*** jtriley has joined #openstack-kolla | 00:04 | |
*** mdnadeem has quit IRC | 00:05 | |
*** mdnadeem has joined #openstack-kolla | 00:06 | |
*** mdnadeem has quit IRC | 00:07 | |
*** mdnadeem has joined #openstack-kolla | 00:08 | |
*** diogogmt has joined #openstack-kolla | 00:17 | |
*** sdake_ is now known as sdake | 00:19 | |
*** ssurana has quit IRC | 00:21 | |
*** ayoung has joined #openstack-kolla | 00:22 | |
*** godleon has joined #openstack-kolla | 00:41 | |
*** beekhof has quit IRC | 00:42 | |
*** beekhof has joined #openstack-kolla | 00:48 | |
*** zhiwei has joined #openstack-kolla | 00:50 | |
*** zhiwei has quit IRC | 00:54 | |
*** phuongnh has joined #openstack-kolla | 00:59 | |
*** daneyon_ has joined #openstack-kolla | 01:14 | |
*** rhallisey has quit IRC | 01:17 | |
*** v1k0d3n has quit IRC | 01:17 | |
*** daneyon_ has quit IRC | 01:19 | |
*** weiyu has joined #openstack-kolla | 01:21 | |
*** zhiwei has joined #openstack-kolla | 01:25 | |
*** jtriley has quit IRC | 01:27 | |
*** jtriley has joined #openstack-kolla | 01:28 | |
*** zhiwei has quit IRC | 01:30 | |
*** zhiwei has joined #openstack-kolla | 01:37 | |
*** asalkeld has joined #openstack-kolla | 01:39 | |
*** sdake has quit IRC | 01:49 | |
*** tuvenen has quit IRC | 01:50 | |
*** elemoine has quit IRC | 01:51 | |
*** unicell1 has quit IRC | 01:53 | |
*** tuvenen has joined #openstack-kolla | 01:55 | |
*** elemoine has joined #openstack-kolla | 01:56 | |
*** sacharya has joined #openstack-kolla | 01:57 | |
*** ravig has joined #openstack-kolla | 02:08 | |
*** inc0 has joined #openstack-kolla | 02:11 | |
*** inc0 has quit IRC | 02:12 | |
*** jtriley has quit IRC | 02:16 | |
*** sdake has joined #openstack-kolla | 02:16 | |
*** sdake has quit IRC | 02:18 | |
*** jtriley has joined #openstack-kolla | 02:37 | |
*** jtriley has quit IRC | 02:43 | |
*** yuanying has quit IRC | 02:51 | |
*** mummer has joined #openstack-kolla | 03:01 | |
*** jackning has joined #openstack-kolla | 03:10 | |
*** diogogmt has quit IRC | 03:10 | |
jackning | Hi,Jeffrey Zhang,how are you! | 03:28 |
---|---|---|
jackning | I find the endponit ipaddress and the database ipaddress are also reference {{ kolla_internal_fqdn }} | 03:28 |
jackning | for example: | 03:28 |
jackning | cinder_admin_endpoint: "{{ admin_protocol }}://{{ kolla_internal_fqdn }}:{{ cinder_api_port }}/v2/%(tenant_id)s" | 03:28 |
jackning | cinder_database_address: "{{ kolla_internal_fqdn }}" | 03:28 |
jackning | It may not consider the endpoint service and database service are not on the same host. | 03:29 |
jackning | In our project,we decide to use company mariadb cluster instead of default mariadb container. | 03:29 |
jackning | I update the kolla code and test success | 03:29 |
jackning | 1) Add a variable in etc/kolla/globals.yml,in my environment | 03:29 |
jackning | database_address: "${my database server address}" | 03:29 |
jackning | 2) Update all database address ,such as nova | 03:29 |
jackning | nova_database_address: "{{ database_address }}" | 03:29 |
jackning | 3).another action ,enable_mariadb: "no" in /etc/kolla/globals.yml | 03:29 |
jackning | What is your opinion? | 03:29 |
*** neophy has joined #openstack-kolla | 03:29 | |
*** mdnadeem has quit IRC | 03:35 | |
*** mdnadeem has joined #openstack-kolla | 03:35 | |
*** yuanying has joined #openstack-kolla | 03:48 | |
*** daneyon_ has joined #openstack-kolla | 03:57 | |
*** daneyon_ has quit IRC | 04:02 | |
*** mdnadeem_ has joined #openstack-kolla | 04:12 | |
*** ravig has quit IRC | 04:22 | |
*** ayoung has quit IRC | 04:22 | |
*** patchbot has quit IRC | 04:22 | |
*** patchbot has joined #openstack-kolla | 04:22 | |
*** ravig has joined #openstack-kolla | 04:23 | |
*** mummer has quit IRC | 04:27 | |
*** jabari has joined #openstack-kolla | 04:31 | |
*** ravig has quit IRC | 04:32 | |
*** jabari has quit IRC | 04:34 | |
*** dave-mccowan has quit IRC | 04:36 | |
*** jabari has joined #openstack-kolla | 04:40 | |
*** jabari has quit IRC | 04:41 | |
*** jabari has joined #openstack-kolla | 04:41 | |
*** sacharya has quit IRC | 04:57 | |
*** unicell has joined #openstack-kolla | 05:05 | |
*** unicell has quit IRC | 05:09 | |
*** unicell has joined #openstack-kolla | 05:10 | |
*** unicell has quit IRC | 05:15 | |
*** gfidente has quit IRC | 05:15 | |
*** unicell has joined #openstack-kolla | 05:24 | |
*** tfukushima has joined #openstack-kolla | 05:41 | |
*** tfukushima has quit IRC | 05:41 | |
*** tfukushima has joined #openstack-kolla | 05:41 | |
*** phuongnh has quit IRC | 05:48 | |
*** phuongnh has joined #openstack-kolla | 05:48 | |
*** jackning has quit IRC | 05:53 | |
*** unicell has quit IRC | 05:53 | |
*** unicell1 has joined #openstack-kolla | 05:53 | |
*** sacharya has joined #openstack-kolla | 05:57 | |
*** sacharya has quit IRC | 06:02 | |
*** unicell has joined #openstack-kolla | 06:05 | |
*** unicell1 has quit IRC | 06:06 | |
*** unicell1 has joined #openstack-kolla | 06:24 | |
*** unicell has quit IRC | 06:25 | |
*** mikelk has joined #openstack-kolla | 06:34 | |
*** SiRiuS__ has joined #openstack-kolla | 06:38 | |
*** daneyon_ has joined #openstack-kolla | 06:40 | |
*** daneyon_ has quit IRC | 06:44 | |
*** coolsvap_ has joined #openstack-kolla | 06:44 | |
openstackgerrit | Jack Ning proposed openstack/kolla: Make endpoint and database address separately https://review.openstack.org/324238 | 06:45 |
*** sdake has joined #openstack-kolla | 06:57 | |
*** Serlex has joined #openstack-kolla | 07:05 | |
*** Mr_Broken has joined #openstack-kolla | 07:07 | |
*** coolsvap_ has quit IRC | 07:13 | |
*** coolsvap_ has joined #openstack-kolla | 07:15 | |
*** fragatina has quit IRC | 07:18 | |
*** fragatina has joined #openstack-kolla | 07:19 | |
*** phuongnh has quit IRC | 07:20 | |
*** fragatina has quit IRC | 07:23 | |
*** fragatina has joined #openstack-kolla | 07:26 | |
*** neophy has quit IRC | 07:29 | |
*** tyrola has joined #openstack-kolla | 07:29 | |
*** fragatina has quit IRC | 07:31 | |
*** sdake has quit IRC | 07:39 | |
*** fragatina has joined #openstack-kolla | 07:39 | |
*** asalkeld has quit IRC | 07:43 | |
*** fragatina has quit IRC | 07:44 | |
*** sacharya has joined #openstack-kolla | 07:59 | |
*** matrohon has joined #openstack-kolla | 07:59 | |
*** coolsvap_ has quit IRC | 07:59 | |
*** sacharya has quit IRC | 08:03 | |
openstackgerrit | Hui Kang proposed openstack/kolla: Add Kuryr ansible role https://review.openstack.org/298894 | 08:03 |
*** mbound has joined #openstack-kolla | 08:08 | |
openstackgerrit | Hui Kang proposed openstack/kolla: Build openvswitch image using source type https://review.openstack.org/309726 | 08:08 |
openstackgerrit | Zhijiang Hu proposed openstack/kolla: Check avahi-daemon is not running when using haproxy in prechecks https://review.openstack.org/323715 | 08:18 |
*** prithiv has joined #openstack-kolla | 08:28 | |
*** jmccarthy has quit IRC | 08:31 | |
*** jmccarthy has joined #openstack-kolla | 08:32 | |
*** mbound has quit IRC | 08:49 | |
*** sdake has joined #openstack-kolla | 09:13 | |
*** daneyon_ has joined #openstack-kolla | 09:22 | |
*** daneyon_ has quit IRC | 09:27 | |
*** sdake has quit IRC | 09:40 | |
*** tfukushima has quit IRC | 09:47 | |
*** sacharya has joined #openstack-kolla | 10:00 | |
*** rhallisey has joined #openstack-kolla | 10:02 | |
*** neophy has joined #openstack-kolla | 10:03 | |
*** sacharya has quit IRC | 10:04 | |
*** pbourke_ has quit IRC | 10:10 | |
*** pbourke_ has joined #openstack-kolla | 10:10 | |
*** rhallisey has quit IRC | 10:21 | |
*** neophy has quit IRC | 10:22 | |
*** alyson_ has quit IRC | 10:27 | |
*** weiyu has quit IRC | 10:33 | |
tyrola | Hey guys, anyone seen this error already in the mitaka branch? 2016-06-02 12:49:00.570 1 ERROR nova.compute.manager OSError: [Errno 2] No such file or directory: '/var/lib/nova/instances‘ | 10:50 |
tyrola | Seems like the bootstrap container does not bootstrap correctly. If I create this directly nova works | 10:50 |
nilk | permission denied for mkdir in extend_start.sh | 10:52 |
nilk | missing sudo rule maybe? | 10:52 |
nilk | hmm it's actually skipping bootstrap | 10:55 |
*** rhallisey has joined #openstack-kolla | 11:00 | |
openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Add bootstrap method to the quickstart doc https://review.openstack.org/324113 | 11:04 |
*** godleon has quit IRC | 11:09 | |
*** alyson_ has joined #openstack-kolla | 11:16 | |
*** dwalsh has joined #openstack-kolla | 11:17 | |
*** Mr_Broken has quit IRC | 11:32 | |
*** prithiv has quit IRC | 11:32 | |
*** Mr_Broken has joined #openstack-kolla | 11:32 | |
*** Mr_Broken has quit IRC | 11:37 | |
*** mikelk has quit IRC | 11:40 | |
*** mikelk has joined #openstack-kolla | 11:41 | |
*** mikelk has quit IRC | 11:49 | |
*** mliima has joined #openstack-kolla | 11:51 | |
*** mliima has quit IRC | 12:00 | |
*** Mr_Broken has joined #openstack-kolla | 12:00 | |
*** sacharya has joined #openstack-kolla | 12:00 | |
*** sacharya has quit IRC | 12:05 | |
*** jtriley has joined #openstack-kolla | 12:06 | |
*** mliima has joined #openstack-kolla | 12:07 | |
*** ravig has joined #openstack-kolla | 12:08 | |
*** prithiv has joined #openstack-kolla | 12:08 | |
*** jtriley has quit IRC | 12:11 | |
mliima | mornign guys | 12:13 |
*** mgoddard_ has joined #openstack-kolla | 12:14 | |
*** mgoddard has quit IRC | 12:14 | |
*** dave-mccowan has joined #openstack-kolla | 12:28 | |
*** v1k0d3n has joined #openstack-kolla | 12:30 | |
*** tyrola has quit IRC | 12:41 | |
*** tyrola has joined #openstack-kolla | 12:42 | |
*** nilk has quit IRC | 12:43 | |
*** nilk has joined #openstack-kolla | 12:47 | |
*** berendt has joined #openstack-kolla | 12:51 | |
*** neophy has joined #openstack-kolla | 12:56 | |
*** zhiwei has quit IRC | 12:56 | |
*** papacz has joined #openstack-kolla | 12:58 | |
*** daneyon_ has joined #openstack-kolla | 12:58 | |
*** daneyon_ has quit IRC | 13:03 | |
*** stvnoyes has quit IRC | 13:04 | |
*** stvnoyes has joined #openstack-kolla | 13:05 | |
*** inc0 has joined #openstack-kolla | 13:09 | |
inc0 | hello | 13:09 |
*** zhiwei has joined #openstack-kolla | 13:11 | |
*** mbound has joined #openstack-kolla | 13:11 | |
inc0 | pbourke_, mandre ping | 13:12 |
pbourke_ | inc0: hey | 13:12 |
mandre | hey inc0 | 13:12 |
inc0 | https://review.openstack.org/#/c/323589/ so I've made this working | 13:12 |
patchbot | inc0: patch 323589 - kolla - Refactor of base Dockerfile to allow custom repo | 13:12 |
inc0 | gates are green, it's functioanl | 13:12 |
inc0 | basically you can customize anything we want to allow with that | 13:12 |
inc0 | would appreciate your feedback | 13:13 |
*** mbound has quit IRC | 13:13 | |
pbourke_ | sure | 13:13 |
*** mbound has joined #openstack-kolla | 13:13 | |
inc0 | also I have idea how to give sdake his DSL without breaking Kolla in the process | 13:14 |
mandre | I've seen, it's great | 13:14 |
mandre | I've also been thinking about how to cleanup the build script, while giving us great flexibility | 13:15 |
mandre | I'll paste an example in a minute | 13:16 |
inc0 | well I'd start by breaking it up to submodules | 13:16 |
*** neophy has quit IRC | 13:18 | |
mandre | inc0: http://paste.openstack.org/show/507282/ | 13:21 |
*** ravig has quit IRC | 13:21 | |
inc0 | yay for not looking for a files on disk | 13:21 |
inc0 | also that would make it whole lot easier to implement this approach to drivers: | 13:22 |
mandre | i'm planning to have multiple source for files | 13:22 |
inc0 | https://review.openstack.org/#/c/323974/6 | 13:22 |
patchbot | inc0: patch 323974 - kolla - DO NOT MERGE: demo of how drivers could work | 13:22 |
inc0 | yeah that would be possible too | 13:22 |
*** diogogmt has joined #openstack-kolla | 13:22 | |
mandre | have a kolla_search_path in kolla.conf for example, so that the operator can provide it's own set of files | 13:22 |
mandre | also, I want to break kolla-build.conf into kolla.conf and kolla-build.conf | 13:23 |
inc0 | or just allow this to consume absolute paths | 13:23 |
inc0 | per container | 13:23 |
mandre | kolla.conf will contain the list of enabled features, among other things | 13:23 |
inc0 | and build repos and such? | 13:23 |
mandre | and is read by the build script and the deployment script | 13:23 |
inc0 | yeah I like that | 13:24 |
inc0 | we could use same config twice which is nice | 13:24 |
inc0 | like base image or install type | 13:24 |
inc0 | also driver for neutron for example | 13:24 |
mandre | yep, that's the idea | 13:24 |
inc0 | I'm with you | 13:24 |
inc0 | this would be cool | 13:25 |
inc0 | I hope we can convince sdake to move in this direction rather than throwing everything away | 13:25 |
mandre | sdake is reasonable, I'm sure we can get to an agreement that satisfies everyone | 13:26 |
inc0 | so the idea I had to merge with sdake dsl is for DSL to create customization files | 13:26 |
inc0 | which is essentially what you did in your example | 13:26 |
inc0 | we won't break anything and we'll have even more control | 13:27 |
mandre | we're in violent agreement | 13:27 |
inc0 | as we'll be able to generate customization files by DSL | 13:27 |
inc0 | and then handcraft what's wrong with them | 13:27 |
inc0 | and it won't break Kolla at all | 13:27 |
mandre | yup, while still keeping the dockerfiles around | 13:27 |
inc0 | yup, this is good idea imho, everyone gets what they want | 13:28 |
mandre | we'll make this work | 13:28 |
inc0 | and it doens't have to wait with implementation at all, basic mechanism is already there and all we need to do is polish it and put a lot of hooks into code | 13:29 |
inc0 | we'll make names of hooks based on a schema, so DSL parser can easily guess it | 13:29 |
inc0 | and will inject correct code into them to achieve results | 13:29 |
mandre | and what I like too, is that your deployment tool will read kolla.conf to look for enabled features and doesn't require specific configuration | 13:31 |
*** cloudnautique has joined #openstack-kolla | 13:31 | |
mandre | and you only build what you deploy | 13:32 |
mandre | obviously | 13:32 |
inc0 | yeah, good point | 13:34 |
inc0 | and we can also put build logic into ansible | 13:35 |
inc0 | if not image -->> build it | 13:35 |
inc0 | or have "build" playbook which will read conf and build stuff you need | 13:36 |
Mech422 | anyone else getting a hash sum mismatch trying to apt-get update apt.dockerproject.org ? | 13:36 |
inc0 | Mech422, try to run --no-cache | 13:36 |
inc0 | build | 13:36 |
inc0 | ahh | 13:36 |
inc0 | you're not building;) | 13:36 |
mandre | yep, haven't though of that, would be a neat feature | 13:36 |
inc0 | did docker break their repo? | 13:36 |
inc0 | well, I'm sure much more would come | 13:37 |
inc0 | features | 13:37 |
Mech422 | inc0: oh - good morning btw :-) Didn't realize anyone was up yet ... | 13:38 |
*** reddydodda has quit IRC | 13:38 | |
Mech422 | that what I get for typing while scrolled back :-P | 13:38 |
inc0 | hehe | 13:38 |
inc0 | I just woken up | 13:38 |
inc0 | so don't expect me to solve hard stuff yet;) | 13:38 |
inc0 | we also had another heavy storm in the middle of night | 13:38 |
Mech422 | inc0: the changes to remove OVS db container seem to work like a charm | 13:40 |
inc0 | good, I'm glad | 13:41 |
Mech422 | inc0: but I notice that (cinder? glance?) is using the wrong address to try and pull cephx keys | 13:41 |
Mech422 | inc0: its using the 'main' address of the box, and not the IP on the 'storage net' | 13:41 |
inc0 | hmm, I thought ansible does that | 13:41 |
inc0 | well, that's good | 13:41 |
inc0 | as OSD communicates with mons using management | 13:42 |
inc0 | the actual storage comes from osd -> client | 13:42 |
Mech422 | inc0: ahh - well nothing was listening on that IP on port 6789...so it fails | 13:42 |
inc0 | and client should communicate over storage net | 13:42 |
Mech422 | inc0: I'm just rebuilding everything to track it down further | 13:42 |
inc0 | technically mons doesn't even have to be connected to storage net | 13:43 |
inc0 | (unless I'm mistaken) | 13:43 |
Mech422 | well, something should probably be listening on the port we try to get the cephx keys from... | 13:43 |
Mech422 | otherwise, we end up with VERY secure storage :-) | 13:44 |
inc0 | well, so keys are distributed by ansible | 13:44 |
*** ravig has joined #openstack-kolla | 13:44 | |
inc0 | or maybe not.. | 13:45 |
inc0 | https://github.com/openstack/kolla/blob/master/ansible/roles/ceph/tasks/distribute_keyrings.yml | 13:45 |
inc0 | so I think it gets cephx from mons and then generate populated keyring | 13:45 |
*** mgoddard__ has joined #openstack-kolla | 13:46 | |
inc0 | and it pushes this keyring to osds | 13:46 |
*** reddydodda has joined #openstack-kolla | 13:46 | |
Mech422 | inc0: http://pastebin.com/ED9xATM2 | 13:46 |
inc0 | ahh | 13:46 |
inc0 | ok yeah, I forgot about that | 13:47 |
inc0 | I guess you've found a bug;)_ | 13:47 |
inc0 | we either should make glance to talk to mons over mgmt network | 13:47 |
Mech422 | It's my techno-tard field... makes code bitrot in realtime :-) | 13:47 |
inc0 | as stor network should be used on monmaps really | 13:47 |
inc0 | are you sure deploying Kolla is such a good idea? Maybe you want to break other stuff?:P | 13:48 |
inc0 | j/k | 13:48 |
Mech422 | inc0: yeah - I have to play with Kolla and give my poor toaster and VCR a chance to recover... | 13:48 |
Mech422 | :-) | 13:48 |
*** mgoddard_ has quit IRC | 13:49 | |
inc0 | toaster, now that's an peculiar place to setup openstack | 13:49 |
inc0 | do you hate your toasts that much? | 13:49 |
Mech422 | virtual toast.... | 13:49 |
Mech422 | all the taste, half the calories! | 13:49 |
openstackgerrit | Christian Berendt proposed openstack/kolla: Document the use of external Ceph journal drives https://review.openstack.org/324539 | 13:49 |
inc0 | or vtoast;) | 13:49 |
Mech422 | inc0: my field must be getting stronger, as it broke dockerproject.org hash too | 13:50 |
Mech422 | inc0: I wonder what the range is now.... | 13:50 |
inc0 | well, since most likely their servers are in US | 13:50 |
inc0 | fairly big I guess | 13:50 |
inc0 | unless it's ttl issue | 13:51 |
pbourke_ | inc0: lol at the jinja branch name | 13:58 |
pbourke_ | inc0: the install_yum macro isn't used anywhere? | 13:59 |
*** sacharya has joined #openstack-kolla | 14:01 | |
inc0 | not yet | 14:04 |
inc0 | I'm playing with ideas now | 14:04 |
inc0 | for refactoring | 14:04 |
inc0 | I think I'll break up customization to one patch and refactorings to another | 14:04 |
cloudnautique | Hello, I am pretty close to having Kolla all running, but RabbitMQ clusterer keeps dying and I can't really track down the issue. Any thoughts on where to look? | 14:04 |
cloudnautique | I'm running CentOS on CentOS | 14:05 |
cloudnautique | running off of master essentially | 14:05 |
inc0 | cloudnautique, it restarts or can't deploy? | 14:05 |
cloudnautique | restarts | 14:05 |
cloudnautique | infinite loop | 14:05 |
*** ayoung has joined #openstack-kolla | 14:05 | |
inc0 | /var/lib/docker/volumes/kolla_log/_data/ | 14:05 |
inc0 | check the logs plz | 14:05 |
*** Mr_Broken has quit IRC | 14:05 | |
cloudnautique | clusterer stopping node now | 14:06 |
inc0 | hmm | 14:06 |
*** sacharya has quit IRC | 14:06 | |
*** Mr_Broken has joined #openstack-kolla | 14:06 | |
inc0 | well, check rabbitmq logs, might have some clue | 14:06 |
cloudnautique | rabbit logs are blank | 14:07 |
cloudnautique | :-\ | 14:07 |
*** wmiller has quit IRC | 14:09 | |
cloudnautique | I'm not terribly familiar with Rabbit MQ, would the configured IP/hostname need to be /etc/hosts | 14:10 |
*** wmiller has joined #openstack-kolla | 14:11 | |
inc0 | yeah | 14:12 |
inc0 | but that should be caught by playbook | 14:12 |
inc0 | cloudnautique, docker logs rabbitmq | 14:12 |
cloudnautique | also, rather boring... copy config files and then executing /usr/sbin/rabbitmq-server | 14:13 |
cloudnautique | followed by Clusterer stopping node now | 14:14 |
cloudnautique | is there a way to crank up verbosity to an 11? :) | 14:14 |
inc0 | hmf...I can't say I'm superb at clusterer as well and I haven't seen anything like it before | 14:14 |
inc0 | can you try to clean everything, rebuild and redeploy? | 14:15 |
inc0 | also try stable/mitaka instead of master plz | 14:15 |
cloudnautique | OK. | 14:18 |
cloudnautique | is kolla-ansible cleanup sufficient? | 14:19 |
*** diogogmt has quit IRC | 14:19 | |
*** ayoung has quit IRC | 14:19 | |
mliima | cloudnautique, yes | 14:19 |
cloudnautique | thank you | 14:20 |
mliima | cloudnautique, aio or multinode? | 14:20 |
cloudnautique | currently AIO. Though I will be breaking that apart to controller/compute | 14:21 |
cloudnautique | at somepoint | 14:21 |
*** diogogmt has joined #openstack-kolla | 14:21 | |
cloudnautique | I AM running this in a slightly abnormal way in that I'm running Ansible from a container to deploy onto the host | 14:21 |
inc0 | shouldnt really matter | 14:22 |
inc0 | ansible is glorified ssh | 14:23 |
pbourke_ | inc0: so can I put all my custom blocks in the one override file? | 14:23 |
inc0 | yeah | 14:24 |
inc0 | in any container | 14:24 |
pbourke_ | sweet | 14:24 |
inc0 | and override file on its own is a jinja | 14:24 |
pbourke_ | need to make that clearer in the review | 14:24 |
inc0 | so you can also break it down with includes if you're more powerish user | 14:24 |
pbourke_ | I'll add a few comments | 14:24 |
inc0 | yeah, obviously, I'll work on it today | 14:24 |
inc0 | I'll break out pseudo refactors for now | 14:25 |
inc0 | to a separate ps | 14:25 |
*** ravig has quit IRC | 14:27 | |
*** dwalsh has quit IRC | 14:34 | |
*** dwalsh has joined #openstack-kolla | 14:35 | |
*** ayoung has joined #openstack-kolla | 14:36 | |
*** david-lyle has joined #openstack-kolla | 14:44 | |
*** diogogmt has quit IRC | 14:44 | |
*** ppowell has joined #openstack-kolla | 14:46 | |
*** zhiwei has quit IRC | 14:48 | |
*** thumpba has joined #openstack-kolla | 14:49 | |
*** ppowell has quit IRC | 14:52 | |
*** Lyncos has left #openstack-kolla | 14:55 | |
*** mark-casey has joined #openstack-kolla | 14:57 | |
*** ppowell has joined #openstack-kolla | 14:59 | |
*** thumpba has quit IRC | 15:00 | |
*** thumpba has joined #openstack-kolla | 15:01 | |
*** jtriley has joined #openstack-kolla | 15:03 | |
*** The_Ball has joined #openstack-kolla | 15:04 | |
*** diogogmt has joined #openstack-kolla | 15:04 | |
The_Ball | I've setup a cinder with kolla a while back, I'm not sure which version I'm on however, what's and easy way of telling? | 15:05 |
*** thumpba has quit IRC | 15:05 | |
*** ppowell has quit IRC | 15:07 | |
*** mgkwill has quit IRC | 15:07 | |
*** huikang has joined #openstack-kolla | 15:08 | |
*** mgkwill has joined #openstack-kolla | 15:09 | |
inc0 | version of cinder? | 15:10 |
inc0 | well in any case, it's the same as kollas | 15:10 |
inc0 | master->master | 15:10 |
inc0 | stable/mitaka->stable/mitaka | 15:11 |
*** thumpba has joined #openstack-kolla | 15:17 | |
*** ppowell has joined #openstack-kolla | 15:19 | |
*** jtriley has quit IRC | 15:19 | |
*** vhosakot has joined #openstack-kolla | 15:26 | |
*** sacharya has joined #openstack-kolla | 15:46 | |
*** sacharya has quit IRC | 15:46 | |
openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla: Add .vimrc to .gitignore https://review.openstack.org/324628 | 15:52 |
*** mgoddard__ has quit IRC | 15:57 | |
*** matrohon has quit IRC | 15:57 | |
*** mgoddard has joined #openstack-kolla | 15:59 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Use debian repos for debian base docker image https://review.openstack.org/288936 | 16:00 |
The_Ball | inc0, oh ok, so if you use the master branch of kolla it will pull the master branch of cinder et al | 16:00 |
inc0 | The_Ball, correct | 16:00 |
The_Ball | neat :) | 16:01 |
inc0 | well if you use binary install it's a bit different as we have only daily builds for centos and builds every few weeks for ubuntu | 16:01 |
inc0 | but source is straight master | 16:01 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Use debian repos for debian base docker image https://review.openstack.org/288936 | 16:04 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make container dind unpin old docker relase https://review.openstack.org/297434 | 16:04 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make heka container build with debian https://review.openstack.org/297737 | 16:05 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make nova-libvirt container build with debian https://review.openstack.org/297169 | 16:07 |
*** ayoung has quit IRC | 16:09 | |
*** ayoung has joined #openstack-kolla | 16:10 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: make rabbitmq container build with debian https://review.openstack.org/297728 | 16:10 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make swift-base container build with debian https://review.openstack.org/297121 | 16:13 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make container kolla-toolbox build with debian https://review.openstack.org/297104 | 16:15 |
*** ravig has joined #openstack-kolla | 16:16 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make container dind unpin old docker relase https://review.openstack.org/297434 | 16:16 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make nova-libvirt container build with debian https://review.openstack.org/297169 | 16:17 |
*** ravig has quit IRC | 16:18 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make heka container build with debian https://review.openstack.org/297737 | 16:18 |
*** mgoddard_ has joined #openstack-kolla | 16:19 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: make rabbitmq container build with debian https://review.openstack.org/297728 | 16:20 |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: add curlrc file to base image https://review.openstack.org/301911 | 16:21 |
*** mgoddard has quit IRC | 16:22 | |
*** ravig has joined #openstack-kolla | 16:22 | |
*** david-lyle has quit IRC | 16:24 | |
*** mgoddard_ has quit IRC | 16:24 | |
*** sacharya has joined #openstack-kolla | 16:27 | |
*** ravig has quit IRC | 16:27 | |
*** ravig has joined #openstack-kolla | 16:33 | |
*** ravig_ has joined #openstack-kolla | 16:34 | |
*** ravig has quit IRC | 16:34 | |
*** Mr_Broken has quit IRC | 16:35 | |
*** Mr_Broken has joined #openstack-kolla | 16:35 | |
*** daneyon_ has joined #openstack-kolla | 16:35 | |
*** ravig has joined #openstack-kolla | 16:36 | |
*** dwalsh has quit IRC | 16:38 | |
*** ravig_ has quit IRC | 16:38 | |
*** daneyon_ has quit IRC | 16:40 | |
*** Mr_Broken has quit IRC | 16:40 | |
*** ravig has quit IRC | 16:41 | |
*** ravig has joined #openstack-kolla | 16:43 | |
*** sdake has joined #openstack-kolla | 16:44 | |
*** ravig has quit IRC | 16:44 | |
*** ravig has joined #openstack-kolla | 16:44 | |
sdake | morning | 16:44 |
mliima | morning sdake | 16:45 |
*** ravig_ has joined #openstack-kolla | 16:46 | |
inc0 | yo sdake, did you get my message | 16:48 |
inc0 | ? | 16:48 |
*** prithiv has quit IRC | 16:48 | |
sdake | inc0 i just rolled out of bed | 16:48 |
sdake | i'll get to my email in a moment | 16:48 |
inc0 | long story short, I've found middle ground for us | 16:49 |
vhosakot | morning! | 16:49 |
inc0 | I can give you DSL and it won't break anything | 16:49 |
inc0 | we'll keep backwards compatibility and all | 16:49 |
inc0 | and you'll be able to do whatever the hell you want at per-container manner | 16:49 |
inc0 | key to that: https://review.openstack.org/#/c/324665/ | 16:49 |
patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 16:49 |
*** ravig has quit IRC | 16:49 | |
inc0 | this is fully functional as it is | 16:50 |
*** jtriley has joined #openstack-kolla | 16:50 | |
inc0 | take a look at this patch and I'll tell you what i have in mind | 16:51 |
*** ravig_ has quit IRC | 16:51 | |
*** ssurana has joined #openstack-kolla | 16:52 | |
*** fragatina has joined #openstack-kolla | 16:57 | |
*** fragatina has quit IRC | 16:58 | |
*** fragatina has joined #openstack-kolla | 16:59 | |
wirehead_ | Okay, let me see if I can cancel the dueling sword order on Amazon. :P | 16:59 |
pbourke_ | id go ahead and order that anyway wirehead_ :D | 16:59 |
inc0 | ah no need, I play with swords in my free time;) fun fact, I'm watching sword review now | 16:59 |
inc0 | https://www.youtube.com/watch?v=OjoWAg1HJKw | 17:00 |
inc0 | when I get proper house I'm going to start collecting swords | 17:00 |
*** unicell1 has quit IRC | 17:02 | |
inc0 | btw pbourke_ I extracted customization part https://review.openstack.org/#/c/324665/ | 17:03 |
patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 17:03 |
*** coolsvap_ has joined #openstack-kolla | 17:07 | |
sdake | inc0 moment still waking up | 17:07 |
inc0 | yeah I figured | 17:07 |
sdake | inc0 reading patch now moment | 17:09 |
*** ayoung has quit IRC | 17:09 | |
*** dwalsh has joined #openstack-kolla | 17:09 | |
*** coolsvap_ has quit IRC | 17:11 | |
sdake | inc0 i left comments | 17:12 |
sdake | one thing to me - is it feels like another bolt-on :) | 17:14 |
inc0 | sdake, and answered | 17:16 |
inc0 | no, if we make it accross the board and combine with refactor, it will be neat and elegant | 17:17 |
inc0 | also you'll get mechanism in place today if you want, as opposed to full release of work for building everything for DSL | 17:18 |
sdake | the second part is appealing to me personally | 17:20 |
sdake | not sure about operators | 17:20 |
sdake | maybe we should ask them what they want | 17:20 |
inc0 | sdake, I can tell them - they want matured and stable api | 17:21 |
sdake | inc0 just read your response | 17:21 |
sdake | its a mashup of dsl and jinja2 iiuc | 17:21 |
inc0 | it is, DSL will be kinda on the side | 17:22 |
sdake | i didn't catch that in the initial review | 17:22 |
inc0 | I'm ok with having it in /tools | 17:22 |
inc0 | but it will be fully functional | 17:22 |
inc0 | minus requirement of making yaml with every-single-package | 17:22 |
inc0 | you just make yaml with overrides | 17:22 |
inc0 | and you can make eparse turn list of repos to override of initial repos, or if you want to have specific repo in for example neutron, you remove old repos and add new repos in block at begginging of neutron container | 17:23 |
sdake | how is the adding/removing doen | 17:23 |
sdake | parser has to parse dockerfiles? | 17:23 |
inc0 | really, all boils down to put hooks in places you need, and there is no cost whatsoever as it might be empty | 17:23 |
inc0 | parser will generate something like this: | 17:24 |
inc0 | (gimme a moment) | 17:24 |
sdake | fwiw I dont give a hoot about the dsl I care about the requirements | 17:25 |
sdake | and being easy to use, which is subjective, so I didn't list it a a requirement :) | 17:25 |
inc0 | sdake, http://paste.openstack.org/show/507344/ | 17:25 |
inc0 | well, with this mechanism you can do whatever the hell you want really | 17:26 |
inc0 | all boils down to put hooks in correct places and generate correct override file | 17:26 |
inc0 | more power users can break single override file into multiple | 17:26 |
sdake | this is sort of the idea nihilfer had | 17:26 |
inc0 | and only point to master one | 17:26 |
inc0 | not quite, but similar | 17:27 |
inc0 | and again, code there is fully functional, it passed the gates before I extractred it from wip patch | 17:27 |
sdake | well this is alot better then your last go - which I saw balooning into 1800 files added to the repository | 17:27 |
inc0 | wip patch | 17:28 |
sdake | yup | 17:28 |
inc0 | it was called wip for a reason | 17:28 |
sdake | i get it, brilliance doesn't happen immediately | 17:28 |
inc0 | I still think we should break up our dockerfiles into multiple files so they wont be this long | 17:28 |
inc0 | but that's a different discussion | 17:28 |
inc0 | anyway, does this approach meet your requirements? You will be able to make every change listed in your spec with this | 17:29 |
inc0 | and again, if you want eparse, wonderful, whatever makes override file easier to write | 17:30 |
inc0 | it won't be super easy at start so we might actually use some tool assistance there | 17:30 |
sdake | inc0 right lets keep the refactor dockerfiles as a separate discussion as that will just complicate the issue | 17:31 |
inc0 | so this patchset is just customization | 17:32 |
sdake | inc0 from your paste, and your wip patch, they dont match | 17:32 |
inc0 | then we'll get to tedious work of adding block wherever needed | 17:32 |
*** unicell has joined #openstack-kolla | 17:32 | |
inc0 | my paste is suggestion how DSL parser can work | 17:32 |
sdake | one problem in the wip is overridess | 17:33 |
inc0 | you parse Elemental yaml into customization file | 17:33 |
sdake | there is no wayt o override what is there | 17:33 |
inc0 | no, it works | 17:33 |
sdake | line 199 | 17:33 |
sdake | apt keys are installed | 17:33 |
sdake | how do I provide different apt keys (or none at all in my scenario) | 17:33 |
inc0 | you wrap this whole thing into a block | 17:34 |
inc0 | and you can override this block | 17:34 |
inc0 | just a matter of adding a block | 17:34 |
sdake | so block endblock replace what is there? | 17:34 |
inc0 | well, we'd need to refactor this whole thing as it's add-key && add-key && other stuff | 17:34 |
inc0 | yeah | 17:34 |
inc0 | if you provide new content, it will | 17:35 |
inc0 | if you won't it will stay default | 17:35 |
inc0 | look at line 196 | 17:35 |
inc0 | defaulu is sources.list but in customization file I replace it | 17:35 |
sdake | https://review.openstack.org/#/c/324665/1//COMMIT_MSG | 17:36 |
patchbot | sdake: patch 324665 - kolla - Mechanism for Dockerfile customization | 17:36 |
sdake | line 17-18 - what does that mean | 17:36 |
inc0 | https://review.openstack.org/#/c/324665/1/etc/kolla/custom-dockerfile-example.j2 line 3 must be there | 17:37 |
patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 17:37 |
inc0 | the extends ... | 17:37 |
*** zhiwei has joined #openstack-kolla | 17:37 | |
inc0 | this will be customization file syntax | 17:37 |
inc0 | you know what's super neat imho? | 17:37 |
inc0 | companies might creaate super sophiscicated (I messed this word didn't I?) override files as properitary content | 17:38 |
inc0 | they can use kolla to ship their stuff with providing a single file or repo:) | 17:38 |
sdake | why have extends and not just parse the data directly in? | 17:38 |
inc0 | thats how jinja works, in jinja case this is really a master template | 17:39 |
inc0 | I render override file not Dockerfile | 17:39 |
sdake | why not just render the raw dockerfile from the yaml? | 17:40 |
sdake | with the endblocks/etc | 17:40 |
sdake | sorry not raw dockefile, dockerfile.j2 | 17:41 |
inc0 | well, this will be completely pointless work imho... | 17:41 |
inc0 | we already have dockerfile.j2 | 17:41 |
inc0 | why rewrite it? | 17:41 |
inc0 | ? | 17:41 |
sdake | i wasn't suggesting that minus adding the blocks | 17:41 |
*** zhiwei has quit IRC | 17:41 | |
inc0 | well, so DSL will render a jinja template really | 17:42 |
*** sean-k-mooney has joined #openstack-kolla | 17:42 | |
inc0 | this customization file is a jinja template | 17:42 |
inc0 | only with blocks and nothing else | 17:42 |
sdake | no with original code, but with blocks/endblocks | 17:42 |
inc0 | I don't get what you mean... | 17:42 |
inc0 | you want to recreate our existing Dockerfiles? | 17:43 |
sdake | i know having trouble expressing my ideas | 17:43 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use a barrier to know all threads are done vs 'unfinished_tasks' https://review.openstack.org/322308 | 17:43 |
sdake | i'm suggesting adding end and start blocks to the dockerfiles that exist today | 17:44 |
sdake | and having the dsl do merge/replace | 17:44 |
sdake | instead of dsl do it all | 17:45 |
inc0 | that would work, sure, but that would require us to support 2 sets of full dockerfile logic | 17:45 |
inc0 | which will be a nightmare | 17:45 |
sdake | no one dockerfile.j2 | 17:45 |
inc0 | ok, but if you change a package you need to change it in dockerfile and yaml | 17:45 |
sdake | no, yaml stays the same | 17:45 |
sdake | sorry | 17:46 |
sdake | dockerfile is changed (in 115 placees) to add these endblocks/block sections for insertion | 17:46 |
sdake | if any customization is desired, yaml parser outputs the block sections | 17:46 |
inc0 | yeah, that was idea, only instead of making block out of everything in dockerfile | 17:46 |
inc0 | we'll add blocks (possibly empty) in lots of places in all of dockerfiles | 17:47 |
inc0 | we'll have hundreds of blocks | 17:47 |
sdake | thousands more like it | 17:47 |
inc0 | that will allow you to inject stuff exactly where you want it | 17:47 |
sdake | but we have a community that can actually pull that off | 17:47 |
inc0 | sure, not an issue if we define proper naming standards at the begginging | 17:47 |
sdake | what i dont understand is why we need the custom-dockerfile-example.j2 | 17:48 |
inc0 | it's an example | 17:48 |
*** thumpba has quit IRC | 17:48 | |
inc0 | how blocks works | 17:48 |
inc0 | we may or may not merge it at the end | 17:48 |
inc0 | but I want to have it there for the sake of clarity how mechanism works | 17:48 |
sdake | i guess i dont understand exactly what your propossing | 17:49 |
sdake | up for a high bandwidth webex call? | 17:49 |
inc0 | let's go for hangouts plz | 17:49 |
sdake | wfm | 17:49 |
sdake | can you get it started | 17:49 |
sdake | give me 5 minutes | 17:49 |
sdake | i dont use hangouts all that much | 17:49 |
harlowja | qq, when i run `kolla-build glance -d -t source` why does that start to build nova and such | 17:50 |
inc0 | harlowja, it will build dependencies | 17:50 |
inc0 | but nova shouldn't be one of them | 17:50 |
harlowja | ok, odd | 17:50 |
inc0 | seems buggy, let me try it | 17:51 |
harlowja | ya, i'm trying it with https://review.openstack.org/322308 | 17:51 |
inc0 | btw I started to review your patches, might looking at gates being red? | 17:51 |
inc0 | https://review.openstack.org/#/c/321317/10 | 17:52 |
patchbot | inc0: patch 321317 - kolla - Use a task model and dumb workers | 17:52 |
inc0 | kk I'll try | 17:52 |
inc0 | hmm...did we change our gates to ansible2? | 17:53 |
sdake | yes | 17:53 |
inc0 | harlowja, did you rebase your patches lately?;) | 17:53 |
harlowja | no, lol | 17:53 |
harlowja | should i | 17:54 |
harlowja | ? | 17:54 |
sdake | you ont have to rebase just have to recheck | 17:54 |
inc0 | please do;) | 17:54 |
*** thumpba has joined #openstack-kolla | 17:54 | |
inc0 | well, you need to rebase so you have changes for ansible2 | 17:54 |
harlowja | k | 17:54 |
inc0 | anyway rebasing is cool every now and then | 17:54 |
harlowja | lol | 17:54 |
sdake | nah a recheck will test master | 17:55 |
sdake | it doesn't need a rebase | 17:55 |
sdake | inra guys corrected me on that | 17:55 |
sdake | and both sam and i verified it | 17:55 |
inc0 | ok, fair enough | 17:55 |
sdake | this behavior is different then the past | 17:55 |
sdake | it used to only check out the patch stream and not master | 17:55 |
coolsvap | sdake, inc0 can you please write in brief somewhere etherpad maybe about the discussion? i am very curious to understand the jinja2 workflow | 17:59 |
inc0 | coolsvap, https://review.openstack.org/#/c/324665/1 | 17:59 |
patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 17:59 |
inc0 | try it out | 17:59 |
coolsvap | inc0, i will | 18:00 |
sdake | inc0 he is talking about our google hangout | 18:00 |
inc0 | kk | 18:00 |
sdake | coolsvap your welcome to come | 18:00 |
inc0 | coolsvap, if you want to join in | 18:00 |
inc0 | just let me know whats your hangouts id | 18:01 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use a barrier to know all threads are done vs 'unfinished_tasks' https://review.openstack.org/322308 | 18:01 |
coolsvap | sdake, i had a long day just came home, inc0 its coolsvap@gmail.com let me jump in | 18:02 |
inc0 | whoa, it's like midnight for you now? | 18:02 |
*** diogogmt has quit IRC | 18:03 | |
sdake | ok inc0 the glue has kicked in lets chata | 18:03 |
coolsvap | inc0, yes | 18:03 |
inc0 | sdake login to hangouts:P | 18:04 |
*** diogogmt has joined #openstack-kolla | 18:04 | |
inc0 | anyone else wants to join in? | 18:04 |
mliima | \o | 18:04 |
mliima | I I I \o/ | 18:05 |
*** jtriley has quit IRC | 18:05 | |
mliima | mauriciolimab@gmail.com inc0 | 18:05 |
rhallisey | inc0, I can come for ~15min | 18:05 |
inc0 | sure | 18:05 |
rhallisey | brb | 18:05 |
*** rhallisey has quit IRC | 18:05 | |
sdake | http://paste.openstack.org/show/507344/ | 18:07 |
*** rhallisey has joined #openstack-kolla | 18:07 | |
*** ravig has joined #openstack-kolla | 18:07 | |
*** SiRiuS__ has quit IRC | 18:08 | |
inc0 | https://hangouts.google.com/call/yhhq3wizxbfvfpxxkdof3fg4tme link to hangouts guys, anyone is welcome to join | 18:08 |
sdake | we will start at 11:25 | 18:09 |
*** david-lyle has joined #openstack-kolla | 18:09 | |
sdake | pleae join now if you can | 18:09 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use an image object, recorder object and status constants https://review.openstack.org/321844 | 18:09 |
*** ravig has quit IRC | 18:09 | |
*** diogogmt has quit IRC | 18:10 | |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Stop using a global logger for all the things https://review.openstack.org/321884 | 18:10 |
sdake | spec for keys needed under discussion in the hangout | 18:11 |
*** diogogmt has joined #openstack-kolla | 18:14 | |
coolsvap | dave-mccowan, sdake https://review.openstack.org/#/c/324665/1 | 18:19 |
patchbot | coolsvap: patch 324665 - kolla - Mechanism for Dockerfile customization | 18:19 |
sdake | https://review.openstack.org/#/c/324665/1 | 18:19 |
patchbot | sdake: patch 324665 - kolla - Mechanism for Dockerfile customization | 18:19 |
*** Serlex has quit IRC | 18:19 | |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use a task model and dumb workers https://review.openstack.org/321317 | 18:21 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use an image object, recorder object and status constants https://review.openstack.org/321844 | 18:21 |
*** dwalsh has quit IRC | 18:21 | |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Stop using a global logger for all the things https://review.openstack.org/321884 | 18:22 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Add non-intrusive ctrl-c handling https://review.openstack.org/321913 | 18:22 |
*** huikang has quit IRC | 18:22 | |
*** huikang has joined #openstack-kolla | 18:23 | |
sean-k-mooney | inc0: is the hangout discussion regarding the custimisation/dsl or something else? | 18:23 |
sean-k-mooney | just reading back the logs now | 18:23 |
coolsvap | sean-k-mooney, related to customization/dsl | 18:24 |
inc0 | sean-k-mooney, yeha | 18:24 |
inc0 | I'm back btw, joining hangouts | 18:24 |
sean-k-mooney | ok cool | 18:24 |
inc0 | https://review.openstack.org/#/c/324665/1 http://paste.openstack.org/show/507344/ | 18:25 |
patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 18:25 |
sean-k-mooney | i might join as i am interested in how this evolves, though not sure what if anything i can add | 18:25 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use a barrier to know all threads are done vs 'unfinished_tasks' https://review.openstack.org/322308 | 18:26 |
*** ravig has joined #openstack-kolla | 18:26 | |
*** ravig_ has joined #openstack-kolla | 18:27 | |
*** ravig has quit IRC | 18:27 | |
*** huikang has quit IRC | 18:28 | |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use a barrier to know all threads are done vs 'unfinished_tasks' https://review.openstack.org/322308 | 18:28 |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use a barrier to know all threads are done vs 'unfinished_tasks' https://review.openstack.org/322308 | 18:28 |
harlowja | ok, seems like that's working now :-P | 18:30 |
*** ravig_ has quit IRC | 18:32 | |
*** SiRiuS__ has joined #openstack-kolla | 18:32 | |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use 'image' not 'item' https://review.openstack.org/324756 | 18:33 |
harlowja | another idea i had, what would u guys think about outputting a tree showing all the images to be built, before the work starts? | 18:34 |
harlowja | i have the following code @ https://github.com/openstack/taskflow/blob/master/taskflow/types/tree.py#L259 that can do this | 18:34 |
harlowja | and/or i can make a mini version like that | 18:35 |
*** david-lyle has quit IRC | 18:35 | |
harlowja | seems like there is some graphviz stuff, but that's probably not easily viewable on a console... | 18:35 |
*** dwalsh has joined #openstack-kolla | 18:43 | |
*** harlowja has quit IRC | 18:44 | |
*** huikang has joined #openstack-kolla | 18:59 | |
*** tyrola has quit IRC | 19:06 | |
*** ravig has joined #openstack-kolla | 19:08 | |
*** thumpba has quit IRC | 19:09 | |
*** thumpba has joined #openstack-kolla | 19:11 | |
*** ravig has quit IRC | 19:13 | |
*** ravig has joined #openstack-kolla | 19:15 | |
*** ravig has quit IRC | 19:16 | |
cloudnautique | inc0 looks like in my rabbitmq-env.sh file the RABBITMQ_NODENAME var was missing @hostname | 19:17 |
*** godleon has joined #openstack-kolla | 19:17 | |
*** huikang has quit IRC | 19:18 | |
inc0 | strange, should be setup in container | 19:18 |
inc0 | did you get it working? | 19:18 |
*** huikang has joined #openstack-kolla | 19:18 | |
*** mummer has joined #openstack-kolla | 19:18 | |
*** tyrola has joined #openstack-kolla | 19:23 | |
*** huikang has quit IRC | 19:23 | |
*** ravig has joined #openstack-kolla | 19:24 | |
*** david-lyle has joined #openstack-kolla | 19:24 | |
dims | folks, has anyone exercised/tested the rabbitmq-clusterer thoroughly? | 19:27 |
inc0 | dims, cloudnautique had issue | 19:29 |
inc0 | I did test it pretty much, but something might be broken | 19:30 |
inc0 | you're second person today asking about it | 19:30 |
dims | inc0 :thanks. lol. who was the first? | 19:30 |
*** Jeffrey4l_ has quit IRC | 19:30 | |
inc0 | cloudnautique, | 19:30 |
dims | ah ok | 19:31 |
dims | cloudnautique : what did you see/find? | 19:31 |
inc0 | I'll get my head around it and start testing | 19:31 |
*** ravig has quit IRC | 19:31 | |
dims | inc0 : thanks! | 19:31 |
*** Jeffrey4l_ has joined #openstack-kolla | 19:31 | |
cloudnautique | in the template for rabbitmq-env.sh the first hostname line seems to be missing the @host name. Once I added it.. the server fired up | 19:32 |
dims | inc0 : i am hearing FUD around rabbitmq-clusterer, so wanted to hear actual experiences if any | 19:32 |
dims | cloudnautique : so seems to be working now? | 19:32 |
*** mark-casey has left #openstack-kolla | 19:33 | |
inc0 | so I did test it heavily during upgrade | 19:33 |
inc0 | putting down, starting up in various order | 19:33 |
inc0 | and it worked really well | 19:33 |
inc0 | but that's about it | 19:33 |
dims | inc0 : ack thanks! | 19:33 |
cloudnautique | yeah, rabbitmq seems good. | 19:33 |
dims | inc0 cloudnautique : thanks for the feedback | 19:34 |
cloudnautique | I had a few other hacks to deploy kolla in master that I'm trying to remember back in the Mitaka branch to get it all up | 19:34 |
cloudnautique | :) | 19:34 |
dims | cool | 19:34 |
*** sdake has quit IRC | 19:35 | |
*** sdake has joined #openstack-kolla | 19:38 | |
*** openstackstatus has quit IRC | 19:39 | |
*** openstack has joined #openstack-kolla | 19:40 | |
*** openstackstatus has joined #openstack-kolla | 19:40 | |
*** ChanServ sets mode: +v openstackstatus | 19:40 | |
*** sdake_ has joined #openstack-kolla | 19:40 | |
*** sdake has quit IRC | 19:43 | |
*** zhiwei has joined #openstack-kolla | 19:46 | |
*** vinkman has joined #openstack-kolla | 19:47 | |
*** Lyncos has joined #openstack-kolla | 19:49 | |
Lyncos | Hi guys | 19:49 |
inc0 | hey Lyncos | 19:50 |
Lyncos | How it's going today ? | 19:50 |
*** zhiwei has quit IRC | 19:50 | |
inc0 | good, we got this customization thing sorted out | 19:51 |
*** ravig has joined #openstack-kolla | 19:51 | |
Lyncos | What thing ? | 19:51 |
inc0 | DSL discusson? have you seen it? | 19:52 |
*** tyrola has quit IRC | 19:52 | |
Lyncos | MAG recapped it to me a bit yesterday | 19:52 |
Lyncos | but I think we missed the conclusion | 19:52 |
inc0 | well, we just concluded it | 19:52 |
inc0 | https://review.openstack.org/#/c/324665/1 | 19:52 |
patchbot | inc0: patch 324665 - kolla - Mechanism for Dockerfile customization | 19:52 |
Lyncos | looking at it | 19:52 |
*** mbound has quit IRC | 19:53 | |
Lyncos | are you good with Heat ? | 19:53 |
inc0 | not that good, rhallisey is tho | 19:53 |
inc0 | and sdake_ created it so he probably knows somethign too | 19:53 |
Lyncos | I would need quick help but it's not kolla related | 19:53 |
*** neophy has joined #openstack-kolla | 19:54 | |
Lyncos | I'm trying to avoid $public_ipv4 being replaced by empty string | 19:54 |
Lyncos | I guess they left already ? | 19:54 |
*** tyrola has joined #openstack-kolla | 19:54 | |
Lyncos | http://paste.openstack.org/show/507365/ line 85 I want $public_ipv4 being written to the file | 19:55 |
Lyncos | escaping dosen't work | 19:55 |
Lyncos | and the Heat channel is no help | 19:55 |
*** dwalsh has quit IRC | 19:55 | |
*** ravig has quit IRC | 19:56 | |
*** ravig has joined #openstack-kolla | 19:57 | |
*** david-lyle has quit IRC | 19:57 | |
*** huikang has joined #openstack-kolla | 19:58 | |
*** tyrola has quit IRC | 19:59 | |
*** ravig has quit IRC | 20:01 | |
rhallisey | Lyncos, see line 88 | 20:01 |
Lyncos | rhallisey: line 88 works... but I found nowhere to find the local ip address | 20:01 |
rhallisey | Lyncos, use line 88 as an example to replace $public_ipv4 | 20:02 |
rhallisey | add to params: | 20:03 |
rhallisey | $public_ipv4: { get_param: <ip_address_variable> } | 20:03 |
rhallisey | Lyncos, http://docs.openstack.org/developer/heat/template_guide/hot_spec.html#str-replace | 20:04 |
Lyncos | rhallisey: the variable dosen't exist... instead what I'm doing .. I spawn the instances... create a local cloud-init file and run it | 20:04 |
*** ravig has joined #openstack-kolla | 20:04 | |
rhallisey | Lyncos, I guess I'm confused, why can't you add it as a parameter? | 20:05 |
Lyncos | Because the IP is provided by nova when the VM is created.. I cannot get the Variable before | 20:06 |
*** ravig has quit IRC | 20:06 | |
Lyncos | A lot of documentation says I just need to put $public_ipv4 and it will be automatically replaced by cloud-init... but that dosen't work on our implementation .. not sure why | 20:07 |
*** ravig has joined #openstack-kolla | 20:07 | |
Lyncos | in that case.. cloud-init make it empty | 20:08 |
sdake_ | Lyncos i don't have an answwre to your question, try #heat | 20:08 |
*** huikang has quit IRC | 20:08 | |
inc0 | lol | 20:08 |
Lyncos | #heat is dead | 20:08 |
inc0 | sdake_, helpful as usual:P | 20:08 |
sdake_ | openstack-heat? | 20:08 |
Lyncos | dead too | 20:08 |
*** huikang has joined #openstack-kolla | 20:09 | |
inc0 | tripleo channel is home of heat experts too | 20:09 |
*** ravig has quit IRC | 20:09 | |
Lyncos | I'll try tripleo .. I hope it is more active than heat | 20:09 |
nilk | inc0: I like that customization approach | 20:11 |
*** ravig has joined #openstack-kolla | 20:11 | |
inc0 | yay | 20:11 |
inc0 | I'm glad we're in violent agreement, good result of a semi-major shitstorm | 20:11 |
Lyncos | I like it too | 20:12 |
*** neophy has quit IRC | 20:12 | |
*** thumpba has quit IRC | 20:12 | |
*** huikang has quit IRC | 20:13 | |
*** ravig has quit IRC | 20:13 | |
rhallisey | Lyncos, sorry ran to the bathroom. I see your issue | 20:13 |
inc0 | yeah, if we put blocks in correct places, you can pretty much do whatever you like without meddling with kolla code | 20:13 |
Lyncos | i think $public_ipv4 is generated by coreos but I don't know how ... | 20:14 |
rhallisey | Lyncos, you could have that resource run by cloud-init | 20:15 |
rhallisey | that have some script that picks up the ip | 20:15 |
*** dwalsh has joined #openstack-kolla | 20:15 | |
*** ravig has joined #openstack-kolla | 20:15 | |
Lyncos | rhallisey: this is what I'm trying to achieve .. my cloud-init is creating another cloud-init file locally on the node with the ip in it but parsing is interpreting "$public_ipv4 variables instead of writing only "$public_ipv4" string to the file | 20:16 |
*** mliima has quit IRC | 20:16 | |
inc0 | Lyncos, sed it up! | 20:17 |
Lyncos | hmmm good idea | 20:17 |
Lyncos | I need to figure out where I did put my sed skills | 20:17 |
Lyncos | :-) | 20:17 |
inc0 | look under google | 20:17 |
Lyncos | hehe | 20:17 |
rhallisey | Lyncos, ya go with sed :) | 20:17 |
Lyncos | I'll go with sed | 20:18 |
Lyncos | I'll figure out a one lineer | 20:18 |
inc0 | hehe | 20:18 |
Lyncos | tnx | 20:18 |
*** rhallisey has quit IRC | 20:19 | |
*** rhallisey has joined #openstack-kolla | 20:20 | |
*** ravig has quit IRC | 20:20 | |
*** ravig has joined #openstack-kolla | 20:20 | |
*** huikang has joined #openstack-kolla | 20:22 | |
*** harlowja has joined #openstack-kolla | 20:24 | |
inc0 | ok guys I've hit the wall | 20:26 |
inc0 | cya all tomorrow | 20:26 |
*** inc0 has quit IRC | 20:26 | |
Lyncos | I'm leaving too | 20:26 |
Lyncos | see ya later | 20:26 |
*** Lyncos has left #openstack-kolla | 20:26 | |
cloudnautique | In the setup/installation docs, is there a note I missed about setting the mounts to Shared on the Docker daemon? | 20:27 |
wirehead_ | cloudnautique: I see it in http://docs.openstack.org/developer/kolla/quickstart.html | 20:29 |
wirehead_ | IT’s in the Kolla-Kubernetes quickstart as well. | 20:29 |
cloudnautique | Thank you :) | 20:30 |
cloudnautique | I completely missed it | 20:30 |
wirehead_ | Hm. That’s probably a doc issue then. Things such as that should be hard to miss. :) | 20:31 |
cloudnautique | It is just as likely my issue :) | 20:31 |
wirehead_ | well, as the self-designated docs-obsessed-programmer in this channel, I view it as a doc issue. :) | 20:32 |
*** huikang has quit IRC | 20:37 | |
*** huikang has joined #openstack-kolla | 20:38 | |
*** huikang has quit IRC | 20:43 | |
sdake_ | rhallisey ping | 20:45 |
rhallisey | sdake_, hey | 20:46 |
*** tyrola has joined #openstack-kolla | 20:47 | |
*** fragatina has quit IRC | 20:47 | |
*** dwalsh has quit IRC | 20:47 | |
*** tyrola has quit IRC | 20:47 | |
*** dwalsh has joined #openstack-kolla | 20:49 | |
openstackgerrit | Ken Wronkiewicz proposed openstack/kolla-kubernetes: Add some hints for working with MariaDB. https://review.openstack.org/324812 | 20:53 |
*** mbound has joined #openstack-kolla | 20:53 | |
*** mbound has quit IRC | 20:58 | |
openstackgerrit | Joshua Harlow proposed openstack/kolla: Use 'image' not 'item' https://review.openstack.org/324756 | 20:59 |
*** ayoung has joined #openstack-kolla | 20:59 | |
*** daneyon_ has joined #openstack-kolla | 21:06 | |
*** ppowell has quit IRC | 21:07 | |
sdake_ | sbezverk ping | 21:07 |
*** rhallisey has quit IRC | 21:08 | |
*** tyrola has joined #openstack-kolla | 21:09 | |
*** daneyon_ has quit IRC | 21:10 | |
*** SiRiuS__ has quit IRC | 21:14 | |
sbezverk | sdake_ pong | 21:19 |
*** ravig has quit IRC | 21:22 | |
*** ayoung has quit IRC | 21:25 | |
harlowja | qq, would people be opposed to making `kolla-ansible` a actual python command vs bash, seems odd that one of the commands isn't python | 21:28 |
harlowja | seems/feels | 21:28 |
harlowja | when it could just be python :-/ | 21:29 |
harlowja | (thus matching the rest of the commands code and such) | 21:29 |
harlowja | i'll covert it if people don't mind | 21:29 |
harlowja | *convert | 21:29 |
*** ravig has joined #openstack-kolla | 21:35 | |
*** huikang has joined #openstack-kolla | 21:38 | |
*** rahuls has joined #openstack-kolla | 21:40 | |
*** huikang has quit IRC | 21:43 | |
*** ravig has quit IRC | 21:44 | |
*** huikang has joined #openstack-kolla | 21:44 | |
*** tyrola has quit IRC | 21:48 | |
dmsimard | rhallisey: have you guys seen this ? https://github.com/ansible/ansible-container | 21:56 |
dmsimard | oh, he's gone | 21:56 |
*** mummer has quit IRC | 21:59 | |
*** zhiwei has joined #openstack-kolla | 22:02 | |
*** zhiwei has quit IRC | 22:06 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Make nova-libvirt container build with debian https://review.openstack.org/297169 | 22:08 |
*** ayoung has joined #openstack-kolla | 22:12 | |
*** sacharya has quit IRC | 22:17 | |
sdake_ | dmsimard my wife has mentioned it to me about 10 times | 22:19 |
sdake_ | dmsimard what she failed to do is explain how it relates to kolla | 22:19 |
dmsimard | sdake_: you have a very technical wife :) | 22:19 |
sdake_ | could you do that? | 22:19 |
sdake_ | she works at ansible | 22:19 |
dmsimard | oh, wow | 22:19 |
dmsimard | lol | 22:19 |
sdake_ | robyn bergeron | 22:19 |
sdake_ | heard the name? | 22:19 |
dmsimard | Oh, really ? | 22:19 |
sdake_ | she was prevously fedora project leader | 22:19 |
dmsimard | ofc I have :) | 22:19 |
*** openstackgerrit has quit IRC | 22:19 | |
*** openstackgerrit has joined #openstack-kolla | 22:20 | |
sdake_ | i'm pretty sure some nights i went to bed I wasn't sure if she would murderme in the night because of that FPL job :) | 22:20 |
dmsimard | sdake_: https://twitter.com/dmsimard/status/735674985067450372 :p | 22:20 |
dmsimard | sdake_: I don't know too much about ansible container, just skimmed the readme | 22:20 |
*** huikang has quit IRC | 22:21 | |
*** huikang has joined #openstack-kolla | 22:22 | |
*** huikang has quit IRC | 22:27 | |
*** huikang has joined #openstack-kolla | 22:30 | |
*** dave-mccowan has quit IRC | 22:30 | |
*** prithiv has joined #openstack-kolla | 22:31 | |
openstackgerrit | Benedikt Trefzer proposed openstack/kolla: Use debian repos for debian base docker image https://review.openstack.org/288936 | 22:32 |
*** huikang has quit IRC | 22:40 | |
*** huikang has joined #openstack-kolla | 22:40 | |
sdake_ | hey guys | 22:41 |
sdake_ | and gals | 22:41 |
sdake_ | fantastic news | 22:41 |
sdake_ | Kolla Midcycle at Ansible HQ (Durhum, NC) July 11th, 12th | 22:41 |
wirehead_ | Just so you know, sdake, I pictured you something like this: http://shockwave75.homestead.com/files/Good_News_Everyone.jpg | 22:42 |
*** huikang has quit IRC | 22:43 | |
*** huikang has joined #openstack-kolla | 22:43 | |
sdake_ | wirehead_ :) | 22:45 |
*** vhosakot has quit IRC | 22:46 | |
*** fragatina has joined #openstack-kolla | 22:46 | |
*** dave-mccowan has joined #openstack-kolla | 22:48 | |
*** cu5 has joined #openstack-kolla | 22:50 | |
*** cu5 has quit IRC | 22:52 | |
*** mummer has joined #openstack-kolla | 22:52 | |
*** cu5 has joined #openstack-kolla | 22:52 | |
*** asalkeld has joined #openstack-kolla | 22:59 | |
*** ayoung has quit IRC | 23:04 | |
rahuls | Hi All, I am trying kolla AIO deployment and right now stuck with an error "ImportError: No module named plugins.action" when I do kolla-build deploy | 23:07 |
*** huikang has quit IRC | 23:08 | |
*** huikang has joined #openstack-kolla | 23:09 | |
rahuls | looking at the code, it tries to import ActionBase from ansible.plugins.action but the ansible folder has no subfolders/modules as plugins | 23:10 |
rahuls | Here is the detailed output from the logs:- | 23:12 |
rahuls | http://pastebin.com/pKXi7UKB | 23:12 |
*** huikang has quit IRC | 23:12 | |
*** huikang has joined #openstack-kolla | 23:12 | |
*** huikang has quit IRC | 23:12 | |
rahuls | my ansible version is 1.9.4 and kolla is 2.0.0 | 23:13 |
openstackgerrit | Ettore Simone proposed openstack/kolla: Enable HAProxy consider MariaDB wsrep_local_state https://review.openstack.org/322200 | 23:15 |
*** godleon has quit IRC | 23:19 | |
*** prithiv has quit IRC | 23:19 | |
*** david-lyle has joined #openstack-kolla | 23:20 | |
sdake_ | rahuls master has been ported to 2.0.z | 23:22 |
sdake_ | rahuls if you want ansible 1.9.z you can use either mitaka or liberty branches | 23:22 |
rahuls | @sdake_ Thank you for the suggestions. Let me give it a try. | 23:23 |
rahuls | I guess those steps or details are missing from the quickstart guide | 23:24 |
*** berendt has quit IRC | 23:28 | |
rahuls | it worked, thanks. Should I file a documentation bug for this? The quickstart guide says Ansible 1.9.4 to < 2.0.0 but nowhere its mentioned that one should checkout stable/mitaka before proceeding | 23:29 |
openstackgerrit | Ettore Simone proposed openstack/kolla: Enable HAProxy consider MariaDB wsrep_local_state https://review.openstack.org/322200 | 23:33 |
*** rahuls has quit IRC | 23:35 | |
*** rahuls has joined #openstack-kolla | 23:36 | |
wirehead_ | rahuls: there’s a bug already open. | 23:37 |
wirehead_ | rahuls: https://review.openstack.org/#/c/321525/ | 23:37 |
patchbot | wirehead_: patch 321525 - kolla - [doc] change Ansible version to >= 2 in quickstart | 23:37 |
*** cu5 has quit IRC | 23:43 | |
*** cu5 has joined #openstack-kolla | 23:43 | |
rahuls | wirehead_: thank you for your response. | 23:44 |
*** cu5 has quit IRC | 23:45 | |
*** sdake has joined #openstack-kolla | 23:54 | |
*** sdake_ has quit IRC | 23:56 | |
*** sdake has quit IRC | 23:57 | |
*** vinkman1 has joined #openstack-kolla | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!