sbezverk | kfox1111: it seems ok I run it about 4 times and everything was ok, this time everything is green. I think it is a good to be merged | 00:06 |
---|---|---|
*** sdake has joined #openstack-kolla | 00:09 | |
*** crushil has joined #openstack-kolla | 00:13 | |
*** sdake has quit IRC | 00:14 | |
kfox1111 | kk | 00:18 |
kfox1111 | heh... | 00:23 |
kfox1111 | tarballs.o.o seems less stable then the hub. :/ | 00:23 |
*** crushil has quit IRC | 00:33 | |
*** sdake has joined #openstack-kolla | 00:39 | |
kfox1111 | sdake: https://review.openstack.org/#/c/426417/ | 00:45 |
kfox1111 | and https://review.openstack.org/#/c/426438/ | 00:45 |
*** rhallisey has quit IRC | 00:53 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: [TEST][MASTER][KOLLA] https://review.openstack.org/426625 | 00:55 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Pull container tallballs into the gate https://review.openstack.org/426598 | 00:56 |
*** lrensing has quit IRC | 00:58 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Gate fix 4 https://review.openstack.org/426609 | 01:01 |
*** goldyfruit has quit IRC | 01:11 | |
*** crushil has joined #openstack-kolla | 01:16 | |
SamYaple | kfox1111: tarballs.o.o was never meant for that size and load. im really shocked infra was ok with you grabbing it to be honest | 01:21 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Pull container tallballs into the gate https://review.openstack.org/426598 | 01:23 |
kfox1111 | sounds like it. | 01:26 |
kfox1111 | they did it for kolla-ansible too. so maybe they are thinking they will scale it up, if it fails over. | 01:27 |
kfox1111 | faillls | 01:27 |
kfox1111 | falls | 01:27 |
kfox1111 | cant... type... | 01:27 |
*** sdake has quit IRC | 01:28 | |
SamYaple | well the thing is, that mirror syncs, so youre likely pulling from other parts of the world sometimes | 01:33 |
kfox1111 | well, the alternative is to always pull from other parts of the world. | 01:34 |
SamYaple | well the alternative is a registry proxy | 01:38 |
SamYaple | one registry setup someone in openstack infra, push images to taht registry when building. and pulling is done through a proxy so it only "pulls" once per region | 01:39 |
SamYaple | thats the least complicated setup at least | 01:39 |
kfox1111 | perhaps, perhaps not. depends on what other infra they have in place. | 01:40 |
kfox1111 | rsyncing a tarball to a bunch of local servers may be generally easier to do then building it up with docker tools. | 01:40 |
*** v1k0d3n has quit IRC | 02:00 | |
*** v1k0d3n has joined #openstack-kolla | 02:02 | |
openstackgerrit | Jeffrey Zhang proposed openstack/kolla: [TEST][MASTER][KOLLA] https://review.openstack.org/426625 | 02:05 |
*** crushil has quit IRC | 02:06 | |
*** unicell has quit IRC | 02:10 | |
*** goldyfruit has joined #openstack-kolla | 02:12 | |
*** sdake has joined #openstack-kolla | 02:20 | |
*** sdake has quit IRC | 02:20 | |
*** sdake has joined #openstack-kolla | 02:33 | |
sbezverk | sdake: ping | 02:48 |
sdake | sup sbezverk | 02:48 |
sbezverk | we have bunch of important PS which require review and get merged. if you have some time can you check them? | 02:48 |
sdake | sbezverk i can review until i leave my parents house | 02:49 |
sdake | which is probably about 20 minutes out | 02:49 |
sbezverk | ok | 02:50 |
*** rwellum has joined #openstack-kolla | 02:50 | |
sdake | sbezverk are the services packages completed? | 02:51 |
sbezverk | sdake: essential yes.. | 02:52 |
sbezverk | sdake: need more testing but for testing need some fixes | 02:53 |
sdake | sbezverk how about an ack on https://review.openstack.org/#/c/425809/ | 02:59 |
openstackgerrit | Merged openstack/kolla-kubernetes: PS adds glance cleanup service https://review.openstack.org/426438 | 03:00 |
openstackgerrit | Merged openstack/kolla-kubernetes: Add a script to build an example cloud yaml https://review.openstack.org/426417 | 03:01 |
sdake | sbezverk please have a look at https://review.openstack.org/#/c/423973/1 | 03:08 |
sdake | note to all, we dont need perfection - the goal is to refactor as we proceed | 03:09 |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes: Change version from 0.4.0 to 0.5.0 https://review.openstack.org/423957 | 03:10 |
sdake | sbezverk ^^ please have a review - i'd like the version changed to 0.5.0 quickly as to avoid the constant complaint about folks having to do rebases - and also avoid the problem of having to do this right prior to relesae | 03:11 |
sdake | ok - time to drive home and rebase the dev env patch | 03:11 |
sdake | bbl | 03:11 |
*** sdake has quit IRC | 03:12 | |
*** sdake has joined #openstack-kolla | 03:13 | |
openstackgerrit | Merged openstack/kolla-kubernetes: Test keystone endpoint list https://review.openstack.org/425809 | 03:18 |
*** sdake has quit IRC | 03:22 | |
openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: PS Adds keystone cleanup service https://review.openstack.org/426400 | 03:24 |
*** rwellum has quit IRC | 03:27 | |
*** hfu has joined #openstack-kolla | 03:31 | |
*** jrobinson is now known as jrobinson-afk | 03:38 | |
*** japestinho has joined #openstack-kolla | 03:42 | |
*** crushil has joined #openstack-kolla | 03:48 | |
*** goldyfruit has quit IRC | 03:50 | |
*** sdake has joined #openstack-kolla | 03:51 | |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes: Support rolling upgrades of horizon https://review.openstack.org/333996 | 03:59 |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes: hange version from 0.4.0 to 0.5.0 https://review.openstack.org/423957 | 03:59 |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes: Change version from 0.4.0 to 0.5.0 https://review.openstack.org/423957 | 03:59 |
*** crushil has quit IRC | 04:05 | |
*** jrobinson-afk is now known as jrobinson | 04:06 | |
*** lamt has joined #openstack-kolla | 04:16 | |
*** sdake has quit IRC | 04:17 | |
cliles | SamYaple - ok so it took me a few beers and some time to figure it out. | 04:17 |
cliles | I'm using nfs for cinder, so 58eee09c15dea81ec51bb366af23681de70ceaa1 changed the users/uids and threw my world for a loop | 04:18 |
*** sdake has joined #openstack-kolla | 04:24 | |
*** mdnadeem has joined #openstack-kolla | 04:29 | |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes: WIP: Add Development Setup Script for use with Kubernetes Helm https://review.openstack.org/405720 | 04:31 |
*** sdake_ has joined #openstack-kolla | 04:36 | |
*** kristian__ has joined #openstack-kolla | 04:38 | |
*** lamt has quit IRC | 04:38 | |
*** sdake has quit IRC | 04:38 | |
*** lamt has joined #openstack-kolla | 04:41 | |
*** adrian_otto has joined #openstack-kolla | 04:42 | |
*** kristian__ has quit IRC | 04:42 | |
*** yuanying has joined #openstack-kolla | 04:48 | |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes: Add information to QSG about reliance on kolla-ansible https://review.openstack.org/420384 | 04:49 |
*** adrian_otto has quit IRC | 04:51 | |
*** sudipto has joined #openstack-kolla | 04:59 | |
*** sudipto_ has joined #openstack-kolla | 04:59 | |
*** lamt has quit IRC | 05:00 | |
*** lamt has joined #openstack-kolla | 05:03 | |
*** sdake_ has quit IRC | 05:15 | |
SamYaple | cliles: yea thats considered an upgrade bug | 05:26 |
SamYaple | cliles: there is a kolla-ansible side to that patch, that hasnt been written | 05:27 |
*** skramaja_ has quit IRC | 05:28 | |
*** saneax-_-|AFK is now known as saneax | 05:29 | |
*** skramaja has joined #openstack-kolla | 05:36 | |
*** unicell has joined #openstack-kolla | 05:44 | |
*** satyar has joined #openstack-kolla | 05:50 | |
*** kristian__ has joined #openstack-kolla | 06:01 | |
*** unicell1 has joined #openstack-kolla | 06:01 | |
*** unicell has quit IRC | 06:02 | |
*** kristian__ has quit IRC | 06:05 | |
*** lamt has quit IRC | 06:15 | |
*** adrian_otto has joined #openstack-kolla | 06:17 | |
*** jrobinson has quit IRC | 06:35 | |
*** adrian_otto has quit IRC | 06:45 | |
openstackgerrit | Sajauddin proposed openstack/kolla: Kolla support for power server https://review.openstack.org/423239 | 06:56 |
*** kristian__ has joined #openstack-kolla | 06:56 | |
satyar | Hi sean-k-mooney | 07:07 |
*** unicell has joined #openstack-kolla | 07:07 | |
satyar | Hi SamYaple | 07:08 |
*** kristian__ has quit IRC | 07:08 | |
*** bmace_ has quit IRC | 07:08 | |
*** unicell1 has quit IRC | 07:08 | |
*** bmace_ has joined #openstack-kolla | 07:08 | |
*** trinaths has joined #openstack-kolla | 07:14 | |
*** mkoderer has joined #openstack-kolla | 07:26 | |
*** sayantani01 has quit IRC | 07:32 | |
*** sayantani01 has joined #openstack-kolla | 07:33 | |
bmace_ | sdake: around? | 07:34 |
*** sayantani01 has quit IRC | 07:41 | |
*** matrohon has joined #openstack-kolla | 07:52 | |
*** prameswar has joined #openstack-kolla | 08:00 | |
*** Serlex has joined #openstack-kolla | 08:10 | |
*** pcaruana has joined #openstack-kolla | 08:15 | |
*** shardy has joined #openstack-kolla | 08:18 | |
*** saneax is now known as saneax-_-|AFK | 08:27 | |
*** msimonin has joined #openstack-kolla | 08:31 | |
*** saneax-_-|AFK is now known as saneax | 08:36 | |
openstackgerrit | Saverio Proto proposed openstack/kolla: Use become: True when updating /etc/hosts https://review.openstack.org/425191 | 08:37 |
*** egonzalez has joined #openstack-kolla | 08:39 | |
pomac | egonzalez: was that keystone thing a bug or just missaligned packages? | 08:40 |
egonzalez | pomac: missaligned packages, the containar was not replazed with the new version while upgrading | 08:40 |
pomac | egonzalez: ahh ok, thanks =) | 08:40 |
egonzalez | keystone db was in ocata and packages in newton | 08:40 |
egonzalez | as per the talk i had with keystone people, in O need to add default_domain_id option | 08:41 |
pomac | and default_domain_id shouldn't have any real effect if you don't need keystone v2 | 08:43 |
*** jrobinson has joined #openstack-kolla | 08:44 | |
egonzalez | is needed for backward compatible if in some point in time any user was created under default domain | 08:44 |
egonzalez | Jeffrey4l_: are we in feature freeze, right? | 08:47 |
pomac | we use the AD to authenitcate users so, we neeed to define a domain, and users need to be authenticted against that - so, we have to define the default_domain_id | 08:47 |
*** mgoddard has joined #openstack-kolla | 08:48 | |
egonzalez | pomac: AD? once I tried to integrate in keystone and broke the AD, how you do it? | 08:49 |
*** jrobinson has quit IRC | 08:49 | |
pomac | egonzalez: readonly | 08:49 |
egonzalez | I modified the AD schema to make it think like a real LDAP. probably not the best method | 08:50 |
*** jrobinson has joined #openstack-kolla | 08:50 | |
pomac | egonzalez: we have insourced our AD and it now works... more like a "standard" ldap than it ever has =) | 08:52 |
egonzalez | anybody know where are ubuntu "trunk" repositories? | 09:01 |
*** saneax is now known as saneax-_-|AFK | 09:01 | |
*** sudipto_ has quit IRC | 09:10 | |
*** saneax-_-|AFK is now known as saneax | 09:11 | |
*** sudipto has quit IRC | 09:11 | |
*** sudipto_ has joined #openstack-kolla | 09:11 | |
*** sudswas__ has joined #openstack-kolla | 09:13 | |
*** saneax is now known as saneax-_-|AFK | 09:15 | |
*** sayantani01 has joined #openstack-kolla | 09:15 | |
*** zioproto has joined #openstack-kolla | 09:16 | |
*** gfidente has joined #openstack-kolla | 09:19 | |
*** gfidente has quit IRC | 09:19 | |
*** gfidente has joined #openstack-kolla | 09:19 | |
*** sayantani01 has quit IRC | 09:20 | |
*** chopmann has joined #openstack-kolla | 09:28 | |
*** JoseMello has joined #openstack-kolla | 09:28 | |
*** dasTor_ has quit IRC | 09:31 | |
*** dasTor has joined #openstack-kolla | 09:32 | |
*** jrobinson has quit IRC | 09:32 | |
*** mgoddard has quit IRC | 09:48 | |
*** jistr has quit IRC | 09:49 | |
*** jistr has joined #openstack-kolla | 09:51 | |
*** athomas has joined #openstack-kolla | 09:57 | |
*** saneax-_-|AFK is now known as saneax | 09:58 | |
*** sayantani01 has joined #openstack-kolla | 10:09 | |
*** sayantani01 has quit IRC | 10:14 | |
*** mkoderer has quit IRC | 10:16 | |
*** arslanq has joined #openstack-kolla | 10:17 | |
arslanq | Hi guys, i am new on kolla and just wanted to try it out. I went through its documentation and i saw that by default it supports OVS as neutron plugin. Does kolla support OVN integration or i would have to go brown field? | 10:20 |
*** mgoddard has joined #openstack-kolla | 10:21 | |
openstackgerrit | Nenad Radojevic proposed openstack/kolla-ansible: External_fqdn for host_href in barbican.conf https://review.openstack.org/426697 | 10:21 |
*** sambetts|afk is now known as sambetts | 10:28 | |
portdirect | arslanq: at the moment there is no OVN support in Kolla, though it should not be hard to support | 10:28 |
arslanq | portdirect: the only issue em looking forward is scaling out the cluster after brown field. | 10:30 |
portdirect | arslanq: I currently run OVN on one of my containerised deployments and used it as the SDN layer for my harbor project | 10:30 |
egonzalez | arslanq: not implemented yet https://blueprints.launchpad.net/kolla/+spec/ovn-controller-neutron | 10:30 |
arslanq | egonzalez: any timelines for that? | 10:31 |
openstackgerrit | Saverio Proto proposed openstack/kolla-ansible: Fix BASEDIR to package Kolla with snapcraft.io https://review.openstack.org/426703 | 10:32 |
*** pbourke has quit IRC | 10:32 | |
*** pbourke has joined #openstack-kolla | 10:32 | |
egonzalez | arslanq: not sure, the guy working on it has not made any more changes since november. | 10:33 |
egonzalez | arslanq: I think is up to you if want work on it | 10:33 |
arslanq | egonzalez: Sure i would love to contribute. first i need to set my timelines accordingly. Thanks anyhow | 10:34 |
*** satyar has quit IRC | 10:36 | |
egonzalez | arslanq: take your time, we have a full release cycle this time ;) | 10:38 |
egonzalez | ask for help if needed | 10:39 |
*** yuanying_ has joined #openstack-kolla | 10:48 | |
*** yuanying has quit IRC | 10:51 | |
*** jmccarthy has joined #openstack-kolla | 11:01 | |
*** prameswar has quit IRC | 11:01 | |
*** trinaths1 has joined #openstack-kolla | 11:02 | |
*** trinaths has quit IRC | 11:02 | |
*** sayantani01 has joined #openstack-kolla | 11:03 | |
zioproto | I am trying to install the current master of kolla and kolla-ansible. But the command kolla-ansible fails with 'runtime/cgo: pthread_create failed: Resource temporarily unavailable' | 11:07 |
zioproto | I do something wrong during the installation ? | 11:07 |
*** sayantani01 has quit IRC | 11:08 | |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Make service_plugins more configurable https://review.openstack.org/426709 | 11:12 |
egonzalez | zioproto: try using tools/kolla-ansible instead of kolla-ansible command, probably made python setup.py, but the command is not real using master | 11:14 |
*** lamt has joined #openstack-kolla | 11:18 | |
*** lamt has quit IRC | 11:19 | |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Make service_plugins and extensions more configurable https://review.openstack.org/426709 | 11:20 |
*** mliima has joined #openstack-kolla | 11:21 | |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Make service_plugins and extensions more configurable https://review.openstack.org/426709 | 11:23 |
*** suka_ has joined #openstack-kolla | 11:23 | |
*** yuanying_ has quit IRC | 11:25 | |
*** dims has joined #openstack-kolla | 11:26 | |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Make service_plugins and extensions more configurable https://review.openstack.org/426709 | 11:28 |
egonzalez | zioproto: BTW, that error seems to be a docker issue | 11:38 |
des_consolado | morning all, I'm working on the image for neutron-dynamic-routing aka neutron-bgp-dragent again today... I'm a bit confused as to why this seems to have two names though | 11:38 |
des_consolado | there is a deb package in the deb repo's for neutron-bgp-dragent, and there is a neutron-dynamic-routing tarballs in the OpenStack tarballs site... | 11:39 |
egonzalez | zioproto: https://github.com/docker/docker/issues/20096 | 11:39 |
des_consolado | the neutron_dynamic_routing code seems to basically be the python library, and the bgp-dragent is a binary that uses this library, it looks like.... | 11:40 |
egonzalez | des_consolado: binary packages are not named equal as source, even between distributions are different | 11:40 |
*** trinaths1 has quit IRC | 11:41 | |
egonzalez | see dashboards names, in deb python-trove-ui, source trove-dashboard, rpm openstack-trove-dashboard | 11:41 |
des_consolado | egonzalez: okay, so I'm just wondering how I should name this for kolla's purposes: kolla/docker/neutron/neutron-bgp-dragent or kolla/docker/neutron/neutron-dynamic-routing | 11:42 |
des_consolado | because we end up using the same name for the service whether binary or source, don't we? | 11:43 |
des_consolado | in terms of the image/container name at least | 11:43 |
egonzalez | des_consolado: i would just name neutron-bgp-agent | 11:43 |
des_consolado | okay, thanks | 11:43 |
egonzalez | all the services are called neutron-*-agent | 11:44 |
egonzalez | des_consolado: as far I can see there is no binary in rpm repos right? | 11:48 |
egonzalez | des_consolado: can you please open a blueprint for the feature? | 11:50 |
*** skuda has joined #openstack-kolla | 11:50 | |
*** rwellum has joined #openstack-kolla | 11:54 | |
*** sayantani01 has joined #openstack-kolla | 11:57 | |
*** skramaja has quit IRC | 12:00 | |
openstackgerrit | Merged openstack/kolla-ansible: Fix a typo in config.yml https://review.openstack.org/426467 | 12:00 |
des_consolado | egonzalez: yeah there's no rpm binary right now. I'll have a look at blueprints and see what they're all about now. | 12:01 |
*** sayantani01 has quit IRC | 12:02 | |
zioproto | egonzalez, but this happens on the operator seed node where Docker is not installed | 12:05 |
des_consolado | I've posted this blueprint here https://blueprints.launchpad.net/kolla/+spec/neutron-bgp-agent hope it is okay, first time using this | 12:18 |
des_consolado | looks like we're just using this https://wiki.openstack.org/wiki/Blueprints#Blueprints_only_lifecycle and not the 'specs' part above | 12:19 |
openstackgerrit | Merged openstack/kolla-ansible: Unify haproxy prechecks condition order https://review.openstack.org/415510 | 12:25 |
openstackgerrit | Paul Bourke (pbourke) proposed openstack/kolla: Remove unneeded oraclelinux-extras.repo https://review.openstack.org/426735 | 12:25 |
*** jpstnh has joined #openstack-kolla | 12:26 | |
skuda | Hello everyone!, I am pretty new to OpenStack, I want to install a cluster on bare metal servers, not a very big cluster, about 10 machines, I am evaluating the available deployment options, I am trying to choose between OpenStack-Ansible and Kolla-Ansible, I think those are both production ready. I did read a little bit about Stackanetes and OpenStack-helm but from what I understood those are not recommended for p | 12:27 |
skuda | roduction yet. | 12:27 |
skuda | I would like to be able to do live migrations of virtual machines, use lxc in nova-compute and I would love to install using a deployment solution that allows me to have HA setup and upgrade "easily" to next OpenStack versions, I would thank any advice on this. | 12:28 |
*** rhallisey has joined #openstack-kolla | 12:29 | |
skuda | I don't have a strong position on lxc vs docker as I see it's one of the bigger differences between Kolla-Ansible and OpenStack-Ansible, what I would really like is to have something working on top of kubernetes or dc/os but this doesn't seem to be production ready yet. | 12:29 |
pbourke | Jeffrey4l_: hi jeffrey, could you have another look at vbel's response on https://review.openstack.org/#/c/400187/ | 12:36 |
*** shardy is now known as shardy_lunch | 12:37 | |
mliima | hi guys, need reviews here | 12:46 |
mliima | https://review.openstack.org/#/c/422157/ | 12:46 |
*** sayantani01 has joined #openstack-kolla | 12:51 | |
*** sayantani01 has quit IRC | 12:56 | |
*** dave-mccowan has joined #openstack-kolla | 13:00 | |
*** chopmann has quit IRC | 13:00 | |
*** proto_ has joined #openstack-kolla | 13:01 | |
*** zioproto has quit IRC | 13:01 | |
*** mgiles has joined #openstack-kolla | 13:05 | |
*** proto_ has quit IRC | 13:06 | |
*** matrohon has quit IRC | 13:09 | |
*** eaguilar has joined #openstack-kolla | 13:10 | |
*** athomas has quit IRC | 13:11 | |
*** matrohon has joined #openstack-kolla | 13:11 | |
*** mgiles has quit IRC | 13:15 | |
*** shardy_lunch is now known as shardy | 13:18 | |
*** athomas has joined #openstack-kolla | 13:20 | |
*** mdnadeem has quit IRC | 13:21 | |
*** sbezverk has quit IRC | 13:27 | |
*** eaguilar has quit IRC | 13:28 | |
*** saneax is now known as saneax-_-|AFK | 13:29 | |
egonzalez | skuda: one of the differences between OSA and kolla is that we deploy openstack services into docker containers, they into lxc containers | 13:30 |
egonzalez | skuda: that does not means that instances will use nova-lxc driver, nor docker driver | 13:31 |
egonzalez | actually we use libvirt for nova | 13:31 |
*** sbezverk has joined #openstack-kolla | 13:33 | |
*** sbezverk has quit IRC | 13:34 | |
*** sbezverk has joined #openstack-kolla | 13:34 | |
*** JoseMello has quit IRC | 13:35 | |
*** saneax-_-|AFK is now known as saneax | 13:37 | |
*** mkoderer has joined #openstack-kolla | 13:38 | |
*** sayantani01 has joined #openstack-kolla | 13:39 | |
skuda | egonzalez: understood, thanks. I would like to use lxc inside nova instances, it doesn't matter too much for me what technology runs OpenStack itself, I mean if the nova compute nodes can start lxc "virtual machines" | 13:43 |
*** sdake has joined #openstack-kolla | 13:43 | |
egonzalez | ATM kolla does not have implemented lxc driver for nova, don't know if OSA does | 13:43 |
*** sayantani01 has quit IRC | 13:44 | |
egonzalez | morning sdake | 13:45 |
sdake | sup egonzalez | 13:45 |
skuda | ahh that's different, and good to know, thanks, would be difficult for me try to adapt the needed playbook or the docker image? | 13:45 |
*** goldyfruit has joined #openstack-kolla | 13:48 | |
*** srwilkers has joined #openstack-kolla | 13:54 | |
*** sdake has quit IRC | 13:58 | |
sbezverk | sdake: in dev setup script, do you have the part of genconfig? | 13:59 |
openstackgerrit | Francisco Edigleison Barbosa proposed openstack/kolla-ansible: Add how configure backend to cinder HNAS iSCSI driver guide https://review.openstack.org/426769 | 14:05 |
*** eaguilar has joined #openstack-kolla | 14:10 | |
*** sdake has joined #openstack-kolla | 14:11 | |
*** sdake has quit IRC | 14:14 | |
openstackgerrit | Merged openstack/kolla-kubernetes: Added option to use --provide with vagrant. https://review.openstack.org/425148 | 14:15 |
*** satyar has joined #openstack-kolla | 14:18 | |
*** jheroux has joined #openstack-kolla | 14:19 | |
*** eaguilar has quit IRC | 14:21 | |
*** lamt has joined #openstack-kolla | 14:25 | |
openstackgerrit | Merged openstack/kolla-kubernetes: PS Adds keystone cleanup service https://review.openstack.org/426400 | 14:25 |
*** sdake has joined #openstack-kolla | 14:27 | |
*** mattmceuen has joined #openstack-kolla | 14:28 | |
*** fguillot has joined #openstack-kolla | 14:29 | |
*** zioproto has joined #openstack-kolla | 14:29 | |
*** sdake has quit IRC | 14:31 | |
*** sayantani01 has joined #openstack-kolla | 14:33 | |
*** athomas has quit IRC | 14:34 | |
*** jtriley has joined #openstack-kolla | 14:35 | |
*** lamt has quit IRC | 14:36 | |
openstackgerrit | Merged openstack/kolla: Fix the murano policy file path https://review.openstack.org/426237 | 14:36 |
openstackgerrit | Merged openstack/kolla: Update 'addUser' to 'createUser' https://review.openstack.org/426503 | 14:37 |
*** sayantani01 has quit IRC | 14:38 | |
openstackgerrit | Merged openstack/kolla-ansible: Fix security group creation https://review.openstack.org/424690 | 14:43 |
openstackgerrit | Merged openstack/kolla-ansible: Support non standard Python path https://review.openstack.org/424515 | 14:43 |
*** sdake has joined #openstack-kolla | 14:45 | |
*** athomas has joined #openstack-kolla | 14:46 | |
*** eaguilar has joined #openstack-kolla | 14:49 | |
*** m1dev has quit IRC | 14:51 | |
openstackgerrit | Merged openstack/kolla: Update elastic repo to https https://review.openstack.org/425539 | 14:51 |
*** sdake_ has joined #openstack-kolla | 14:54 | |
Pavo | morning gents | 14:54 |
Pavo | question | 14:54 |
Pavo | can the deployer be on the external vip network and able to deploy? | 14:54 |
Pavo | or does it have to be on the internal vip network? | 14:55 |
*** lamt has joined #openstack-kolla | 14:57 | |
*** sdake has quit IRC | 14:57 | |
*** eaguilar has quit IRC | 15:00 | |
*** rmart04 has joined #openstack-kolla | 15:00 | |
*** crushil has joined #openstack-kolla | 15:01 | |
sdake_ | pavo I am not certain it matters | 15:02 |
sdake_ | pavo however the true test is to do it and see | 15:02 |
Pavo | yeah I was thinking the same | 15:03 |
sdake_ | Pavo the key is the deployment node must have ssh access to all of the nodes | 15:03 |
Pavo | as long as the deployer has access to the target nodes it shouldn't matter I would think | 15:03 |
sdake_ | right | 15:03 |
sdake_ | :) | 15:03 |
*** TxGirlGeek has joined #openstack-kolla | 15:04 | |
*** eaguilar has joined #openstack-kolla | 15:05 | |
*** shardy has quit IRC | 15:13 | |
*** adrian_otto has joined #openstack-kolla | 15:14 | |
openstackgerrit | Merged openstack/kolla: Create deliverables.yaml https://review.openstack.org/424784 | 15:18 |
*** adrian_otto has quit IRC | 15:20 | |
*** adrian_otto has joined #openstack-kolla | 15:24 | |
*** saneax is now known as saneax-_-|AFK | 15:30 | |
*** trinaths has joined #openstack-kolla | 15:30 | |
*** trinaths1 has joined #openstack-kolla | 15:34 | |
*** sayantani01 has joined #openstack-kolla | 15:34 | |
*** adrian_otto has quit IRC | 15:37 | |
*** proto_ has joined #openstack-kolla | 15:37 | |
*** adrian_otto has joined #openstack-kolla | 15:38 | |
*** trinaths has quit IRC | 15:38 | |
*** zioproto has quit IRC | 15:40 | |
*** sayantani01 has quit IRC | 15:49 | |
*** sayantani01 has joined #openstack-kolla | 15:50 | |
*** Serlex has left #openstack-kolla | 15:53 | |
openstackgerrit | Merged openstack/kolla-ansible: Add how configure backend to cinder HNAS iSCSI driver guide https://review.openstack.org/426769 | 15:55 |
*** proto_ has quit IRC | 15:55 | |
*** eaguilar has quit IRC | 15:59 | |
*** zioproto has joined #openstack-kolla | 15:59 | |
*** inc0 has joined #openstack-kolla | 16:04 | |
*** sdake_ has quit IRC | 16:06 | |
*** sdake has joined #openstack-kolla | 16:09 | |
openstackgerrit | Steven Dake proposed openstack/kolla: Tidy up deliverables.yaml file https://review.openstack.org/426822 | 16:09 |
sayantani01 | Hi sdake Can you please take a look at https://review.openstack.org/#/c/426352/ when you have time :) | 16:12 |
inc0 | good morning | 16:12 |
sdake | sayantani01 just to be clear, I am not the docs liason | 16:12 |
sdake | sayantani01 but happy to take a look :) | 16:12 |
sayantani01 | Thanks :) | 16:13 |
sdake | sayantani01 mind i ask how you produced that work? | 16:13 |
sdake | did you just copy the files over and do a git commit? | 16:13 |
sdake | sayantani01 have to admit reviewing all that would take days to do properly | 16:14 |
sdake | sayantani01 so need to rely on the workflow you used to do the job | 16:14 |
*** shardy has joined #openstack-kolla | 16:14 | |
sdake | sayantani01 so would like to know which workflow you used :) | 16:14 |
sayantani01 | I used notepad++ to check the difference between the docs | 16:14 |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible: Remove hardcode /etc/kolla path https://review.openstack.org/426827 | 16:14 |
sdake | sayantani01 - ok as long as your committed to fixing complaints that may arise from errors in the change, I'd be happy to workflow it :) | 16:15 |
sayantani01 | sdake Yes, I will be happy to take that up :) | 16:15 |
sdake | inc0 morning dude | 16:15 |
sdake | inc0 can you review https://review.openstack.org/426822 plz | 16:15 |
*** unicell has quit IRC | 16:18 | |
openstackgerrit | Merged openstack/kolla-ansible: Update Kolla-Ansible documents https://review.openstack.org/426352 | 16:18 |
*** rmart04 has quit IRC | 16:20 | |
*** rmart04 has joined #openstack-kolla | 16:22 | |
openstackgerrit | Merged openstack/kolla-ansible: Fix placeholder for 'service_auth' section in neutron_lbaas.conf https://review.openstack.org/422566 | 16:24 |
*** rmart04 has quit IRC | 16:26 | |
kfox1111 | morning. | 16:32 |
*** chopmann has joined #openstack-kolla | 16:33 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Gate fix 4 https://review.openstack.org/426609 | 16:35 |
*** rmart04 has joined #openstack-kolla | 16:36 | |
*** kbaegis has joined #openstack-kolla | 16:37 | |
kfox1111 | sdake: hows the dev env going? | 16:37 |
*** rmart04 has quit IRC | 16:37 | |
sdake | kfox1111 well not swell - i took the weekend off, to you know, catch up on hosuework and stuff :) | 16:37 |
sdake | kfox1111 i have a bunch of info to get the folks working on halcyon | 16:38 |
kfox1111 | kk | 16:38 |
kfox1111 | think its finishable by the ptg? | 16:39 |
*** msimonin has quit IRC | 16:40 | |
sdake | kfox1111 which part? | 16:40 |
sdake | a dev env? | 16:40 |
sdake | god I hope so, I have a live demo of kolla-kubernetes at a meetup in 2 weeks | 16:40 |
sdake | and i don't want to show a pretty picture of gerrit ;) | 16:41 |
kfox1111 | yeah. a devenv. | 16:41 |
kfox1111 | heh. yeah. | 16:41 |
sdake | and say "trust me, it works, in the gate" :) | 16:41 |
sdake | not ideal | 16:41 |
sdake | even if it is clunky - people like to see how the sausage is made | 16:42 |
kfox1111 | +1 | 16:43 |
*** kbaegis has quit IRC | 16:43 | |
openstackgerrit | jianyi proposed openstack/kolla-ansible: Add zun ansible role https://review.openstack.org/416179 | 16:45 |
*** rmart04 has joined #openstack-kolla | 16:46 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Check to see if Horizon is working in the gate. https://review.openstack.org/426025 | 16:47 |
des_consolado | hey how does nova just work inside a container? I see that there are these named volumes, nova_compute, nova_libvirt_qemu, and that on the host there are a couple of processes running, namely qemu-kvm, kvm-irqfd-clean and kvm-pit... | 16:48 |
openstackgerrit | sayantani proposed openstack/kolla-ansible: Update Bifrost documentation https://review.openstack.org/426841 | 16:48 |
*** sdake_ has joined #openstack-kolla | 16:49 | |
inc0 | sdake, no, they don't, and I speak as an citizen of country famous from their sausages....however that might sound | 16:50 |
egonzalez | des_consolado: basically libvirt/nova-c share namespaces with the host | 16:51 |
inc0 | egonzalez, afair nova-c doesn't share pid namespace | 16:51 |
inc0 | libvirt does tho | 16:51 |
inc0 | so vms really run outside pid namespace | 16:51 |
inc0 | des_consolado, as for all the kvm features, kernel is shared in all the containers | 16:52 |
des_consolado | are there some useful commands I can use to look at this that you guys use? | 16:52 |
inc0 | and kvm is kernel feature | 16:52 |
des_consolado | I was just looking at the process tree with ps's f option | 16:52 |
egonzalez | inc0: yeah, you are right | 16:52 |
des_consolado | and I know ip netns | 16:52 |
des_consolado | and docker inspect, of course... | 16:52 |
inc0 | des_consolado, so do ps aux inside container and outside | 16:53 |
*** adrian_otto has quit IRC | 16:53 | |
des_consolado | but what about all these other namespaces? I kind of understand network namespaces at least to have the simple idea... I have no idea what's happening under the hood with the other namespaces tho | 16:53 |
*** sdake has quit IRC | 16:53 | |
inc0 | you'll see difference in most of the containers | 16:53 |
inc0 | but ps aux in libvirt container will show all the host processes | 16:53 |
*** trinaths1 has quit IRC | 16:54 | |
*** matrohon has quit IRC | 16:59 | |
*** kbaegis has joined #openstack-kolla | 17:00 | |
sdake_ | inc0 right - its a metaphor dude :) | 17:01 |
*** adrian_otto has joined #openstack-kolla | 17:01 | |
inc0 | what is? pid namespace? | 17:02 |
*** japestinho_ has joined #openstack-kolla | 17:02 | |
openstackgerrit | sayantani proposed openstack/kolla-ansible: Update Multinode deployment of kolla guide https://review.openstack.org/426852 | 17:02 |
inc0 | nah, I think it's pretty literal | 17:02 |
sdake_ | inc0 sausage making | 17:02 |
inc0 | I got that one;) | 17:02 |
sdake_ | ;) | 17:02 |
inc0 | just...well...it's pretty nasty | 17:02 |
inc0 | don't dig up too much;) | 17:02 |
*** rmart04 has quit IRC | 17:02 | |
sdake_ | right - sort of like creating software :) | 17:02 |
*** kbaegis has quit IRC | 17:02 | |
kfox1111 | heh. yeah. you don't wana know how its made, only that it is good. :) | 17:03 |
inc0 | exactly | 17:03 |
kfox1111 | https://review.openstack.org/#/c/426609/ please | 17:03 |
*** jpstnh has quit IRC | 17:03 | |
des_consolado | so why is it that you can see all those other processes from the libvirt container? I just looked at all that playbooks in tasks/ for the nova role... | 17:03 |
des_consolado | nothing jumped out as being something that would do anything with binding extra more potent shit onto the libvirt container | 17:04 |
*** kbaegis has joined #openstack-kolla | 17:04 | |
inc0 | des_consolado, libvirt is one container that doesn't separate pids | 17:04 |
kfox1111 | des_consolado: look for something that says something like --hostpid | 17:05 |
inc0 | if it would, every upgrade would mean destruction of vms - you don't want that | 17:05 |
des_consolado | ah pid_mode | 17:05 |
inc0 | yup | 17:05 |
kfox1111 | inc0: I kind of assumed upgrading libvirt would do that anyway, | 17:05 |
kfox1111 | as the destruction of the contiainer kills all children? | 17:05 |
inc0 | kfox1111, negative, vms stay untouched | 17:06 |
kfox1111 | or am I mistaken and hostpid works differnetly? | 17:06 |
*** sdake has joined #openstack-kolla | 17:06 | |
inc0 | so this is funny and scary | 17:06 |
inc0 | it does work like you describe if you run qemu | 17:06 |
inc0 | if you run qemu+kvm, vms stays alive | 17:06 |
kfox1111 | uh... why is that? | 17:06 |
inc0 | probably because of vms being kernel processes | 17:06 |
inc0 | with kvm | 17:06 |
inc0 | but I have nothing to prove that, only that it works like this | 17:07 |
SamYaple | kfox1111: libvirt launches the qemu process in a different cgroup | 17:07 |
*** kbaegis has quit IRC | 17:07 | |
SamYaple | and host pid namespace | 17:07 |
kfox1111 | ah... | 17:07 |
kfox1111 | and docker doesn't cleanup child cgroups? | 17:08 |
SamYaple | well, its not a child cgroup | 17:08 |
SamYaple | but in general, with host things it wont clean them up if you don't | 17:08 |
kfox1111 | hmm... ok. | 17:08 |
inc0 | SamYaple, but remember? it works differently if you run just qemu | 17:09 |
*** chopmann has quit IRC | 17:09 | |
*** sdake_ has quit IRC | 17:09 | |
inc0 | it actually kills vms then | 17:09 |
SamYaple | inc0: it doesnt launch it in a differente cgroup then | 17:09 |
kfox1111 | did we ever slide the kubectl command into any of the toolbox containers? | 17:11 |
kfox1111 | I've got a bug that might be easy to fix if so. | 17:11 |
kfox1111 | looks like kubetoolbox has it. though an old version. | 17:13 |
kfox1111 | hmm... there's not a 4.x version on the hub though. | 17:14 |
inc0 | I'm pretty sure it was in fencing container | 17:15 |
inc0 | need +2 on this https://review.openstack.org/#/c/426105/ | 17:16 |
*** msimonin has joined #openstack-kolla | 17:18 | |
*** zioproto has quit IRC | 17:21 | |
*** kbaegis has joined #openstack-kolla | 17:21 | |
*** msimonin has quit IRC | 17:21 | |
*** msimonin has joined #openstack-kolla | 17:21 | |
*** msimonin has quit IRC | 17:22 | |
*** msimonin has joined #openstack-kolla | 17:22 | |
*** msimonin has quit IRC | 17:22 | |
*** msimonin has joined #openstack-kolla | 17:22 | |
*** msimonin has quit IRC | 17:23 | |
*** msimonin has joined #openstack-kolla | 17:23 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Check to see if Horizon is working in the gate. https://review.openstack.org/426025 | 17:25 |
*** saneax-_-|AFK is now known as saneax | 17:25 | |
*** msimonin has quit IRC | 17:25 | |
*** msimonin has joined #openstack-kolla | 17:25 | |
*** msimonin has quit IRC | 17:26 | |
*** msimonin has joined #openstack-kolla | 17:26 | |
kfox1111 | sbezverk: portdirect: sdake: inc0: srwilkers: if we have some support packages for testing, should we make a helm/tests rather then put them in helm/services or helm/microservices? | 17:26 |
*** msimonin has quit IRC | 17:27 | |
*** eaguilar has joined #openstack-kolla | 17:27 | |
kfox1111 | that might indicate to folks that they shouldn't use them for production. (selenium/ceph mon/ost) | 17:27 |
sbezverk | kfox1111: what do you mean support packages? | 17:28 |
openstackgerrit | sayantani proposed openstack/kolla-ansible: Fix pushing images to local registry documentation https://review.openstack.org/426858 | 17:28 |
sbezverk | kfox1111: maybe helm/tools then? | 17:28 |
kfox1111 | like, for the horizon test, I made a selenium package. not really something that should end up in our shipped kolla repo I think. | 17:29 |
kfox1111 | same with ceph for testing. | 17:29 |
kfox1111 | the altenative might be to make sure test is in the title? | 17:29 |
*** egonzalez has left #openstack-kolla | 17:29 | |
kfox1111 | test-selenium or test-ceph-mon ? | 17:29 |
kfox1111 | tools might work too. | 17:29 |
sbezverk | kfox1111: helm/tools/tests/selenium ? | 17:29 |
srwilkers | kfox1111, i think it'd be appropriate to use helm/tools how sbezverk is mentioning | 17:30 |
sbezverk | we could have other items under tools, so separate from tests make sense I think | 17:30 |
kfox1111 | a thrid option is helm/tools/test-selenium | 17:30 |
sbezverk | kfox1111: would be good to keep tests in a single location | 17:31 |
sbezverk | othewise they will mixup with other things | 17:31 |
kfox1111 | +1 | 17:31 |
kfox1111 | I guess the question is, will we ever ship the test stuff too? | 17:32 |
sbezverk | then let's not put it in helm | 17:32 |
kfox1111 | I guess maybe if we want a dev env that works without needing to do a git checkout. | 17:32 |
sbezverk | we already have tests and tools | 17:32 |
sbezverk | create tests/helm/selenium | 17:32 |
kfox1111 | hmm... that would work too. | 17:33 |
openstackgerrit | sayantani proposed openstack/kolla: Fix pushing images to local registry https://review.openstack.org/426338 | 17:36 |
*** eaguilar has quit IRC | 17:39 | |
*** eaguilar has joined #openstack-kolla | 17:41 | |
*** adrian_otto has quit IRC | 17:42 | |
*** crushil has quit IRC | 17:43 | |
*** rmart04 has joined #openstack-kolla | 17:45 | |
*** vhosakot has joined #openstack-kolla | 17:52 | |
*** newmember has quit IRC | 17:54 | |
*** mgoddard has quit IRC | 17:57 | |
*** eaguilar has quit IRC | 17:58 | |
*** adrian_otto has joined #openstack-kolla | 17:59 | |
*** rmart04 has quit IRC | 17:59 | |
*** gfidente is now known as gfidente|afk | 18:00 | |
*** kbaegis has quit IRC | 18:02 | |
*** adrian_otto has quit IRC | 18:02 | |
*** harlowja has joined #openstack-kolla | 18:02 | |
*** absubram has joined #openstack-kolla | 18:04 | |
*** kbaegis has joined #openstack-kolla | 18:07 | |
*** eaguilar has joined #openstack-kolla | 18:09 | |
*** rmart04 has joined #openstack-kolla | 18:12 | |
*** rmart04 has quit IRC | 18:12 | |
*** athomas has quit IRC | 18:15 | |
*** sudipto_ has quit IRC | 18:15 | |
*** sudswas__ has quit IRC | 18:15 | |
*** unicell has joined #openstack-kolla | 18:15 | |
sdake | kfox1111 ++ | 18:22 |
sdake | kfox1111 i think helm/tests sounds good, or alternately tests/helm | 18:22 |
sdake | kfox1111 i think we dont want to mix up the charts to deploy openstack with the charts to test kolla-kubernetes in the same dirs | 18:23 |
*** crushil has joined #openstack-kolla | 18:24 | |
kfox1111 | what about taggingn the test packages explicitly in the name? | 18:25 |
*** jrich523 has quit IRC | 18:26 | |
*** crushil has quit IRC | 18:30 | |
*** crushil has joined #openstack-kolla | 18:31 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 18:32 |
*** saneax is now known as saneax-_-|AFK | 18:33 | |
*** tonanhngo has joined #openstack-kolla | 18:34 | |
*** crushil has quit IRC | 18:38 | |
*** sayantani01 has quit IRC | 18:42 | |
*** mgoddard has joined #openstack-kolla | 18:44 | |
*** tyrola has quit IRC | 18:46 | |
*** tyrola has joined #openstack-kolla | 18:47 | |
openstackgerrit | Merged openstack/kolla: Tidy up deliverables.yaml file https://review.openstack.org/426822 | 18:56 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 18:57 |
*** jrich523 has joined #openstack-kolla | 18:57 | |
*** sayantani01 has joined #openstack-kolla | 19:04 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 19:05 |
jascott1 | kfox1111 you are trying to validate all templates in the gate? | 19:06 |
kfox1111 | jascott1: yeah. we use to do that with kollakube, but stopped doing it when helm came about. | 19:07 |
kfox1111 | its problematic, cause features regress sometimes. | 19:07 |
kfox1111 | for example, I want to require all containers have an explicit imagePullPolicy. | 19:07 |
kfox1111 | I could go in and add them all.b | 19:07 |
kfox1111 | but if someone else adds a new template, they may forget. | 19:07 |
jascott1 | ah gotcha | 19:08 |
kfox1111 | but if I put in a gate check, it would catch the problem right away. | 19:08 |
jascott1 | +1 for consistency check | 19:08 |
kfox1111 | helm's making it rather hard though. :/ | 19:08 |
kfox1111 | ideally, I'd like to use native helm's templating as a plugin can get out of sync. | 19:09 |
kfox1111 | but that requires the server to be up, which is a heavy lift. | 19:09 |
jascott1 | right | 19:09 |
kfox1111 | helm-template plugin workss, but still requires building from source, which requries root to install go.... etc. | 19:09 |
jascott1 | why not build it and stick it somewhere? | 19:10 |
kfox1111 | yeah, going to have to try that next. but ideally, the plugin should get a release. | 19:10 |
kfox1111 | or better yet, merged into helm. | 19:10 |
jascott1 | ya | 19:10 |
kfox1111 | that particular use case is pretty importantn I think. | 19:10 |
jascott1 | fightin YAML dragons | 19:11 |
kfox1111 | if a user can't ask for the temlpate rendered for inspection easily, its a problem. | 19:11 |
kfox1111 | and the dry-run/debug flags don't produce proper yaml. | 19:11 |
jemcevoy | I am seeing two strange problems since I has a network glitch last week. Kibana stopped collecting data and horizon will not draw network diagrams... Anyone have an idea what may be wrong? | 19:12 |
*** vhosakot has quit IRC | 19:12 | |
*** eaguilar has quit IRC | 19:14 | |
kfox1111 | jascott1: https://github.com/kubernetes/helm/issues/1887 | 19:19 |
jascott1 | cool | 19:20 |
*** matrohon has joined #openstack-kolla | 19:20 | |
kfox1111 | jascott1: did you see the last comment in: https://github.com/kubernetes/helm/issues/1883 ? | 19:21 |
kfox1111 | curious what you think of that one. | 19:22 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 19:23 |
*** skuda has quit IRC | 19:24 | |
jascott1 | kfox1111 im concerned they wont go for it | 19:24 |
kfox1111 | jascott1: any particular reason? | 19:24 |
kfox1111 | it would really help us out. :/ | 19:25 |
jascott1 | seems like quite a bit of bits for our case | 19:25 |
jascott1 | if others need it and speak up then may be different idk | 19:25 |
kfox1111 | I think it would be a pretty simple ps in the end? | 19:25 |
kfox1111 | I think as people start to have more of their own common libs, it will become more requested. | 19:26 |
jascott1 | yeah i dont think it would difficult | 19:26 |
kfox1111 | how long that takes though, is hard to say. | 19:26 |
kfox1111 | I kind of want to see if we can take parts of kolla-common and upstream them as a common helm lib. | 19:26 |
kfox1111 | the kolla_val_get stuff and the common_service template for example are very generic. | 19:27 |
jascott1 | adding more files and convention is where i would expect resistance | 19:27 |
kfox1111 | I may end up using them for some of my other non kolla packages. | 19:27 |
kfox1111 | ah. | 19:27 |
kfox1111 | well, if it really comes to it, | 19:27 |
kfox1111 | they could be put in values under values.yaml | 19:28 |
kfox1111 | parent_values: | 19:28 |
kfox1111 | a: .... | 19:28 |
kfox1111 | b: ................... | 19:28 |
kfox1111 | where a and b are the chunks taht can be included by the parent. | 19:28 |
jascott1 | right i suspect that might be easier sell | 19:28 |
* kfox1111 shrugs. | 19:28 | |
kfox1111 | I think the file based way is cleaner. | 19:28 |
jascott1 | idk hopefully we can get TS to give us up/down on it | 19:28 |
kfox1111 | but don't really care. | 19:28 |
jascott1 | im doing tags now but will look into this | 19:30 |
kfox1111 | thanks. | 19:31 |
*** sdake has quit IRC | 19:32 | |
*** mgoddard has quit IRC | 19:34 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 19:34 |
kfox1111 | jascott1: you going to the ptg? | 19:43 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 19:44 |
*** rwellum has quit IRC | 19:46 | |
*** mkoderer has quit IRC | 19:46 | |
*** matrohon has quit IRC | 19:47 | |
*** sdake has joined #openstack-kolla | 19:47 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 19:51 |
*** MasterOfBugs has joined #openstack-kolla | 19:52 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 19:58 |
*** jheroux has quit IRC | 20:00 | |
*** shardy has quit IRC | 20:01 | |
*** haplo37 has quit IRC | 20:01 | |
*** TxGirlGeek has quit IRC | 20:02 | |
*** crushil has joined #openstack-kolla | 20:07 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:08 |
*** tonanhngo has quit IRC | 20:09 | |
*** haplo37 has joined #openstack-kolla | 20:10 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:18 |
*** oomichi has joined #openstack-kolla | 20:18 | |
*** TxGirlGeek has joined #openstack-kolla | 20:22 | |
*** eaguilar has joined #openstack-kolla | 20:24 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:24 |
*** sambetts is now known as sambetts|afk | 20:25 | |
*** kbaegis has quit IRC | 20:30 | |
*** kbaegis has joined #openstack-kolla | 20:30 | |
*** newmember has joined #openstack-kolla | 20:30 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:31 |
*** sayantani01 has quit IRC | 20:32 | |
jascott1 | kfox1111 I am going to PTG, yourself? | 20:36 |
kfox1111 | yeah. so I'll see you there. :) | 20:36 |
jascott1 | excellente' | 20:36 |
kfox1111 | maybe we can talk out some more of the helm issues. | 20:37 |
jascott1 | sounds fun to me :) | 20:37 |
kfox1111 | cool. :) | 20:37 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:38 |
kfox1111 | oh.... lame... hmm... | 20:46 |
kfox1111 | http://logs.openstack.org/77/426877/13/check/gate-kolla-kubernetes-python27-ubuntu-xenial/f221376/console.html | 20:46 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:49 |
jascott1 | wrong path? | 20:49 |
kfox1111 | trying to find the path, as its in a tox venv. | 20:50 |
kfox1111 | trying an alternate aproach now... | 20:51 |
jascott1 | glad you know the arcane ways of gate magic | 20:51 |
kfox1111 | heh. only know about tox due to coding the gate. had to learn averything from scratch. :/ | 20:53 |
kfox1111 | closer... | 20:56 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 20:56 |
*** rwellum has joined #openstack-kolla | 20:56 | |
*** sayantani01 has joined #openstack-kolla | 20:57 | |
*** kbaegis has quit IRC | 21:00 | |
*** kbaegis has joined #openstack-kolla | 21:04 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 21:05 |
kfox1111 | its rare changes are being made to the begining of the gate. its kind of nice. :) | 21:05 |
*** sdake has quit IRC | 21:06 | |
*** sdake has joined #openstack-kolla | 21:08 | |
*** skuda has joined #openstack-kolla | 21:09 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 21:11 |
bmace_ | sdake: have you tried to run the dev script against the halycon-vagrant-kubernetes env yet? | 21:12 |
*** kbaegis has quit IRC | 21:12 | |
*** crushil has quit IRC | 21:12 | |
*** kbaegis has joined #openstack-kolla | 21:16 | |
*** crushil has joined #openstack-kolla | 21:21 | |
*** eaguilar has quit IRC | 21:23 | |
*** sayantani01 has quit IRC | 21:29 | |
*** msimonin has joined #openstack-kolla | 21:30 | |
*** Jeffrey4l__ has joined #openstack-kolla | 21:34 | |
*** Jeffrey4l_ has quit IRC | 21:35 | |
*** sayantani01 has joined #openstack-kolla | 21:39 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 21:41 |
bmace_ | sdake: kfox1111: looks like a lot of variables specific to peoples environments are slipping into that dev container review. i think all the defaults in there should be based on the halycon-vagrant-kubernetes env and people can tweak it from there for their own environments, imo | 21:44 |
*** satyar has quit IRC | 21:45 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 21:47 |
sbezverk | kfox1111: man that cloud.yaml is so useful!!! I am doing per service installation, it is so easy to keep all vals together :) | 21:47 |
*** newmember has quit IRC | 21:48 | |
*** jtriley_ has joined #openstack-kolla | 21:49 | |
*** gfidente|afk is now known as gfidente | 21:51 | |
kfox1111 | sbezverk: cool. :) | 21:51 |
*** sayantani01 has quit IRC | 21:51 | |
kfox1111 | that was what I was hoping would happen. :) | 21:51 |
sbezverk | kfox1111: check this out : http://paste.openstack.org/show/596940/ | 21:52 |
*** jtriley has quit IRC | 21:52 | |
sbezverk | kfox1111: why internal service register request tries to use keystone-public?!?!? | 21:52 |
*** tonanhngo has joined #openstack-kolla | 21:53 | |
sbezverk | it is supposed to use internal within the cluster | 21:53 |
kfox1111 | thats weird... | 21:53 |
sbezverk | kfox1111: unless something got screwed by genconfig | 21:54 |
sbezverk | as I needed to regenerate all config files | 21:55 |
*** jtriley_ has quit IRC | 21:55 | |
kfox1111 | looks like defaultDnsName is keystone-public. | 21:55 |
kfox1111 | we may want different defaultDnsNames for public vs admin. | 21:56 |
kfox1111 | I guess I never tested it with a private -admin endpoint after writing it. | 21:56 |
sbezverk | kfox1111: yeah, cause registering service during bootstrap is failing now | 21:57 |
*** sayantani01 has joined #openstack-kolla | 21:57 | |
kfox1111 | yeah. I think the gate is always forcing external now. | 21:57 |
sbezverk | kfox1111: strange though it is working at the gate | 21:57 |
sbezverk | kfox1111: do you want to take care of it? ir you are busy? | 21:58 |
kfox1111 | I'm right here in the code. I can take an initial stab at it. | 21:58 |
sbezverk | kfox1111: appreciate it | 21:59 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: Fix keystone endpoint creation defaults. https://review.openstack.org/426948 | 22:01 |
kfox1111 | sbezverk: see if that fixes it. | 22:01 |
sbezverk | kfox1111: let me try it | 22:02 |
*** breitz has quit IRC | 22:03 | |
*** breitz has joined #openstack-kolla | 22:03 | |
*** rhallisey has quit IRC | 22:05 | |
sbezverk | kfox1111: now I am not sure it is that, because on glance cleanup I see the same error.. | 22:09 |
sbezverk | let me double check configs | 22:10 |
kfox1111 | we probably should break up that job into 3 at some point. | 22:12 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 22:12 |
openstackgerrit | Sam Yaple proposed openstack/kolla: Revert "Removed percona from ubuntu repos" https://review.openstack.org/426953 | 22:13 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 22:19 |
*** crushil has quit IRC | 22:20 | |
sbezverk | kfox1111: using "dns_name" as a variable name I think a bit confusing. because it is not always dns host name will be used, in some cases just public ip addresses, especially for private installation | 22:21 |
sbezverk | using public_address I think would be more appropriet | 22:22 |
sbezverk | cause it could be dns host or just ip address, what do you think? | 22:22 |
*** crushil has joined #openstack-kolla | 22:24 | |
*** tyrola has quit IRC | 22:24 | |
kfox1111 | yeah. I see your point. | 22:25 |
kfox1111 | admin_address/public_address then? | 22:25 |
*** tyrola has joined #openstack-kolla | 22:25 | |
kfox1111 | or we split it to 3 jobs, and just call it address? | 22:25 |
*** unicell has quit IRC | 22:27 | |
*** mgoddard has joined #openstack-kolla | 22:27 | |
sbezverk | kfox1111: yeah | 22:29 |
sbezverk | I think it will be more generic | 22:29 |
sbezverk | kfox1111: creating each endpoint separately is overkill I think but it is up to you | 22:30 |
*** crushil has quit IRC | 22:31 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 22:34 |
kfox1111 | we do it everywhere else. | 22:37 |
kfox1111 | keystone's the last holdout. | 22:38 |
kfox1111 | may be good just for consistency if nothing else. | 22:38 |
sbezverk | kfox1111: in this sure | 22:38 |
sbezverk | but I still see issue :( kfox1111 | 22:38 |
sbezverk | http://paste.openstack.org/show/596943/ | 22:38 |
kfox1111 | hmm. | 22:39 |
sbezverk | now it uses public ip, but instead from inside I has to use keystone-admin or keystone-internal | 22:39 |
*** unicell has joined #openstack-kolla | 22:39 | |
kfox1111 | the extenal address should work from internal... | 22:40 |
kfox1111 | is the nat'ing in the cloud working? | 22:40 |
sbezverk | I guess the problem here is we cannot give one solution for two use cases of openstack command | 22:40 |
kfox1111 | should be able to. in theory external should always work. | 22:41 |
kfox1111 | the questino is, why isnt it working now? | 22:41 |
kfox1111 | whats the kubectl descripbe svc keystone-admin have? | 22:41 |
sbezverk | http://paste.openstack.org/show/596944/ | 22:42 |
sbezverk | and I am positive it was working last friday | 22:42 |
sbezverk | kfox1111: my expectation was keystone-admin/keystone-internal names are used within the cluster since they are resolved by kube dns | 22:44 |
sbezverk | but for public authentication I give public ip address | 22:44 |
sbezverk | I do not need to run openstach service commands from outside so I do not care about external admin access | 22:45 |
sbezverk | it is very basic config I think | 22:45 |
kfox1111 | while the keystone-admin/keystone-public endpoints should be resolvable, nothing will be using those names, as they will look at whats in keystone. | 22:46 |
kfox1111 | sbezverk: then, I think just set admin_port_external=false | 22:47 |
kfox1111 | should be internal only then. | 22:47 |
sbezverk | kfox1111: Oh I thought it is false by default | 22:48 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 22:48 |
kfox1111 | not sure anymore. | 22:48 |
kfox1111 | maybe it got set to true. | 22:48 |
*** adrian_otto has joined #openstack-kolla | 22:48 | |
*** pramodrj07 has joined #openstack-kolla | 22:49 | |
*** MasterOfBugs has quit IRC | 22:51 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 22:57 |
*** mgoddard has quit IRC | 22:59 | |
sbezverk | kfox1111: still the same thing :( | 22:59 |
sbezverk | kfox1111: if you have time I can setup webex to poke around.. | 23:01 |
*** apuimedo has joined #openstack-kolla | 23:01 | |
*** jrobinson has joined #openstack-kolla | 23:02 | |
*** saneax-_-|AFK is now known as saneax | 23:03 | |
*** unicell has quit IRC | 23:05 | |
sbezverk | kfox1111: I think these two are reversed: | 23:05 |
sbezverk | {{- $defaultPublicDnsName := printf "%s-admin" $elementName }} | 23:05 |
sbezverk | {{- $defaultAdminDnsName := printf "%s-public" $elementName }} | 23:05 |
kfox1111 | doh. your right. | 23:06 |
kfox1111 | go ahead and fix the ps if you got a minute. | 23:06 |
sbezverk | let me test it first in my test bed | 23:06 |
*** TxGirlGeek has quit IRC | 23:07 | |
kfox1111 | k | 23:08 |
kfox1111 | trying to figure out why helm template's breaking. :/ | 23:08 |
kfox1111 | really weird. | 23:08 |
*** sdake has quit IRC | 23:09 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:10 |
*** gfidente is now known as gfidente|afk | 23:13 | |
*** adrian_otto has quit IRC | 23:13 | |
*** jrist has quit IRC | 23:13 | |
*** unicell has joined #openstack-kolla | 23:14 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:17 |
openstackgerrit | sayantani proposed openstack/kolla: Minor documentation change https://review.openstack.org/426968 | 23:23 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:23 |
*** lrensing has joined #openstack-kolla | 23:26 | |
*** sayantani01 has quit IRC | 23:27 | |
*** jrist has joined #openstack-kolla | 23:29 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:30 |
*** TxGirlGeek has joined #openstack-kolla | 23:35 | |
*** TxGirlGeek has quit IRC | 23:38 | |
inc0 | guys, deploy gates are red | 23:40 |
kfox1111 | for what repo? | 23:40 |
inc0 | https://review.openstack.org/#/c/426822/ | 23:40 |
inc0 | kolal | 23:40 |
inc0 | kolla | 23:40 |
inc0 | kolla-ansible rly | 23:40 |
kfox1111 | k. | 23:41 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:41 |
sbezverk | kfox1111: check this out, I think it is helm bug http://paste.openstack.org/show/596952/ | 23:45 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:45 |
kfox1111 | looking... | 23:45 |
sbezverk | variable is set to 'false' but if true condition is executed | 23:45 |
*** lrensing has quit IRC | 23:47 | |
kfox1111 | hmm... | 23:47 |
kfox1111 | actualy, I think that one's my bad too. | 23:47 |
kfox1111 | its getting pulled in as a string. | 23:47 |
kfox1111 | and "false" can evaluate to true in gotl when its a string. :/ | 23:48 |
sbezverk | oh it should be raw then? | 23:48 |
kfox1111 | yeah. :/ | 23:48 |
kfox1111 | sorry. | 23:48 |
sbezverk | np | 23:48 |
sbezverk | let me quickly change it to prove | 23:48 |
openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla: Revert "Removed percona from ubuntu repos" https://review.openstack.org/426953 | 23:49 |
*** sdake has joined #openstack-kolla | 23:50 | |
openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla: Revert "Removed percona from ubuntu repos" https://review.openstack.org/426953 | 23:50 |
inc0 | SamYaple, ^ | 23:51 |
inc0 | guys I need eyes on this patchset, ubuntu is totally broken on multinode without it | 23:51 |
sdake | inc0 your removing xtradb support? | 23:51 |
inc0 | no | 23:51 |
sdake | inc0 which raises the question i've asekd multiple times, whwat is xtradb | 23:51 |
inc0 | reverting patch that removed repo for it | 23:52 |
sdake | oh | 23:52 |
sbezverk | kfox1111: ok get_raw has different parameters :( | 23:52 |
inc0 | we thought packages are in main ubuntu repo | 23:52 |
inc0 | but they turned out to be very old | 23:52 |
kfox1111 | sbezverk: yeah. | 23:52 |
openstackgerrit | Steve Baker proposed openstack/kolla: Don't bootstrap working mariadb https://review.openstack.org/426975 | 23:52 |
kfox1111 | sdake: a source closer to upstream but not the distro's packages. | 23:53 |
Pavo | have a mattermost server up for kolla if you guys are interested | 23:53 |
Pavo | http://ddi.hopto.org:8080/signup_user_complete/?id=3ufq6eprabyx3eqaqkbgwoitow | 23:53 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes: WIP: Helm template testing https://review.openstack.org/426877 | 23:54 |
sbezverk | kfox1111: yep, with get_raw it is working properly | 23:58 |
kfox1111 | ok. cool. | 23:59 |
kfox1111 | we're getting so many config options now, its hard to test render them all in the existing gate. | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!