| *** kprice789 has quit IRC | 00:14 | |
| *** luksky has quit IRC | 01:04 | |
| *** gyee has quit IRC | 01:13 | |
| *** scarlisle has quit IRC | 01:22 | |
| *** goldyfruit_ has quit IRC | 01:43 | |
| *** tinwood has quit IRC | 02:10 | |
| *** tinwood has joined #openstack-ansible | 02:12 | |
| *** udesale has joined #openstack-ansible | 04:40 | |
| *** raukadah is now known as chkumar|rover | 05:57 | |
| *** southquist has joined #openstack-ansible | 06:00 | |
| *** elenalindq has joined #openstack-ansible | 06:36 | |
| *** udesale_ has joined #openstack-ansible | 06:46 | |
| *** udesale has quit IRC | 06:47 | |
| *** luksky has joined #openstack-ansible | 07:14 | |
| *** jawad_axd has joined #openstack-ansible | 07:26 | |
| *** yolanda has quit IRC | 07:32 | |
| *** ebbex has quit IRC | 07:55 | |
| *** udesale_ has quit IRC | 07:58 | |
| *** udesale_ has joined #openstack-ansible | 07:58 | |
| *** cshen has joined #openstack-ansible | 08:08 | |
| *** sshnaidm|afk is now known as sshnaidm | 08:12 | |
| *** rpittau|afk is now known as rpittau | 08:14 | |
| *** udesale_ has quit IRC | 08:28 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 08:29 |
|---|---|---|
| *** dmsimard5 has joined #openstack-ansible | 08:31 | |
| *** dmsimard has quit IRC | 08:33 | |
| *** dmsimard5 is now known as dmsimard | 08:33 | |
| *** tosky has joined #openstack-ansible | 08:46 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible master: manila: refactor playbooks https://review.opendev.org/704942 | 08:50 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 08:55 |
| *** luksky has quit IRC | 08:58 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 09:02 |
| *** DanyC has joined #openstack-ansible | 09:02 | |
| mnaser | so looks like we're installing uwsgi everywhere | 09:02 |
| mnaser | which means it is living on compute nodes and in my case manila in that wip patch | 09:03 |
| mnaser | https://github.com/openstack/ansible-role-uwsgi/blob/master/tasks/uwsgi_install.yml | 09:03 |
| mnaser | there's no conditionals on that | 09:03 |
| mnaser | nor is there any here: https://github.com/openstack/openstack-ansible-os_nova/blob/master/tasks/main.yml#L184-L192 | 09:03 |
| *** DanyC has quit IRC | 09:04 | |
| *** DanyC has joined #openstack-ansible | 09:04 | |
| mnaser | even tho uwsgi is empty, we still create a user and create a python_venv | 09:05 |
| mnaser | we dont deploy any uwsgi files or create systemd units but.. | 09:05 |
| noonedeadpunk | oh....... I yes, that's true.... | 09:06 |
| noonedeadpunk | I think it's easier to patch uwsgi role | 09:07 |
| noonedeadpunk | let me do that | 09:07 |
| mnaser | that's the whole reason we invented it :D | 09:07 |
| mnaser | one change and we fix it all | 09:07 |
| *** DanyC has quit IRC | 09:09 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/ansible-role-uwsgi master: Run uwsgi tasks only when uwsgi_services defined https://review.opendev.org/704944 | 09:12 |
| noonedeadpunk | mnaser: I think this is exactly what I missed ^ | 09:13 |
| mnaser | noonedeadpunk: is "when: <empty-list>" eval to false? i always forget lol | 09:15 |
| mnaser | if thats the case, then that seems good | 09:15 |
| *** DanyC has joined #openstack-ansible | 09:15 | |
| mnaser | noonedeadpunk: we can probably even skip the include_vars too honestly | 09:15 |
| noonedeadpunk | in this empty dict, but yes - I'v tested this as I always forget as well lol) | 09:15 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 09:15 |
| mnaser | noonedeadpunk: if you have a compelling reason not to skip the include_vars im up for it | 09:16 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/ansible-role-uwsgi master: Run uwsgi tasks only when uwsgi_services defined https://review.opendev.org/704944 | 09:17 |
| noonedeadpunk | yeah, good call | 09:18 |
| *** ivve has joined #openstack-ansible | 09:19 | |
| mnaser | i'm going to +1 my patches when ive tested them locally as we dont have much manila ci right now | 09:19 |
| noonedeadpunk | actually we could probably test things with https://review.opendev.org/#/c/675934/16 since it passes metal jobs | 09:21 |
| noonedeadpunk | btw, don't you face issues with undefined things like manila_default_store here https://opendev.org/openstack/openstack-ansible/src/branch/master/playbooks/os-manila-install.yml#L39-L52 | 09:23 |
| noonedeadpunk | As I saw you left these bind mounts while for me playbook was just failing | 09:25 |
| mnaser | noonedeadpunk: yeah i had to manually define it to "generic" or something | 09:25 |
| mnaser | i need to get back to that later and figure out what/how | 09:25 |
| mnaser | or what it really meant | 09:25 |
| noonedeadpunk | the thing is it's not used anywhere http://codesearch.openstack.org/?q=manila_default_store&i=nope&files=&repos= | 09:25 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 09:25 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: refactor: service group names https://review.opendev.org/704945 | 09:25 |
| noonedeadpunk | So I just dropped this out, but if you'll figure out that they are really required for smth... | 09:26 |
| mnaser | noonedeadpunk: yes i actually noticed they werent used for anything too so i was confused, so i might be in favour of dropping them too | 09:27 |
| noonedeadpunk | what I'm thinking is maybe we will merge at least some CI before refactoring? | 09:31 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: refactor: service group names https://review.opendev.org/704945 | 09:31 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 09:31 |
| mnaser | if you have something passing im up for it ya | 09:31 |
| noonedeadpunk | at least centos and ubuntu metal jobs with ganesha pass | 09:33 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Add manila tempest testing https://review.opendev.org/674653 | 09:34 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add manila CI https://review.opendev.org/675934 | 09:34 |
| mnaser | noonedeadpunk: added a few things to the review | 09:35 |
| *** udesale has joined #openstack-ansible | 09:35 | |
| *** luksky has joined #openstack-ansible | 09:35 | |
| noonedeadpunk | yeah | 09:36 |
| noonedeadpunk | fixing | 09:36 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: refactor: service group names https://review.opendev.org/704945 | 09:36 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 09:36 |
| jrosser | o/ morning | 09:38 |
| mnaser | o/ | 09:39 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 09:41 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Revert "Mark ubuntu distro install as non voting" https://review.opendev.org/704946 | 09:41 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Revert "Mark ubuntu distro install as non voting" https://review.opendev.org/704946 | 09:41 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add manila CI https://review.opendev.org/675934 | 09:42 |
| jrosser | noonedeadpunk: mnaser: this stack of 3 patches fix up how we install tempest on master in light of the dropping of py2, they're passing now so ready for reviews | 09:43 |
| jrosser | https://review.opendev.org/#/c/704567/9 | 09:43 |
| mnaser | jrosser: reviewing | 09:43 |
| jrosser | i propose to backport 704567 to at least train (and possibly stein) to make the ubuntu distro jobs more sensible and resilient to future changes in tempest | 09:44 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add manila CI https://review.opendev.org/675934 | 09:50 |
| mnaser | jrosser: reviewed -- and most of https://review.opendev.org/#/q/topic:osa/manila-generic+(status:open+OR+status:merged) is ok right now | 09:57 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: cleanup: config values https://review.opendev.org/704951 | 10:06 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Always use virtualenv and pip to install tempest on Ubuntu https://review.opendev.org/704567 | 10:06 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Always use virtualenv and pip to install tempest on Ubuntu https://review.opendev.org/704567 | 10:16 |
| openstackgerrit | Magnus Bergman proposed openstack/openstack-ansible-haproxy_server master: Enable ip_nonlocal_bind for IPv6 in addition to IPv4 https://review.opendev.org/704955 | 10:20 |
| *** spatel has joined #openstack-ansible | 10:29 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible master: Add manila_service_region https://review.opendev.org/704957 | 10:32 |
| *** spatel has quit IRC | 10:34 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest master: Always use virtualenv and pip to install tempest on Ubuntu https://review.opendev.org/704567 | 10:35 |
| *** pcaruana has quit IRC | 10:46 | |
| *** spatel has joined #openstack-ansible | 10:47 | |
| jawad_axd | Hi all ! Question is, from instances, download/upload speed is around 10MB/s while from underlying host is 80MB/s. How can I increase/tune bandwidth parameters? I am using Stein,linux bridge,vxlans setup.Any pointers? | 10:50 |
| *** zbr has quit IRC | 10:52 | |
| *** udesale has quit IRC | 10:53 | |
| jrosser | jawad_axd: https://www.youtube.com/watch?v=uItivBAzz8s | 11:12 |
| *** rpittau is now known as rpittau|bbl | 11:14 | |
| *** cshen has quit IRC | 11:16 | |
| jawad_axd | @jrosser Thanks | 11:18 |
| *** cshen has joined #openstack-ansible | 11:23 | |
| *** zbr has joined #openstack-ansible | 11:23 | |
| *** csmart has quit IRC | 11:30 | |
| jrosser | jawad_axd: seeing what you say in the nova channel i assumed you were talking about simply network throughput | 11:32 |
| jrosser | try to test things that independently test your network and storage if you are concerned about the performance of "wget this big file onto my ceph backed disk" | 11:34 |
| *** zbr has quit IRC | 11:35 | |
| noonedeadpunk | mnaser: looking at https://review.opendev.org/#/c/704575/1/tasks/magnum_post_install.yml one more time I have a question - we will run only one time for the play with change you've offered. But we will also upload image every time when role is launched? | 11:35 |
| *** zbr has joined #openstack-ansible | 11:36 | |
| jawad_axd | @jrosser Yes. I guess I need to test on different levels..between instances, on same network, same hosts, different hosts etc..then disk. I am going to do it now. Thanks | 11:36 |
| *** cshen has quit IRC | 11:40 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: [WIP] Add image download for manila https://review.opendev.org/704972 | 11:42 |
| *** alti_17 has joined #openstack-ansible | 11:43 | |
| *** cshen has joined #openstack-ansible | 11:47 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Mark ubuntu distro install as non voting https://review.opendev.org/704750 | 11:47 |
| *** pcaruana has joined #openstack-ansible | 11:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server master: Enable ip_nonlocal_bind for IPv6 in addition to IPv4 https://review.opendev.org/704955 | 11:52 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Use contraints for tempest plugins https://review.opendev.org/703979 | 12:00 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest stable/train: Mark ubuntu distro install as non voting https://review.opendev.org/704978 | 12:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest stable/train: Use contraints for tempest plugins https://review.opendev.org/704980 | 12:03 |
| mnaser | noonedeadpunk: no because os_image is smart enough to do a sha of the file locally and compare it to the remote one | 12:19 |
| *** pcaruana has quit IRC | 12:21 | |
| *** DanyC has quit IRC | 12:24 | |
| *** DanyC has joined #openstack-ansible | 12:25 | |
| *** DanyC has quit IRC | 12:29 | |
| *** DanyC has joined #openstack-ansible | 12:30 | |
| *** DanyC has quit IRC | 12:31 | |
| noonedeadpunk | mnaser: than we probably should simplify octavia image upload process | 12:32 |
| mnaser | noonedeadpunk: i tihnk the reason we do that is because os_image doesnt support tags :( | 12:32 |
| noonedeadpunk | I mean maybe we can delegate image upload to os_image and just afterwards set a tag, instead of getting old id, dropping it, and uploading with command | 12:36 |
| *** DanyC has joined #openstack-ansible | 12:36 | |
| noonedeadpunk | (or maybe we can patch os_image one day:)) | 12:36 |
| mnaser | noonedeadpunk: i like that idea of delegate image upload to os_image | 12:43 |
| mnaser | because we can actually use register: | 12:43 |
| mnaser | and just run the tag | 12:43 |
| mnaser | i mean openstack modules live inside opendev now :p | 12:44 |
| noonedeadpunk | yeah, so offering change to os_image maybe also the way easier, but not sure about backporting.... | 12:45 |
| noonedeadpunk | it's new feature anyway | 12:45 |
| *** ansmith has quit IRC | 12:45 | |
| noonedeadpunk | mnaser: btw, CI passed https://review.opendev.org/#/c/675934/ so we can use it in some way to check refactoring | 12:46 |
| mnaser | noonedeadpunk: +1d | 12:49 |
| noonedeadpunk | can you also take a look at https://review.opendev.org/#/c/674653/ as it's the last unmerged dependency | 12:51 |
| mnaser | noonedeadpunk: done | 12:56 |
| *** rh-jelabarre has joined #openstack-ansible | 12:56 | |
| *** goldyfruit_ has joined #openstack-ansible | 12:59 | |
| mnaser | noonedeadpunk: i will rebase all my stuff ontop of yours | 13:01 |
| noonedeadpunk | mnaser: can I rebase your patches on top of 675934 so that they were tested with CI as well? | 13:01 |
| mnaser | .... lol =P | 13:01 |
| noonedeadpunk | yeah, ok) | 13:02 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Add manila CI https://review.opendev.org/675934 | 13:07 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: refactor: service group names https://review.opendev.org/704945 | 13:07 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 13:07 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: cleanup: config values https://review.opendev.org/704951 | 13:07 |
| mnaser | noonedeadpunk: there | 13:07 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_keystone stable/train: Federated openid support using auth_mod_openidc https://review.opendev.org/705001 | 13:08 |
| noonedeadpunk | gshippey: why ^ is not for master? | 13:09 |
| mnaser | noonedeadpunk: can i rebase your image download wip stuff (i can test it off locally first of all and also it has some things that break with teh key changes i made) | 13:10 |
| noonedeadpunk | yes, sure | 13:10 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_keystone master: Federated openid support using auth_mod_openidc https://review.opendev.org/705002 | 13:10 |
| noonedeadpunk | I also see a bug in it... | 13:11 |
| mnaser | noonedeadpunk: what is it, i can fix that too? | 13:11 |
| noonedeadpunk | os_distro: "{{ item.distro }}" | 13:11 |
| noonedeadpunk | https://review.opendev.org/#/c/704972/1/tasks/manila_image.yml | 13:11 |
| gshippey | @noonedeadpunk fixed! | 13:11 |
| noonedeadpunk | like properties are probably not needed here | 13:11 |
| mnaser | noonedeadpunk: ah yea | 13:12 |
| noonedeadpunk | (anyway key is not defined) | 13:12 |
| noonedeadpunk | as well as item.public | 13:12 |
| noonedeadpunk | So probably I should rebase it? | 13:12 |
| mnaser | noonedeadpunk: i can fix those | 13:12 |
| noonedeadpunk | and fix stuff | 13:13 |
| mnaser | also we probably need to upload it to the service project | 13:13 |
| mnaser | vs looks like it uploads to admin one right now | 13:13 |
| noonedeadpunk | yeah, it does upload to admin | 13:13 |
| noonedeadpunk | I was kinda in progress of writing when realizied that it's time reallocate from home, and everything was done on home laptop) | 13:14 |
| noonedeadpunk | so just pushed not to loose progress lol | 13:14 |
| mnaser | oh lol | 13:14 |
| mnaser | yeah i can pick this up as i have an env im going back and forth on right now | 13:14 |
| noonedeadpunk | ok, than I'll focus on fixing lxc deployments with ganesha | 13:16 |
| mnaser | ++ | 13:17 |
| noonedeadpunk | item.file is undefined as well(( | 13:18 |
| jrosser | is there any doc on the architecture you are aiming at here? | 13:18 |
| jrosser | i've been reading the manila docs to try to figure out what you're doing | 13:19 |
| jrosser | but they are a bit "all choices are available" | 13:19 |
| noonedeadpunk | I think we're doing a bit different things like the docs said :p I'm trying to do in CI smth like https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/13/html-single/cephfs_via_nfs_back_end_guide_for_the_shared_file_system_service/index | 13:20 |
| noonedeadpunk | CephFS via NFS | 13:20 |
| mnaser | jrosser: yeah its very "do as you wish" : | 13:20 |
| mnaser | :( | 13:20 |
| jrosser | whats the manila image doing? | 13:21 |
| jrosser | the share-server right? | 13:21 |
| mnaser | jrosser: thats when you want to use the 'generic' driver | 13:21 |
| mnaser | (which im working to implement in this case) | 13:21 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Add image download for manila https://review.opendev.org/704972 | 13:22 |
| mnaser | kinda like octavia service-model thing | 13:22 |
| jrosser | right yes proper multitenant | 13:22 |
| mnaser | the cephfs via nfs is pretty multitenant too as it uses ganesha and that has some sort of proper setup too | 13:22 |
| mnaser | afaik, | 13:22 |
| jrosser | have had some folk ask for shared nfs recently so this is interesting | 13:22 |
| jrosser | you end up sharing a network though? | 13:23 |
| jrosser | so the different project instances can see each other <- thats maybe my misunderstanding of the ganesha setup | 13:23 |
| mnaser | with generic manila plugs the vm into the tenant straight | 13:23 |
| noonedeadpunk | mnaser: I think it's used for tempest as well for ganesha deployment | 13:23 |
| noonedeadpunk | out of what I see in tempest logs... | 13:24 |
| noonedeadpunk | or maybe it's jsut my setup.... | 13:25 |
| noonedeadpunk | mnaser: seems all patches since https://review.opendev.org/#/c/704935/9 are in conflict because of its dependency | 13:27 |
| mnaser | noonedeadpunk: i think they arent necesarily in conflict but bc that change failed | 13:28 |
| noonedeadpunk | so they depend on https://review.opendev.org/#/c/704942/1 and https://review.opendev.org/#/c/674653/ which are conflicting | 13:29 |
| *** cshen has quit IRC | 13:29 | |
| *** rpittau|bbl is now known as rpittau | 13:29 | |
| mnaser | noonedeadpunk: thats a different repo tho :p | 13:29 |
| noonedeadpunk | yes, but manila patches depends on these two | 13:30 |
| noonedeadpunk | and zuul probably can't use these dependecies as they're conflicting | 13:31 |
| mnaser | yep | 13:31 |
| mnaser | i rechedk the first one which failed | 13:31 |
| mnaser | 2nd one seems ok | 13:31 |
| noonedeadpunk | I mean that probably 704942 should be rebased on top of the 674653 | 13:31 |
| *** rh-jelabarre has quit IRC | 13:36 | |
| *** ansmith has joined #openstack-ansible | 13:38 | |
| *** cshen has joined #openstack-ansible | 13:38 | |
| djhankb | Hello again, I'm still working on my S->T Upgrade and am running into a problem with Swift on the venv build- if anyone would be able to possibly be able to point me in the right direction: http://paste.openstack.org/show/788973/ it would be greatly appreciated. | 13:42 |
| jrosser | djhankb: what operating system is that you are using? | 13:51 |
| djhankb | Debian 9 | 13:52 |
| jrosser | and would this be the swift service virtualenv, on a controller node, with LXC? | 13:56 |
| djhankb | that is correct, I am running "setup-openstack.yml" on TASK [python_venv_build : Install python packages into the venv] | 13:57 |
| djhankb | it also fails on the swift node itself | 13:57 |
| jrosser | can you try something | 13:58 |
| djhankb | of course | 13:58 |
| jrosser | the ansible role is in /etc/ansible/roles/os_swift | 13:58 |
| jrosser | i am thinking there is something missing here https://github.com/openstack/openstack-ansible-os_swift/blob/master/vars/debian.yml#L24-L28 | 13:59 |
| jrosser | could you add "libsystemd-dev" as an extra item in that list | 13:59 |
| djhankb | and this would be on the deployment node correct? | 14:00 |
| jrosser | thats right | 14:01 |
| jrosser | i'm not sure on debian which package provides libsystemd-journal.pc | 14:02 |
| *** pcaruana has joined #openstack-ansible | 14:03 | |
| djhankb | I'll check that out here | 14:04 |
| jrosser | what is probably happening here is that the packages which supply those missing .pc files in your log are provided by another ansible role in the CI tests | 14:05 |
| jrosser | which run on bare metal without containers, so you can get "pollution" between what one role installs and the next one requires | 14:05 |
| jrosser | afaik we don't actually have a CI job which tests swift in an LXC setup | 14:05 |
| jrosser | just as reference, here is what happens in os_glance https://github.com/openstack/openstack-ansible-os_glance/blob/master/vars/debian.yml#L27-L29 | 14:07 |
| *** sshnaidm is now known as sshnaidm|afk | 14:07 | |
| *** bjolo has joined #openstack-ansible | 14:08 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Add Manila key generation and distribution https://review.opendev.org/705019 | 14:14 |
| jrosser | djhankb: libsystemd-journal.pc may not actually be required..... https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=795147 | 14:17 |
| openstack | Debian bug 795147 in libsystemd-journal-dev "libsystemd-journal-dev is missing in Debian Stretch" [Normal,Open] | 14:17 |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Add Manila key generation and distribution https://review.opendev.org/705019 | 14:20 |
| *** heikkine has joined #openstack-ansible | 14:22 | |
| *** sshnaidm|afk is now known as sshnaidm | 14:32 | |
| *** pcaruana has quit IRC | 14:34 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_manila master: Add Manila key generation and distribution https://review.opendev.org/705019 | 14:38 |
| *** rh-jelabarre has joined #openstack-ansible | 14:49 | |
| *** jawad_axd has quit IRC | 14:50 | |
| ioni | i had an issue with octavia not sure if because i have some sort of frankenstein with magnum/octavia/heat from train and the rest from stein | 14:50 |
| ioni | octavia-lib was too old in venv for octavia | 14:50 |
| *** jawad_axd has joined #openstack-ansible | 14:50 | |
| ioni | it was octavia-lib 1.1.1 and octavia 5.0.2devsomething | 14:50 |
| ioni | i had to manually update octavia-lib to latest | 14:51 |
| ioni | it was an upgrade from stein to train for octavia | 14:51 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest stable/train: Use contraints for tempest plugins https://review.opendev.org/704980 | 14:53 |
| *** jawad_ax_ has joined #openstack-ansible | 14:54 | |
| *** jawad_a__ has joined #openstack-ansible | 14:55 | |
| *** jawad_axd has quit IRC | 14:55 | |
| jrosser | ioni: did you make sure the constraints were pointing to the right train for the components you updated? | 14:56 |
| jrosser | right train / right branch | 14:56 |
| ioni | jrosser, yes | 14:57 |
| jrosser | interesting | 14:57 |
| ioni | i did not test to see if is working without any venv, like starting from scratch | 14:58 |
| ioni | i just fixed it manually and get over it | 14:58 |
| *** udesale has joined #openstack-ansible | 14:58 | |
| *** jawad_ax_ has quit IRC | 14:58 | |
| noonedeadpunk | mnaser: https://review.opendev.org/#/c/705019/3 is exactly what I was testing localy before pushing | 14:58 |
| noonedeadpunk | but also public key distibution to nfs hosts | 15:00 |
| *** jawad_a__ has quit IRC | 15:00 | |
| *** pcaruana has joined #openstack-ansible | 15:02 | |
| *** trident has quit IRC | 15:06 | |
| ioni | noonedeadpunk, nice to see that you guys work on manilla | 15:07 |
| ioni | i do want to use it in the near future | 15:08 |
| ioni | with ceph | 15:08 |
| jrosser | ^ same, i think i am most interested in the generic model with per project nfs servers, octavia style | 15:09 |
| *** southquist has quit IRC | 15:09 | |
| jrosser | there is some networking fun to be had there for sure | 15:10 |
| ioni | i don't know if used it wrong but i did activate legacy policy on octavia :) | 15:10 |
| ioni | do you guys want to drop that? | 15:11 |
| jrosser | likewise | 15:11 |
| jrosser | i'd like to understand what to do instead becasue unless you do that, out of the box it's not usuable by end users | 15:12 |
| ioni | i mostly look at manilla for shared volume for kubernets | 15:12 |
| *** trident has joined #openstack-ansible | 15:17 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/train: Use nova_placement_service_password for placement_service_password https://review.opendev.org/704842 | 15:18 |
| *** chkumar|rover is now known as raukadah | 15:24 | |
| djhankb | jrosser: I think your fix got me past where I was however I am now seeing this one which I thought was interesting: ERROR: Error [Errno 2] No such file or directory: 'git' while executing command git clone -q https://opendev.org/openstack/swift /tmp/pip-install-wj3geysk/swift | 15:39 |
| jrosser | djhankb: i would expect that to come in from here | 15:43 |
| *** spatel has quit IRC | 15:43 | |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_keystone master: Federated openid support using auth_mod_openidc https://review.opendev.org/705002 | 15:44 |
| jrosser | djhankb: https://github.com/openstack/openstack-ansible-os_swift/blob/master/vars/debian.yml#L25 | 15:46 |
| *** cshen has quit IRC | 15:52 | |
| *** spatel has joined #openstack-ansible | 15:53 | |
| *** gyee has joined #openstack-ansible | 16:04 | |
| djhankb | jrosser: is that /etc/ansible/roles/os_swift/vars/debian.yml used to build the LXC containers as well? The git-core package was not installed automatically for some reason | 16:05 |
| *** ivve has quit IRC | 16:10 | |
| jrosser | that should be on the venv build host (repo container) if i remember correctly | 16:11 |
| jrosser | i.e part of the python_venv_build role tasks is to install those _devel_ packages onto the venv build host | 16:12 |
| djhankb | gotcha, I don't completely understand the underpinnings of what all is going on here, so I'm trying to educate myself as much as possible as I've been building out my labs. | 16:14 |
| jrosser | run just the os_swift_install playbook | 16:15 |
| djhankb | roger that | 16:15 |
| jrosser | try setting '-e venv_rebuild=yes' on the end of the command | 16:16 |
| jrosser | setting this https://github.com/openstack/ansible-role-python_venv_build/blob/master/defaults/main.yml#L107 | 16:16 |
| jrosser | and watch what/where it does the python_venv_build role tasks from within that | 16:16 |
| *** jawad_axd has joined #openstack-ansible | 16:23 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: manila: refactor playbooks https://review.opendev.org/704942 | 16:24 |
| *** alti_17 has quit IRC | 16:26 | |
| *** jawad_axd has quit IRC | 16:27 | |
| *** admin0 has quit IRC | 16:28 | |
| *** admin0 has joined #openstack-ansible | 16:30 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Add manila_service_region https://review.opendev.org/704957 | 16:31 |
| *** udesale has quit IRC | 16:43 | |
| *** jawad_axd has joined #openstack-ansible | 16:43 | |
| *** jawad_axd has quit IRC | 16:48 | |
| *** jawad_axd has joined #openstack-ansible | 17:04 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: refactor: service group names https://review.opendev.org/704945 | 17:06 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Start using uWSGI role https://review.opendev.org/704935 | 17:07 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: cleanup: config values https://review.opendev.org/704951 | 17:07 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add image download for manila https://review.opendev.org/704972 | 17:07 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add Manila key generation and distribution https://review.opendev.org/705019 | 17:08 |
| *** jawad_axd has quit IRC | 17:08 | |
| spatel | Folks, I am trying setup LDAP auth with keystone and doing keystone_ldap: stanza and running ansible playbook "openstack-ansible os-keystone-install.yml --tags keystone-config" but its not touching any config | 17:15 |
| spatel | does "--tags keystone-config" is correct way to handle this ? | 17:16 |
| noonedeadpunk | on stein tags do not work | 17:18 |
| noonedeadpunk | on train they probably should.... | 17:18 |
| spatel | noonedeadpunk: really???? i am running stein | 17:18 |
| spatel | noonedeadpunk: you are saying i have to run full playbook? | 17:19 |
| spatel | just worried if i break something :O | 17:19 |
| noonedeadpunk | yeah, or replace in main.yml mainly include_tasks with import_tasks where it's applicable | 17:20 |
| noonedeadpunk | I think this is commit https://opendev.org/openstack/openstack-ansible-os_keystone/commit/a2783e4e3f84c2681093c90d5fd2daa0f91dac77 | 17:21 |
| spatel | noonedeadpunk: oh! so i can just make change in line 160 instead of everyone right? | 17:24 |
| spatel | include_tasks: keystone_ldap_setup.yml | 17:24 |
| spatel | to | 17:24 |
| spatel | import_tasks: keystone_ldap_setup.yml | 17:24 |
| *** jawad_axd has joined #openstack-ansible | 17:25 | |
| noonedeadpunk | if you need to run only these tasks than yes | 17:25 |
| spatel | Yes just single task at present.. do i need to do same on keystone_uwsgi.yml file or should i leave it alone | 17:25 |
| noonedeadpunk | There might be a ceveat if some variable which was set during runtime (like with set_fact) is used inside ldap stuff | 17:26 |
| noonedeadpunk | hum, wondering how this works... https://opendev.org/openstack/openstack-ansible-os_keystone/src/branch/master/tasks/main.yml#L135 | 17:27 |
| noonedeadpunk | like import with variable in imported filename shouldn't work.... | 17:28 |
| spatel | hmm... i think i should run full playbook in that case... | 17:29 |
| *** jawad_axd has quit IRC | 17:29 | |
| spatel | this is too much editing... | 17:29 |
| noonedeadpunk | you can try at least | 17:29 |
| noonedeadpunk | change L160 as you said | 17:30 |
| spatel | lets try with L160 and see if i am lucky | 17:30 |
| spatel | wish openstack-ansible has dry run option before commit change | 17:31 |
| *** evrardjp has joined #openstack-ansible | 17:33 | |
| spatel | noonedeadpunk: that works... | 17:34 |
| *** jawad_axd has joined #openstack-ansible | 17:45 | |
| *** jawad_axd has quit IRC | 17:50 | |
| *** DanyC has quit IRC | 17:51 | |
| spatel | noonedeadpunk: Thank you!!! so much... | 17:52 |
| redkrieg | Is it possible to run two different ceph clusters under openstack-ansible? The goal is to use one for glance and one for cinder. | 17:53 |
| *** errr has quit IRC | 17:56 | |
| *** errr has joined #openstack-ansible | 17:56 | |
| *** rpittau is now known as rpittau|afk | 18:02 | |
| *** jawad_axd has joined #openstack-ansible | 18:06 | |
| *** jawad_axd has quit IRC | 18:11 | |
| redkrieg | or a tiered system using crush maps? | 18:16 |
| *** sshnaidm is now known as sshnaidm|afk | 18:20 | |
| *** luksky has quit IRC | 18:25 | |
| *** jawad_axd has joined #openstack-ansible | 18:27 | |
| jrosser | redkrieg: is there a reason to use multiple clusters? glance and cinder currently use different pools anyway so you can put those on different device classes if you wish | 18:29 |
| jrosser | so you can do hdd for glance and ssd for cinder, as an example | 18:30 |
| jrosser | but I think you need to think very very carefully, because what I described there might be a bad idea performance wise | 18:30 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add image download for manila https://review.opendev.org/704972 | 18:31 |
| jrosser | Ceph is smart enough to do a no-op snapshot of a glance image and use that in cinder | 18:31 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add Manila key generation and distribution https://review.opendev.org/705019 | 18:31 |
| *** jawad_axd has quit IRC | 18:31 | |
| jrosser | so you might be surprised in a bad way about th performance you get with “slower” storage for glance | 18:32 |
| redkrieg | jrosser: thanks for the advice. are device classes the only way to differentiate? apparently the reasoning behind keeping the services on different hardware is primarily for capacity planning | 18:34 |
| jrosser | one Ceph cluster could have hosts full of ssd / nvme / hdd on the most appropriate hardware for each | 18:35 |
| jrosser | you can define your own device classes, in inbuilt hdd/ssd just get you going for free | 18:36 |
| jrosser | then you can choose which pools are placed on which device classes | 18:36 |
| jrosser | all of this can happen within one cluster | 18:36 |
| redkrieg | okay, sounds like we'll be using ssd for cinder and hdd for glance. you're saying that reads for the base image will go to the hdd osds and the reads for everything else on the volume will be on ssd? | 18:37 |
| jrosser | I believe so but you should double check | 18:38 |
| jrosser | you can also have multiple Ceph back ends in cinder | 18:38 |
| jrosser | I make hdd and ssd volume types available too | 18:39 |
| redkrieg | thanks, that helps a lot. is openstack-ansible able to handle assigning device classes to the cinder/glance pools or is that something I'll have to configure manually afterward? | 18:39 |
| jrosser | up to you really | 18:39 |
| jrosser | osa is a thin wrapper around Ceph-ansible | 18:39 |
| jrosser | so it’s a choice of you using Ceph-a single directly and deploying your own storage cluster, or putting the same data in osa and having everything there | 18:40 |
| jrosser | argh phone typos | 18:41 |
| redkrieg | awesome, I'll have to look in to that more. thanks again for the pointers, I know what direction to look now. | 18:41 |
| jrosser | really depends on how you want to manage the ceph | 18:41 |
| jrosser | separate concern, or integrated | 18:41 |
| jrosser | personally I separate | 18:41 |
| redkrieg | integrated is always my preference only because I like to have configuration management in a state where I can simply run it again on fresh installs and get the same setup as my production cluster | 18:42 |
| jrosser | there is some value in decoupling the version of Ceph deployed by a particular osa release and what Ceph-ansible can do natively - otherwise you might be backed in a corner down the line | 18:44 |
| jrosser | look at the requirements for Ceph release vs. Ceph-ansible version. vs required ansible version | 18:45 |
| redkrieg | no direct upgrade path in ceph? | 18:45 |
| jrosser | there’s a not terribly nice mapping with the osa release version vs ansible version, for example | 18:45 |
| jrosser | requiring the same a single version for both severely limits your options | 18:46 |
| jrosser | *ansible | 18:46 |
| *** ivve has joined #openstack-ansible | 18:47 | |
| *** jawad_axd has joined #openstack-ansible | 18:47 | |
| *** jawad_axd has quit IRC | 18:52 | |
| *** sshnaidm|afk is now known as sshnaidm|off | 18:53 | |
| spotz | noonedeadpunk: You mind if I clean up some wording in https://review.opendev.org/#/c/705019/5/defaults/main.yml? | 19:00 |
| noonedeadpunk | yeah, sure feel free:) Anyway it's mnaser's patch :p | 19:01 |
| spotz | hehehe | 19:02 |
| jrosser | spotz: o/ thanks for the heads up on the London ops meeting a while back - gshippey and I went a long and it was great | 19:04 |
| spotz | jrosser: great news! Wish I'd been there | 19:04 |
| openstackgerrit | Amy Marrich (spotz) proposed openstack/openstack-ansible-os_manila master: Add Manila key generation and distribution https://review.opendev.org/705019 | 19:05 |
| *** DanyC has joined #openstack-ansible | 19:09 | |
| *** luksky has joined #openstack-ansible | 19:14 | |
| *** DanyC has quit IRC | 19:39 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [WIP] Adjustments for LXC CI https://review.opendev.org/705107 | 19:58 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add LXC CI job https://review.opendev.org/705108 | 20:00 |
| djhankb | jrosser: Just a heads-up, adding 'libsystemd-dev' to 'swift_devel_distro_packages' in /etc/ansible/roles/os_swift/vars/debian.yml and then running the full os-swift-install.yml with '-e venv_rebuild=yes' completed successfully | 20:08 |
| jrosser | djhankb: that’s great | 20:10 |
| jrosser | would you like to make a patch? | 20:10 |
| djhankb | I would love to, however I'm not sure of the exact process | 20:10 |
| jrosser | if you are interested in learning that then this is a very small change which is a great way to start out | 20:11 |
| jrosser | otherwise I can do it for you | 20:11 |
| *** hamzy_ has joined #openstack-ansible | 20:11 | |
| djhankb | of course I would be, if you would be willing to point me in the right direction. | 20:11 |
| jrosser | have a read of this https://docs.openstack.org/infra/manual/developers.html | 20:12 |
| jrosser | first patch you submit involves setting up a bunch of stuff but once that is done it’s much easier for 2nd/3rd..... | 20:13 |
| *** hamzy has quit IRC | 20:14 | |
| jrosser | djhankb: there is a sandbox repo that you can test this all out in https://docs.openstack.org/infra/manual/sandbox.html | 20:16 |
| djhankb | awesome, I already have most of that set up- UbuntuOne, Git SSH Keys, etc. That sandbox looks helpful too. I've used Git a lot on my own personal stuff but not really at all when it comes to shared repos and pull requests and stuff. | 20:18 |
| jrosser | ok, so really the new thing here is the gerrit workflow and git-review which is a little different to what you do on github | 20:19 |
| jrosser | it’s getting a little late here, spotz are you around to help out getting to grips with making patches if needed ^ ? | 20:21 |
| djhankb | Ok cool, I'll give this a go. Thank you so much for your time and patience with helping me out today. | 20:22 |
| spotz | You mean how to submit, etc? That I can do:) | 20:23 |
| djhankb | Thanks spotz | 20:24 |
| spotz | djhankb: If I don't respond right away please bear with me I'm on a conference call | 20:25 |
| jrosser | awesome! thanks for taking the time to contribute djhankb | 20:25 |
| djhankb | I'm a 20+ year sysadmin/network engineer - learning the workflows around coding have been challenging but fun and rewarding. | 20:25 |
| djhankb | spotz: not a problem, I have another thing I'm going to need to wrap up before I can pivot over to the patch | 20:27 |
| *** jawad_axd has joined #openstack-ansible | 20:31 | |
| *** andymccr has quit IRC | 20:35 | |
| *** jawad_axd has quit IRC | 20:35 | |
| spotz | djhankb: Ok off the call, if you haven't yet checkout https://docs.openstack.org/contributors/code-and-documentation/index.html I'm going to get you a link to the PDF I used to use for Git and Gerrit Lunch and Learn in case you still need to get accounts | 20:45 |
| djhankb | oh that would be awesome | 20:46 |
| spotz | djhankb: https://github.com/spotz/ghc/blob/master/gitghcpdf.pdf | 20:46 |
| spotz | Setting people up to contribute is my jam:) | 20:46 |
| djhankb | It's nice to help those help themselves! | 20:47 |
| spotz | I love handson workshops:) | 20:51 |
| *** jawad_axd has joined #openstack-ansible | 20:51 | |
| *** jawad_axd has quit IRC | 20:56 | |
| cjloader | spotz is great! | 21:04 |
| spotz | Thanks cjloader:) | 21:04 |
| *** elenalindq has quit IRC | 21:11 | |
| *** jawad_axd has joined #openstack-ansible | 21:12 | |
| *** jawad_axd has quit IRC | 21:16 | |
| *** ansmith has quit IRC | 21:24 | |
| djhankb | ok spotz - you still there? | 21:28 |
| *** jawad_axd has joined #openstack-ansible | 21:32 | |
| spotz | djhankb: yeep | 21:33 |
| *** jawad_axd has quit IRC | 21:37 | |
| *** rh-jelabarre has quit IRC | 21:38 | |
| djhankb | Cool - So I've got my account all in order, ssh keys added, git configured, and I'm at the point of creating a branch (I think) and am curious if I need to be referencing a specific Bug# or if we just give a good description of the issue | 21:40 |
| djhankb | There wouldn't be a bug# because I just hit it this morning | 21:40 |
| djhankb | (The way your documentation reads is a bit intimidating for the novice!) ;-) | 21:41 |
| *** spatel has quit IRC | 21:46 | |
| *** jawad_axd has joined #openstack-ansible | 21:53 | |
| *** jawad_axd has quit IRC | 21:57 | |
| djhankb | Ok, so once I have the branch created, I simply make my change, commit and then run "git review" ? | 22:04 |
| *** ansmith has joined #openstack-ansible | 22:07 | |
| openstackgerrit | Henry Bonath proposed openstack/openstack-ansible-os_swift master: add libsystemd-dev to swift_devel_distro_packages https://review.opendev.org/705125 | 22:13 |
| djhankb | ooh there it goes! (sorry I am excited) | 22:13 |
| djhankb | spotz: Now that it's been added to Gerrit - is that all I need to do at this point? | 22:17 |
| spotz | djhankb: this yours https://review.opendev.org/#/c/705125/ ? | 22:18 |
| spotz | djhankb: Keep in mind the PDF is for a workshiop with helpers in the room. Sorry had to do real work | 22:19 |
| djhankb | spotz: yes that was me | 22:20 |
| *** jawad_axd has joined #openstack-ansible | 22:21 | |
| spotz | djhankb: Ok so the code itself looks good and some people would review it at this point. I tend to watch to see if it passes tests and gets verified before reviewing. Now if it fails you need to check the logs, sometimes it's not your patch but might need a recheck | 22:22 |
| spotz | Now depending on the load verification can be quick or take a while | 22:24 |
| *** jawad_axd has quit IRC | 22:25 | |
| djhankb | ok right on, I just want to make sure I'm following the proper procedure. This was a simple thing as you could see. | 22:27 |
| spotz | Those are my favorites:) | 22:33 |
| *** jawad_axd has joined #openstack-ansible | 22:42 | |
| *** jawad_axd has quit IRC | 22:47 | |
| *** ivve has quit IRC | 23:07 | |
| *** errr has quit IRC | 23:10 | |
| *** errr has joined #openstack-ansible | 23:10 | |
| *** luksky has quit IRC | 23:21 | |
| *** jawad_axd has joined #openstack-ansible | 23:23 | |
| *** jawad_axd has quit IRC | 23:28 | |
| *** jawad_axd has joined #openstack-ansible | 23:44 | |
| *** jawad_axd has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!