| *** mmalchuk has joined #openstack-ansible | 00:05 | |
| *** gyee has quit IRC | 00:06 | |
| *** mmalchuk has quit IRC | 00:09 | |
| *** guilhermesp has quit IRC | 00:25 | |
| *** guilhermesp has joined #openstack-ansible | 00:25 | |
| *** mmalchuk has joined #openstack-ansible | 00:35 | |
| *** mmalchuk has quit IRC | 00:39 | |
| *** mmalchuk has joined #openstack-ansible | 00:48 | |
| *** mmalchuk has quit IRC | 00:52 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible stable/pike: Revert "Work around nested virt issues on OVH test nodes" https://review.openstack.org/589309 | 00:55 |
|---|---|---|
| openstackgerrit | Logan V proposed openstack/openstack-ansible stable/queens: Revert "Work around nested virt issues on OVH test nodes" https://review.openstack.org/589310 | 00:55 |
| *** gregoryo has joined #openstack-ansible | 00:55 | |
| *** mmalchuk has joined #openstack-ansible | 01:01 | |
| *** mmalchuk has quit IRC | 01:06 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_neutron master: Cleanup files and templates using smart sources https://review.openstack.org/588647 | 01:09 |
| *** mmalchuk has joined #openstack-ansible | 01:14 | |
| *** mmalchuk has quit IRC | 01:19 | |
| *** mmalchuk has joined #openstack-ansible | 01:53 | |
| *** mmalchuk has quit IRC | 01:57 | |
| *** yolanda has quit IRC | 02:03 | |
| *** mmalchuk has joined #openstack-ansible | 02:05 | |
| *** mmalchuk has quit IRC | 02:10 | |
| *** lbragstad has quit IRC | 02:18 | |
| *** mmalchuk has joined #openstack-ansible | 02:18 | |
| *** mmalchuk has quit IRC | 02:22 | |
| *** spatel has joined #openstack-ansible | 02:34 | |
| *** cCain has joined #openstack-ansible | 02:43 | |
| cCain | o/ | 02:44 |
| *** coolj has joined #openstack-ansible | 02:56 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Mark bionic gate jobs as non-voting https://review.openstack.org/589161 | 03:08 |
| *** mmalchuk has joined #openstack-ansible | 03:09 | |
| *** mmalchuk has quit IRC | 03:14 | |
| *** jhesketh_ is now known as jhesketh | 03:14 | |
| *** mmalchuk has joined #openstack-ansible | 03:22 | |
| *** mmalchuk has quit IRC | 03:27 | |
| *** dave-mccowan has quit IRC | 03:31 | |
| *** mmalchuk has joined #openstack-ansible | 03:34 | |
| *** spatel has quit IRC | 03:37 | |
| *** mmalchuk has quit IRC | 03:38 | |
| *** udesale has joined #openstack-ansible | 03:48 | |
| *** mmalchuk has joined #openstack-ansible | 03:52 | |
| *** mmalchuk has quit IRC | 03:56 | |
| *** mmalchuk has joined #openstack-ansible | 04:05 | |
| *** mmalchuk has quit IRC | 04:09 | |
| *** poopcat has quit IRC | 04:09 | |
| *** Bhujay has joined #openstack-ansible | 04:13 | |
| *** Bhujay has quit IRC | 04:30 | |
| *** pdeters has joined #openstack-ansible | 04:59 | |
| *** shyambiradar has joined #openstack-ansible | 05:12 | |
| *** guilhermesp has quit IRC | 05:14 | |
| *** mmalchuk has joined #openstack-ansible | 05:33 | |
| *** shyambiradar has quit IRC | 05:39 | |
| *** shyambiradar has joined #openstack-ansible | 05:39 | |
| *** nicolasbock has joined #openstack-ansible | 05:46 | |
| *** hwoarang has joined #openstack-ansible | 05:57 | |
| *** cCain has quit IRC | 05:59 | |
| *** fghaas has joined #openstack-ansible | 06:05 | |
| *** dxiri has quit IRC | 06:08 | |
| *** DanyC has joined #openstack-ansible | 06:13 | |
| hwoarang | good morning | 06:18 |
| prometheanfire | moin | 06:22 |
| prometheanfire | hwoarang: you going to the ptg? | 06:22 |
| hwoarang | nope | 06:22 |
| prometheanfire | boo | 06:26 |
| prometheanfire | don't want a visit from thomas? | 06:26 |
| prometheanfire | https://media0dk-a.akamaihd.net/39/57/7518bfcf2a4996b3dd5d34c20c4d1101.jpg | 06:27 |
| hwoarang | haha i've heard it will not be there this year :) | 06:29 |
| hwoarang | evrardjp: good morning. can you help me understand your comment on https://review.openstack.org/#/c/584765/ so i can 'upgrade' your vote? :) | 06:30 |
| evrardjp | good morning | 06:30 |
| evrardjp | hwoarang: let's not lose CI time, but if you need to update this, we now define the voting: false or voting: true arguments in projects.yaml not in jobs.yaml | 06:31 |
| hwoarang | oh ok | 06:31 |
| evrardjp | project* | 06:31 |
| hwoarang | ok will do that from now on | 06:32 |
| *** pcaruana has joined #openstack-ansible | 06:36 | |
| prometheanfire | evrardjp: hwoarang I've been bugged about https://review.openstack.org/556586 so passing the bug on to you like any good office env | 06:37 |
| jrosser | i have a couple moreof these ready to go if anyone has a moment https://review.openstack.org/#/q/topic:bionic-check+(status:open) | 06:39 |
| jrosser | evrardjp: also interested in your comments on this https://review.openstack.org/#/c/589146/ after the discussion at the last meeting | 06:40 |
| *** yolanda has joined #openstack-ansible | 06:40 | |
| * hwoarang grabs a coffee and will be back for reviews | 06:41 | |
| *** Bhujay has joined #openstack-ansible | 06:42 | |
| *** DanyC has quit IRC | 06:43 | |
| *** vnogin has quit IRC | 06:47 | |
| *** hamzaachi has joined #openstack-ansible | 06:53 | |
| *** luksky has joined #openstack-ansible | 06:54 | |
| *** phasespace has joined #openstack-ansible | 06:56 | |
| phasespace | When upgrading between major versions, is it possible to skip one version? I.e. is it possible to upgrade from ocata to queens directly, or should we first upgrade to pike, then queens? | 06:56 |
| prometheanfire | phasespace: gernerally possible to skip a single version, ymmv | 06:57 |
| *** shyambiradar has quit IRC | 07:00 | |
| *** shyambiradar has joined #openstack-ansible | 07:00 | |
| phasespace | Ok, thanks. Is a major upgrade usually fraught with difficulty? Are there any particular recommendations on how to proceed? | 07:01 |
| phasespace | I.e. if I'm on an old version of ocata, should I do a minor upgrade before the major upgrade? | 07:02 |
| prometheanfire | a single upgrade is generally ok, it's best to do them one at a time | 07:03 |
| prometheanfire | and best to be up to date before the major | 07:03 |
| prometheanfire | all extra work of course | 07:03 |
| phasespace | I can deal with more work if it's safer | 07:04 |
| *** gkadam has joined #openstack-ansible | 07:04 | |
| phasespace | Thank you very much, prometheanfire | 07:04 |
| prometheanfire | np | 07:08 |
| prometheanfire | like I said, ymmv, ianal, etc. You need to test to know how your infra works | 07:09 |
| *** shyambiradar has quit IRC | 07:13 | |
| evrardjp | prometheanfire: I will have a look | 07:16 |
| prometheanfire | thanks | 07:16 |
| phasespace | What are your experiences with major upgrades? Did it go mostly without a hitch, or did most of you experience some issues that needed to be resolved? | 07:18 |
| prometheanfire | going from N to N+1 works fine in my experience, but that was mostly simple testing | 07:23 |
| *** shyambiradar has joined #openstack-ansible | 07:23 | |
| *** phasespace has quit IRC | 07:26 | |
| *** phasespace has joined #openstack-ansible | 07:26 | |
| jrosser | evrardjp: could you explain for me a bit more when we would see an lxc2->lxc3 upgrade? | 07:35 |
| *** luksky has quit IRC | 07:41 | |
| evrardjp[m] | queens 16.04 -> rocky 16.04 -> rocky 18.04 | 07:42 |
| *** tosky has joined #openstack-ansible | 07:42 | |
| jrosser | but we can't do rocky 16.04 can we? | 07:44 |
| jrosser | oh sorry yes we can | 07:44 |
| *** rmart04 has joined #openstack-ansible | 07:44 | |
| evrardjp[m] | rocky 16 to 18 will be up to the deployer but we'll use a different code to generate the lxc config | 07:44 |
| evrardjp[m] | which will put containers down | 07:45 |
| jrosser | and that would involve an OS reinstallation | 07:45 |
| evrardjp[m] | it's acceptable imo because system upgrades anyway | 07:45 |
| evrardjp[m] | but the code itself was not necessary if it was everything generated with lxc2 and if platform is lxc3 do the config upgrade using the tooling. | 07:46 |
| jrosser | which would youo prefer? | 07:46 |
| evrardjp[m] | i trust your judgment | 07:47 |
| evrardjp[m] | you have taken time to get this in, scratching it would be costly. | 07:47 |
| *** shyambiradar has quit IRC | 07:47 | |
| evrardjp[m] | s/scratching/abandonning/ | 07:48 |
| *** pdeters has quit IRC | 07:55 | |
| Tahvok | Does it mean after upgrading to rocky we would need to insert another controller with ubuntu 18 and then transfer one by one all other controllers to 18 as well? | 08:07 |
| Tahvok | I mean, if you want to go the safe path and not be left with 2 controllers while doing an upgrade to the third.. | 08:07 |
| *** DanyC has joined #openstack-ansible | 08:17 | |
| *** DanyC has quit IRC | 08:18 | |
| prometheanfire | evrardjp: we support rocky on 16.04? | 08:19 |
| prometheanfire | thought that migration was supposed to take place in queens (nice if so) | 08:19 |
| evrardjp | prometheanfire: rocky will be 16.04 for source installs only. | 08:20 |
| *** DanyC has joined #openstack-ansible | 08:20 | |
| evrardjp | rocky is the jump platform,like Newton was | 08:20 |
| evrardjp | Tahvok: in that case I'd add 3 controllers, do the 18.04 upgrade, kill the 16.04 controllers, but that depens on the users I guess. | 08:21 |
| prometheanfire | k | 08:21 |
| evrardjp | Tahvok: let me rephrase | 08:22 |
| evrardjp | if you have 3 16.04 controllers, you can upgrade them (if you're running from source) to 16.04 rocky. | 08:23 |
| evrardjp | then, kill one those, you still have a cluster working. | 08:23 |
| evrardjp | then you add 3 new 18.04 controllers, join to the system. | 08:23 |
| evrardjp | then kill the remaining 2 controllers. | 08:24 |
| evrardjp | that's how I'd do it. But it's not straightforward as usual. | 08:24 |
| evrardjp | The other option is to live upgrade. | 08:24 |
| evrardjp | if you live upgrade and convert the containers to run on lxc3, they would still be just fine | 08:24 |
| evrardjp | you can then flush facts and re-deploy | 08:25 |
| evrardjp | maybe destroy on the way to get new 18.04 containers. | 08:25 |
| * prometheanfire prefers expand/contract, but that takes extra infra | 08:25 | |
| evrardjp | yeah. | 08:26 |
| evrardjp | I find it cleaner too prometheanfire | 08:26 |
| evrardjp | it's harder though | 08:26 |
| prometheanfire | can be, ya | 08:27 |
| prometheanfire | though side effects of in place upgrade can hit you at odd times | 08:27 |
| *** electrofelix has joined #openstack-ansible | 08:33 | |
| hwoarang | the lxc3 is a surprise i think | 08:35 |
| *** aludwar has quit IRC | 08:36 | |
| hwoarang | website says "For production environment, try to stick to LXC 1.0.x or 2.0.x as these are the long term, stable releases which we will support until April 2019 (1.0.x) or April 2021 (2.0.x)." | 08:36 |
| hwoarang | they even have ubuntu repos for these releases. would it be completely wrong to stick with 2.0.X on bionic? | 08:37 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Replace tempest with healthchecks https://review.openstack.org/587544 | 08:37 |
| *** aludwar has joined #openstack-ansible | 08:37 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible master: [docs] Add documentation for 'install_method' variable https://review.openstack.org/589398 | 08:37 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible master: [docs] Add documentation for 'install_method' variable https://review.openstack.org/589398 | 08:39 |
| jrosser | hwoarang: it doesnt look like there is a lxc2 for bionic though? https://launchpad.net/~ubuntu-lxc/+archive/ubuntu/lxc-stable | 08:42 |
| hwoarang | hmm true :/ | 08:44 |
| hwoarang | misleading info on the website i suppose | 08:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Mark bionic gate jobs as non-voting https://review.openstack.org/589160 | 08:45 |
| *** shyambiradar has joined #openstack-ansible | 08:47 | |
| Tahvok | Why add 3 controllers instead of adding only one, then reinstalling the rest one by one? | 08:49 |
| Tahvok | Only look at infra side, you would need less machines in that case... | 08:50 |
| *** luksky has joined #openstack-ansible | 08:51 | |
| jrosser | hwoarang: could you take a look at https://review.openstack.org/#/c/587532/ and https://review.openstack.org/#/c/587531/ if you have a moment? | 08:52 |
| hwoarang | sure | 08:54 |
| *** priteau has joined #openstack-ansible | 08:55 | |
| evrardjp | hwoarang: jrosser let me ask on charms channel | 08:59 |
| Pradiprwt | Hi Everyone, Is anyone aware of this issue >> fatal: [infra1_utility_container-550bfa83]: FAILED! => {"failed": true, "msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"} | 09:06 |
| hwoarang | Pradiprwt: what distro? looks like you are running on a distro that is not known to OSA | 09:07 |
| hwoarang | Pradiprwt: what playbook? we need some more info | 09:07 |
| openstackgerrit | Merged openstack/openstack-ansible master: zuul: Add aio_lxc jobs for distribution installations https://review.openstack.org/584765 | 09:07 |
| Pradiprwt | I am using ubuntu 16.04 | 09:07 |
| Pradiprwt | hwoarang: verbose output >> http://paste.openstack.org/show/727476/ | 09:08 |
| hwoarang | Pradiprwt: what OSA version? | 09:09 |
| *** gregoryo has quit IRC | 09:09 | |
| Pradiprwt | 16.0.16 | 09:09 |
| Pradiprwt | pike | 09:09 |
| Tahvok | Pradiprwt, hwoarang: had exactly this problem 2 weeks ago | 09:11 |
| Tahvok | The only think that helped was to recreate the container | 09:11 |
| evrardjp | Pradiprwt: are your facts up to date? | 09:11 |
| evrardjp | Pradiprwt: it looks like you didn't have the proper facts. | 09:12 |
| Pradiprwt | evrardjp: yes, saturday i have stated deployement | 09:12 |
| evrardjp | Create a play with setup: as argument, and run it with openstack-ansible and it should gather those facts. | 09:12 |
| evrardjp | not sure what you mean there | 09:13 |
| jrosser | oh thats not good - re-running boostrap-ansible.sh shouldnt stacktrace should it? | 09:15 |
| Pradiprwt | should I re-create the container ? | 09:16 |
| evrardjp | jrosser: nope -- could you paste somewhere? | 09:18 |
| evrardjp | Setuptools? | 09:18 |
| evrardjp | that looks fun! | 09:19 |
| jrosser | evrardjp: http://paste.openstack.org/show/727477/ | 09:19 |
| evrardjp | jrosser: remove your pip.conf : p | 09:20 |
| evrardjp | but wait | 09:20 |
| evrardjp | is that done with bootstrap-ansible.sh ? | 09:20 |
| jrosser | yes, see the first line | 09:21 |
| evrardjp | that's interesting | 09:21 |
| evrardjp | you have the osa venv activate? | 09:21 |
| evrardjp | activated* ? | 09:21 |
| jrosser | ? | 09:21 |
| evrardjp | ok I think we miss an --isolated flag on L197 | 09:22 |
| evrardjp | let me fix that quick | 09:22 |
| jrosser | i'm just following the aio quickstart - that says nothing about venv | 09:22 |
| evrardjp | oh but we have. | 09:23 |
| evrardjp | whivh version is that? | 09:23 |
| jrosser | https://docs.openstack.org/openstack-ansible/latest/user/aio/quickstart.html | 09:24 |
| evrardjp | oh well I guess you have an old milestone? | 09:24 |
| evrardjp | mmm | 09:25 |
| evrardjp | I don't think this can happen on following the docs on a clean node | 09:25 |
| evrardjp | did you follow: https://docs.openstack.org/openstack-ansible/latest/user/aio/quickstart.html#rebuilding-an-aio ? | 09:25 |
| jrosser | ahha - that'll be it | 09:26 |
| jrosser | though i thought i could pull latest roles with bootstrap-ansible.sh on its own | 09:26 |
| jrosser | ^ which is what i wanted to do - just pull in the lastest merges to the roles | 09:27 |
| evrardjp | it would, but you deployed already | 09:28 |
| evrardjp | so it means you still have the pip.conf | 09:28 |
| jrosser | yeah, i see that now :) | 09:28 |
| evrardjp | weirdly that should get ignored because --isolated. | 09:29 |
| evrardjp | which code do you run? | 09:29 |
| evrardjp | precisely | 09:29 |
| jrosser | this is master | 09:29 |
| evrardjp | sha? | 09:29 |
| jrosser | bd00e7c1ab8e8e7e04d1d505deb611fac5c7fa51 | 09:29 |
| evrardjp | that's very weird | 09:30 |
| Pradiprwt | evrardjp: How to create a play with setup: as argument, to gather facts ? | 09:32 |
| evrardjp | jrosser: L193 is super scary: http://paste.openstack.org/show/727477/ | 09:34 |
| *** guilhermesp has joined #openstack-ansible | 09:35 | |
| Pradiprwt | i have deleted the container but giving same error while recreating >> http://paste.openstack.org/show/727478/ | 09:37 |
| evrardjp | jrosser: ok | 09:39 |
| evrardjp | Pradiprwt: please gather facts like I suggested. | 09:40 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_almanach master: Switch to using project-templates https://review.openstack.org/587531 | 09:40 |
| evrardjp | jrosser: ok could you do something for me on your interesting node ? | 09:40 |
| jrosser | sure | 09:40 |
| jrosser | oh well i did just move the pip.conf aside and reboostrap | 09:41 |
| evrardjp | (assuming you still have the pip.conf that is preventing stuff) | 09:41 |
| jrosser | i still have the pip.conf though just renamed | 09:41 |
| evrardjp | could you put it back and try this command ? | 09:41 |
| evrardjp | virtualenv --python=/usr/bin/python2 --clear --never-download /opt/ansible-runtime | 09:41 |
| Pradiprwt | evrardjp: I don't know how to do that, can you please give me some guide or refrence | 09:42 |
| evrardjp | Pradiprwt: I told you above. | 09:42 |
| jrosser | evrardjp: http://paste.openstack.org/show/727479/ | 09:42 |
| evrardjp | hahah it works | 09:42 |
| evrardjp | jrosser: great! | 09:43 |
| evrardjp | #victorydance | 09:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Ensure the octavia flavor is private https://review.openstack.org/589239 | 09:43 |
| evrardjp | let me check what this code does, and since when it's shipped :D | 09:43 |
| evrardjp | jrosser: could you try something else real quick too? | 09:44 |
| jrosser | sure no problem | 09:45 |
| evrardjp | 1) re-running this without the --never-download, see if it fails | 09:45 |
| jrosser | stactrace | 09:45 |
| noonedeadpunk | odyssey4me are you here? | 09:45 |
| evrardjp | 2) re-running with --no-pip --no-setuptools --no-wheel | 09:45 |
| *** fghaas has quit IRC | 09:46 | |
| evrardjp | after step 2, you need to run | 09:46 |
| jrosser | hold on - does 2) have --never-download or not? | 09:46 |
| evrardjp | /opt/ansible-runtime/bin/pip install --isolated --upgrade pip setuptools wheel | 09:46 |
| evrardjp | 2) does not have never download | 09:46 |
| odyssey4me | evrardjp jrosser I'm guessing this is a new complexity after merging the change to use packages? | 09:47 |
| evrardjp | odyssey4me: not really | 09:47 |
| evrardjp | it's an upgrade case | 09:47 |
| odyssey4me | ah ok, well good thing we catch it now :) | 09:47 |
| evrardjp | virtualenv is taking the pip.conf | 09:47 |
| evrardjp | in upgrades we remove the pip.conf | 09:47 |
| evrardjp | I think there is a way to avoid this | 09:47 |
| evrardjp | with virtualenv | 09:47 |
| evrardjp | (during virtualenv bootstrap) | 09:48 |
| odyssey4me | is this for queens, or rocky? | 09:48 |
| evrardjp | rocky | 09:48 |
| jrosser | evrardjp: http://paste.openstack.org/show/727481/ | 09:48 |
| evrardjp | I guess we can consider it a regression | 09:48 |
| evrardjp | jrosser: as expected. | 09:49 |
| evrardjp | great. | 09:49 |
| evrardjp | you've confirmed things for me, that's perfect! | 09:49 |
| evrardjp | let me check those options | 09:49 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible master: [docs] Add documentation for 'install_method' variable https://review.openstack.org/589398 | 09:49 |
| odyssey4me | yeah, you can't do no-pip and expect pip to be there... and we don't want to use get-pip.py to put it there any more | 09:50 |
| *** shyambiradar has quit IRC | 09:50 | |
| evrardjp | odyssey4me: oh fyi: https://virtualenv.pypa.io/en/stable/reference/#creating-your-own-bootstrap-scripts | 09:50 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_almanach master: Switch to using project-templates https://review.openstack.org/587531 | 09:50 |
| evrardjp | odyssey4me: yeah so my idea was that virtualenv ships with those packages anyway | 09:50 |
| evrardjp | pip setuptools etc. | 09:50 |
| odyssey4me | yes it does, and we resolved to use those in your patch | 09:51 |
| evrardjp | but virtualenv itself is a command that respects pip and tries to download instead of using what's shipped inside | 09:51 |
| evrardjp | so --never-download uses its own code | 09:51 |
| *** shyambiradar has joined #openstack-ansible | 09:51 | |
| evrardjp | I will check which version introduced it (I think it's a behavior change based on versions of virtualenv, it was default before, and is not the default anymore) | 09:52 |
| odyssey4me | ah yes, I see that - I think we did use that once | 09:52 |
| odyssey4me | I vaguely recall that perhaps one of the distributions didn't support it... I guess we'll see | 09:53 |
| *** rmart04 has quit IRC | 09:53 | |
| odyssey4me | oh no, I'm mixing it up with the '--always-copy' flag | 09:54 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_keystone/blob/master/tasks/keystone_install_source.yml#L97-L98 | 09:55 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_congress master: Switch to using project-templates https://review.openstack.org/587532 | 09:57 |
| evrardjp | odyssey4me: so never download in the code seem to have been through a series of default configurations | 09:57 |
| odyssey4me | evrardjp hwoarang jrosser so, I'm a little confused about https://review.openstack.org/#/c/587532/4/zuul.d/project.yaml | 09:57 |
| evrardjp | in 13.1.2 (minimum version of suse), it seems never download exists but gets ignored as it will not download. However latest version of virtualenv is different, so I am checking where and why it reappeared. | 09:58 |
| odyssey4me | by using the template, and then overriding the config to make jobs non-voting, we're essentially making a bunch of jobs run in gate which aren't voting - which is a waste of CI resources | 09:58 |
| jrosser | this is my fault, sorry | 09:58 |
| odyssey4me | With these repositories, does it not make sense to rather not use the template? | 09:58 |
| odyssey4me | Perhaps rather just implement jobs in check that are non-voting with a note to use the template when they are? | 09:59 |
| jrosser | converting to the template has bought in the bionic jobs where previously they were missing | 09:59 |
| jrosser | and then a bunch of collatoral because the other distros are all brok for that role | 09:59 |
| odyssey4me | jrosser: yeah, understood | 10:00 |
| odyssey4me | perhaps we should be changing the templates - something like a template per distro? | 10:00 |
| odyssey4me | sorry, you've been hacking away at this and my questions are coming from the side | 10:00 |
| jrosser | however - for the vast majority of them the brok is actually quite small | 10:00 |
| jrosser | missing vars and the like for opensuse seems quite common | 10:01 |
| jrosser | i really don't mind which way this goes tbh | 10:02 |
| odyssey4me | yep, there's never been the effort made to get those roles to support more distributions - many of them are new, and many are just fringe roles | 10:02 |
| evrardjp | odyssey4me: https://github.com/pypa/virtualenv/commit/f50df26c44311390351d0a76c79cfc136241ece5 and https://github.com/pypa/virtualenv/commit/3d7361ff2e31472cb69d00150fbdf5a3c9af2a0d -- suse has 13.1.2 which doesn't include this code, so no defaults were applied because old code auto never downloads. | 10:02 |
| evrardjp | interesting. | 10:02 |
| odyssey4me | in that patch in particular, you could just be adding 'openstack-ansible-functional-ubuntu-bionic' as a job in check and gate and move on - it's simple and focused on the intention of your work which is to get bionic tested and working | 10:02 |
| odyssey4me | the conversion to use templates sems to me a distraction | 10:03 |
| odyssey4me | *seems | 10:03 |
| jrosser | i can do that | 10:03 |
| noonedeadpunk | odyssey4me: lxc_containers_create fails on this step https://github.com/openstack/openstack-ansible-lxc_container_create/blob/b2020d5c10b7def43490fc192eb70969bf80517a/tasks/lxc_container_config.yml#L71, if it's run not from root user. If I run test by root, everything is okey. And the interesting thing is, that containers created by non-root user doesn't have any ethernet interface (probably they are cleaned by "Remove legacy | 10:03 |
| evrardjp | odyssey4me: I think we shouldn't make it non voting | 10:03 |
| noonedeadpunk | network config" step). But why it's happening only for non-root user - is a question for me now. | 10:03 |
| odyssey4me | well, I'd like thoughts on the matter from hwoarang and evrardjp | 10:03 |
| evrardjp | we jsut break stuff and say "please fix it if that matters to you | 10:03 |
| evrardjp | hahah | 10:03 |
| odyssey4me | evrardjp: that's the point - by forcing the template to be used we're adding a bunch of tests to gate which shouldn't be there | 10:04 |
| evrardjp | odyssey4me: I don't mean changing the project. | 10:04 |
| odyssey4me | I'm thinking it'd be better to just add the bionic job and leave converting the role to use the templates out of it | 10:04 |
| evrardjp | odyssey4me: I think new works should be in their own template. | 10:05 |
| evrardjp | so for example, bringing bionic support would be in its own template. | 10:05 |
| evrardjp | the point of using a template is to make sure consistency happens (to avoid the horizon problem we had earlier) | 10:05 |
| evrardjp | (as a reminder: horizon was voting for bionic, but keystone was not, and a patch on keystone merged that broke bionic) | 10:06 |
| *** kaiokmo has quit IRC | 10:06 | |
| odyssey4me | ok, I have no idea what you mean | 10:06 |
| evrardjp | haha | 10:07 |
| evrardjp | let me rephrase that | 10:07 |
| odyssey4me | doing a template for a single release seems silly to me - it's a single job | 10:07 |
| evrardjp | while I think setting up jobs in each repo for new work is fine, and was the way to go, maybe we should think differently | 10:07 |
| odyssey4me | doing it for a single distro seems fine to me - xenial & bionic, opensuse x and y | 10:07 |
| odyssey4me | centos of course is as old as the hills and doesn't have another release | 10:07 |
| odyssey4me | but meh, I don't think how we use the templates should be part of jrosser's work to get bionic done... it's a distraction | 10:08 |
| jrosser | my priority right now is to get bionic roles voting | 10:08 |
| evrardjp | agreed, if ppl respect his work. | 10:08 |
| odyssey4me | it can be improved or re-modeled after as a seperate body of work | 10:08 |
| odyssey4me | so for now, I suggest, just adding the single job to any repo that doesn't use the template | 10:09 |
| evrardjp | jrosser: ^ | 10:09 |
| jrosser | ok well it's mostly all done, and via the template | 10:09 |
| jrosser | those which arent merged i can change | 10:09 |
| odyssey4me | yep, works for me | 10:09 |
| evrardjp | jrosser: those who fail shouldn't be done using the template | 10:10 |
| odyssey4me | the shortest possible path to success | 10:10 |
| evrardjp | yeah. | 10:10 |
| jrosser | speaking of which this needs moving on https://review.openstack.org/#/c/589158/ | 10:10 |
| jrosser | https://review.openstack.org/#/c/586756/ | 10:10 |
| jrosser | likewise unless we want to edit them | 10:11 |
| jrosser | i was asked to get the broken stuff non-voting *before* we switch the majority to voting so this is critical path | 10:11 |
| odyssey4me | to me https://review.openstack.org/#/c/586756/10/zuul.d/project.yaml makes sense if someone's going to pick up making that role with with opensuse/centos very soon | 10:12 |
| odyssey4me | if not, we should not be making it use the template | 10:12 |
| odyssey4me | that's far too many jobs in gate which aren't voting | 10:12 |
| jrosser | as that looks just like missing vars i agree | 10:12 |
| evrardjp | steps were simple before the templates: we added non voting jobs, we made them voting. | 10:13 |
| evrardjp | and added them to gates. | 10:13 |
| evrardjp | we should keep the same path in the future. | 10:13 |
| evrardjp | and when it's voting on all our "core" repos part of an aio deploy, then move this to template. | 10:13 |
| evrardjp | then when IN the template, it means it's an expected thing for minimum roles in the future. That means part of the maturity checking. | 10:14 |
| evrardjp | it seems like a story when you do it like that. | 10:14 |
| evrardjp | it helps clarify what is the maturity level of a role | 10:14 |
| evrardjp | if the role doesn't pass that maturity level, then we downgrade it, according to existing procedures. | 10:15 |
| evrardjp | we evince those roles from the integrated build, and they become "community extras" | 10:15 |
| odyssey4me | fair enough - enough pontificating, what's the actionable outcome for jrosser ;) | 10:15 |
| evrardjp | odyssey4me: we need a clear story for docs. | 10:16 |
| evrardjp | that looks acceptable to me. | 10:16 |
| evrardjp | For jrosser, everything that's not ready for template should not use template. | 10:16 |
| evrardjp | like TODO(jrosser): re-use template when x is finally passing. | 10:17 |
| evrardjp | but I am not 100% aware of the current state, so I trust the current status, and we'll clarify in the long run | 10:17 |
| odyssey4me | the exception being if the role is only not passing bionic because that's the subject of this body of work | 10:17 |
| evrardjp | jrosser: do you have an update on the etherpad on what's using templates? | 10:17 |
| evrardjp | odyssey4me: so because we introduced (my bad) template too soon, we need to deal with this. | 10:18 |
| jrosser | no, but you can see here https://review.openstack.org/#/q/topic:bionic-check+(status:open+OR+status:merged) | 10:18 |
| *** jamesdenton has quit IRC | 10:18 | |
| jrosser | thats what has changed, and mostly incorrectly | 10:18 |
| *** electrofelix has quit IRC | 10:18 | |
| odyssey4me | so, for example, I think https://review.openstack.org/589158 is perfectly ok - because I know that state will not last long given that bionic is the target of jrosser's work | 10:19 |
| jrosser | congress is totall madness though for example | 10:19 |
| odyssey4me | if the move to using templates has merged already, then I guess we carry a TODO once bionic is done to go back and either make those other platforms work, or switch the job config back | 10:19 |
| jrosser | i had planned to pass across this again after bionic and pick up the low hanging fruit of the other distros | 10:20 |
| odyssey4me | thanks jrosser - appreciate the work you're putting into this! | 10:20 |
| evrardjp | yeah that's very appreciated. | 10:21 |
| jrosser | no worries - feels like im missing a bunch of assumed knowledge/process though :/ | 10:21 |
| evrardjp | interesting, almanach is passing suse? | 10:21 |
| evrardjp | https://review.openstack.org/#/c/587531/6/zuul.d/project.yaml | 10:22 |
| odyssey4me | jrosser: it may not be assumed, it may also be that we're constantly learning - but if there is something that you find is assumed or newly learned, we should document it | 10:22 |
| odyssey4me | almanach does almost nothing, so that's not all that surprising | 10:22 |
| jrosser | oh something has gone all a bit wrong there? | 10:25 |
| jrosser | let me fix that | 10:26 |
| *** olivierbourdon38 has joined #openstack-ansible | 10:26 | |
| *** electrofelix has joined #openstack-ansible | 10:26 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_magnum master: Switch to using project-templates https://review.openstack.org/588380 | 10:31 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_almanach master: Add CI jobs for bionic https://review.openstack.org/587531 | 10:32 |
| hwoarang | odyssey4me: leaving for lunch now so I will check backlog a bit later if that's OK | 10:33 |
| hwoarang | i saw some highlights there | 10:34 |
| *** shyambiradar has quit IRC | 10:34 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_congress master: Add CI jobs for bionic https://review.openstack.org/589423 | 10:36 |
| *** jamesdenton has joined #openstack-ansible | 10:37 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_magnum master: Switch to using project-templates https://review.openstack.org/588380 | 10:38 |
| openstackgerrit | Maksim Malchuk proposed openstack/openstack-ansible master: Add an ability to use custom PyPi index for deploy https://review.openstack.org/577652 | 10:44 |
| *** sm806 has joined #openstack-ansible | 10:47 | |
| *** shyambiradar has joined #openstack-ansible | 10:49 | |
| *** vnogin has joined #openstack-ansible | 10:49 | |
| *** fghaas has joined #openstack-ansible | 10:54 | |
| *** shyambiradar has quit IRC | 10:54 | |
| *** bradm has quit IRC | 10:56 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-os_neutron master: Adds networking-ovn support https://review.openstack.org/584067 | 11:03 |
| *** galaxyblr has joined #openstack-ansible | 11:04 | |
| *** ansmith has quit IRC | 11:06 | |
| jrosser | is this more broken log collection? https://github.com/openstack/openstack-ansible/blame/master/scripts/scripts-library.sh#L204 | 11:08 |
| jrosser | because this http://logs.openstack.org/07/586607/8/check/openstack-ansible-deploy-aio_lxc-ubuntu-bionic/f28ca34/job-output.txt.gz#_2018-08-07_07_15_13_795384 | 11:09 |
| *** cshen has joined #openstack-ansible | 11:14 | |
| odyssey4me | jrosser: yeah, wrong function name again: https://github.com/openstack/openstack-ansible/blob/master/scripts/scripts-library.sh#L142 | 11:14 |
| *** udesale has quit IRC | 11:15 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/queens: Fix gate log collection https://review.openstack.org/589427 | 11:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Fix gate log collection https://review.openstack.org/589428 | 11:17 |
| jrosser | i'm just fixing this next one | 11:17 |
| *** Bhujay has quit IRC | 11:18 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Fix log collection https://review.openstack.org/589430 | 11:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_congress master: Add CI jobs for bionic https://review.openstack.org/589423 | 11:22 |
| odyssey4me | jrosser: quick commit subject suggestion there | 11:31 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Fix log compression https://review.openstack.org/589430 | 11:34 |
| *** vnogin has quit IRC | 11:36 | |
| *** shyambiradar has joined #openstack-ansible | 11:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Add qdrouterd user to mq setup https://review.openstack.org/589234 | 11:41 |
| odyssey4me | could I get another set of eyes on https://review.openstack.org/589234 please as I'd like to roll that into the changes I'm making to the roles | 11:43 |
| *** guilhermesp has quit IRC | 11:45 | |
| *** guilhermesp has joined #openstack-ansible | 11:46 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: Allow tags to be used for MQ tasks https://review.openstack.org/589437 | 11:46 |
| *** vnogin has joined #openstack-ansible | 11:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_almanach master: Add CI jobs for bionic https://review.openstack.org/587531 | 11:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican master: Allow tags to be used for MQ tasks https://review.openstack.org/589438 | 11:49 |
| jrosser | the artefact sync at the end of a job is very slow - coming up on 10mins - is that normal? | 11:50 |
| odyssey4me | jrosser: for roles, I noticed that the journal extraction is taking much longer | 11:51 |
| jrosser | it took 5 mins to copy the ara sqlite | 11:51 |
| odyssey4me | ever since it was added, the log collection is much, much slower | 11:51 |
| odyssey4me | for the integrated build we don't have that though | 11:51 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer master: Allow tags to be used for MQ tasks https://review.openstack.org/589440 | 11:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cloudkitty master: Allow tags to be used for MQ tasks https://review.openstack.org/589441 | 11:55 |
| *** vnogin has quit IRC | 11:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_congress master: Allow tags to be used for MQ tasks https://review.openstack.org/589442 | 11:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cloudkitty master: Allow tags to be used for MQ tasks https://review.openstack.org/589441 | 11:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_congress master: Use generic vars file for ubuntu https://review.openstack.org/589215 | 11:58 |
| *** Emine has joined #openstack-ansible | 12:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Ensure the octavia flavor is private https://review.openstack.org/589239 | 12:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Ensure the octavia network setup is in the right project https://review.openstack.org/589240 | 12:01 |
| jrosser | odyssey4me: TASK [Run log collection script] -> 20mins on centos! | 12:02 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Mark bionic gate jobs as non-voting https://review.openstack.org/589158 | 12:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Ensure the octavia network setup is in the right project https://review.openstack.org/589240 | 12:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Add retries to service role creation https://review.openstack.org/589245 | 12:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Use interface instead of endpoint_type https://review.openstack.org/589246 | 12:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Clean up tests a bit https://review.openstack.org/589259 | 12:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Add retries to all os_* module tasks https://review.openstack.org/589445 | 12:14 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests master: Change bionic to be voting https://review.openstack.org/587478 | 12:16 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible master: [docs] Add documentation for 'install_method' variable https://review.openstack.org/589398 | 12:17 |
| savvas | GM everyone | 12:23 |
| *** Bhujay has joined #openstack-ansible | 12:24 | |
| savvas | I can't seem to get Octavia to run properly, probably missing something: http://paste.openstack.org/show/727497/ | 12:26 |
| jamesdenton | savvas what version are you working with? | 12:29 |
| savvas | 17.0.7 I think | 12:29 |
| savvas | latest Queens stable | 12:29 |
| savvas | yes 17.0.7 | 12:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift master: Revert "Unblock gate: Provide default rabbitmq password" https://review.openstack.org/589450 | 12:31 |
| jamesdenton | savvas In the neutron server container... can you check the bridge mappings in ml2_conf.ini? | 12:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate master: Allow tags to be used for MQ tasks https://review.openstack.org/589452 | 12:36 |
| savvas | jamesdenton: http://paste.openstack.org/show/727501/ | 12:37 |
| jamesdenton | thx | 12:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Allow tags to be used for MQ tasks https://review.openstack.org/589454 | 12:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat master: Allow tags to be used for MQ tasks https://review.openstack.org/589455 | 12:42 |
| savvas | Any thoughts jamesdenton? The playbook will also fail on a syntax error if I uncomment the octavia-housekeeping under group binding | 12:45 |
| *** diliprenkila has joined #openstack-ansible | 12:46 | |
| jamesdenton | savvas was just looking at the playbooks to see if something unexpected was being drawn in. not seeing it. we should have some octavia folks around in a little bit, if not already | 12:47 |
| openstackgerrit | Merged openstack/openstack-ansible-os_congress master: Add CI jobs for bionic https://review.openstack.org/589423 | 12:47 |
| jamesdenton | but it may be worthwhile to file a bug | 12:47 |
| jamesdenton | on both issues | 12:47 |
| savvas | right, so we are in agreement that this should not be happening and that it is not a configuration issue | 12:50 |
| *** mmalchuk has quit IRC | 12:50 | |
| *** phasespace has quit IRC | 12:50 | |
| *** DimGR has quit IRC | 12:50 | |
| savvas | I wasn't sure if I was reading the docs right, but I can't make sense of it. | 12:50 |
| *** mmalchuk_ has joined #openstack-ansible | 12:50 | |
| *** xteknoid has quit IRC | 12:53 | |
| jamesdenton | Do you happen to have the full log you can post? I can't really tell if the physical_network name was obscured by Ansible or if ******** was the actual name it was attempting to use | 12:53 |
| *** shyambiradar has quit IRC | 12:54 | |
| noonedeadpunk | odyssey4me: for what reason masakari_venv_download was changed here https://review.openstack.org/#/c/584629/10/defaults/main.yml ? As with this change test fails on venv download attempt | 12:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: Allow tags to be used for MQ tasks https://review.openstack.org/589437 | 12:56 |
| odyssey4me | noonedeadpunk: when the role joins the integrated build, the venv must not be downloaded for the monitors | 12:57 |
| odyssey4me | noonedeadpunk: otherwise it will have the wrong venv | 12:58 |
| *** shyambiradar has joined #openstack-ansible | 13:01 | |
| *** ansmith has joined #openstack-ansible | 13:03 | |
| savvas | jamesdenton: obscured by Ansible, not sure why it does that | 13:08 |
| *** jamesdenton has quit IRC | 13:10 | |
| *** shyambiradar has quit IRC | 13:10 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic master: Allow tags to be used for MQ tasks https://review.openstack.org/589461 | 13:10 |
| *** lbragstad has joined #openstack-ansible | 13:11 | |
| *** jamesdenton has joined #openstack-ansible | 13:11 | |
| *** ArchiFleKs has quit IRC | 13:16 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Reduce log level on CentOS by default https://review.openstack.org/589463 | 13:20 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests master: Reduce log level on CentOS by default https://review.openstack.org/589464 | 13:22 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set CentOS kernel logging to quiet in gates https://review.openstack.org/589463 | 13:23 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone master: Allow tags to be used for MQ tasks https://review.openstack.org/589466 | 13:24 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set CentOS kernel logging to 'quiet' in CI https://review.openstack.org/589463 | 13:24 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set CentOS kernel logging to 'quiet' in AIO https://review.openstack.org/589463 | 13:25 |
| openstackgerrit | Maksim Malchuk proposed openstack/openstack-ansible stable/queens: Add an ability to use custom PyPi index for deploy https://review.openstack.org/577653 | 13:29 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set kernel logging to 'quiet' in AIOs https://review.openstack.org/589463 | 13:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum master: Allow tags to be used for MQ tasks https://review.openstack.org/589468 | 13:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_masakari master: Allow tags to be used for MQ tasks https://review.openstack.org/589470 | 13:36 |
| *** ArchiFleKs has joined #openstack-ansible | 13:41 | |
| *** galaxyblr has quit IRC | 13:44 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Allow tags to be used for MQ tasks https://review.openstack.org/589473 | 13:44 |
| errr | is it possible to use openstack-ansible-plugins is a project that isnt part of OSA? I was looking and dont see it in pip so I wondered if it had deps that would keep it from working if I wasnt using OSA | 13:44 |
| odyssey4me | errr: it's usable outside of OSA as it has no real deps - what specifically are you keen on re-using? | 13:45 |
| errr | odyssey4me: well you know how I love to hate federation.. Im currently doing an rpc-r gig where I have to get okta working on it | 13:46 |
| *** ArchiFleKs has quit IRC | 13:46 | |
| errr | so I want to use the keystone module | 13:46 |
| odyssey4me | errr: ah ok, specifically for the federation mapping imports - or something else? | 13:47 |
| odyssey4me | most of what the keystone module does is now available in ansible upstream modules | 13:47 |
| odyssey4me | we're actively working on phasing that module out | 13:47 |
| *** ArchiFleKs has joined #openstack-ansible | 13:47 | |
| errr | yeah, Ive already robbed from the playbooks to get the shibboleth stuff all moved over for this one off thing, but now I need to do t the keystone config with setting up mapping and all that | 13:48 |
| errr | well this is literally a 1 off and I just have to get it working by thursday | 13:48 |
| odyssey4me | errr: ah ok - at some point I'll move onto the federation bits and remove the use of the module, as those things were needed because the federation implementation was done prior to there being any openstack/keystone client capability to do what it does | 13:49 |
| odyssey4me | now there is, so I'll likely switch it up to simply write out a json file from the config, then import it using the openstack client | 13:49 |
| errr | hey, so okta now provides free developer accounts with saml2 support so maybe a test could be made using it so things dont break when changed? | 13:50 |
| savvas | jamesdenton: https://bugs.launchpad.net/openstack-ansible/+bug/1785824 | 13:50 |
| openstack | Launchpad bug 1785824 in openstack-ansible "Octavia fails on octavia-housekeeping group binding" [Undecided,New] | 13:50 |
| *** jamesdenton has quit IRC | 13:52 | |
| errr | or maybe some way to automate out some testshib tests or something.. | 13:52 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Allow tags to be used for MQ tasks https://review.openstack.org/589480 | 13:52 |
| *** jamesdenton has joined #openstack-ansible | 13:53 | |
| odyssey4me | errr: yeah, would be nice - one day I'll do it, but if anyone else is keen and has the know-how we'd welcome some sort of CI test | 13:53 |
| *** mmalchuk_ has quit IRC | 13:53 | |
| errr | sadly I just get dumped on to short term contracts and have just enough time to get a bug fixed when I am working on this stuff | 13:54 |
| odyssey4me | I've been thinking I might just do it in external CI some day, that way it could validate against MS Federation, OKTA and others. | 13:54 |
| errr | we really need to get sales to make this a priority to product so one of us will get time assigned to make this happen. | 13:55 |
| *** ArchiFleKs has quit IRC | 13:56 | |
| *** ArchiFleKs has joined #openstack-ansible | 13:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Allow tags to be used for MQ tasks https://review.openstack.org/589485 | 13:57 |
| cloudnull | mornings | 13:57 |
| errr | hi cloudnull | 13:57 |
| *** vnogin has joined #openstack-ansible | 13:57 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_sahara master: Allow tags to be used for MQ tasks https://review.openstack.org/589489 | 14:01 |
| *** vnogin has quit IRC | 14:02 | |
| *** ArchiFleKs has quit IRC | 14:02 | |
| *** ArchiFleKs has joined #openstack-ansible | 14:03 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tacker master: Allow tags to be used for MQ tasks https://review.openstack.org/589490 | 14:04 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_neutron master: Cleanup files and templates using smart sources https://review.openstack.org/588647 | 14:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_trove master: Allow tags to be used for MQ tasks https://review.openstack.org/589491 | 14:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_watcher master: Allow tags to be used for MQ tasks https://review.openstack.org/589492 | 14:08 |
| *** eglute has joined #openstack-ansible | 14:10 | |
| odyssey4me | cloudnull: really happy to see you pick up on that work - that's been on my TODO for ages, and just haven't managed to get there | 14:11 |
| cloudnull | ++ | 14:11 |
| odyssey4me | it'd be super to get that into rocky if we can | 14:11 |
| cloudnull | totally agree. saves a lot of operations and work for future | 14:12 |
| *** vnogin has joined #openstack-ansible | 14:12 | |
| odyssey4me | I was also thinking that we just symlink the config folder to the venv - that automatically makes rolling back an option to develop in the future for minor upgrades. | 14:12 |
| cloudnull | ++ that was my thoughts as well | 14:13 |
| cloudnull | additionally if /openstack/venvs is a bind mount the config is totally outside the container. | 14:13 |
| jrosser | bionic -> voting if folks agree it is time https://review.openstack.org/#/c/587478/ | 14:14 |
| odyssey4me | cloudnull: FYI, not sure if this will give you ideas you love or hate, but I did keystone a while back: https://github.com/openstack/openstack-ansible-os_keystone/commit/8367e442304a16687be602d4a23762d3408b9010 | 14:14 |
| cloudnull | ++ I took that and what mnaser did for cinder and made a pattern that we can apply to most, if not all, roles. https://review.openstack.org/#/q/topic:smart-sources+(status:open+OR+status:merged) | 14:15 |
| cloudnull | I did those few just to let folks see the pattern | 14:15 |
| cloudnull | its a fairly easy change to apply across the board | 14:16 |
| odyssey4me | makes sense to me, although I'm tied up right now - so others will need to review and see if it makes sense for them | 14:17 |
| *** Bhujay has quit IRC | 14:18 | |
| odyssey4me | my only concern after skimming it is that it feels complex | 14:18 |
| *** nicolasbock has quit IRC | 14:18 | |
| odyssey4me | I know it cuts down a lot of what we carry, and potentially makes the roles more multi-series friendly, but there's a cost there in the complexity of it. Unfortunately I can't afford the time to really think through whether I have any ideas to simplify it. | 14:19 |
| odyssey4me | If anything, it's better than that awful sources-branch-updater. | 14:19 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zun master: Allow tags to be used for MQ tasks https://review.openstack.org/589494 | 14:22 |
| errr | odyssey4me: I was just looking at https://docs.ansible.com/ansible/2.3/list_of_cloud_modules.html#openstack and I think mapping and federation are missing from the modules they have | 14:26 |
| *** gkadam_ has joined #openstack-ansible | 14:27 | |
| odyssey4me | errr: yes, they are, but it's not necessary to have them | 14:27 |
| errr | odyssey4me: how so? The commands in https://docs.openstack.org/keystone/latest/advanced-topics/federation/federated_identity.html#mapping-combinations have you run those commands | 14:28 |
| errr | do you just fall to `command:` and use the cli client? | 14:29 |
| odyssey4me | errr: yep - spit out the json from our source config, then import with the openstack cli | 14:29 |
| errr | ah | 14:29 |
| *** gkadam has quit IRC | 14:29 | |
| *** yolanda has quit IRC | 14:31 | |
| cloudnull | sorry in meetings :) | 14:31 |
| * cloudnull the meetings will continue until moral improves | 14:32 | |
| errr | no they wont. they will never end no matter what | 14:32 |
| errr | better you than me though :P | 14:32 |
| *** hwoarang has quit IRC | 14:37 | |
| *** yolanda has joined #openstack-ansible | 14:37 | |
| openstackgerrit | Amy Marrich (spotz) proposed openstack/openstack-ansible master: [docs] Add documentation for 'install_method' variable https://review.openstack.org/589398 | 14:41 |
| *** priteau has quit IRC | 14:50 | |
| *** FrankZhang has joined #openstack-ansible | 14:51 | |
| *** kaiokmo has joined #openstack-ansible | 14:57 | |
| *** Emine has quit IRC | 14:58 | |
| evrardjp | folks I won't be able to stay long during the meeting of today | 15:06 |
| *** nicolasbock has joined #openstack-ansible | 15:07 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Basic implementation of masakari-monitors https://review.openstack.org/584629 | 15:07 |
| *** pcaruana has quit IRC | 15:11 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Clean up tests a bit https://review.openstack.org/589259 | 15:12 |
| *** cshen has quit IRC | 15:14 | |
| noonedeadpunk | odyssey4me: I don't have any idea, why tests are failing, when they are run with non-root user... Moreover, it's strange, that they don't fail for root, as containers doesn't have network interfaces at all on this step. I may suggest, that ansible reaches container from root by nsenter/lxc-attach, but I don't fully understand, how it works... | 15:16 |
| *** spatel has joined #openstack-ansible | 15:17 | |
| evrardjp | noonedeadpunk: we are using nsenter now for our connection plugin | 15:17 |
| evrardjp | odyssey4me: ping? | 15:18 |
| noonedeadpunk | evrardjp: and how is it launched/triggered? Could you point me to this plugin (if it's public one)? | 15:19 |
| evrardjp | https://github.com/openstack/openstack-ansible-plugins/blob/master/connection/ssh.py | 15:19 |
| *** KeithMnemonic has joined #openstack-ansible | 15:20 | |
| evrardjp | it's a hijack of ssh plugin, so we don't have to explicitly say it's 'our' connection plugin | 15:20 |
| odyssey4me | evrardjp: in meetings right now | 15:20 |
| evrardjp | ok | 15:21 |
| evrardjp | prometheanfire: are you in meetings too? | 15:21 |
| evrardjp | I'd need to know something about requirements | 15:21 |
| prometheanfire | sure | 15:23 |
| *** Bhujay has joined #openstack-ansible | 15:23 | |
| prometheanfire | in meetings, but multitasking :D | 15:23 |
| *** priteau has joined #openstack-ansible | 15:26 | |
| *** DanyC has quit IRC | 15:27 | |
| *** DanyC has joined #openstack-ansible | 15:27 | |
| *** DanyC has quit IRC | 15:31 | |
| *** gyee has joined #openstack-ansible | 15:32 | |
| *** diliprenkila has quit IRC | 15:35 | |
| *** dxiri has joined #openstack-ansible | 15:38 | |
| *** v1k0d3n has joined #openstack-ansible | 15:44 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Fix log compression https://review.openstack.org/589430 | 15:45 |
| *** olivierbourdon38 has quit IRC | 15:46 | |
| *** mmalchuk has joined #openstack-ansible | 15:50 | |
| evrardjp | ok | 15:53 |
| evrardjp | who is chairing today? | 15:53 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: [TEST] bionic https://review.openstack.org/586607 | 15:59 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: [TEST] bionic https://review.openstack.org/586607 | 16:00 |
| *** Bhujay has quit IRC | 16:00 | |
| *** luksky has quit IRC | 16:02 | |
| evrardjp | ok I guess I am chairing for 30 minutes then | 16:02 |
| evrardjp | cloudnull, DimGR, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, scarlisle, luckyinva, ntt, javeriak, spotz, vdo, jmccrory, alextricity25, jasondotstar, admin0, michaelgugino, ametts, bgmccollum, darrenc, JRobinson__, colinmcnamara, thorst, | 16:02 |
| evrardjp | adreznec, eil397, qwang,nishpatwa_, cathrichardson, drifterza, hwoarang, cshen, ullbeking, Tahvok, mnaser, nicolasbock, jrosser | 16:02 |
| evrardjp | Meeting time :) | 16:03 |
| spotz | Ok you didn't get booted evrardjp.... jungleboyj swears pings get you booted | 16:03 |
| prometheanfire | neat | 16:03 |
| evrardjp | #startmeeting openstack_ansible_meeting | 16:04 |
| openstack | Meeting started Tue Aug 7 16:04:02 2018 UTC and is due to finish in 60 minutes. The chair is evrardjp. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:04 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:04 |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:04 |
| evrardjp | spotz: seems I am still here. | 16:04 |
| spotz | evrardjp: yep:) | 16:04 |
| evrardjp | #topic rollcall | 16:04 |
| evrardjp | o/ | 16:05 |
| evrardjp | #topic Last week highlights | 16:06 |
| evrardjp | spotz: there seem to be something weird with the bot though. | 16:06 |
| spotz | o/ | 16:06 |
| evrardjp | it doesn't seem to ack my commands | 16:06 |
| evrardjp | weirdly | 16:06 |
| evrardjp | let's continue and hope things are logged | 16:07 |
| spotz | Yeah noticed that in the woo meeting | 16:07 |
| evrardjp | evrardjp and odyssey4me thanked jrosser for the work done in Bionic | 16:07 |
| *** spatel has quit IRC | 16:07 | |
| evrardjp | odyssey4me: reminded that there is more work to be done | 16:08 |
| evrardjp | hwoarang reminded that leap 15 work is blocked due to mariadb, lxc, and sudo | 16:08 |
| evrardjp | evrardjp reminded that RC1 is this week | 16:08 |
| evrardjp | and evrardjp asked to fill the etherpad https://etherpad.openstack.org/p/osa-stein-ptg | 16:09 |
| bgmccollum | o/ | 16:09 |
| evrardjp | is there any other last week highlights? | 16:09 |
| evrardjp | if not let's go to bugs | 16:09 |
| evrardjp | #topic bugtriage | 16:09 |
| evrardjp | please see our usual etherpad: https://etherpad.openstack.org/p/osa-bugtriage | 16:09 |
| evrardjp | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785824 | 16:09 |
| openstack | Launchpad bug 1785824 in openstack-ansible "Octavia fails on octavia-housekeeping group binding" [Undecided,New] | 16:09 |
| evrardjp | I am asking more details about the openstack_user_config, as this seems an inventory issue | 16:10 |
| johnsom | This was just a mis-configured host networking I think... Ping the user, I'm pretty sure they resolved that issue on their own | 16:11 |
| evrardjp | johnsom: I wouldn't be surprised about a user configuration issue. | 16:12 |
| evrardjp | let's move on | 16:12 |
| evrardjp | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785651 | 16:12 |
| openstack | Launchpad bug 1785651 in openstack-ansible "P->Q Upgrade - Nova cert shows as down" [Undecided,New] | 16:12 |
| evrardjp | I guess we need some cleanup of haproxy from P to Q as a service looks removed? | 16:13 |
| evrardjp | I am not aware of what nova-cert does, and if it really was included in nova-api . Proves my internal knowledge of nova. | 16:13 |
| evrardjp | :p | 16:13 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/queens: Fix log compression https://review.openstack.org/589536 | 16:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Fix log compression https://review.openstack.org/589537 | 16:14 |
| evrardjp | https://github.com/openstack/nova/commit/d6d5c6be0cc0738bb7d67ca8391e5cc4f0a419f2 | 16:14 |
| evrardjp | I'd say confirmed and medium, as we need to remove things in haproxy, which we haven't done. | 16:15 |
| bgmccollum | And the containers | 16:15 |
| bgmccollum | And the service entry in Nova? | 16:16 |
| evrardjp | well the thing is... I don't see anything in haproxy configuration | 16:16 |
| evrardjp | so I guess it's probably only the containers and service entries | 16:17 |
| bgmccollum | Maybe the inventory too? I went thought this myself, and forget all the places I had to clean up... | 16:18 |
| evrardjp | we can mark it as confirmed then | 16:18 |
| evrardjp | first hand experience there :p | 16:18 |
| cloudnull | o/ | 16:19 |
| evrardjp | nova_cert is still in the inventory | 16:19 |
| evrardjp | omg | 16:19 |
| evrardjp | in rocky | 16:19 |
| evrardjp | that's fun. | 16:19 |
| evrardjp | confirmed medium? | 16:20 |
| evrardjp | it doesn't hurt it's just ugly so I'd say "fix when convenient" is appropriate | 16:21 |
| bgmccollum | its not hurting anything, its just the ocata-era nova-cert starts up, then fails because of object version mismatch or somethings...and quietly goes away | 16:21 |
| evrardjp | ok | 16:21 |
| evrardjp | let's move on | 16:21 |
| evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1785646 | 16:21 |
| openstack | Launchpad bug 1785646 in openstack-ansible "Cleanup of haproxy files could be more obvious" [Undecided,New] | 16:21 |
| evrardjp | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785646 | 16:21 |
| evrardjp | confirmed wishlist? | 16:22 |
| evrardjp | it's a bug pending to happen basically | 16:22 |
| evrardjp | anyone agrees? | 16:23 |
| bgmccollum | another doesn't hurt? | 16:23 |
| evrardjp | bgmccollum: not sure what you mean | 16:23 |
| bgmccollum | it doesn't break anything if the service is gone from the openstack side, but the haproxy config still exists... | 16:24 |
| evrardjp | that's true | 16:24 |
| evrardjp | so confirmed wishlist? | 16:24 |
| bgmccollum | sure | 16:24 |
| evrardjp | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785592 | 16:25 |
| openstack | Launchpad bug 1785592 in openstack-ansible "dynamic inventory doesn't handle is_metal: false->true change" [Undecided,New] | 16:25 |
| evrardjp | (I have to go soon though) | 16:25 |
| evrardjp | cloudnull: could you have a look at this ^ ? | 16:25 |
| cloudnull | sure | 16:25 |
| cloudnull | assign that one to me | 16:25 |
| jrosser | that felt very very similar to one we patched previously | 16:26 |
| evrardjp | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785517 | 16:26 |
| openstack | Launchpad bug 1785517 in openstack-ansible "iptables checksum-fill causing kernel warning error stack" [Undecided,New] | 16:26 |
| *** vnogin has quit IRC | 16:27 | |
| evrardjp | that sounds like invalid | 16:27 |
| evrardjp | cloudnull: can you confirm? | 16:27 |
| bgmccollum | This is the kernel-deubg slowdown issue, right? | 16:27 |
| cloudnull | evrardjp that kernel one seems valid | 16:28 |
| *** DanyC has joined #openstack-ansible | 16:28 | |
| evrardjp | yes but not in that sense I'd say | 16:28 |
| cloudnull | cent7 seems to have debug logging enabled by default | 16:28 |
| *** spatel has joined #openstack-ansible | 16:28 | |
| bgmccollum | Exacerbated by the checksumming? | 16:28 |
| evrardjp | cloudnull: I have submitted a patch for that | 16:28 |
| cloudnull | the checksumming is something that we should probably have a look at. | 16:28 |
| cloudnull | we may not even need that anymore | 16:28 |
| *** electrofelix has quit IRC | 16:29 | |
| evrardjp | that's probably true | 16:29 |
| *** devx has quit IRC | 16:29 | |
| evrardjp | but that can be an user configuration, and with my patch explaining the change for centos, we should be good, right? | 16:29 |
| evrardjp | so I should mark this as invalid? | 16:30 |
| evrardjp | or I keep this for the checksum investigation? | 16:30 |
| * cloudnull hasn't seen your patch | 16:30 | |
| cloudnull | however it sounds like it covers this | 16:30 |
| cloudnull | or if i have seen it I dont remember | 16:30 |
| bgmccollum | other distros with kernel-debug not enabled do not exhibit this problem | 16:31 |
| cloudnull | or maybe they are and we're just not seeing it ? | 16:31 |
| evrardjp | cloudnull: I think that's the case | 16:31 |
| cloudnull | as in we're not seeing the logs | 16:31 |
| evrardjp | we are not seeing it | 16:31 |
| *** devx has joined #openstack-ansible | 16:31 | |
| cloudnull | ++ | 16:31 |
| cloudnull | I would believe that | 16:31 |
| evrardjp | and it doesn't hurt that much on iops | 16:31 |
| bgmccollum | i though the slowdown was the massive amount of logging happening | 16:32 |
| *** DanyC has quit IRC | 16:32 | |
| evrardjp | yes | 16:32 |
| * bgmccollum shrugs | 16:32 | |
| evrardjp | that's waht I'd say | 16:32 |
| cloudnull | however I'd also believe its just a crusty old cent7 issue :) | 16:32 |
| evrardjp | so if no logging, better behavior | 16:32 |
| evrardjp | haha | 16:32 |
| *** lbragstad[m] has quit IRC | 16:32 | |
| evrardjp | cloudnull: for fun we should set it to 7 1 4 7 everywhere | 16:32 |
| evrardjp | in the meantime I mark this as fix released | 16:32 |
| bgmccollum | for science | 16:32 |
| cloudnull | ++ | 16:33 |
| evrardjp | bgmccollum: for science! | 16:33 |
| *** DanyC has joined #openstack-ansible | 16:33 | |
| evrardjp | fix comitted* | 16:33 |
| evrardjp | I have to go | 16:33 |
| evrardjp | can someone take over from here? | 16:33 |
| spotz | Let me find the link evrardjp | 16:34 |
| evrardjp | spotz: https://etherpad.openstack.org/p/osa-bugtriage | 16:34 |
| evrardjp | we are there: | 16:34 |
| evrardjp | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785386 | 16:34 |
| openstack | Launchpad bug 1785386 in openstack-ansible "Integration of Swift and Manila with Openstack-ansible having Ceph backend?" [High,New] | 16:34 |
| evrardjp | I have to run | 16:34 |
| evrardjp | ttyl everyone | 16:35 |
| evrardjp | #chair spotz | 16:35 |
| spotz | go I got it | 16:35 |
| openstack | Current chairs: evrardjp spotz | 16:35 |
| evrardjp | thanks spotz :) | 16:35 |
| cloudnull | later evrardjp | 16:35 |
| spotz | Ok so looking at it what are the opinions? | 16:35 |
| spotz | Sounds kinda wishlist to me | 16:36 |
| bgmccollum | Not too familiar with Ceph, but shouldn't it "just work" if the rados gateway is deployed? | 16:36 |
| *** DanyC has quit IRC | 16:36 | |
| spotz | I'm taking it as they'd like a config option? | 16:37 |
| *** DanyC has joined #openstack-ansible | 16:37 | |
| spotz | Now they shouldn't have changed it to high importance but another issue:) | 16:37 |
| spotz | odyssey4me cloudnull jrosser? | 16:37 |
| jrosser | well quite a few of these things seem to fall into the "user story" category maybe | 16:38 |
| jrosser | as in its probably all possible given the right config | 16:38 |
| bgmccollum | is manilla even an OSA deployable project? | 16:38 |
| spotz | If it can be done already it's a docs issue. But not sure | 16:38 |
| *** gkadam_ has quit IRC | 16:38 | |
| odyssey4me | it makes absolutely no sense to 'integrate swift' with ceph | 16:38 |
| cloudnull | I think andymccr was working on that role? | 16:39 |
| odyssey4me | and manila is not something we have a role for | 16:39 |
| bgmccollum | well there you go...wish list? | 16:39 |
| spotz | Or won't fix | 16:39 |
| odyssey4me | well, as the question is posed there - it can be answered | 16:39 |
| odyssey4me | There is no manila role in OSA. We would welcome a contribution. | 16:39 |
| spotz | Ok changing to wishlist, and I'll mention the no manila role. Even if it doesn't make sense could they config ceph and swift? | 16:40 |
| odyssey4me | And Swift is not something that you integrate on top of Ceph. Please look at Ceph RGW if you want a swift API with a ceph back-end. | 16:40 |
| bgmccollum | I think that is the spirit of what he is asking... | 16:41 |
| bgmccollum | But without a Manila role, there is nothing to more to provide. | 16:41 |
| jrosser | bitrotted stuff here https://github.com/bkreitch/openstack-ansible-os_manila | 16:42 |
| spotz | reply - There is not currently a Manila role within OpenStack-Ansible. We would readily accept a contribution of this role. Swift is not something that is integrated on top of Ceph. Please look at Ceph RGW if you want a swift API with a ceph back-end. | 16:42 |
| bgmccollum | sounds good to me | 16:43 |
| spotz | jrosser: But we can't support someone else's and I don't think we want to point folks to non-supported repos? | 16:43 |
| odyssey4me | interesting, that role might be nice to import and iterate | 16:43 |
| cloudnull | https://github.com/andymcc/openstack-ansible-os_manila | 16:43 |
| jrosser | spotz: of course, looks like someones had a bash though | 16:43 |
| cloudnull | oh sorry. | 16:43 |
| spotz | Now Andy's is another story:), but anyone know the status? | 16:44 |
| odyssey4me | andymccr's is clearly nothing more than scaffolding | 16:44 |
| odyssey4me | and andymccr is unlikely to continue work on it given his work changes | 16:44 |
| spotz | Ok I went with wish-list and what I posted in channel let me get the next one | 16:44 |
| spotz | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785365 | 16:45 |
| openstack | Launchpad bug 1785365 in openstack-ansible "LXC container network slow because of kernel debug mesg" [Undecided,New] | 16:45 |
| spotz | 15 minute warning FYI | 16:45 |
| spotz | cloudnull: you have any thoughts? | 16:46 |
| cloudnull | that was the same issue evrardjp touched on a bit ago | 16:46 |
| cloudnull | maybe duplicate | 16:46 |
| spotz | I'll make a note in the etherpad for JP to take a look. Let me get the next one | 16:47 |
| *** spatel has quit IRC | 16:47 | |
| spotz | #link https://bugs.launchpad.net/openstack-ansible/+bug/1785068 | 16:47 |
| openstack | Launchpad bug 1785068 in openstack-ansible "os_glance : Ensure glance service" [Undecided,New] | 16:47 |
| spotz | Looks like he's included info anout his setup and config settings | 16:48 |
| jrosser | external and internal vip same -> bzzzzt | 16:48 |
| jrosser | that never ends well | 16:49 |
| *** DanyC has quit IRC | 16:49 | |
| *** DanyC_ has joined #openstack-ansible | 16:49 | |
| spotz | ok told him that marked invalid and asked him to retry | 16:50 |
| spotz | #link https://bugs.launchpad.net/openstack-ansible/+bug/1784949 | 16:50 |
| openstack | Launchpad bug 1784949 in openstack-ansible "Queen issue adding compute nodes " [Undecided,New] | 16:50 |
| spotz | Next:) | 16:50 |
| spotz | Has anyone done an upgrade to queens and run into this? | 16:55 |
| *** Emine has joined #openstack-ansible | 16:55 | |
| odyssey4me | not me, but this issue is not unique to queens or an upgrade | 16:55 |
| spotz | True bug or something we should document? | 16:56 |
| bgmccollum | Sounds like he didn't run the setup-hosts.yml playbook? | 16:56 |
| bgmccollum | Or not... | 16:56 |
| bgmccollum | :. | 16:56 |
| *** DanyC_ has quit IRC | 16:56 | |
| odyssey4me | the issue is that the tasks which do the linking seem to have not ran - there's not enough here to diagnose anything | 16:56 |
| *** DanyC has joined #openstack-ansible | 16:56 | |
| *** cshen has joined #openstack-ansible | 16:56 | |
| *** tosky has quit IRC | 16:57 | |
| jrosser | if we can point to where the venv should get pulled in and ask for confirmation thats been run, and debug output if it doesnt do anything? | 16:57 |
| odyssey4me | we pushed this up recently: https://review.openstack.org/#/q/Id962efe16c425424715409f071c4a304f8416001 | 16:57 |
| jrosser | "I found i don't have /openstack/venvs/nova-17.0.8 directory" <- the whole thing is missing though? | 16:58 |
| odyssey4me | so all that stuff is done after the venv is downloaed and extracted | 16:58 |
| spotz | Well May 31st, I would think he had that patch? | 16:58 |
| odyssey4me | which means that somehow this person got to that task without the venv being there | 16:58 |
| spotz | Ok let me ask him for his steps and maybe he missed a step? | 16:59 |
| spotz | This is going to be the last one we get to today | 16:59 |
| odyssey4me | so basically there should be a failed task somewhere around the venv deployment, or the task was skipped for whatever reason - either way, the reporter need to provide more info | 16:59 |
| spotz | ok asked for steps and for them to look for failed tasks. And we're out of time | 17:00 |
| spotz | #endmeeting | 17:00 |
| odyssey4me | these tasks have to run, and if they didn't then the person will need to see why: https://github.com/openstack/openstack-ansible-os_nova/blob/stable/queens/tasks/nova_install.yml#L74-L127 | 17:01 |
| openstack | Meeting ended Tue Aug 7 17:00:59 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:01 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2018/openstack_ansible_meeting.2018-08-07-16.04.html | 17:01 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2018/openstack_ansible_meeting.2018-08-07-16.04.txt | 17:01 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2018/openstack_ansible_meeting.2018-08-07-16.04.log.html | 17:01 |
| jrosser | i've been having a look at why some of the jobs are running slow - this appears to be swapping http://logs.openstack.org/30/589430/2/check/openstack-ansible-deploy-aio_lxc-ubuntu-xenial/bea65a6/logs/openstack/instance-info/dstat.html | 17:01 |
| spotz | thanks all | 17:02 |
| odyssey4me | jrosser: yeah, swapping's to be expected given how much we're putting into such a small host | 17:02 |
| jrosser | i see in my aio that lots of the services have their process/threds reduced | 17:03 |
| jrosser | but thats not consistent | 17:03 |
| odyssey4me | jrosser: ah, then we should fix that | 17:04 |
| jrosser | tons and tons of msqld, and more nova-scheduler, glance-api, nova-compute than you might expect and zillions of pypiserver | 17:04 |
| jrosser | though pypiserver is small so probly doesnt matter | 17:05 |
| jrosser | htop sorted by memory usage is useful | 17:05 |
| odyssey4me | hmm, I don't think we have a tunable for the number of pypiserver threads | 17:06 |
| odyssey4me | the others there should be tunables already, we're obviously just missing applying those right | 17:06 |
| jrosser | pypi server is only small | 17:06 |
| odyssey4me | I've actually been wanting to look into using uwsgi's cheaper algo for auto-scaling rather than using a fixed set of threads. | 17:07 |
| *** DanyC_ has joined #openstack-ansible | 17:07 | |
| jrosser | mysql, nova-api-os-compute, nova-compute, glance-api would be a good place to start | 17:07 |
| odyssey4me | I did a short test and it seemed to work, but I figured that's something we should rather implement very early in a cycle so that we get maximum test time. | 17:07 |
| jrosser | on other news my bionic aio is getting to tempest and failing 1 out of 114, so close | 17:08 |
| * jrosser travels | 17:08 | |
| *** DanyC_ has quit IRC | 17:09 | |
| *** DanyC_ has joined #openstack-ansible | 17:10 | |
| *** DanyC has quit IRC | 17:11 | |
| odyssey4me | ok, finally done with meetings | 17:13 |
| odyssey4me | xgerman_: could you revisit your review on https://review.openstack.org/#/c/589246/ and that patch chain - I think all the bases are covered for most of the patches now | 17:14 |
| *** ThomasWhite has quit IRC | 17:15 | |
| odyssey4me | yep, the last patch's tests just passed - just waiting for the results | 17:15 |
| *** fghaas has quit IRC | 17:16 | |
| odyssey4me | evrardjp: it looks like gnocchi now has a minimum requirement for setuptools>=30.3.0 | 17:20 |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible-os_cinder master: Setup oslo.messaging extra packages for optional drivers https://review.openstack.org/582632 | 17:21 |
| *** DanyC has joined #openstack-ansible | 17:22 | |
| *** DanyC_ has quit IRC | 17:24 | |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible-os_neutron master: Setup oslo.messaging extra packages for optional drivers https://review.openstack.org/583021 | 17:25 |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible-os_nova master: Setup oslo.messaging extra packages for optional drivers https://review.openstack.org/584450 | 17:26 |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible-os_keystone master: Setup oslo.messaging extra packages for optional drivers https://review.openstack.org/584479 | 17:32 |
| *** spatel has joined #openstack-ansible | 17:32 | |
| spatel | I am trying to add "mon_allow_pool_delete: True" in user_variable but it seems not working | 17:32 |
| spatel | i finish my playbook but it didn't appended that option in mon config | 17:33 |
| *** fghaas has joined #openstack-ansible | 17:39 | |
| *** markvoelker_ has quit IRC | 17:45 | |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible-os_heat master: Setup oslo.messaging extra packages for optional drivers https://review.openstack.org/584513 | 17:47 |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible master: [WIP] Add qdrouterd role for messaging backend deployment https://review.openstack.org/570261 | 17:52 |
| *** hamzaachi has quit IRC | 18:00 | |
| *** Leo_m has joined #openstack-ansible | 18:01 | |
| Leo_m | hi, I destroyed the octavia container but when I try to rebuild calling "# openstack-ansible setup-hosts.yml --limit octavia_all" I'm getting a failure fatal: [aio1_octavia_server_container-4c328689]: FAILED! => {"failed": true, "msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"} with the deleted container | 18:03 |
| Leo_m | any ideas? | 18:03 |
| prometheanfire | can someone take a look at this jinja? | 18:07 |
| prometheanfire | enabled_hardware_types = {{ filtered_ironic_drivers.hardware_type | join(',') }} is returning enabled_hardware_types = i,p,m,i | 18:07 |
| prometheanfire | instead of ipmi | 18:07 |
| *** poopcat has joined #openstack-ansible | 18:07 | |
| prometheanfire | I think perhaps I need to insert a list between filtered_ironic_drivers.hardware_type and the value in the datastucture | 18:08 |
| prometheanfire | https://github.com/openstack/openstack-ansible-os_ironic/commit/fc1e4f78dc9b5e415ad71a4664638bddfa190fd2#diff-04f1b8cc0a91d47ed7905aad52cc46deR19 seems to be causing it, perhaps the filter needs updating | 18:10 |
| spatel | Leo_m: try to delete aio1_octavia_server_container-4c328689 in inventory and see | 18:10 |
| odyssey4me | Leo_m: include lxc_hosts in that limit | 18:10 |
| odyssey4me | Leo_m: do not do what spatel just suggested | 18:10 |
| mnaser | guilhermesp: so pretty much i'mm at https://review.openstack.org/#/c/584604/ | 18:10 |
| guilhermesp | looking | 18:10 |
| prometheanfire | doesn't this create a list? https://github.com/openstack/openstack-ansible-os_ironic/commit/fc1e4f78dc9b5e415ad71a4664638bddfa190fd2#diff-6d62b73b3f46cfa11f6466d08cf45303R186 | 18:10 |
| *** cshen has quit IRC | 18:11 | |
| mnaser | that's where i'm stuck so far, if you look at the logs, you'll notice that it's failing to find cinder-rtstool, that's because cinder-rtstool is located in /openstack/venv/cinder-testing/bin/cinder-rtstool, but $PATH does not include that inside the service | 18:11 |
| odyssey4me | prometheanfire: ah, that's because filtered_ironic_drivers.hardware_type is a string, not a list | 18:11 |
| guilhermesp | that's a good start to me | 18:11 |
| mnaser | guilhermesp: so whats needed right now is to make sure that /openstack/venv/cinder-testing/bin is inside $PATH | 18:12 |
| prometheanfire | odyssey4me: ya, I get that part, but the link I just did should have it as a list from my shitty jinja reading/writing | 18:12 |
| mnaser | and that's where i didn't have time to follow up on :X | 18:12 |
| guilhermesp | k mnaser I'm going to focus on it asap | 18:12 |
| guilhermesp | don't worry :P | 18:12 |
| mnaser | cool, lemme know if you need a hand, i know cloudnull tried his hand at adding it to $PATH but it didnt go through | 18:13 |
| *** fghaas has quit IRC | 18:13 | |
| guilhermesp | great, I give you guys updates when I put my hands on it | 18:13 |
| *** luksky has joined #openstack-ansible | 18:14 | |
| odyssey4me | prometheanfire: I have a suspicion - just verifying | 18:14 |
| Leo_m | odyssey4me: openstack-ansible setup-hosts.yml --limit lxc_hosts ? | 18:15 |
| Leo_m | or what do you mean | 18:15 |
| prometheanfire | odyssey4me: thanks | 18:15 |
| spatel | Leo_m: i think he is saying add both in limit | 18:15 |
| Leo_m | how would it be? | 18:16 |
| odyssey4me | prometheanfire: do you have a sample output of what filtered_ironic_drivers contains? | 18:16 |
| spatel | Leo_m: --limit octavia_all --limit lxc_hosts something like that | 18:16 |
| spatel | or may be --limit octavia_all, lxc_hosts | 18:16 |
| odyssey4me | Leo_m: --limit octavia_all,lxc_hosts | 18:17 |
| odyssey4me | whenever you limit to a container set, also always include the hosts because some tasks are delegated to the hosts | 18:17 |
| Leo_m | --limit octavia_all --limit lxc_hosts is going, thx!! | 18:17 |
| *** cshen has joined #openstack-ansible | 18:18 | |
| prometheanfire | odyssey4me: no :| | 18:18 |
| prometheanfire | odyssey4me: would that be in a gate job? | 18:19 |
| odyssey4me | prometheanfire: I think there's a bug in https://github.com/openstack/openstack-ansible-os_ironic/blob/master/vars/main.yml#L182-L189 because it looks to me like it will always update the same dict, so there'll only ever be one result | 18:19 |
| *** diliprenkila has joined #openstack-ansible | 18:19 | |
| odyssey4me | prometheanfire: nope, but if you had a test system running you'd be able to get it | 18:19 |
| prometheanfire | odyssey4me: we do, would it be in cached facts? | 18:19 |
| diliprenkila | Hi all, I am hitting this error while checking namespaces on network nodes | 18:19 |
| diliprenkila | root@ctrl1:~# ip netns RTNETLINK answers: Invalid argument RTNETLINK answers: Invalid argument RTNETLINK answers: Invalid argument RTNETLINK answers: Invalid argument qdhcp-60ea65d9-bf4c-4fd1-a8a9-8cae22e28662 RTNETLINK answers: Invalid argument qrouter-3755419c-2a20-4ed0-9012-af24863f2447 RTNETLINK answers: Invalid argument snat-3755419c-2a20-4ed0-9012-af24863f2447 | 18:20 |
| odyssey4me | prometheanfire: ansible -m debug -e var=filtered_ironic_drivers ironic_all | 18:20 |
| prometheanfire | k | 18:20 |
| prometheanfire | "msg": "Hello world!" | 18:21 |
| odyssey4me | prometheanfire: from within the /opt/openstack-ansible/playbooks folder, to ensure you get the inventory included | 18:21 |
| prometheanfire | lol | 18:21 |
| prometheanfire | ya, did that | 18:21 |
| bgmccollum | isn't it -a not -e ? | 18:22 |
| diliprenkila | ip netns RTENETLINK invalid argument https://etherpad.openstack.org/p/Bv9e3EB4gt | 18:22 |
| odyssey4me | prometheanfire: yes, bgmccollum is right | 18:22 |
| prometheanfire | "filtered_ironic_drivers": "VARIABLE IS NOT DEFINED!" | 18:22 |
| diliprenkila | on the network nodes | 18:22 |
| *** cshen has quit IRC | 18:23 | |
| odyssey4me | prometheanfire: hmm, oh of course it's not a group var so you'd have to do it in a task in the role | 18:23 |
| prometheanfire | add a debug task then? | 18:23 |
| mnaser | evrardjp: thanks for hosting today. | 18:28 |
| mnaser | i'm sorry, the past few days have been overwhelmingly busy | 18:28 |
| mnaser | i'm almost out of it | 18:28 |
| odyssey4me | prometheanfire: yep, it seems clear that the way the var is put together isn't right, so you'll have to fire up a test system and work it out | 18:29 |
| prometheanfire | k | 18:29 |
| odyssey4me | or put together a test playbook on your own machine with the same starting data and figure it out | 18:29 |
| prometheanfire | did so, running the debug play now | 18:29 |
| odyssey4me | that'd be alot quicker I suppose | 18:29 |
| guilhermesp | mnaser: hope to help you out in order to make thing lighter for ya :D | 18:29 |
| prometheanfire | https://gist.github.com/prometheanfire/a62b68b27341505c500ad80617060eca | 18:30 |
| prometheanfire | so it looks like it's not generating a list for each key | 18:30 |
| prometheanfire | guess I play around with that filter | 18:31 |
| odyssey4me | prometheanfire: yes, so this is very broken: https://github.com/openstack/openstack-ansible-os_ironic/blob/master/vars/main.yml#L182-L189 | 18:33 |
| bgmccollum | should "set concat_drivers = {}" be "set concat_drivers = []" if you are expecting a list? | 18:34 |
| odyssey4me | that is constantly updating the same dict - so the last one wins | 18:34 |
| odyssey4me | yes, except I don't think it's a list that's expected | 18:34 |
| bgmccollum | ok | 18:34 |
| odyssey4me | given the way it's used here: https://github.com/openstack/openstack-ansible-os_ironic/blob/fc1e4f78dc9b5e415ad71a4664638bddfa190fd2/templates/ironic.conf.j2#L19-L24 | 18:35 |
| odyssey4me | I suspect that this would all be made far simpler by just using jmespath instead | 18:35 |
| prometheanfire | odyssey4me: no clue what that is | 18:35 |
| odyssey4me | prometheanfire: check here http://jmespath.org/ - put the data in and figure out the query to get the result, then use that with the json_query filter via ansible | 18:36 |
| prometheanfire | that could be better, yes | 18:37 |
| odyssey4me | so if you get https://github.com/openstack/openstack-ansible-os_ironic/blob/master/vars/main.yml#L182-L189 to build your list of various settings | 18:38 |
| odyssey4me | then use json_query to get the values you need | 18:38 |
| *** madorn has quit IRC | 18:41 | |
| *** madorn has joined #openstack-ansible | 18:43 | |
| *** diliprenkila has quit IRC | 18:44 | |
| odyssey4me | prometheanfire: this test data works for me: https://gist.github.com/odyssey4me/f2c2065a5496b09e219d5663ab304e98 | 18:48 |
| odyssey4me | so if you get that filtered list to actually be a list of the settings, then change the .conf entries to use json_query to grab the list of attributes, you have the right combo | 18:49 |
| *** openstackgerrit has quit IRC | 18:49 | |
| odyssey4me | of course you could probably do away with the filtered list by using jmespath from the get go - do one to filter the list, then one to get the attributes | 18:50 |
| *** fghaas has joined #openstack-ansible | 18:56 | |
| prometheanfire | that's the plan | 19:00 |
| prometheanfire | right now I'm figuring out how to multiselect from ironic driver types based on the list of ironic_drivers_enabled | 19:00 |
| *** fghaas has quit IRC | 19:04 | |
| *** cshen has joined #openstack-ansible | 19:08 | |
| spatel | jamesdenton: are you there? I migrating haproxy to F5 and i am seeing haproxy using SSL cert so do i need to move my cert to F5 ? | 19:11 |
| spatel | bind 10.30.2.9:6080 ssl crt /etc/ssl/private/haproxy.pem | 19:11 |
| jamesdenton | spatel are you using a real cert? or unsigned private cert? | 19:12 |
| spatel | I have real cert yes | 19:12 |
| jamesdenton | yes, then you'll need to create the profile on the F5 and attach it to the virtual server | 19:13 |
| spatel | I have wildcard certs already installed for other services i want to use that | 19:13 |
| jamesdenton | then you should be able to reuse that existing profile | 19:14 |
| spatel | so all i need to do attach cert to VS_SERVER | 19:14 |
| jamesdenton | well, the cert/key should be defined in an SSL profile, and that SSL profile is attached to VS_SERVER | 19:15 |
| spatel | yup! | 19:16 |
| spatel | just wanted to make sure this cert is not specific to openstack which we need to bring on F5 | 19:16 |
| spatel | also i am not using SPICE console so i shouldn't use this config in F5 | 19:17 |
| spatel | bind 10.30.2.9:6080 ssl crt /etc/ssl/private/haproxy.pem | 19:17 |
| spatel | ignore that | 19:17 |
| spatel | reate ltm monitor http /OSTACK/OSTACK_MON_HTTP_NOVA_SPICE_CONSOLE { defaults-from http destination *:6082 recv "200 OK" send "HEAD /spice_auto.html HTTP/1.1\r\nHost: rpc\r\n\r\n" } | 19:17 |
| spatel | I am using novac | 19:17 |
| spatel | I need to hand edit that and change port too | 19:17 |
| jamesdenton | yes, you'll want to edit that | 19:19 |
| Leo_m | spatel: where is the container inventory? I may need to delete like you mention, my recreated octavia container is not picking up the new lbaas cidr from the updated openstack_user_config.yml | 19:19 |
| kaiokmo | jrosser: hi, since https://review.openstack.org/#/c/589189/ is merged, can you set https://review.openstack.org/#/c/588464/ to abandoned, please? | 19:22 |
| *** throwsb1 has joined #openstack-ansible | 19:23 | |
| *** vollman has quit IRC | 19:25 | |
| jrosser | kaiokmo: would you prefer i did it? you should see a button on gerrit i think | 19:33 |
| jrosser | cloudnull: this feels suspicious, restart systemd-networkd followed by unable to ssh.... what do you think? http://logs.openstack.org/07/586607/10/check/openstack-ansible-deploy-aio_lxc-ubuntu-bionic/9ddb187/job-output.txt.gz#_2018-08-07_17_27_11_985240 | 19:35 |
| cloudnull | jrosser quite possibly | 19:36 |
| cloudnull | it should leave the interface alone given that its already up | 19:40 |
| cloudnull | but I guess if its trying to reconfigure teh interface that could be the cause | 19:41 |
| cloudnull | it also could be a situation where the interface is just not up/ready | 19:41 |
| kaiokmo | jrosser: ah, my mistake. thanks :) | 19:42 |
| odyssey4me | jrosser cloudnull if I could get your eyes on https://review.openstack.org/589248 It'd be grand - it's a small enough patch | 19:43 |
| cloudnull | ++ | 19:43 |
| odyssey4me | also cloudnull https://review.openstack.org/589064 | 19:43 |
| odyssey4me | and https://review.openstack.org/588508 | 19:43 |
| cloudnull | done. | 19:44 |
| * cloudnull goes back to lurking while on conference calls | 19:44 | |
| odyssey4me | tyvm | 19:44 |
| prometheanfire | odyssey4me: I'm learning to hate jmespath | 19:45 |
| *** openstackgerrit has joined #openstack-ansible | 19:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift master: Revert "Unblock gate: Provide default rabbitmq password" https://review.openstack.org/589450 | 19:46 |
| prometheanfire | it doesn't even look like you can pass two sets to it so it can do comparisons | 19:46 |
| *** hamzaachi has joined #openstack-ansible | 19:46 | |
| prometheanfire | so that logic has to remain outside of it | 19:46 |
| odyssey4me | prometheanfire: yeah, it works best with lists of things - the trouble you have is that it's just a dict of dicts | 19:47 |
| prometheanfire | ya, no examples for that | 19:47 |
| odyssey4me | so pre-process it into a list, then use jmespath to get the things you want | 19:47 |
| prometheanfire | not sure that's valid either, the datastucture as is makes sense, not sure it's able to be converted without loosing data | 19:48 |
| odyssey4me | well, you already are with the filtered list - it's just no generating a list, it's overwriting itself | 19:49 |
| odyssey4me | so change it to actually make a list, then use jmespath to extract the attributes form the list and give you a list as a result | 19:49 |
| prometheanfire | wat | 19:49 |
| spatel | Leo_m: folks won't agreed with me but thats how do workaround | 19:49 |
| spatel | Leo_m: do /var/opt/openstack-ansible/scripts/inventory-manage.py -l | 19:50 |
| spatel | /var/opt/openstack-ansible/scripts/inventory-manage.py -r <container_name> | 19:50 |
| odyssey4me | prometheanfire: check the test playbook in https://gist.github.com/odyssey4me/f2c2065a5496b09e219d5663ab304e98 | 19:54 |
| odyssey4me | prometheanfire: try it out and see the results | 19:55 |
| prometheanfire | how did you generate your test data? | 19:55 |
| prometheanfire | that's the part I can't do | 19:55 |
| odyssey4me | prometheanfire: I used ansible to spit the var out in json | 19:56 |
| prometheanfire | which var | 19:56 |
| * odyssey4me is lazy | 19:56 | |
| odyssey4me | https://gist.github.com/odyssey4me/f2c2065a5496b09e219d5663ab304e98#file-test-playbook-yml | 19:56 |
| odyssey4me | the first issue is that the filtered_ironic_drivers in the role overwrites itself | 19:57 |
| prometheanfire | ok | 19:57 |
| odyssey4me | it produces a single dict, not a list of dicts | 19:57 |
| prometheanfire | well, you changed concat_drivers to a list | 19:57 |
| prometheanfire | which helped | 19:58 |
| odyssey4me | the second is that once that's fixed (see the play), you can't extract them in the conf file as it stands now because that doesn't work with a list of dicts... so the solution is to use a json_query to get a resulting list of attributes | 19:58 |
| odyssey4me | then some of them are duplicated, so we filter that through unique, and do the concatenation | 19:58 |
| odyssey4me | there might be a smart way to get rid of filtered_ironic_drivers, but meh | 19:59 |
| odyssey4me | I spent around 15 mins trying with just jmespath, and concluded that it wasn't the right tool | 19:59 |
| prometheanfire | odyssey4me: this I can work with, the change to list is what I was missing | 20:00 |
| odyssey4me | yep | 20:00 |
| jrosser | odyssey4me: can i trade you this one? https://review.openstack.org/#/c/587478/ check carefully given its another project templates one :) | 20:02 |
| odyssey4me | jrosser: rock and roll | 20:03 |
| *** fghaas has joined #openstack-ansible | 20:03 | |
| jrosser | \o/ /me relieved the working stuff stands a better change of staying that way | 20:04 |
| odyssey4me | jrosser: yep, thanks - great work getting that done in time for the RC! | 20:04 |
| jrosser | ideally i'd like a nv integrated build too thats not a total car crash | 20:05 |
| odyssey4me | now if only these centos jobs would behave | 20:05 |
| *** dxiri has quit IRC | 20:08 | |
| jrosser | cloudnull: i was condsidering putting in a wait_for to check the ssh was available, but it's restarted in a handler so wasnt quite sure where to put that | 20:08 |
| spatel | jamesdenton: what is script for look like some external monitor, do i need that? https://raw.githubusercontent.com/dpham-rs/rpc-openstack/master/scripts/f5-monitor.sh | 20:12 |
| spatel | it says "F5 External monitor script for Liberty deployments" | 20:13 |
| jamesdenton | no, you don't need it | 20:13 |
| spatel | cool! | 20:13 |
| spatel | thanks! | 20:13 |
| jamesdenton | We used it to get valid tokens from keystone and then hit services w/ a token | 20:13 |
| logan- | nice work fixing the bionic gpg stuff jrosser | 20:17 |
| *** ansmith has quit IRC | 20:17 | |
| *** fghaas has quit IRC | 20:18 | |
| odyssey4me | cores, I have plenty of reviews ready to look at if you're able: https://goo.gl/k9cCfF | 20:19 |
| *** mmalchuk_ has joined #openstack-ansible | 20:21 | |
| jrosser | logan-: thanks :) it just got so ridiculously complicated - then i stopped trying to fix it and looked at what it actually did | 20:22 |
| *** mmalchuk has quit IRC | 20:24 | |
| openstackgerrit | Matthew Thode proposed openstack/openstack-ansible-os_ironic master: fix ironic driver data structure usage https://review.openstack.org/589630 | 20:28 |
| prometheanfire | odyssey4me: ^ | 20:28 |
| odyssey4me | prometheanfire: looks legit, have you got a test env to try it out on? | 20:29 |
| prometheanfire | yep, tested already | 20:30 |
| prometheanfire | conf output is valid | 20:30 |
| prometheanfire | relocating and lunching now | 20:30 |
| odyssey4me | prometheanfire: hooray! | 20:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Ensure the octavia flavor is private https://review.openstack.org/589239 | 20:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Ensure the octavia network setup is in the right project https://review.openstack.org/589240 | 20:33 |
| odyssey4me | heh, check it out - zuul can now do in-line comments if you give it the right info: https://zuul-ci.org/docs/zuul/user/jobs.html#leaving-file-comments | 20:33 |
| odyssey4me | an example: https://review.openstack.org/#/c/589305/1/.zuul.yaml | 20:34 |
| prometheanfire | neat | 20:39 |
| -openstackstatus- NOTICE: Due to a bug, Zuul has been unable to report on cherry-picked changes over the last 24 hours. This has now been fixed; if you encounter a cherry-picked change missing its results (or was unable to merge), please recheck now. | 20:41 | |
| *** nicolasbock has quit IRC | 20:45 | |
| jrosser | logan-: this could use a vars file for bionic https://github.com/logan2211/ansible-etcd/tree/master/vars | 20:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build master: Use a venv to build the python wheels https://review.openstack.org/589633 | 20:54 |
| *** mmalchuk has joined #openstack-ansible | 20:57 | |
| *** mmalchuk_ has quit IRC | 21:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build master: Use a venv to build the python wheels https://review.openstack.org/589633 | 21:02 |
| *** FrankZhang has quit IRC | 21:05 | |
| jrosser | odyssey4me: "ERROR cinder.cmd.volume [-] Configuration for cinder-volume does not specify "enabled_backends". Using DEFAULT section to configure drivers is not supported since Ocata." thats likley whats breaking bionic | 21:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build master: Simplify pip/setuptools/wheel prep for venvs https://review.openstack.org/589643 | 21:18 |
| *** dxiri has joined #openstack-ansible | 21:23 | |
| *** Leo_m has quit IRC | 21:40 | |
| *** Leo_m has joined #openstack-ansible | 21:46 | |
| *** hamzaachi has quit IRC | 21:46 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum master: Execute service setup against a delegated host using Ansible built-in modules https://review.openstack.org/582296 | 21:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum master: Execute service setup against a delegated host using Ansible built-in modules https://review.openstack.org/582296 | 21:47 |
| *** cshen has quit IRC | 21:50 | |
| *** Leo_m has quit IRC | 21:51 | |
| *** Leo_m has joined #openstack-ansible | 21:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Add retries to service role creation https://review.openstack.org/589245 | 21:59 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/pike: Add tempest test blacklist capability https://review.openstack.org/588508 | 22:05 |
| *** DanyC has quit IRC | 22:05 | |
| spatel | jamesdenton: I am seeing following pool but that doesn't make any sense | 22:09 |
| spatel | create ltm virtual /OSTACK/OSTACK_VS_GALERA_READ { destination 172.28.0.9:3307 ip-protocol tcp mask 255.255.255.255 pool /OSTACK/OSTACK_POOL_GALERA_READ profiles replace-all-with { /Common/fastL4 { } } mirror enabled source-address-translation { pool /OSTACK/OSTACK_SNATPOOL type snat } } | 22:09 |
| spatel | Port 3307 ??? | 22:09 |
| spatel | I don't think we need that | 22:09 |
| *** priteau has quit IRC | 22:09 | |
| *** cshen has joined #openstack-ansible | 22:10 | |
| *** spatel has quit IRC | 22:13 | |
| *** cshen has quit IRC | 22:14 | |
| *** throwsb1 has quit IRC | 22:24 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Use interface instead of endpoint_type https://review.openstack.org/589246 | 22:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Add retries to all os_* module tasks https://review.openstack.org/589445 | 22:31 |
| *** luksky has quit IRC | 22:37 | |
| *** cCain has joined #openstack-ansible | 22:41 | |
| *** bradm has joined #openstack-ansible | 22:43 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Do not install linux-image-extra by default https://review.openstack.org/589065 | 22:50 |
| openstackgerrit | Merged openstack/openstack-ansible stable/queens: Do not install linux-image-extra by default https://review.openstack.org/589064 | 22:50 |
| *** ansmith has joined #openstack-ansible | 22:55 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Remove deprecated python-software-properties https://review.openstack.org/584609 | 23:07 |
| *** priteau has joined #openstack-ansible | 23:14 | |
| *** pmannidi has joined #openstack-ansible | 23:18 | |
| *** vnogin has joined #openstack-ansible | 23:28 | |
| *** vnogin has quit IRC | 23:33 | |
| *** Leo_m has quit IRC | 23:46 | |
| *** Leo_m_ has joined #openstack-ansible | 23:46 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Update retention process to use rollup https://review.openstack.org/589674 | 23:52 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
