 |
SamYaple | dhcp-agent shouldn't be touching tap on host at all | 08:08 |
Jeffrey4l | i thinks so. but this is what i saw. will double check again. | 08:08 |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm Keystone K8s Services https://review.openstack.org/408963 | 08:10 |
*** pcaruana has joined #openstack-kolla | 08:10 | |
n0isyn0ise | looking more deeply at my problem with keystone-base container image, I got the following error : INFO:kolla.image.build.keystone-base:error: unpacking of archive failed on file /usr/sbin/suexec: cpio: cap_set_file | 08:19 |
*** matrohon has joined #openstack-kolla | 08:26 | |
*** Pavo has quit IRC | 08:29 | |
*** sayanta__ has quit IRC | 08:30 | |
openstackgerrit | zhubingbing proposed openstack/kolla: add fluent-plugin-grok-parser in fluentd https://review.openstack.org/410089 | 08:31 |
*** egonzalez90 has joined #openstack-kolla | 08:31 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Keystone service endpoint creation (keystone specific) https://review.openstack.org/408843 | 08:34 |
*** Pavo has joined #openstack-kolla | 08:34 | |
*** huzhengchuan has quit IRC | 08:36 | |
*** huzhengchuan has joined #openstack-kolla | 08:38 | |
*** msimonin has quit IRC | 08:43 | |
*** jascott1_ has quit IRC | 08:51 | |
SamYaple | portdirect: Jeffrey4l: no ubuntu/debian static uid/gid mapping | 08:54 |
portdirect | SamYaple: yeah - i found an old (very old!) doc from back in the day that defined uid/gid and noted that - I'll see if i can dig it up again | 08:55 |
SamYaple | portdirect: for ubuntu/debian? | 08:55 |
SamYaple | because i just asked the package maintainers | 08:55 |
portdirect | SamYaple: sorry I explained it badly just then | 08:56 |
portdirect | heres the commit: https://git.openstack.org/cgit/openstack/openstack-manuals/commit/?id=f46adca7488d8a383b4f6adb86bf3e93ebfcd6b0 | 08:56 |
*** huzhengchuan has quit IRC | 08:56 | |
SamYaple | cool | 08:57 |
*** rmart04 has joined #openstack-kolla | 08:57 | |
*** yingjun has quit IRC | 08:58 | |
*** athomas has joined #openstack-kolla | 09:02 | |
*** shardy has joined #openstack-kolla | 09:04 | |
SamYaple | portdirect: what did yo uthink about the goals for kolla-salt? | 09:06 |
portdirect | +2 | 09:07 |
portdirect | like them :) | 09:07 |
*** msimonin has joined #openstack-kolla | 09:07 | |
SamYaple | cool. that was my original purpose. when i first created my ansible+docker deployment tool thing. That is what i was doing | 09:07 |
*** msimonin1 has joined #openstack-kolla | 09:08 | |
SamYaple | portdirect: copmlete with static uid/gid ;) | 09:08 |
SamYaple | https://github.com/SamYaple/yaodu/blob/master/ansible/roles/docker_build/templates/ubuntu/nova/Dockerfile.j2#L23 | 09:08 |
portdirect | lol, nice indentation too :) | 09:09 |
SamYaple | i like pretties | 09:09 |
portdirect | nice name too :) was going to ask earlier but things were pretty hectic, how did you get a 40mb ubuntu keystone ? | 09:11 |
SamYaple | it wasnt with yaodu here, this was recently | 09:11 |
*** msimonin has quit IRC | 09:11 | |
portdirect | yeah - i figured, you must have used pretty much every trick in the book to get there no? | 09:12 |
SamYaple | purge all non-useful things (man pages, systemd, gcc, etc) and then basically `docker export` to flatten it. then take all of the flattened images and compute a common layer based on files taht exist in every image | 09:12 |
SamYaple | it takes longer but produces much better, much smaller images | 09:12 |
portdirect | yeah - i used to flatten all the aline stuf using that trick we both explored but got bored or waiting :) | 09:13 |
portdirect | susprised that it got that small though just stripping out packages | 09:14 |
*** matrohon has quit IRC | 09:14 | |
*** huzhengchuan has joined #openstack-kolla | 09:14 | |
SamYaple | indeed. a whole lotta cruft | 09:14 |
SamYaple | libvirt will be the largest container | 09:15 |
portdirect | I really want to get some time to look at the kolla source containers, it makes me a very sad panda that gcc and freinds are left in them :/ | 09:15 |
SamYaple | well the first step in my patented-3-step-plan is to have a final layer ontop of all the images that deletes al lthe cruft | 09:16 |
SamYaple | the images remain the same size, but the final layer only has 40MB of data | 09:16 |
SamYaple | no gcc | 09:16 |
SamYaple | once that is working and solid, then we can recompute layers | 09:17 |
portdirect | sounds good - i think quite a few people would be interested in that moving forward | 09:17 |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Keystone service endpoint creation (keystone specific) https://review.openstack.org/408843 | 09:21 |
*** matrohon has joined #openstack-kolla | 09:23 | |
*** gomarivera has joined #openstack-kolla | 09:23 | |
*** Serlex has joined #openstack-kolla | 09:25 | |
*** gomarivera has quit IRC | 09:28 | |
*** yatin_on_leave has quit IRC | 09:33 | |
*** gfidente has joined #openstack-kolla | 09:34 | |
*** elynn has quit IRC | 09:36 | |
*** dasTor has joined #openstack-kolla | 09:38 | |
*** dasTor_ has joined #openstack-kolla | 09:39 | |
*** dasTor_ has quit IRC | 09:39 | |
*** sp_ has quit IRC | 09:40 | |
*** nrado has joined #openstack-kolla | 09:46 | |
*** yuanying has quit IRC | 09:47 | |
*** yatin_on_leave has joined #openstack-kolla | 09:47 | |
*** sp_ has joined #openstack-kolla | 09:52 | |
*** yatin_on_leave has quit IRC | 09:54 | |
*** papacz has quit IRC | 09:56 | |
*** sp_ has quit IRC | 09:57 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm common user creation jobs https://review.openstack.org/408864 | 10:00 |
*** JoseMello has joined #openstack-kolla | 10:00 | |
*** papacz has joined #openstack-kolla | 10:04 | |
*** duonghq has quit IRC | 10:05 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Keystone service endpoint creation (keystone specific) https://review.openstack.org/408843 | 10:06 |
*** yatin_on_leave has joined #openstack-kolla | 10:07 | |
*** sp_ has joined #openstack-kolla | 10:09 | |
*** papacz has quit IRC | 10:09 | |
dupengfei | Anybody , Who can help me, what helm/microservice is in kolla-kubernetes. | 10:09 |
*** papacz has joined #openstack-kolla | 10:10 | |
*** JoseMello has quit IRC | 10:10 | |
*** tovin07__ has quit IRC | 10:12 | |
*** JoseMello has joined #openstack-kolla | 10:12 | |
*** yatin_on_leave has quit IRC | 10:13 | |
*** sp_ has quit IRC | 10:14 | |
*** sdake has joined #openstack-kolla | 10:15 | |
*** ccesario has joined #openstack-kolla | 10:16 | |
sdake | morning | 10:17 |
*** zhubingbing has quit IRC | 10:21 | |
*** yatin_on_leave has joined #openstack-kolla | 10:26 | |
*** sp_ has joined #openstack-kolla | 10:27 | |
sdake | pbourke around? | 10:28 |
pbourke | sdake: morning | 10:28 |
sdake | sup dude | 10:29 |
*** Pavo has quit IRC | 10:29 | |
sdake | can you ack a review plz | 10:29 |
sdake | https://review.openstack.org/#/c/409965/2 | 10:29 |
openstackgerrit | Merged openstack/kolla-ansible: Fix kolla-genpwd and kolla-mergepwd https://review.openstack.org/409909 | 10:29 |
openstackgerrit | Merged openstack/kolla: Add kafka image https://review.openstack.org/406874 | 10:29 |
pbourke | sdake: done thanks | 10:29 |
sdake | oool thanks | 10:29 |
*** yatin_on_leave has quit IRC | 10:31 | |
*** Pavo has joined #openstack-kolla | 10:33 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm common user creation jobs https://review.openstack.org/408864 | 10:38 |
portdirect | o/ | 10:38 |
sdake | sup portdirect | 10:40 |
sdake | say portdirect did you see i beat on launchpad a bit | 10:40 |
portdirect | refactoring - fun times :) | 10:40 |
portdirect | yeah - looks ok i think but to be honest not had a chance to check properly | 10:41 |
portdirect | i'll have a nosey now | 10:41 |
sdake | ya refactoring launchpad ftw ;) | 10:42 |
sdake | i was clicking so many things i lost track;) | 10:42 |
sdake | you were finished with the dev env docs correct? | 10:42 |
*** sp_ has quit IRC | 10:43 | |
*** yatin_on_leave has joined #openstack-kolla | 10:43 | |
portdirect | yeah - its seems to be working, would like to to get a local gate for people to use but thats a bit too much of a stretch at the moment - I can talk people thorugh a getto rigged one | 10:43 |
sdake | i marked them done | 10:43 |
portdirect | nice - cheers - soon as we have the microservices done I'll get them updated to reflect where we are | 10:44 |
portdirect | suppose i talked myself into helm-services :) | 10:47 |
*** gomarivera has joined #openstack-kolla | 10:48 | |
*** gomarivera has quit IRC | 10:52 | |
portdirect | dupengfei: hey | 10:54 |
dupengfei | hello | 10:54 |
*** sp_ has joined #openstack-kolla | 10:54 | |
dupengfei | portdirect: hello | 10:55 |
dupengfei | portdirect: What is the current state of helm in kolla-k8s? | 10:55 |
dupengfei | portdirect: Developing, ready, or nearly ready? | 10:55 |
portdirect | dupengfei: it is at an early stage of development | 10:55 |
portdirect | we are currently converting the microservices from the original kolla-k8s implementation to helm packages | 10:56 |
egonzalez90 | SamYaple: which backend do you prefer for designate, powerdns or bind? | 10:57 |
portdirect | dupengfei: if you wanted to get invovled it would be great :) | 10:57 |
dupengfei | portdirect: Now, we maintain 2 versions, normal and helm ? | 10:57 |
sdake | next on deck is -> https://launchpad.net/kolla-kubernetes/+milestone/0.5.0 | 10:58 |
portdirect | dupengfei: no the original is being depricated, but you currently have systems in production running the orginal implementaion? | 10:58 |
dupengfei | portdirect: Yes, pleasure. | 10:58 |
dupengfei | portdirect: OH MY GOD!!!! depricated? | 10:58 |
sdake | portdirect i dont knwo if there are original systems in production with the code base in place or not | 10:58 |
portdirect | sdake, dupengfei: has i think? | 10:59 |
sdake | dupengfei we haven't done a 1.0.0 release of kolla-ansible | 10:59 |
sdake | rather kolla-kubernetes | 10:59 |
dupengfei | portdirect: We are testing beta version using kolla-kubernetes. | 10:59 |
portdirect | ok - thats better :) you had me worried in your pm | 10:59 |
dupengfei | portdirect: We have a lot of bugs to fix, | 10:59 |
sdake | dupengfei its sort of not automated ;) | 11:00 |
portdirect | there will be a lot of bugs at the moment, as we transition from one arch to another | 11:01 |
dupengfei | portdirect: I was not active in open-source community, the change is too huge. | 11:02 |
mliima_ | sdake, ? | 11:02 |
dupengfei | portdirect: What is the major | 11:02 |
dupengfei | portdirect: what is the major place to discuss such changes ? | 11:03 |
portdirect | dupengfei: the mailing list would be the most appropriate place i think | 11:03 |
sdake | mliima_ sup? | 11:04 |
dupengfei | portdirect: This room or launchpad. | 11:04 |
dupengfei | portdirect: Oh I know. | 11:04 |
mliima_ | well, you wrote mliima | 11:04 |
sdake | dupengfei irc works, ml works, launchpad tracker (discussion) works | 11:04 |
sdake | mliima_ oh i was asking you to ack that patch pbourke acked about 8 hrs ago :) | 11:04 |
SamYaple | egonzalez90: bind9 is the stable go to | 11:04 |
sdake | mliima_ so all good | 11:04 |
sdake | mliima_ btw how you been :) | 11:05 |
SamYaple | egonzalez90: but if you can implement both, yay? | 11:05 |
mliima_ | fine sdake hehe | 11:05 |
sdake | i dont see you often because i am working on keeping a normal sleeping schedule | 11:05 |
sdake | I sort of passed out at 8:30 last night and woke up around 3:30 today | 11:05 |
dupengfei | portdirect: If I submit bugfixs or bps to former directory, can be merged or not ? | 11:05 |
mliima_ | still working on "clnt_create-RPC-Unknown-host" error | 11:05 |
mliima_ | :? | 11:05 |
pbourke | the tools/ dir in kolla / kolla-ansible is a mess | 11:06 |
mliima_ | :/ | 11:06 |
sdake | pbourke ten to agree | 11:06 |
pbourke | we need to blow away one of them | 11:06 |
*** yatin_on_leave has quit IRC | 11:06 | |
pbourke | before the divergence becomes untraceable | 11:06 |
SamYaple | ugh why did the ceph repo get removed in the ubuntu image | 11:06 |
SamYaple | now we won't get updates | 11:06 |
portdirect | dupengfei: could you tell us what bugs you have found - we have never used newton with kolla-k8s that I'm aware of, but I may well be mistaken in that | 11:06 |
*** sp_ has quit IRC | 11:07 | |
sdake | dupengfei if you file bugs against kolla-kubernetes the core team will have a look | 11:07 |
sdake | dupengfei - although the code bse is under heavy active developement and we haven't committed to maintaining a release until ew branch | 11:08 |
sdake | dupengfei the branch happens at 1.0.0 | 11:08 |
sdake | dupengfei that said, we o want to fix the bugs present | 11:08 |
*** Bico_Fino has joined #openstack-kolla | 11:12 | |
*** zhugaoxiao has joined #openstack-kolla | 11:13 | |
openstackgerrit | Sam Yaple proposed openstack/kolla: Add monasca images https://review.openstack.org/406422 | 11:13 |
openstackgerrit | dupengfei proposed openstack/kolla: Not to pull the base image when building. https://review.openstack.org/408940 | 11:16 |
dupengfei | portdirect: lot of bugs, ie: | 11:17 |
dupengfei | portdirect: cinder-lvm is not usefull, because, the /etc/iscsi is not mounted | 11:18 |
*** yatin_on_leave has joined #openstack-kolla | 11:18 | |
*** wanghongxu has quit IRC | 11:19 | |
portdirect | dupengfei: it should be, as it is specified in the existing template,(https://github.com/openstack/kolla-kubernetes/blob/master/services/cinder/cinder-volume-lvm-pod.yml.j2#L114-L115), but please note at the moment we do not test with lvm, but ceph for storage | 11:21 |
*** sp_ has joined #openstack-kolla | 11:22 | |
portdirect | if you file a bug we will try and look into it but it may be some time - have you found a fix that works for you? | 11:22 |
*** yatin_on_leave has quit IRC | 11:23 | |
dupengfei | portdirect: Yes, works. I will submit a bugfix about it. | 11:23 |
portdirect | great, but please expect the erticture to change before the 1.0.0 release | 11:24 |
portdirect | *architecture | 11:24 |
dupengfei | portdirect: What do you mean ? expect the architecture ? | 11:25 |
*** jascott1_ has joined #openstack-kolla | 11:26 | |
portdirect | we are moving away from the custom templating engine toward helm and a more kubernetes native workflow | 11:26 |
*** sp_ has quit IRC | 11:27 | |
dupengfei | portdirect: Should I submit to helm or older custom ? | 11:27 |
dupengfei | portdirect: or both? | 11:27 |
portdirect | if you can submit to helm that would be winderful - the more people we get invovled the better, especially if you are using (or planning to use) kolla-kubernetes. | 11:28 |
dupengfei | portdirect: I will spend 1-2 days to read and test helm architecture. | 11:28 |
portdirect | *wonderfull | 11:28 |
dupengfei | portdirect: OK, I will fix bugs on both. | 11:29 |
dupengfei | portdirect: I am nearly get off work, good bye | 11:30 |
*** sdake_ has joined #openstack-kolla | 11:30 | |
portdirect | dupengfei: bye :) | 11:30 |
openstackgerrit | Sam Yaple proposed openstack/kolla: Add monasca images https://review.openstack.org/406422 | 11:31 |
*** jascott1_ has quit IRC | 11:32 | |
*** sdake has quit IRC | 11:34 | |
*** zhangshuai has joined #openstack-kolla | 11:34 | |
*** yatin_on_leave has joined #openstack-kolla | 11:35 | |
sdake_ | back to bed | 11:36 |
sdake_ | bbl | 11:36 |
*** sdake_ has quit IRC | 11:36 | |
*** sp_ has joined #openstack-kolla | 11:40 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm common user creation jobs https://review.openstack.org/408864 | 11:41 |
*** nrado has left #openstack-kolla | 11:41 | |
*** sdake has joined #openstack-kolla | 11:45 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm common user creation jobs https://review.openstack.org/408864 | 11:49 |
*** JoseMello has quit IRC | 11:54 | |
*** JoseMello has joined #openstack-kolla | 11:55 | |
*** sdake has quit IRC | 11:55 | |
*** zhugaoxiao has quit IRC | 11:56 | |
*** zhugaoxiao has joined #openstack-kolla | 11:56 | |
*** JoseMello has quit IRC | 12:00 | |
*** JoseMello has joined #openstack-kolla | 12:02 | |
*** sdake has joined #openstack-kolla | 12:07 | |
*** Bico_Fino has quit IRC | 12:09 | |
*** sdake has quit IRC | 12:10 | |
openstackgerrit | Pete Birley proposed openstack/kolla: Remove unneeded PythonMySQL libs from OpenStack Base Image https://review.openstack.org/399042 | 12:11 |
*** williamcaban has joined #openstack-kolla | 12:11 | |
*** vincent_vdk has left #openstack-kolla | 12:17 | |
*** williamcaban has quit IRC | 12:22 | |
*** williamcaban has joined #openstack-kolla | 12:22 | |
openstackgerrit | Wei Cao proposed openstack/kolla-ansible: Add freezer ansible role https://review.openstack.org/402224 | 12:25 |
*** rhallisey has joined #openstack-kolla | 12:25 | |
*** sdake has joined #openstack-kolla | 12:25 | |
*** sdake has quit IRC | 12:26 | |
*** jascott1_ has joined #openstack-kolla | 12:28 | |
*** shardy is now known as shardy_lunch | 12:29 | |
*** Pavo has quit IRC | 12:29 | |
*** zhubingbing has joined #openstack-kolla | 12:32 | |
*** jascott1_ has quit IRC | 12:33 | |
*** Pavo has joined #openstack-kolla | 12:33 | |
*** mliima__ has joined #openstack-kolla | 12:37 | |
*** mliima_ has quit IRC | 12:40 | |
*** eaguilar has joined #openstack-kolla | 12:43 | |
*** krtaylor has joined #openstack-kolla | 12:44 | |
*** duonghq has joined #openstack-kolla | 12:46 | |
duonghq | evening guys | 12:46 |
*** mliima_ has joined #openstack-kolla | 12:46 | |
*** nrado has joined #openstack-kolla | 12:47 | |
duonghq | portdirect, sbezverk_ Is anybody working on service layer in Helm? | 12:47 |
openstackgerrit | zhubingbing proposed openstack/kolla-ansible: WIP add fluentd role https://review.openstack.org/407392 | 12:47 |
rhallisey | SamYaple, https://www.percona.com/blog/2014/09/01/galera-replication-how-to-recover-a-pxc-cluster/ | 12:49 |
rhallisey | SamYaple, search for pc.recovery | 12:49 |
*** gomarivera has joined #openstack-kolla | 12:49 | |
*** huzhengchuan has quit IRC | 12:49 | |
SamYaple | rhallisey: yea what about it? | 12:49 |
*** mliima__ has quit IRC | 12:49 | |
rhallisey | SamYaple, it explains the auto recovery | 12:50 |
rhallisey | that galera has built in now | 12:50 |
SamYaple | rhallisey: hold on let me get you a link | 12:50 |
SamYaple | thats been in there for a long itme | 12:50 |
*** JoseMello has quit IRC | 12:51 | |
*** sdake has joined #openstack-kolla | 12:51 | |
duonghq | sdake, due to nobody take any items on helm-services, I guess that I can take any items to start working on? | 12:53 |
SamYaple | rhallisey: eh i can't find link, that has been a thing for a while though | 12:53 |
rhallisey | nice | 12:53 |
SamYaple | rhallisey: its not perfect though, and recovery still needs some user interevention | 12:53 |
SamYaple | rhallisey: you just dont loose data (which we dont currenlty either) | 12:53 |
rhallisey | SamYaple, what user intervention specifically | 12:53 |
SamYaple | so there are two options | 12:54 |
SamYaple | what i posted yesterday where it starts but doesnt wait for the primary | 12:54 |
*** gomarivera has quit IRC | 12:54 | |
SamYaple | or the --wrsrep-recovery (or the related ? option at the endo f the gcomm string) | 12:54 |
SamYaple | and, i would point out, it really doesnt work well unfortunately | 12:55 |
SamYaple | ive only had it work once :( | 12:55 |
mliima_ | guys, why when I ran showmount -e <ip> inside cinder-volume container fails and from my host it works fine :/ | 12:56 |
mliima_ | i get this error in cinder-volume container clnt_create: RPC: Unknown host | 12:56 |
SamYaple | thats assuming i have my facts straight rhallisey. i love to be proven wrong :) | 12:56 |
rhallisey | SamYaple, ok neat. I'm looking at it because I want to understand the workflow | 12:56 |
SamYaple | to be hoenst, i havent looked at it since 10.0.x | 12:57 |
*** shardy_lunch is now known as shardy | 12:57 | |
SamYaple | i think 10.1.x is used now | 12:57 |
SamYaple | rhallisey: http://lists.askmonty.org/pipermail/commits/2016-May/009384.html | 12:59 |
SamYaple | apparenly its suppose to work in 10.1 but its broken | 12:59 |
SamYaple | good find! | 12:59 |
SamYaple | i guess i didnt understand that as well as i thought | 12:59 |
*** fguillot has joined #openstack-kolla | 13:00 | |
rhallisey | I haven't tried it personally, but it's curious | 13:02 |
Jeffrey4l | fyi, we introduced dumb-init in container. so maybe this is related. | 13:02 |
rhallisey | SamYaple, one of the things I wanted to do with kubernetes is have a recovery workflow like this live as part of the cluster. | 13:04 |
rhallisey | so have a bunch of python that will assist in making galera come back in each of the 7 scenarios they outline | 13:04 |
SamYaple | yea for kubernetes youd have to do something like that | 13:05 |
sdake | duonghq dont ask for permission ask for forgivness :) | 13:05 |
rhallisey | ya it's really the only way to make a stateful app like a database seem more cloud native | 13:05 |
sdake | duonghq in other words, yes | 13:06 |
Jeffrey4l | recently, we have a env with 3 node shutdown duo to power issue. but the mariadb cluster is auto recovered when we power on the servers. it never happened before. | 13:06 |
sdake | duonghq however, portdirect and sbezverk_ are working on that part a bit - and may have some work sitting on their harddisk unfinished | 13:06 |
sdake | duonghq or ssd | 13:06 |
*** dave-mccowan has joined #openstack-kolla | 13:06 | |
sdake | or whatever they use to store data :) | 13:06 |
SamYaple | Jeffrey4l: yea thats what i was reading. its supposed to work that way in 10.1 | 13:06 |
rhallisey | Jeffrey4l, I was reading about it here: https://www.percona.com/blog/2014/09/01/galera-replication-how-to-recover-a-pxc-cluster/ | 13:06 |
* sdake dating himself with harddisk :) | 13:06 | |
rhallisey | though that may be old | 13:07 |
Jeffrey4l | cool. | 13:07 |
rhallisey | Jeffrey4l, search for pc.recovery | 13:07 |
SamYaple | rhallisey: the article is old, but it wasnt implemented until 10.1 mariadb | 13:07 |
rhallisey | it talks about the lights out recovery workflow | 13:07 |
Jeffrey4l | hrm it is a old article. i read it before | 13:07 |
rhallisey | k I'll see if I can get a new one | 13:07 |
rhallisey | but I'd image the workflow is similar | 13:08 |
Jeffrey4l | pc.recovery should be a old parameter. and exist all the time. no idea why it do not work as expect before. maybe there are some issue? | 13:09 |
*** duonghq has quit IRC | 13:15 | |
*** srwilkers has joined #openstack-kolla | 13:19 | |
*** fguillot has quit IRC | 13:20 | |
*** duonghq has joined #openstack-kolla | 13:20 | |
*** fguillot has joined #openstack-kolla | 13:20 | |
*** duonghq has joined #openstack-kolla | 13:20 | |
* duonghq back, something wrong w/ my network | 13:21 | |
matrohon | hi kolla | 13:23 |
matrohon | is there any way to easily backport https://bugs.launchpad.net/kolla/+bug/1634223 ? | 13:24 |
openstack | Launchpad bug 1634223 in kolla "Cron logrotation for Elasticsearch is missing" [Medium,Fix released] | 13:24 |
*** markus-k has joined #openstack-kolla | 13:25 | |
openstackgerrit | Mathieu Rohon proposed openstack/kolla: Add logrotate configuration for elasticsearch https://review.openstack.org/410217 | 13:26 |
*** JoseMello has joined #openstack-kolla | 13:26 | |
matrohon | ^ let's see what happens there | 13:28 |
Jeffrey4l | matrohon, sure. +2ed | 13:29 |
*** duonghq has quit IRC | 13:30 | |
matrohon | Jeffrey4l, do you think I can even try to backport to mitaka? or is there any changes that would prevent this patch to easily merge in stable/mitaka? | 13:31 |
Jeffrey4l | matrohon, it is a bug since mitaka. so i think it should be backport to mitaka. | 13:32 |
matrohon | Jeffrey4l, ok let's try, thanks | 13:32 |
Jeffrey4l | matrohon, thanks for point this ;) | 13:32 |
*** JoseMello has quit IRC | 13:32 | |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 13:33 |
srwilkers | good morning o/ | 13:33 |
matrohon | Jeffrey4l, gerrit refuse to cheerypick this patch in mitaka, I'll try to do it anually | 13:33 |
*** JoseMello has joined #openstack-kolla | 13:33 | |
Jeffrey4l | matrohon, yep. there is a big change between mitaka and newton. so backport is not that easy. | 13:34 |
Jeffrey4l | sup srwilkers | 13:34 |
matrohon | Jeffrey4l, no hurry for our "prod" deployment, so I thinks we'd better first upgrade to newton | 13:35 |
Jeffrey4l | matrohon, you have prod env? very cool | 13:35 |
matrohon | Jeffrey4l, but on another deployment, es is very chatty, and it has filled the entire root disk :( | 13:36 |
matrohon | Jeffrey4l, not critical prod, but it's in use as is, yep | 13:36 |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 13:37 |
srwilkers | not much Jeffrey4l how are you? | 13:37 |
Jeffrey4l | matrohon, you may need delete the old files in es. | 13:38 |
Jeffrey4l | there are script in github to do so. | 13:38 |
Jeffrey4l | srwilkers, the same ;) | 13:38 |
matrohon | Jeffrey4l, ack | 13:39 |
egonzalez90 | Jeffrey4l: ping | 13:40 |
Jeffrey4l | egonzalez90, pong | 13:41 |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Zaqar ansible role https://review.openstack.org/407760 | 13:43 |
*** jascott1_ has joined #openstack-kolla | 13:47 | |
*** huzhengchuan has joined #openstack-kolla | 13:48 | |
*** mliima__ has joined #openstack-kolla | 13:48 | |
*** jheroux has joined #openstack-kolla | 13:51 | |
*** mliima_ has quit IRC | 13:51 | |
*** jascott1_ has quit IRC | 13:52 | |
*** yatin_on_leave has quit IRC | 13:53 | |
*** mliima_ has joined #openstack-kolla | 13:53 | |
*** egonzalez90 has quit IRC | 13:54 | |
*** nrado has left #openstack-kolla | 13:55 | |
*** mliima has joined #openstack-kolla | 13:55 | |
*** mliima__ has quit IRC | 13:56 | |
*** mliima__ has joined #openstack-kolla | 13:57 | |
*** mliima_ has quit IRC | 13:58 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Add lost roles field in octavia https://review.openstack.org/410225 | 13:58 |
*** ayoung has joined #openstack-kolla | 13:58 | |
*** mliima_ has joined #openstack-kolla | 13:59 | |
*** mliima has quit IRC | 14:00 | |
*** goldyfruit has joined #openstack-kolla | 14:02 | |
*** v1k0d3n has joined #openstack-kolla | 14:02 | |
*** mliima has joined #openstack-kolla | 14:02 | |
*** mliima__ has quit IRC | 14:03 | |
srwilkers | morning v1k0d3n o/ | 14:03 |
*** lamt has joined #openstack-kolla | 14:03 | |
v1k0d3n | morning srwilkers | 14:03 |
awiddersheim | Jeffrey4l: I also noticed you didn't put the enable_octavia in kolla's example globals.yml file | 14:04 |
*** mliima_ has quit IRC | 14:04 | |
*** v1k0d3n has left #openstack-kolla | 14:04 | |
Jeffrey4l | awiddersheim, oops | 14:04 |
awiddersheim | not a big deal | 14:05 |
*** yatin_on_leave has joined #openstack-kolla | 14:05 | |
Jeffrey4l | stupid mistake ;( | 14:07 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Add lost roles field in octavia https://review.openstack.org/410225 | 14:09 |
*** sdake_ has joined #openstack-kolla | 14:12 | |
awiddersheim | Jeffrey4l: still need to fix egonzalez90's comment as well I think | 14:14 |
Jeffrey4l | yes. thanks. | 14:15 |
*** sdake has quit IRC | 14:15 | |
*** yatin_on_leave has quit IRC | 14:16 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Add lost roles field in octavia https://review.openstack.org/410225 | 14:16 |
*** msimonin has joined #openstack-kolla | 14:16 | |
*** msimonin1 has quit IRC | 14:16 | |
openstackgerrit | Merged openstack/kolla-kubernetes: Helm database creation for Nova https://review.openstack.org/409474 | 14:17 |
openstackgerrit | Merged openstack/kolla-kubernetes: Keystone service endpoint creation (keystone specific) https://review.openstack.org/408843 | 14:17 |
*** HyperJohnGraham has joined #openstack-kolla | 14:20 | |
*** tovin07__ has joined #openstack-kolla | 14:20 | |
openstackgerrit | Li Yingjun proposed openstack/kolla-kubernetes: Fix different repo issue for documentation https://review.openstack.org/405054 | 14:20 |
*** dmsimard has quit IRC | 14:21 | |
*** dmsimard has joined #openstack-kolla | 14:21 | |
*** gomarivera has joined #openstack-kolla | 14:21 | |
*** mdnadeem has quit IRC | 14:22 | |
srwilkers | morning sdake_ | 14:22 |
sdake_ | sup srwilkers | 14:22 |
sdake_ | getting kids to school | 14:22 |
*** eaguilar has quit IRC | 14:25 | |
*** gomarivera has quit IRC | 14:26 | |
openstackgerrit | Li Yingjun proposed openstack/kolla-kubernetes: Fix different repo issue for documentation https://review.openstack.org/405054 | 14:27 |
sbezverk_ | good morning | 14:28 |
*** dmsimard has quit IRC | 14:28 | |
*** dmsimard has joined #openstack-kolla | 14:28 | |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 14:28 |
*** dmsimard has quit IRC | 14:28 | |
srwilkers | hey sbezverk_ :) | 14:28 |
*** Pavo has quit IRC | 14:29 | |
sbezverk_ | srwilkers: getting very close to close glance ;-) | 14:29 |
*** saneax is now known as saneax-_-|AFK | 14:29 | |
*** dmsimard has joined #openstack-kolla | 14:30 | |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 14:31 |
srwilkers | sbezverk_, should hopefully have the rest up in patchsets today :) sorry it took longer than anticipated. nasty bug going around the office took me out of commission for a few days | 14:32 |
*** msimonin has quit IRC | 14:32 | |
*** Pavo has joined #openstack-kolla | 14:34 | |
sbezverk_ | srwilkers: no worries at all :) | 14:34 |
*** msimonin has joined #openstack-kolla | 14:35 | |
*** adrian_otto has joined #openstack-kolla | 14:36 | |
sbezverk_ | srwilkers: have you seen 1.5.0 is out of the door now :-) | 14:37 |
openstackgerrit | Surya Prakash Singh proposed openstack/kolla-ansible: Implement Keystone zero-downtime upgrade https://review.openstack.org/405215 | 14:39 |
*** adrian_otto1 has joined #openstack-kolla | 14:39 | |
*** adrian_otto has quit IRC | 14:40 | |
goldyfruit | guys, some of you are using the cgroup_enable=memory and swapaccount=1 kernel arguments with Kolla ? | 14:43 |
kfox1111 | morning. | 14:43 |
goldyfruit | kfox1111: morning | 14:43 |
*** srwilkers has quit IRC | 14:44 | |
sbezverk_ | kfox1111: good morning | 14:47 |
*** jascott1_ has joined #openstack-kolla | 14:48 | |
kfox1111 | srwilker-: commented on a bug in the glance svc review. | 14:52 |
*** jtriley has joined #openstack-kolla | 14:52 | |
*** jascott1_ has quit IRC | 14:53 | |
sdake_ | hey folks, i have a few minutes, I wanted to point out that launchpad has been updated to more accurately reflect our kolla-kubernetes deliverable plan: https://launchpad.net/kolla-kubernetes/ocata | 14:54 |
sdake_ | dates are aligned with ocata milestone dates for the moment | 14:54 |
sdake_ | however - since the deliverable is release:independent - they are subject to the whims of the broader community :) | 14:55 |
*** tonanhngo has joined #openstack-kolla | 14:55 | |
*** eaguilar has joined #openstack-kolla | 14:57 | |
sdake_ | next kid to school | 14:58 |
sdake_ | bbl | 14:58 |
kfox1111 | sdake_: nice. thanks. :) | 15:00 |
sbezverk_ | kfox1111: do you have time to have chat about "service" level pacjage? | 15:01 |
*** sayantan_ has joined #openstack-kolla | 15:01 | |
*** eaguilar_ has joined #openstack-kolla | 15:02 | |
*** mgiles has joined #openstack-kolla | 15:03 | |
*** eaguilar has quit IRC | 15:04 | |
*** jascott1_ has joined #openstack-kolla | 15:04 | |
*** TxGirlGeek has joined #openstack-kolla | 15:05 | |
openstackgerrit | zhubingbing proposed openstack/kolla-ansible: WIP add fluentd role https://review.openstack.org/407392 | 15:06 |
*** lrensing has joined #openstack-kolla | 15:07 | |
kfox1111 | portdirect: https://review.openstack.org/#/c/408864/7 is looking very nice. couple of minior things. | 15:08 |
*** jascott1_ has quit IRC | 15:08 | |
openstackgerrit | Merged openstack/kolla-kubernetes: Helm Keystone K8s Services https://review.openstack.org/408963 | 15:16 |
*** wwriverrat has quit IRC | 15:29 | |
*** srwilkers has joined #openstack-kolla | 15:31 | |
srwilkers | sbezverk_, yeah, i think it dropped yesterday right? | 15:31 |
srwilkers | re: 1.5.0 that is | 15:31 |
*** bmace has quit IRC | 15:35 | |
sbezverk_ | srwilkers: it seems this morning, I check last night EST time, it was not there | 15:36 |
srwilkers | oh nice | 15:36 |
sbezverk_ | srwilkers: need to build new VM to run 1.5.0 aio | 15:37 |
srwilkers | agreed | 15:37 |
openstackgerrit | Li Yingjun proposed openstack/kolla-kubernetes: Fix different repo issue for documentation https://review.openstack.org/405054 | 15:38 |
*** rhallisey has quit IRC | 15:38 | |
sbezverk_ | srwilkers: running out of core in my server, need new CPU with 22 cores ;-) | 15:38 |
srwilkers | sbezverk_, i mean, early christmas present? ;) | 15:38 |
sbezverk_ | srwilkers: no.. the price tag on these is huge, need to wait when Cisco give them to me ;-) | 15:39 |
srwilkers | sbezverk_, haha. i dont foresee AT&T being so generous for me. one can hope | 15:40 |
* srwilkers crosses fingers | 15:40 | |
SamYaple | i have 3 x 8 core servers at my house that i do all my testing on | 15:41 |
SamYaple | they werent bought for me either :( | 15:41 |
kfox1111 | sbezverk_: gota get ready for work. maybe in about 45 min? | 15:42 |
sbezverk_ | kfox1111: sounds good, ping me when you have time slot available | 15:42 |
kfox1111 | k. I'll ping you as soon as I get there. | 15:42 |
sbezverk_ | SamYaple: I am trying to minimize my electricity bill ;-) so less servers, more core | 15:43 |
SamYaple | sbezverk_: i run highly available stuff at my house, minimum three :P | 15:44 |
sdake_ | me either SamYaple :( | 15:44 |
sdake_ | well, I suppose my paycheck paid for them | 15:44 |
sdake_ | so - in a sense, they were bought for me :) | 15:44 |
srwilkers | SamYaple, HA everywhere, houses included. clearly need a minimum of three houses | 15:45 |
sbezverk_ | SamYaple: If I would do that, I would have to spend my paycheck on eletrcicity only or close ;-) | 15:45 |
*** gomarivera has joined #openstack-kolla | 15:45 | |
SamYaple | sbezverk_: yea its not cheap. my computers double my bill | 15:46 |
SamYaple | i have to budget it in | 15:46 |
*** jhalebli has joined #openstack-kolla | 15:47 | |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 15:47 |
*** saneax-_-|AFK is now known as saneax | 15:48 | |
*** adrian_otto has joined #openstack-kolla | 15:48 | |
*** adrian_otto1 has quit IRC | 15:49 | |
*** gomarivera has quit IRC | 15:52 | |
*** gomarivera has joined #openstack-kolla | 15:53 | |
*** hieulq_ has joined #openstack-kolla | 15:56 | |
*** gomarivera has quit IRC | 15:57 | |
*** TxGirlGeek has quit IRC | 16:00 | |
*** TxGirlGeek has joined #openstack-kolla | 16:00 | |
*** diablo_rojo has joined #openstack-kolla | 16:01 | |
*** zhubingbing has quit IRC | 16:02 | |
*** b_bezak has joined #openstack-kolla | 16:02 | |
*** TxGirlGeek has quit IRC | 16:04 | |
*** yingjun has joined #openstack-kolla | 16:04 | |
*** yingjun has quit IRC | 16:04 | |
*** TxGirlGeek has joined #openstack-kolla | 16:04 | |
*** zhubingbing has joined #openstack-kolla | 16:04 | |
*** yingjun has joined #openstack-kolla | 16:04 | |
*** yingjun has quit IRC | 16:08 | |
*** pcaruana has quit IRC | 16:09 | |
*** eaguilar_ has quit IRC | 16:09 | |
*** eaguilar has joined #openstack-kolla | 16:10 | |
*** huzhengchuan has quit IRC | 16:11 | |
*** matrohon has quit IRC | 16:12 | |
*** wwriverrat has joined #openstack-kolla | 16:17 | |
*** rmart04 has quit IRC | 16:17 | |
*** inc0 has joined #openstack-kolla | 16:18 | |
inc0 | Hello | 16:18 |
mliima | hello inc0 | 16:20 |
SamYaple | hey inc0 | 16:21 |
SamYaple | speak of the devil | 16:21 |
inc0 | uh.. what's up? | 16:23 |
SamYaple | nothing. was just chatting about you in infra | 16:23 |
inc0 | should I be concerned? | 16:23 |
-openstackstatus- NOTICE: Launchpad SSO is not currently working, so logins to our services like review.openstack.org and wiki.openstack.org are failing; the admins at Canonical are looking into the issue but there is no estimated time for a fix yet. | 16:24 | |
*** msimonin has quit IRC | 16:24 | |
*** ChanServ changes topic to "Launchpad SSO is not currently working, so logins to our services like review.openstack.org and wiki.openstack.org are failing; the admins at Canonical are looking into the issue but there is no estimated time for a fix yet." | 16:24 | |
inc0 | I'll be back in business early next week | 16:24 |
*** mewald has joined #openstack-kolla | 16:24 | |
inc0 | I'm in California now | 16:25 |
SamYaple | inc0: nah its fine. you just came up. dont worry :) | 16:25 |
inc0 | ok cool | 16:25 |
*** eaguilar has quit IRC | 16:26 | |
inc0 | Sierra Nevada :) | 16:26 |
SamYaple | having fun inc0? | 16:27 |
*** vbel has joined #openstack-kolla | 16:28 | |
inc0 | hell yeah | 16:28 |
inc0 | 3 National Parks so far | 16:28 |
*** Pavo has quit IRC | 16:29 | |
SamYaple | woah. we have 3 national parks? | 16:29 |
vbel | hi everyone, is it just me not able to connect to review.openstack.org , suddenly asking for openid (which also does not work) | 16:29 |
inc0 | would be 2 more if roads would be open | 16:29 |
SamYaple | vbel: yea its down | 16:30 |
inc0 | vbel infra just said that it's down | 16:30 |
SamYaple | vbel: infra knows | 16:30 |
vbel | thanks guys! | 16:30 |
*** Serlex has quit IRC | 16:33 | |
*** Pavo has joined #openstack-kolla | 16:34 | |
kfox1111 | back. | 16:35 |
kfox1111 | sbezverk_: I'm free for a bit. | 16:35 |
*** sacharya has joined #openstack-kolla | 16:38 | |
sdake_ | sup inc0 | 16:38 |
*** sdake_ is now known as sdake | 16:38 | |
sdake | if you have a chance check out sedona ;) | 16:38 |
*** b_bezak has quit IRC | 16:38 | |
*** b_bezak has joined #openstack-kolla | 16:39 | |
srwilkers | hey inc0 :) | 16:39 |
srwilkers | morning kfox1111 :) | 16:39 |
inc0 | sdake we were in Sedona | 16:40 |
inc0 | gorgeous | 16:40 |
sdake | in arizona of course ;) | 16:40 |
inc0 | totally coming back there | 16:40 |
inc0 | yeah we spent 2 days there | 16:41 |
inc0 | now mammoth lakes | 16:41 |
*** b_bezak has quit IRC | 16:43 | |
*** jhalebli has quit IRC | 16:46 | |
sdake | oh misread that | 16:46 |
*** zhubingbing has quit IRC | 16:46 | |
*** mkoderer has quit IRC | 16:46 | |
*** vbel has quit IRC | 16:48 | |
qwang | morning | 16:49 |
qwang | inc0: how are you doing? | 16:49 |
inc0 | great:) exploring USA | 16:50 |
qwang | it'll be an unforgettable trip :) | 16:50 |
inc0 | it already is, one more week left | 16:51 |
openstackgerrit | Ethan Lynn proposed openstack/kolla-ansible: Add missing become=yes on precheck stage https://review.openstack.org/410315 | 16:51 |
qwang | inc0: cool. enjoy ~~ | 16:52 |
openstackgerrit | Vladislav Belogrudov proposed openstack/kolla-ansible: Re-import Murano core library on upgrade https://review.openstack.org/410316 | 16:52 |
*** bmace has joined #openstack-kolla | 16:58 | |
*** eaguilar has joined #openstack-kolla | 16:59 | |
*** Jeffrey4l has quit IRC | 17:00 | |
*** sacharya_ has joined #openstack-kolla | 17:00 | |
*** Jeffrey4l has joined #openstack-kolla | 17:00 | |
*** ChanServ changes topic to "Have support questions? Please give ask.openstack.org a try. New to Kolla? Please read the documentation here: http://docs.openstack.org/developer/kolla/; Kolla IRC meetngs on Wednesdays @ 16:00 UTC - see agenda @ https://goo.gl/OXB0DL - IRC channel is *LOGGED* @ http://goo.gl/3mzZ7b" | 17:01 | |
-openstackstatus- NOTICE: Canonical admins have resolved the issue with login.launchpad.net, so authentication should be restored now. | 17:01 | |
*** egonzalez90 has joined #openstack-kolla | 17:01 | |
egonzalez90 | Is someone having issues login to gerrit? | 17:02 |
*** sacharya has quit IRC | 17:02 | |
*** HyperJohnGraham has quit IRC | 17:04 | |
kfox1111 | srwilkers: morning. | 17:06 |
*** lrensing has quit IRC | 17:07 | |
*** lrensing has joined #openstack-kolla | 17:07 | |
*** strigazi is now known as strigazi_AFK | 17:07 | |
*** inc0 has quit IRC | 17:08 | |
*** matrohon has joined #openstack-kolla | 17:08 | |
*** jhaleblian has joined #openstack-kolla | 17:09 | |
*** sdake has quit IRC | 17:10 | |
jhaleblian | ne1 avail to field a question on kolla-k8s? checked ask but can't find my issue there. | 17:12 |
srwilkers | kfox1111, hows it going? | 17:13 |
kfox1111 | its going. you? | 17:14 |
*** msimonin has joined #openstack-kolla | 17:14 | |
kfox1111 | srwilkers: http://logs.openstack.org/87/409487/12/check/gate-kolla-kubernetes-deploy-centos-binary-ceph-multi-nv/6656b9a/logs/svc/kolla-glance-api.txt | 17:14 |
kfox1111 | there are two endpoints listed there but I think the default is 1? | 17:14 |
kfox1111 | so I think the selector is matching wrong. | 17:14 |
*** tovin07__ has quit IRC | 17:14 | |
srwilkers | kfox1111, its going. i was actually looking through that too | 17:16 |
kfox1111 | k. let me know if you need any help. :) | 17:16 |
srwilkers | kfox1111, definitely man. thanks :) | 17:18 |
*** jascott1_ has joined #openstack-kolla | 17:18 | |
*** jascott1_ has quit IRC | 17:19 | |
*** matrohon has quit IRC | 17:19 | |
srwilkers | oh | 17:21 |
srwilkers | i see | 17:21 |
srwilkers | passing the type in with the wrong variable name used in the common service | 17:21 |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 17:22 |
*** fragatina has joined #openstack-kolla | 17:22 | |
*** fragatina has quit IRC | 17:23 | |
*** diogogmt has joined #openstack-kolla | 17:23 | |
*** diablo_rojo has quit IRC | 17:23 | |
kfox1111 | ah. | 17:23 |
*** fragatina has joined #openstack-kolla | 17:23 | |
*** jascott1_ has joined #openstack-kolla | 17:24 | |
srwilkers | helps to read ;) | 17:26 |
*** fragatina has quit IRC | 17:26 | |
srwilkers | hey jascott1 o/ | 17:26 |
*** egonzalez90 has quit IRC | 17:26 | |
*** fragatina has joined #openstack-kolla | 17:26 | |
jascott1_ | hey srwilkers hows things? | 17:30 |
srwilkers | jascott1_, things are good. office is quiet. how about you? | 17:30 |
jascott1_ | all good... looking forward to another snow tomorrow | 17:31 |
*** mewald has quit IRC | 17:31 | |
*** sdake has joined #openstack-kolla | 17:32 | |
kfox1111 | jascott1_: https://review.openstack.org/#/c/409959 is looking really close to merge. | 17:33 |
*** athomas has quit IRC | 17:33 | |
kfox1111 | it would be nice to add some testing, but I don't think it can be tested until there are createdb/managedb things added first. | 17:33 |
jascott1_ | so we are testing against 1.4.6 ? I thought we were using 1.5b2 | 17:34 |
kfox1111 | I'm ok merging it without tests so long as the next ps to merge adds those and tests launching. | 17:34 |
jascott1_ | helm/tiller issues still exist? | 17:34 |
kfox1111 | yeah. gate's 1.4.6 | 17:34 |
kfox1111 | looks like 1.5 just came out. :) | 17:34 |
kfox1111 | should try and upgrade the gate to it. | 17:35 |
jascott1_ | ya techonosophos said helm 2.1 was going to drop too | 17:35 |
kfox1111 | nice. | 17:35 |
*** fragatina has quit IRC | 17:35 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Helmize neutron db stuff https://review.openstack.org/410336 | 17:36 |
jascott1_ | he mentioned it yesterday but dont see it yet :/ | 17:36 |
kfox1111 | know if helm 2.0.2 supports k8s 1.5, or do we need to wait? | 17:36 |
jascott1_ | well we had some problems but not sure if its already shaken out | 17:37 |
kfox1111 | oh... I guess we would have to patch petsets too to go to 1.5. | 17:37 |
sdake | kfox1111 only one way to find out ;-) | 17:37 |
jascott1_ | :D | 17:37 |
srwilkers | :) | 17:37 |
kfox1111 | oh.... | 17:38 |
kfox1111 | actually.... hmm... | 17:38 |
kfox1111 | we're pulling from their yum repo's. | 17:39 |
*** eaguilar_ has joined #openstack-kolla | 17:39 | |
kfox1111 | so when 1.5 lands there, it will automatically start getting used. | 17:39 |
kfox1111 | so we may need to upgrade kubeadm though, and maybe helm. | 17:39 |
kfox1111 | and then fix the petsets. | 17:39 |
kfox1111 | so the gate may break today at some point. | 17:39 |
kfox1111 | kubeadm is getting pulled from the repos too... so thats getting upgraded too... | 17:40 |
kfox1111 | k. well, if anyone sees some odd gate breakage, pleas elet me know asap. I'll work on fixing it. | 17:40 |
kfox1111 | meeting. brb. | 17:40 |
*** eaguilar has quit IRC | 17:40 | |
*** fragatina has joined #openstack-kolla | 17:41 | |
*** fragatina has quit IRC | 17:42 | |
*** fragatina has joined #openstack-kolla | 17:42 | |
*** unicell has quit IRC | 17:46 | |
*** JoseMello has quit IRC | 17:46 | |
*** fragatina has quit IRC | 17:48 | |
*** jascott__ has joined #openstack-kolla | 17:48 | |
*** jascott1_ has quit IRC | 17:51 | |
*** srwilkers is now known as srwilkers_lunch | 17:52 | |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helm-ize Heat API and CFN https://review.openstack.org/409959 | 17:52 |
*** gomarivera has joined #openstack-kolla | 17:54 | |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helmize Heat-engine https://review.openstack.org/410048 | 17:56 |
*** gomarivera has quit IRC | 17:59 | |
openstackgerrit | Justin Scott proposed openstack/kolla-kubernetes: WIP Helmize Heat-engine https://review.openstack.org/410048 | 17:59 |
*** tonanhngo has quit IRC | 18:04 | |
*** pbourke has quit IRC | 18:04 | |
*** pbourke has joined #openstack-kolla | 18:04 | |
sayantan_ | Hi everyone | 18:05 |
*** nrado has joined #openstack-kolla | 18:05 | |
jascott__ | hi sayantan_ | 18:06 |
sayantan_ | Hi jascott_ how is it going? | 18:06 |
nrado | Hi there, did someone try to deploy barbican? The barbican_keystone_listener is throwing errors in my case: | 18:08 |
nrado | Could not load 'simple_certificate_event': cannot import name certificate_manager | 18:08 |
nrado | cannot import name certificate_manager | 18:08 |
nrado | ImportError: cannot import name certificate_manager | 18:08 |
nrado | Traceback (most recent call last): | 18:08 |
nrado | File "/usr/bin/barbican-keystone-listener", line 6, in <module> | 18:08 |
nrado | from barbican.cmd.keystone_listener import main | 18:08 |
nrado | ImportError: cannot import name main | 18:08 |
*** krtaylor has quit IRC | 18:09 | |
*** JoseMello has joined #openstack-kolla | 18:11 | |
sayantan_ | I am working on the "Building Container Images" document. I am trying to figure out if the kolla-build.conf file is not modified, from where are the default options for building images invoked? | 18:12 |
openstackgerrit | Qin Wang (qwang) proposed openstack/kolla-kubernetes: WIP - Hel-ming horizon components https://review.openstack.org/408975 | 18:13 |
*** unicell has joined #openstack-kolla | 18:13 | |
jascott__ | sayantan_ good. planning the werk, twerking the plan | 18:14 |
openstackgerrit | Merged openstack/kolla-ansible: Add lost roles field in octavia https://review.openstack.org/410225 | 18:14 |
sayantan_ | haha jascott_ good luck ! | 18:15 |
rafael_sisweb | Hey guys. I'm playing into openstack world and already deploy an simple set of controller + compute + storage nodes using kolla. | 18:16 |
rafael_sisweb | Each of this nodes has only network interface to my internal net. 10.10.0.0/24 ... | 18:16 |
rafael_sisweb | Into "/etc/kolla/globals.yml" i'd set internal_vip_address to an unused IP of this net.. Grea it's works. | 18:16 |
rafael_sisweb | To kolla_external_vip_address i'd set to an unsed IP of my public ip range. Works to.. i can access my dashboard an apis through this public IP. | 18:16 |
rafael_sisweb | But i'd like to understand how and where this IP is attached? None of interfaces of my controller or kolla node has that. | 18:16 |
*** hieulq_ has quit IRC | 18:18 | |
*** klindgren has joined #openstack-kolla | 18:19 | |
sdake | nrado please file a bug | 18:19 |
*** JoseMello has quit IRC | 18:20 | |
*** klindgren_ has quit IRC | 18:20 | |
*** diogogmt has quit IRC | 18:22 | |
*** JoseMello has joined #openstack-kolla | 18:24 | |
*** Pavo has quit IRC | 18:29 | |
portdirect | afternoon kolla \o/ | 18:29 |
*** fragatina has joined #openstack-kolla | 18:29 | |
*** Pavo has joined #openstack-kolla | 18:29 | |
jascott__ | portdirect o/ | 18:29 |
portdirect | kfox1111: just oing through scrollback, but it would be awesome if we shifted to k8s 1.5 - i could run that through the gate, and we would need to use a patched tiller for a few days, but we should be ok and it would really help for things like statefullsets and pvc support | 18:31 |
portdirect | hey jascott__ :) | 18:31 |
sayantan_ | Hi portdirect , sdake , I am trying to work on the building container images document and am trying to figure out how are the default options invoked if the kolla-build.conf is not modified. | 18:34 |
sayantan_ | Any help would be great | 18:34 |
sayantan_ | :) | 18:34 |
portdirect | sayantan_: got a link? | 18:34 |
sdake | in a meeting sorry can't help for about 3 or 4 hours - i'd reach out to jeffrey4l - he is a pro on that work | 18:34 |
portdirect | ^^ | 18:35 |
sayantan_ | portdirect http://docs.openstack.org/developer/kolla/image-building.html | 18:35 |
sayantan_ | sdake thanks :) | 18:36 |
sbezverk_ | portdirect: it way too new | 18:36 |
sayantan_ | jeffrey41, you there? | 18:36 |
sbezverk_ | portdirect: running it in local test bed is fun but rely the gate on i do noit think so | 18:37 |
openstackgerrit | Qin Wang (qwang) proposed openstack/kolla-kubernetes: WIP - Hel-ming horizon components https://review.openstack.org/408975 | 18:38 |
sayantan_ | portdirect the reason I ask is, if we are using the pip install workflow, should we also document how the kolla-build.conf file is generated? Or just document the way how operators can use options to modify the base using -b etc | 18:39 |
portdirect | sayantan_: i think so, i can give you a link that shows how id do it if that helps? | 18:40 |
portdirect | sayantan_: https://github.com/portdirect/dockerfiles/blob/master/kolla-builder/Dockerfile#L47-L49 | 18:40 |
sayantan_ | And is there a way to view the options available with the kolla-build command? | 18:41 |
sayantan_ | Thanks portdirect | 18:41 |
sayantan_ | Let me check the link. | 18:41 |
jascott__ | anyone put thought into the helm keywords? should it be - openstack, - heat, - heat-api or - openstack, - heat, - api ? | 18:42 |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Enable horizon deployment in gate https://review.openstack.org/410057 | 18:42 |
jascott__ | im seeing the second | 18:42 |
srwilkers_lunch | hey portdirect o/ | 18:42 |
*** srwilkers_lunch is now known as srwilkers | 18:42 | |
portdirect | jascott__: theres a can of worms! | 18:42 |
portdirect | jascott__: i started with the former, but have moved to the 2nd, whihc seems to be the way most otehr ps's have used | 18:43 |
jascott__ | cool i will roll with that and we can one mind it later | 18:43 |
portdirect | jascott__: i think the 2nd makes most sense when used with helm search | 18:43 |
portdirect | jascott1: +1 :) | 18:43 |
*** shardy is now known as shardy_afk | 18:43 | |
jascott__ | ya thats what i was thinking (but did not commit) | 18:43 |
portdirect | hey srwilkers o/ | 18:43 |
srwilkers | hows it going? | 18:45 |
portdirect | sbezverk_: fair points, but i really feel we should move over as soon as possible, it makes big (and good) changes to the api that affect us - meaning that what we are working on now will be incompatible with the current k8s at the time of our next release | 18:46 |
jascott__ | portdirect +1 | 18:47 |
srwilkers | portdirect, +1 | 18:47 |
portdirect | srwilkers: not bad man, good lunch :) | 18:47 |
srwilkers | portdirect, ate way too much | 18:48 |
portdirect | 'tis the season for excess | 18:48 |
srwilkers | portdirect, i saw the registration for boston opened up. are you going to the PTG and/or boston summit? | 18:50 |
portdirect | oh - need to check, probably not the PTG :( though I'd be up for virtual attendance if that was somehow possible? will almost certainly make the summit | 18:52 |
*** tonanhngo has joined #openstack-kolla | 18:53 | |
srwilkers | awesome. trying to determine whether they're going to send our team to the summit or not. afaik, ill be at the PTG, but still want to make the summit. may just have to take personal time for that if need be | 18:56 |
portdirect | srwilkers: need to try and see if there is any posibility for finincial support fot the PTG? the foundation used to support individual contributors going to the the summit, so I'll see if theres any possibility of trying to get some help, as I'd really like to attend if at all possible. | 18:57 |
*** tonanhngo has quit IRC | 18:58 | |
sayantan_ | portdirect another concern. When I install kolla using pip, there is no kolla/etc directory. It is kolla/etc_examples directory. And the configuration file is generated in kolla/etc/kolla by oslo-config-generator | 18:59 |
sayantan_ | http://paste.openstack.org/raw/592255/ | 18:59 |
*** fragatina has quit IRC | 19:01 | |
*** krtaylor has joined #openstack-kolla | 19:01 | |
*** HyperJohnGraham has joined #openstack-kolla | 19:01 | |
*** fragatina has joined #openstack-kolla | 19:01 | |
*** fragatina has quit IRC | 19:02 | |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm chart for Cinder API service https://review.openstack.org/410362 | 19:02 |
sayantan_ | So, would it be advisable to include in the docs to change the etc to etc_examples in etc_examples/oslo-config-generator/kolla-build.conf so that the config file is generated correctly ? Or is there a way we can modify that? | 19:02 |
*** fragatina has joined #openstack-kolla | 19:02 | |
kfox1111 | back. | 19:03 |
portdirect | srwilkers: dang! just missed the cut of date for round one of the travel support program :( thought there is another round in jan, so 0-'ll apply for that and see how it goes :) https://www.openstack.org/ptg/#tab_travel | 19:03 |
*** diogogmt has joined #openstack-kolla | 19:03 | |
*** tonanhngo has joined #openstack-kolla | 19:03 | |
kfox1111 | I got general permission to go to the ptg, so mostly just detail work now. | 19:04 |
srwilkers | portdirect, i set a reminder on my calendar to poke you the week before just incase | 19:04 |
srwilkers | thats awesome kfox1111 | 19:04 |
*** diogogmt has quit IRC | 19:04 | |
mliima | guys, anybody knows why showmount -e fails in cinder-volume cotnainer? | 19:05 |
portdirect | srwilkers: nice - thanks man :) thats great kfox1111 | 19:05 |
mliima | ubuntu 16.04 | 19:05 |
*** tonanhngo has quit IRC | 19:05 | |
*** diogogmt has joined #openstack-kolla | 19:05 | |
portdirect | sayantan_: I'm afrail Jeffrey4l, would be much better placed to help you there | 19:05 |
portdirect | *afraid | 19:05 |
kfox1111 | portdirect: hope you get to make it. | 19:06 |
*** tonanhngo has joined #openstack-kolla | 19:06 | |
sayantan_ | Thanks portdirect | 19:06 |
portdirect | kfox1111: cheers dude | 19:07 |
qwang | my ps failed at gate. complaining about unfound key but I don't see a problem. http://logs.openstack.org/75/408975/12/check/gate-kolla-kubernetes-deploy-ubuntu-binary-ceph-nv/542c9b0/console.html#_2016-12-13_18_55_38_780802 | 19:07 |
sayantan_ | ping Jeffrey4l | 19:07 |
kfox1111 | qwang: helm's error logs can be horible at times. :/ | 19:08 |
kfox1111 | I'll take a look. | 19:08 |
portdirect | sayantan_: i think its the middle of the night for Jeffrey4l :( | 19:08 |
qwang | kfox1111: thank you so much | 19:09 |
sayantan_ | oh lol. Didn't realize that | 19:10 |
*** msimonin has quit IRC | 19:14 | |
sbezverk_ | qwang: I suggest to run helm install --dry-run --debug right before you do it in the gate and then go through carefully generated spec. You will fine where the problem is.. | 19:15 |
sbezverk_ | s/fine/find/ | 19:16 |
qwang | sbezverk_: thanks. Will do it | 19:16 |
*** msimonin has joined #openstack-kolla | 19:16 | |
kfox1111 | qwang: so, I started commenting out some stuff by changing the {{- include .... to #{- include in the file. | 19:18 |
kfox1111 | at some point, it stops complaining. | 19:18 |
kfox1111 | so, I'd guess its a missing required variable definition. | 19:18 |
kfox1111 | its complaining if either common_volume_mounts or common_containers are enabled. so something there. | 19:19 |
jascott__ | i thought it might be there is no name for containerPort, is that not required? | 19:20 |
sbezverk_ | --dry-run will give answers to all misteries ;-) | 19:20 |
kfox1111 | sbezverk_: not all. :/ | 19:21 |
portdirect | qwang: sbezverk_'s suggestion is really good - the dev env will give you a running helm/k8s env to test against localy if you dont already have the ability to do that | 19:21 |
kfox1111 | sbezverk_: its really horible with templates sometimes. :/ | 19:21 |
portdirect | sbezverk_: only with debug! | 19:21 |
kfox1111 | sbezverk_: sometimes the only way I've debugged issues in helm is to start commenting out includes. | 19:21 |
portdirect | sbezverk_: the number of times i've run it without and scratch my head for 20 seconds.... | 19:21 |
jascott__ | imo all these approaches are required | 19:22 |
kfox1111 | but yeah, using --dry-run --debug and then iterating until it produces valid yaml is the trick. | 19:22 |
qwang | thank you all. I have a direction now | 19:22 |
*** JoseMello has quit IRC | 19:23 | |
kfox1111 | qwang: I'm still looking too. | 19:24 |
portdirect | i found it i think | 19:24 |
kfox1111 | portdirect: yeah? | 19:24 |
portdirect | indenetation at ports? also no name though I wnt to check locally to make sure | 19:24 |
kfox1111 | portdirect: could be an issue, but shouldn't cause it to fail a render. | 19:25 |
kfox1111 | its complaining at the templating layer... | 19:25 |
kfox1111 | so, it for sure fails if common_volume_mounts is included... | 19:26 |
kfox1111 | but, it only has two vars used: .netHostTrue and .Values.enable_resolvee_conf_net_host_workaround. | 19:27 |
kfox1111 | those look good. | 19:27 |
kfox1111 | it indents 12.... | 19:27 |
kfox1111 | ah. thats it. | 19:27 |
kfox1111 | hmm. no, not entirely.... | 19:28 |
kfox1111 | replacing 12 with 10 on the indents gets most of the way. | 19:29 |
srwilkers | just curious, why is there a port there anyway? maybe this is my ignorance showing, but we didn't have an entry for a port there with the jinja2 template. is there something im missing? | 19:29 |
kfox1111 | cause the horizon-config section after the common_volume_mounts is intented 10. | 19:30 |
*** fragatina has quit IRC | 19:30 | |
*** fragatina has joined #openstack-kolla | 19:30 | |
kfox1111 | srwilkers: I think port in that way is advisory? not sure. probably should either remove it, or change everything to match. | 19:30 |
*** msimonin has quit IRC | 19:30 | |
kfox1111 | port does make sense when used along with net=host sometimes. but not sure if its useful otherwise. | 19:31 |
kfox1111 | there we go. found the other issue. | 19:31 |
portdirect | qwang: http://paste.openstack.org/show/592257/ | 19:32 |
kfox1111 | 10, 10, and 6 | 19:32 |
kfox1111 | will comment in the review. | 19:32 |
*** diablo_rojo has joined #openstack-kolla | 19:33 | |
kfox1111 | qwang: I added the fixs to the review. its basically an indent issue. | 19:34 |
kfox1111 | "We are releasing Kubernetes v1.5.1, later today (Dec. 13)"... | 19:35 |
kfox1111 | fun. | 19:35 |
portdirect | srwilkers: the port is optional with some networking layers, but now policy is enforced by many CNI plugins it is essential to ensure compatibility accross all back ends | 19:35 |
*** msimonin has joined #openstack-kolla | 19:35 | |
srwilkers | portdirect, okay, makes sense. thanks | 19:35 |
*** jascott__ has quit IRC | 19:35 | |
*** another_larsks has joined #openstack-kolla | 19:37 | |
*** jmccarthy has quit IRC | 19:38 | |
qwang | You guys are amazing. Thank y'all | 19:38 |
portdirect | np hope the paste helped :) | 19:39 |
*** jmccarthy has joined #openstack-kolla | 19:39 | |
qwang | I'm out for lunch. Will update the patch after lunch:) | 19:39 |
*** JoseMello has joined #openstack-kolla | 19:40 | |
*** eaguilar_ has quit IRC | 19:43 | |
kfox1111 | qwang: np. glad to help. :) | 19:44 |
*** mliima has quit IRC | 19:45 | |
*** another_larsks has quit IRC | 19:45 | |
*** another_larsks has joined #openstack-kolla | 19:46 | |
*** nrado has quit IRC | 19:46 | |
mgiles | For the folks working on kubernetes kfox1111 sbezverk_ portdirect sdake - Just an FYI I'll be out on vaca for the next week. | 19:46 |
*** diablo_rojo has quit IRC | 19:46 | |
mgiles | Nova microservice stuff is pretty much done other than keystone user/endpoints. I can do them when I get back or somebody else can pick them up... | 19:46 |
kfox1111 | mgiles: ok. thanks for the heads up. | 19:47 |
sdake | mgiles enjoy :) | 19:47 |
kfox1111 | mgiles: have a good vacation. :) | 19:47 |
mgiles | Thanks, I will | 19:47 |
portdirect | mgiles: np -thanks for the good work! that was a toughie for sure you picked up :) | 19:47 |
mgiles | I'm sure there'll be lots of progress during the week. Can't wait to see where you all are in a week! | 19:48 |
kfox1111 | portdirect: +1 :) | 19:48 |
*** common has joined #openstack-kolla | 19:48 | |
jascott1 | hey srwilkers are you taking over heat? | 19:48 |
srwilkers | jascott1, nah man, thats you. i had put my name on that last week but you've got it covered | 19:48 |
srwilkers | feel free to change it on the bp | 19:48 |
jascott1 | did i edit an old version of page or something? i swear I put my name on there | 19:49 |
jascott1 | and put it INPROGRESS and everting | 19:49 |
kfox1111 | jascott1: I think there was a race condition on web page edits? | 19:50 |
jascott1 | no history on bps huh | 19:50 |
kfox1111 | yeah. | 19:50 |
openstackgerrit | Merged openstack/kolla-kubernetes: WIP Helm-ize Heat API and CFN https://review.openstack.org/409959 | 19:50 |
*** msimonin has quit IRC | 19:51 | |
srwilkers | sorry man, i didnt want to give off that impression. when i threw my name at it, it was only because it was unassigned at the time | 19:51 |
*** mewald has joined #openstack-kolla | 19:52 | |
kfox1111 | no worries. :) | 19:53 |
portdirect | srwilkers: needs tests though :) | 19:53 |
*** JoseMello has quit IRC | 19:53 | |
jascott1 | srwilkers sorry! same here | 19:55 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Helmize neutron db stuff https://review.openstack.org/410336 | 19:55 |
kfox1111 | +1 for heat tests. :) | 19:56 |
kfox1111 | needs createdb/managedb stuff before it can be tested though. | 19:56 |
kfox1111 | at least to see if the api's will launch at all. | 19:56 |
kfox1111 | the next set of tests can be added when heat engine's added. | 19:56 |
kfox1111 | to see if it can launch a stack. | 19:56 |
kfox1111 | so, I think the next piece that needs working on is the create/manage db's for heat. | 19:58 |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm chart for Cinder API service https://review.openstack.org/410362 | 19:58 |
*** jgriffith is now known as jgriffith_AutoAw | 19:58 | |
*** msimonin has joined #openstack-kolla | 19:59 | |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 20:00 |
kfox1111 | srwilkers: was the glance one just a rebase? | 20:01 |
kfox1111 | its got heat stuff in it. | 20:01 |
*** jgriffith_AutoAw is now known as jgriffith | 20:01 | |
kfox1111 | indended for a different review? | 20:02 |
*** another_larsks has left #openstack-kolla | 20:02 | |
*** HyperJohnGraham has quit IRC | 20:05 | |
kfox1111 | hehe. | 20:07 |
kfox1111 | 410336,2 and 410362,2 are both running in zuul and kolla-kubernetes related. | 20:08 |
portdirect | kfox1111: ever feel like this? https://cdn.meme.am/cache/instances/folder651/500x/55187651.jpg | 20:08 |
kfox1111 | not confusing at all. :) | 20:08 |
kfox1111 | portdirect: heh. every day of my life.... ;) | 20:08 |
portdirect | better fix my ps's again... :) | 20:09 |
*** matrohon has joined #openstack-kolla | 20:09 | |
*** dupengfei has quit IRC | 20:12 | |
*** jgriffith is now known as jgriffith_AutoAw | 20:17 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm common user creation jobs https://review.openstack.org/408864 | 20:23 |
*** eaguilar has joined #openstack-kolla | 20:28 | |
openstackgerrit | Pete Birley proposed openstack/kolla-kubernetes: Helm common user creation jobs https://review.openstack.org/408864 | 20:28 |
*** Pavo has quit IRC | 20:29 | |
sbezverk_ | srwilkers: hey man how did you manage to get html rendered in the PS log ;-)?? | 20:33 |
*** Pavo has joined #openstack-kolla | 20:33 | |
*** jgriffith_AutoAw is now known as jgriffith | 20:34 | |
srwilkers | sbezverk_, uh oh, whatd i break? ;p | 20:37 |
sbezverk_ | srwilkers: I do not know but I have nevern seen in the log what I see in yours | 20:37 |
* portdirect thinks of the possibilities | 20:37 | |
portdirect | where? | 20:37 |
sbezverk_ | http://logs.openstack.org/87/409487/14/check/gate-kolla-kubernetes-deploy-centos-binary-ceph-multi-nv/c8ee49b/console.html#_2016-12-13_20_23_01_371772 | 20:38 |
srwilkers | http://logs.openstack.org/87/409487/14/check/gate-kolla-kubernetes-deploy-centos-binary-ceph-nv/a9ad258/console.html#_2016-12-13_20_19_27_172030 | 20:38 |
srwilkers | huh | 20:38 |
srwilkers | never seen that | 20:38 |
srwilkers | i think i know whats going on | 20:40 |
*** shardy_afk is now known as shardy | 20:40 | |
portdirect | srwilkers: no external ip for glance svc? | 20:42 |
srwilkers | portdirect, yep | 20:42 |
srwilkers | portdirect, fixing that now | 20:42 |
sbezverk_ | srwilkers: if you do not mind let's get glance merged today, if you need any help please let me know. | 20:43 |
srwilkers | sbezverk_, agreed. if theres any part you want to take, go for it. as i said, i wont stand in your way :) | 20:45 |
portdirect | yeah we need to review the whole svc situation, not sure that the way we currently rely on setting everything up to an external vip, kinda seems like a big step back from what we could do with k8s. we should probably shift to using internal endpoints int the config at some point, as this is exactly what they were desgigned for AFAIK | 20:46 |
sbezverk_ | portdirect: the failure here happens because by default glance client uses public endpoint | 20:53 |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 20:53 |
portdirect | sbezverk_: exactly | 20:54 |
sbezverk_ | portdirect: if we can override glance client setting to use internal or admin, it should work as is | 20:54 |
portdirect | yes, once we have all the microservices over i think it would be good to look at the config situation, as this is (VIP/public svc) is somthing thats concened me for some time | 20:55 |
sbezverk_ | portdirect: at one point we were looking into it, but by some reason did not complete it | 20:55 |
portdirect | I think it would make sense to put an ingress controller deamonset in and use that for public api/svc access | 20:56 |
portdirect | would also give us tls for free :) | 20:57 |
portdirect | (at least at the edge) | 20:57 |
sbezverk_ | portdirect: that was the original plan, just did not have time to implement it | 20:59 |
portdirect | sbezverk_: sweet :) | 21:00 |
openstackgerrit | Steve Wilkerson proposed openstack/kolla-kubernetes: Helm chart for Cinder API service https://review.openstack.org/410362 | 21:03 |
*** dgonzalez has quit IRC | 21:14 | |
*** gfidente has quit IRC | 21:21 | |
*** JRobinson__ has joined #openstack-kolla | 21:22 | |
*** shardy has quit IRC | 21:25 | |
*** HyperJohnGraham has joined #openstack-kolla | 21:26 | |
*** imcsk8_ is now known as imcsk8 | 21:27 | |
*** Jeffrey4l has quit IRC | 21:35 | |
*** msimonin has quit IRC | 21:36 | |
*** msimonin has joined #openstack-kolla | 21:36 | |
*** msimonin has quit IRC | 21:36 | |
srwilkers | portdirect, you around? | 21:36 |
srwilkers | sbezverk_, ping | 21:38 |
sbezverk_ | srwilkers: pong | 21:39 |
srwilkers | sbezverk_, nevermind, answered my own question :) | 21:42 |
portdirect | yup | 21:44 |
srwilkers | heading back to the homestead sbezverk_ -- will have patch sets pushed up for glance-api pod, and the database bits this evening unless someone pushes one up before me | 21:45 |
sbezverk_ | srwilkers: cool | 21:47 |
*** Jeffrey4l has joined #openstack-kolla | 21:47 | |
jascott1 | srwilkers funny i started on glance | 21:51 |
jascott1 | :) | 21:51 |
*** srwilkers has quit IRC | 21:53 | |
*** lrensing has quit IRC | 21:56 | |
*** krtaylor has quit IRC | 22:00 | |
*** jtriley has quit IRC | 22:03 | |
*** rmart04 has joined #openstack-kolla | 22:06 | |
*** Pavo has quit IRC | 22:10 | |
*** sacharya_ has quit IRC | 22:10 | |
*** sacharya has joined #openstack-kolla | 22:11 | |
*** fguillot has quit IRC | 22:11 | |
*** goldyfruit has quit IRC | 22:12 | |
*** JoseMello has joined #openstack-kolla | 22:12 | |
*** HyperJohnGraham has quit IRC | 22:13 | |
*** jhaleblian has quit IRC | 22:13 | |
openstackgerrit | Qin Wang (qwang) proposed openstack/kolla-kubernetes: WIP - Hel-ming horizon components https://review.openstack.org/408975 | 22:13 |
*** mgiles has quit IRC | 22:14 | |
*** rmart04 has quit IRC | 22:15 | |
*** gomarivera has joined #openstack-kolla | 22:16 | |
*** Pavo has joined #openstack-kolla | 22:16 | |
*** matrohon has quit IRC | 22:16 | |
*** TxGirlGeek has quit IRC | 22:18 | |
*** gomarivera has quit IRC | 22:20 | |
*** HyperJohnGraham has joined #openstack-kolla | 22:23 | |
*** Pavo has quit IRC | 22:23 | |
*** Pavo has joined #openstack-kolla | 22:26 | |
*** dgonzalez has joined #openstack-kolla | 22:27 | |
*** Pavo has quit IRC | 22:30 | |
*** lamt has quit IRC | 22:31 | |
*** lamt has joined #openstack-kolla | 22:33 | |
*** tonanhngo_ has joined #openstack-kolla | 22:34 | |
*** Pavo has joined #openstack-kolla | 22:34 | |
*** v1k0d3n has joined #openstack-kolla | 22:36 | |
*** tonanhngo has quit IRC | 22:37 | |
*** srwilkers has joined #openstack-kolla | 22:38 | |
kfox1111 | back | 22:39 |
*** eaguilar has quit IRC | 22:40 | |
srwilkers | wb kfox1111 | 22:41 |
kfox1111 | thx. :) | 22:43 |
*** jheroux has quit IRC | 22:45 | |
*** mewald has quit IRC | 22:47 | |
*** adrian_otto has quit IRC | 22:50 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Helmize neutron db stuff https://review.openstack.org/410336 | 22:52 |
*** HyperJohnGraham_ has joined #openstack-kolla | 23:01 | |
*** HyperJohnGraham has quit IRC | 23:02 | |
*** HyperJohnGraham_ is now known as HyperJohnGraham | 23:02 | |
jascott1 | are we changing to k8s 1.5? | 23:05 |
openstackgerrit | Qin Wang (qwang) proposed openstack/kolla-kubernetes: WIP - Hel-ming horizon components https://review.openstack.org/408975 | 23:05 |
kfox1111 | jascott1: if we do nothing to the gate, it will happen by itself. and petsets will break. | 23:06 |
kfox1111 | I think the question is, will the currently released helm support 1.5? | 23:06 |
jascott1 | ya I see Jan 7 as their release date on GH but i swear techonosophos said they were going to release at same time | 23:07 |
kfox1111 | ugg... | 23:07 |
kfox1111 | this is why I don't trust helm yet for orchestration "A substantial bug with helm upgrade -f has been fixed" | 23:08 |
kfox1111 | I think we should still be working on it, but for production use, scary. | 23:08 |
jascott1 | kfox1111 are you in helm channel on slack? | 23:09 |
kfox1111 | is 2.0.2 work with 1.5 or do we need to try and pin to 1.4 until helms upgraded? | 23:09 |
kfox1111 | jascott1: yeah. but don't always watch it. | 23:09 |
openstackgerrit | Merged openstack/kolla-kubernetes: Helm charts for glance services https://review.openstack.org/409487 | 23:09 |
jascott1 | And 2.1 will now go out _today_ | 23:09 |
jascott1 | I re-asked him just now | 23:09 |
kfox1111 | k. | 23:09 |
kfox1111 | will 2.1 work with 1.4? | 23:10 |
openstackgerrit | sayantani proposed openstack/kolla: Update Quick Start Guide for singlenode deployment https://review.openstack.org/404993 | 23:10 |
sbezverk_ | kfox1111: I would really prefer to stick with 1.4.X for this release.. | 23:10 |
jascott1 | asking | 23:10 |
sbezverk_ | fighting with 1.5.X issues does not sounds fun to me atm | 23:10 |
kfox1111 | sbezverk_: thats what the gate's for though, right? if the gate passes, do we think we will have issues? | 23:11 |
jascott1 | kfox1111 he said yes 2.1 will work with k8s 1.4 | 23:11 |
kfox1111 | jascott1: ok. so I'll bump helm version as soon as its released. | 23:11 |
kfox1111 | sbezverk_: the way the gate is currently, it will hit 1.5 at some point without changes. so either we have to commit something to fix 1.5 issues, or commit something that pins back down to 1.4. | 23:12 |
kfox1111 | not sure which will end up being less work. | 23:12 |
jascott1 | but one choice is future forward | 23:13 |
jascott1 | petsets to statefulsets is a pain but i think we should just bite the bullet | 23:13 |
*** lrensing has joined #openstack-kolla | 23:14 | |
*** HyperJohnGraham has quit IRC | 23:14 | |
jascott1 | about how many do we have? | 23:14 |
kfox1111 | jascott1: its just in a few places. a sed can take care of it pretty easy. | 23:15 |
openstackgerrit | sayantani proposed openstack/kolla: Update Quick Start Guide for singlenode deployment https://review.openstack.org/404993 | 23:16 |
*** JoseMello has quit IRC | 23:16 | |
kfox1111 | yeah, I think I agree. better to bite the bullet sooner rather then later, if 1.5 doesn't pose too many extra issues. | 23:16 |
jascott1 | its all moving pretty fast I think we just have to keep up as much as possible | 23:16 |
sbezverk_ | kfox1111: I am talking about 1.5.0 issues which we cannot fix and will need to wait for kube community to fix.. | 23:17 |
kfox1111 | sbezverk_: yeah. I'm thinking we try and see in the gate? if it has issues, we quickly roll back and pin at 1.4. as we can't support a broken 1.5.1? | 23:19 |
*** HyperJohnGraham has joined #openstack-kolla | 23:19 | |
kfox1111 | if all the gates pass, I'm not sure 1.5.1 will have issues then, as we tested it all. if it fails, we shouldn't merge the review. | 23:20 |
kfox1111 | in parallel to that, the moment the gate breaks, we pin it to 1.4 if the other patch isnt merged? | 23:20 |
kfox1111 | that way we have a stable gate until a 1.5.x stable release is tested and working? | 23:21 |
jascott1 | helm 2.0x will continue to support only 1.4 | 23:21 |
jascott1 | im not sure entirely what he meant by that as far as backporting | 23:21 |
openstackgerrit | sayantani proposed openstack/kolla: Update Quick Start Guide for singlenode deployment https://review.openstack.org/404993 | 23:21 |
jascott1 | i would expect not much | 23:21 |
kfox1111 | jascott1: I do thin kwe should get to 2.1 as soon as possible. | 23:21 |
kfox1111 | jascott1: if they are having severe upgrade bugs, then better to stay as curent as possible. :/ | 23:22 |
jascott1 | can we copy the gate and run both versions for a while/ | 23:22 |
jascott1 | ? | 23:22 |
jascott1 | I guess we should jsut try 1.5 in the gate before we go creating two versions | 23:22 |
kfox1111 | jascott1: possible. but it does start to explode combinatorialy. I was hoping to run both 2.0.2 and 3.0.1 gates son. | 23:22 |
kfox1111 | soon | 23:23 |
kfox1111 | so 4 * 2 * 2 | 23:23 |
sayantan_ | Hi srwilkers | 23:23 |
jascott1 | what kind of guarantees do we make for helm/k8s? | 23:23 |
srwilkers | hey sayantan_ how are you? | 23:23 |
sayantan_ | I am good :) | 23:23 |
sayantan_ | How are you? | 23:24 |
srwilkers | good to hear :) | 23:24 |
sayantan_ | Made the last changes you suggested to the last patch | 23:24 |
sayantan_ | Would be great if you could take a look | 23:24 |
sayantan_ | Thanks for all the comments | 23:24 |
*** Pavo has quit IRC | 23:24 | |
sayantan_ | :) | 23:24 |
srwilkers | Looking through now :) | 23:25 |
sayantan_ | I am yet to update the operators guide which has the pip workflow. Plan to do it tonight. | 23:25 |
sayantan_ | Also, openstack docs team wants to add the Kolla stuff to the openstack install guides | 23:26 |
*** Pavo has joined #openstack-kolla | 23:26 | |
sayantan_ | Once this is through, this will also be added to the main openstack docs (Will have to upload another patch to do that though) | 23:28 |
*** tonanhngo_ has quit IRC | 23:28 | |
*** tonanhngo has joined #openstack-kolla | 23:29 | |
*** mewald has joined #openstack-kolla | 23:31 | |
srwilkers | Sounds good sayantan_. Added a few more comments then it looks good to me :) | 23:31 |
srwilkers | Thanks for working through that | 23:31 |
sayantan_ | Np :) | 23:31 |
jascott1 | anyone have thoughts on compatibility policy wrt helm and k8s? | 23:33 |
*** Pavo has quit IRC | 23:33 | |
*** HyperJohnGraham has quit IRC | 23:35 | |
*** Pavo has joined #openstack-kolla | 23:35 | |
kfox1111 | jascott1: what do you mean? | 23:35 |
jascott1 | what is the answer to: does kolla-k8s support helm version x or k8s version y? | 23:36 |
kfox1111 | portdirect: sorry to do this to you again, but https://review.openstack.org/#/c/408864/ seems to need another rebase. :/ | 23:36 |
jascott1 | kfox1111 maybe this is already written down and I missed it | 23:37 |
portdirect | kfox1111: what id i ever do to you? | 23:37 |
* portdirect sobbs uncontrollably | 23:37 | |
portdirect | :) | 23:37 |
portdirect | np | 23:37 |
kfox1111 | jascott1: we've bet a lot on helm recently, and helms very very new... :/ especially with stuff like https://github.com/kubernetes/helm/releases showing such major issues. | 23:37 |
kfox1111 | jascott1: better to track newer helm I think? | 23:37 |
kfox1111 | portdirect: sorry. :/ | 23:37 |
portdirect | kfox1111: all good - it just means team kolla's cranking out the commits :) | 23:38 |
kfox1111 | yeah. at this rate, we may be able to finish the conversion in the next week or so. | 23:39 |
kfox1111 | a lot of good stuff going in. :) | 23:39 |
kfox1111 | portdirect: maybe break the commit up into a few smaller ones? we can merge each as soon as it goes green. | 23:39 |
jascott1 | kfox1111 i agree. im not sure how to articulate but seems like a question that will come up plus it helps to know what we are shooting for | 23:40 |
srwilkers | yeah, smaller commits seem to be the way to go, especially down to svc/pod/endpoint/etc granularity. some of the reviews get pretty gnarly when they're bunched | 23:41 |
kfox1111 | jascott1: +1 | 23:43 |
portdirect | jascott1, thats an excelent question - I'm currenlty going through a chat with the infra crew to get my wiki account up and running (as i got stuck somewhere) but could someone put it on the ajenda? I've always assumed current k8s reease is where we should be but pehaps not? | 23:43 |
jascott1 | kfox1111 I guess we should look at helm's compatibility policy which from appearance is latest helm works with latest k8s? | 23:43 |
*** Pavo has quit IRC | 23:44 | |
kfox1111 | jascott1: portdirect: yeah. ideally once we get closer to a stable release, we souldn't chace the version so closely. but if keeping current doesn't slow us down too much, staing current is probably better? | 23:44 |
srwilkers | i think thats a good approach jascott1. i think it'll be easier once kolla-k8s isnt in the middle of an architecture change too. seems the helm adoption coupled with a new k8s release is a bit of a unique situation | 23:44 |
portdirect | jascott1: they are trying to i think, but the are still building master against the 1.41 client (unlees the pr to fix that has merged in the last 12 hours) | 23:44 |
kfox1111 | srwilkers: yeah. | 23:45 |
srwilkers | itll be easier to plan once we're closer to a stable release | 23:45 |
portdirect | jascott1: and it merged 6 hours ago! https://github.com/kubernetes/helm/pull/1657, happy days | 23:46 |
kfox1111 | yeah. if helm doesn't have a release soon, and k8s 1.5 hits, we for sure should pin at k8s 1.4 until at least helm is updated. | 23:46 |
jascott1 | portdirect idk but 2.1 is supposed to drop today | 23:46 |
portdirect | of what? | 23:46 |
jascott1 | helm 2.1 | 23:46 |
portdirect | helm 2.1 is jan the 7th, i thought? https://github.com/kubernetes/helm/milestone/22 | 23:46 |
jascott1 | compatible with 1.5 | 23:47 |
jascott1 | no they split it | 23:47 |
*** Pavo has joined #openstack-kolla | 23:47 | |
portdirect | oh nice :) | 23:47 |
jascott1 | so sayeth technosophose | 23:47 |
jascott1 | almost got it | 23:47 |
portdirect | well he would know - need to get back on to him re the configmap/secrets as files stuff, it would really help us out :) | 23:48 |
jascott1 | ya he didnt say wen 2.2 is landing | 23:48 |
jascott1 | 13 mins ago | 23:48 |
jascott1 | https://github.com/kubernetes/helm/releases | 23:48 |
jascott1 | 2.1 | 23:48 |
jascott1 | sooo guess we will see about compatibility soon enough as the gate updates ;) | 23:50 |
*** JRobinson__ has quit IRC | 23:51 | |
kfox1111 | jascott1: making a pr | 23:51 |
portdirect | nice - I've been running both locally and I think It should go over without any hicups :) | 23:52 |
*** tyrola has joined #openstack-kolla | 23:52 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Switch to helm 2.1.0 https://review.openstack.org/410464 | 23:53 |
kfox1111 | now the only question will be, when will k8s 1.5 land in the repo's. | 23:53 |
*** JRobinson__ has joined #openstack-kolla | 23:54 | |
*** mewald has quit IRC | 23:54 | |
portdirect | kubeadm can set up the beta's fine | 23:55 |
openstackgerrit | sayantani proposed openstack/kolla: Update Quick Start Guide for singlenode deployment https://review.openstack.org/404993 | 23:56 |
kfox1111 | yeah. but we just yum/apt-get install some stuff and tell it to go. | 23:56 |
kfox1111 | so whatever is in the repo will get used. | 23:56 |
kfox1111 | the default still seems to be 1.4 | 23:56 |
*** sacharya has quit IRC | 23:57 | |
kfox1111 | yeah. a yum clean all. yum upgrade shows kubectl/kubelet at 1.4.4-3 still. | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!