| *** david-lyle has quit IRC | 00:03 | |
| *** Hosam has joined #openstack-ansible | 00:04 | |
| *** david-lyle has joined #openstack-ansible | 00:07 | |
| spotz | cloudnull: good luck:) | 00:08 |
|---|---|---|
| *** acormier has joined #openstack-ansible | 00:10 | |
| *** hamzy has joined #openstack-ansible | 00:10 | |
| *** cjloader_ has joined #openstack-ansible | 00:10 | |
| *** thorst has joined #openstack-ansible | 00:11 | |
| *** markvoelker has joined #openstack-ansible | 00:13 | |
| *** thorst has quit IRC | 00:15 | |
| *** acormier has quit IRC | 00:19 | |
| *** thorst has joined #openstack-ansible | 00:21 | |
| *** acormier has joined #openstack-ansible | 00:26 | |
| *** thorst has quit IRC | 00:28 | |
| *** acormier has quit IRC | 00:32 | |
| *** karimb has quit IRC | 00:32 | |
| *** MasterOfBugs has quit IRC | 00:34 | |
| *** david-lyle has quit IRC | 00:36 | |
| *** jpstnh has joined #openstack-ansible | 00:37 | |
| *** zerick has quit IRC | 00:40 | |
| *** zerick has joined #openstack-ansible | 00:42 | |
| *** rmelero has quit IRC | 00:42 | |
| *** rmelero has joined #openstack-ansible | 00:42 | |
| *** cjloader_ has quit IRC | 00:43 | |
| *** cmart has quit IRC | 00:45 | |
| *** winggundamth has joined #openstack-ansible | 00:46 | |
| *** rmelero has quit IRC | 00:46 | |
| *** npawelek has quit IRC | 00:49 | |
| *** shashank_t_ has joined #openstack-ansible | 00:50 | |
| *** shashank_t_ has quit IRC | 00:51 | |
| *** thorst has joined #openstack-ansible | 00:52 | |
| *** thorst has quit IRC | 00:52 | |
| *** japestinho has joined #openstack-ansible | 00:54 | |
| *** npawelek has joined #openstack-ansible | 00:54 | |
| *** cNilesh has joined #openstack-ansible | 00:58 | |
| *** SerenaFeng has joined #openstack-ansible | 01:00 | |
| *** cuongnv has joined #openstack-ansible | 01:03 | |
| *** marst has joined #openstack-ansible | 01:07 | |
| *** shashank_t_ has joined #openstack-ansible | 01:09 | |
| *** vnogin has joined #openstack-ansible | 01:10 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: [WIP] Test Ansible 2.3.0.0-0.4.rc4 https://review.openstack.org/446173 | 01:13 |
| *** cathrich_ has joined #openstack-ansible | 01:14 | |
| *** cathrichardson has quit IRC | 01:14 | |
| *** vnogin has quit IRC | 01:15 | |
| *** acormier has joined #openstack-ansible | 01:16 | |
| *** lucasxu has joined #openstack-ansible | 01:17 | |
| *** Mahe has quit IRC | 01:17 | |
| *** acormier has quit IRC | 01:18 | |
| *** Mahe has joined #openstack-ansible | 01:18 | |
| *** lucasxu has quit IRC | 01:20 | |
| *** shashank_t_ has quit IRC | 01:22 | |
| *** thorst has joined #openstack-ansible | 01:26 | |
| *** thorst has quit IRC | 01:26 | |
| *** acormier has joined #openstack-ansible | 01:26 | |
| *** SerenaFeng has quit IRC | 01:29 | |
| *** SerenaFe_ has joined #openstack-ansible | 01:29 | |
| *** lucasxu has joined #openstack-ansible | 01:30 | |
| *** shashank_t_ has joined #openstack-ansible | 01:35 | |
| *** shashank_t_ has quit IRC | 01:40 | |
| *** cjloader_ has joined #openstack-ansible | 01:41 | |
| *** vnogin has joined #openstack-ansible | 01:56 | |
| *** shashank_t_ has joined #openstack-ansible | 01:59 | |
| *** shashank_t_ has quit IRC | 02:03 | |
| *** cuongnv has quit IRC | 02:06 | |
| *** acormier has quit IRC | 02:10 | |
| *** cjloader_ has quit IRC | 02:14 | |
| *** chas has joined #openstack-ansible | 02:16 | |
| *** cuongnv has joined #openstack-ansible | 02:16 | |
| *** vishwanathj has joined #openstack-ansible | 02:17 | |
| *** vishwanathj has quit IRC | 02:19 | |
| *** vishwanathj has joined #openstack-ansible | 02:20 | |
| *** chas has quit IRC | 02:21 | |
| *** Hosam has quit IRC | 02:25 | |
| *** Hosam has joined #openstack-ansible | 02:26 | |
| *** Hosam has quit IRC | 02:31 | |
| *** Jack_Iv has joined #openstack-ansible | 02:41 | |
| *** vnogin has quit IRC | 02:41 | |
| *** cjloader_ has joined #openstack-ansible | 02:45 | |
| *** Jack_Iv has quit IRC | 02:45 | |
| *** pmannidi has quit IRC | 02:47 | |
| *** pmannidi has joined #openstack-ansible | 02:47 | |
| *** cjloader_ has quit IRC | 02:49 | |
| *** gouthamr has quit IRC | 03:01 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Use machinectl to manage image caches https://review.openstack.org/454384 | 03:07 |
| *** acormier has joined #openstack-ansible | 03:13 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts master: Use machinectl to manage image caches https://review.openstack.org/454384 | 03:15 |
| *** zerick has quit IRC | 03:16 | |
| cloudnull | welp. it was a good idea ^ ubuntu is faster [newer systemd], centos is roughly the same (maybe a little slower) [older systemd]. | 03:17 |
| cloudnull | maybe i'll change the title to WIP, regardless I'd love comments on it and if it's worth pursuing further. | 03:18 |
| *** david-lyle has joined #openstack-ansible | 03:35 | |
| *** ianychoi has quit IRC | 03:36 | |
| *** jrobinson has quit IRC | 03:38 | |
| *** acormier has quit IRC | 03:40 | |
| *** david-lyle has quit IRC | 03:42 | |
| *** ianychoi has joined #openstack-ansible | 03:42 | |
| *** jrobinson has joined #openstack-ansible | 03:48 | |
| *** cjloader_ has joined #openstack-ansible | 03:54 | |
| *** SerenaFe_ has quit IRC | 03:57 | |
| *** fyu1 has joined #openstack-ansible | 03:59 | |
| *** fyu has quit IRC | 03:59 | |
| *** fyu1 is now known as fyu | 04:02 | |
| *** markvoelker has quit IRC | 04:11 | |
| *** SerenaFeng has joined #openstack-ansible | 04:12 | |
| *** jamielennox is now known as jamielennox|away | 04:18 | |
| *** cjloader_ has quit IRC | 04:20 | |
| *** adreznec has quit IRC | 04:26 | |
| *** zerick has joined #openstack-ansible | 04:26 | |
| *** adreznec has joined #openstack-ansible | 04:27 | |
| *** marst has quit IRC | 04:28 | |
| *** udesale has joined #openstack-ansible | 04:29 | |
| *** SerenaFeng has quit IRC | 04:30 | |
| *** lucasxu has quit IRC | 04:33 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Converge neutron agents onto Baremetal https://review.openstack.org/454450 | 05:02 |
| *** markvoelker has joined #openstack-ansible | 05:12 | |
| *** chas has joined #openstack-ansible | 05:16 | |
| openstackgerrit | Jirayut Nimsaeng proposed openstack/openstack-ansible-galera_server master: Fix causing conditional check error on mysql_started variable by adding galera-bootstrap tag https://review.openstack.org/454452 | 05:16 |
| *** markvoelker has quit IRC | 05:17 | |
| *** cjloader_ has joined #openstack-ansible | 05:18 | |
| openstackgerrit | Jirayut Nimsaeng proposed openstack/openstack-ansible-galera_server master: Fix conditional check error when run galera-bootstrap tag https://review.openstack.org/454452 | 05:19 |
| *** chas has quit IRC | 05:20 | |
| mpranjic | morning | 05:34 |
| *** SerenaFeng has joined #openstack-ansible | 05:38 | |
| *** foutatoro has quit IRC | 05:38 | |
| *** Jack_Iv has joined #openstack-ansible | 05:38 | |
| *** Jack_Iv has quit IRC | 05:39 | |
| *** Jack_Iv has joined #openstack-ansible | 05:40 | |
| *** Jack_Iv has quit IRC | 05:41 | |
| *** Jack_Iv has joined #openstack-ansible | 05:41 | |
| *** esberglu has joined #openstack-ansible | 05:49 | |
| *** cjloader_ has quit IRC | 05:52 | |
| *** esberglu has quit IRC | 05:53 | |
| *** cjloader_ has joined #openstack-ansible | 05:56 | |
| *** cjloader_ has quit IRC | 05:58 | |
| *** cjloader_ has joined #openstack-ansible | 05:58 | |
| *** arslanq has quit IRC | 06:00 | |
| *** cjloader_ has quit IRC | 06:03 | |
| *** markus_z has joined #openstack-ansible | 06:32 | |
| *** pcaruana has joined #openstack-ansible | 06:34 | |
| *** SerenaFeng has quit IRC | 06:48 | |
| *** SerenaFeng has joined #openstack-ansible | 06:52 | |
| *** markus_z has quit IRC | 07:00 | |
| *** karimb has joined #openstack-ansible | 07:08 | |
| *** Jack_Iv has quit IRC | 07:09 | |
| *** markvoelker has joined #openstack-ansible | 07:13 | |
| *** Jack_Iv has joined #openstack-ansible | 07:15 | |
| *** markus_z has joined #openstack-ansible | 07:15 | |
| *** jrobinson has quit IRC | 07:15 | |
| *** markvoelker has quit IRC | 07:17 | |
| *** Jack_Iv has quit IRC | 07:19 | |
| *** karimb has quit IRC | 07:30 | |
| *** shardy has joined #openstack-ansible | 07:34 | |
| *** chas has joined #openstack-ansible | 07:40 | |
| *** chas has quit IRC | 07:41 | |
| *** chas has joined #openstack-ansible | 07:41 | |
| *** Jack_Iv has joined #openstack-ansible | 07:43 | |
| *** chas has quit IRC | 07:44 | |
| *** chas has joined #openstack-ansible | 07:44 | |
| *** dalees has quit IRC | 07:45 | |
| *** chas has quit IRC | 07:49 | |
| *** foutatoro has joined #openstack-ansible | 07:50 | |
| *** pbandark has joined #openstack-ansible | 07:54 | |
| *** karimb has joined #openstack-ansible | 08:00 | |
| *** manheim has joined #openstack-ansible | 08:04 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova stable/ocata: nova user can read kernel for libguestfs on Ubuntu https://review.openstack.org/454269 | 08:05 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova stable/newton: nova user can read kernel for libguestfs on Ubuntu https://review.openstack.org/454270 | 08:05 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-security master: Do not update grub if grub not used https://review.openstack.org/454261 | 08:12 |
| *** manheim has quit IRC | 08:14 | |
| *** manheim has joined #openstack-ansible | 08:15 | |
| *** Jack_Iv has quit IRC | 08:24 | |
| *** Jack_Iv has joined #openstack-ansible | 08:24 | |
| openstackgerrit | Hengqing Hu proposed openstack/openstack-ansible master: Ensure proxy check is short circuit https://review.openstack.org/454604 | 08:27 |
| *** markus_z has quit IRC | 08:32 | |
| *** esberglu has joined #openstack-ansible | 08:32 | |
| openstackgerrit | Hengqing Hu proposed openstack/openstack-ansible master: Set a sane default haproxy_bind_on_non_local https://review.openstack.org/454606 | 08:34 |
| *** esberglu has quit IRC | 08:36 | |
| mpranjic | andymccr: hi! you still need help with Operations guide? | 08:50 |
| andymccr | mpranjic: hey! :) yes definitely. lemme find a link - we're looking for reviews/read-throughs and content | 08:56 |
| *** dalees has joined #openstack-ansible | 08:56 | |
| mpranjic | andymccr: great! eager to help out. | 08:57 |
| *** Jack_Iv has quit IRC | 09:00 | |
| *** Jack_Iv has joined #openstack-ansible | 09:02 | |
| andymccr | mpranjic: found them, so there is a wiki entry with some planning/sections that are needed etc which is here: https://wiki.openstack.org/wiki/OpenStackAnsible#Documentation_planning - and then asettle who works on a tonne of our docs is after reviews on what is actually there right now too: (https://docs.openstack.org/developer/openstack-ansible/draft-operations-guide/index.html) | 09:02 |
| mpranjic | *click* | 09:03 |
| *** karimb has quit IRC | 09:07 | |
| *** karimb has joined #openstack-ansible | 09:07 | |
| *** Jack_Iv_ has joined #openstack-ansible | 09:08 | |
| *** Jack_Iv has quit IRC | 09:10 | |
| *** karimb has quit IRC | 09:12 | |
| *** dixiaoli has joined #openstack-ansible | 09:12 | |
| hughsaunders | anyone seen "AttributeError: 'Task' object has no attribute 'first_available_file'" error from config_template? | 09:13 |
| hughsaunders | hmm looks like it was deprecated and removed https://github.com/openstack/openstack-ansible-plugins/commit/2f6cac2cf6ddebf462e21c70da7f60fa334b5f1e | 09:16 |
| andymccr | hughsaunders: what version are you using? | 09:16 |
| hughsaunders | andymccr: newton | 09:17 |
| andymccr | does it need a backport then? | 09:17 |
| andymccr | i thought we used 2.1 in newton though so maybe that should be ok | 09:18 |
| *** yolanda has quit IRC | 09:19 | |
| hughsaunders | Possibly, will have a look. It may be that something I'm doing is dropping 2.2 and thats getting used :/ | 09:19 |
| andymccr | hughsaunders: yeah double checked i think its 2.1.5.0 on stable/newton as of now | 09:19 |
| *** yolanda has joined #openstack-ansible | 09:19 | |
| openstackgerrit | Hengqing Hu proposed openstack/openstack-ansible master: Set a sane default haproxy_bind_on_non_local https://review.openstack.org/454606 | 09:24 |
| *** pmannidi has quit IRC | 09:25 | |
| asettle | mpranjic: lemme know if you have any questions! | 09:26 |
| asettle | Just as some context - it should be for OSA operations *only* not just OpenStack operations. But there is some of that content in there we have to move out | 09:26 |
| *** esberglu has joined #openstack-ansible | 09:29 | |
| mpranjic | asettle: sure thing. I will most likely have some questions soon | 09:30 |
| asettle | mpranjic: I'll be around :) | 09:30 |
| mpranjic | asettle: thx :) | 09:31 |
| *** esberglu has quit IRC | 09:34 | |
| *** chas has joined #openstack-ansible | 09:45 | |
| *** chas has quit IRC | 09:50 | |
| *** electrofelix has joined #openstack-ansible | 09:50 | |
| *** jcookeman has joined #openstack-ansible | 09:54 | |
| jcookeman | Hi channel. I'd like to best advice on creating a new instance that boots from a volume which will be the root of the instance. If the volume is not bootable it will boot from the image and create one. I see the boot_from_vol option, but doesn't that create a volume same as the hostname? Can I make it an already provisioned volume? Can I make the name whatever I want? | 09:59 |
| mpranjic | asettle: I should focus on reviews not merged, done or complete, right? this is the first in the list (bottom-up): https://review.openstack.org/#/c/452847/ | 10:11 |
| asettle | mpranjic: well, might be good to review the guide itself at this point. But here'll be a few that are not merged that might require a review too :) | 10:13 |
| asettle | Like this one: https://review.openstack.org/453984 | 10:13 |
| mpranjic | *click* | 10:14 |
| *** cuongnv has quit IRC | 10:14 | |
| *** markus_z has joined #openstack-ansible | 10:14 | |
| asettle | mpranjic: just updated the table, that's the only open review right now. But if you have the time to start reading through the guide - patching things up that you see needs help, etc - raising bugs if necessary | 10:16 |
| asettle | Also, if you look at that table and it isn't assigned to anyone and you think you can help fill in some gaps - please do so :D | 10:16 |
| mpranjic | will do. just to get my head around :) | 10:17 |
| mpranjic | so.. I can also fix things in that review, I presume | 10:17 |
| mpranjic | ? | 10:17 |
| asettle | mpranjic: comment in the PR if you think it's in error :) but if you can add to it, no harm done | 10:18 |
| openstackgerrit | Mario Pranjic proposed openstack/openstack-ansible master: [docs] Add backup information to the OSA Ops Guide Corrected capital S in OpenStack-Ansible Change-Id: Id15335234bdaf612f6d56166d2086c2ab02c7508 Implements: blueprint create-ops-guide https://review.openstack.org/453984 | 10:22 |
| mpranjic | asettle: is this a proper way? sorry for doing baby steps. getting used to the interface | 10:25 |
| asettle | mpranjic: more or less! So, we have commit message rules - basically that each sentence needs a new line. Let me show you: | 10:26 |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible master: [docs] Add backup information to the OSA Ops Guide https://review.openstack.org/453984 | 10:26 |
| asettle | ^^ | 10:26 |
| *** udesale has quit IRC | 10:26 | |
| asettle | mpranjic: normally, if it's a small change like you just did, you won't need to add anything to the commit message. It's just expected that you were 'fixiing a small nit'. I would also recommend (for etiquette sake) replying and mentioning that you pushed a patch with that small fix. | 10:27 |
| *** stuartgr has joined #openstack-ansible | 10:27 | |
| *** cNilesh has quit IRC | 10:27 | |
| asettle | If it's a bigger fix, it's usually best to comment with your concerns, and wait for the original person to reply and fix it :) | 10:27 |
| mpranjic | asettle: thnx :) | 10:29 |
| mpranjic | now.. trying to see my original commmit message ... | 10:30 |
| asettle | Ah! | 10:30 |
| asettle | Lemme show youuuu that fun trick | 10:30 |
| asettle | https://review.openstack.org/#/c/453984/2..3//COMMIT_MSG | 10:30 |
| asettle | (see the 2..3 in the middle of the URL there? That allows me to see the diff between the second commit - yours, and the third commit - mine) | 10:31 |
| asettle | mpranjic: ^ | 10:31 |
| mpranjic | nice :) | 10:31 |
| mpranjic | so I lacked newlines between my message and the rest | 10:32 |
| mpranjic | noted | 10:32 |
| mpranjic | one more thing: in patch 2 and 3 I don't see the comments from patch 1. That is expected? | 10:33 |
| asettle | Yes, that is expected. The new patchset will override the comments on the other patches. | 10:34 |
| asettle | That's okay, you can go back and look by using hte URL to determine what patchset you want to look at | 10:34 |
| asettle | So, for example: https://review.openstack.org/#/c/453984/ (add 1 at the end here) and you can see patch set 1 and teh comments with it | 10:35 |
| asettle | It's all iterative, so you never 'lose' content | 10:35 |
| asettle | Or comments | 10:35 |
| mpranjic | I see | 10:36 |
| mpranjic | I am used to phabricator, so I am getting used to this right now but it will be awesome :) | 10:36 |
| asettle | mpranjic: this is a (long) but good set of guidelines for devs: https://docs.openstack.org/infra/manual/developers.html | 10:37 |
| asettle | Goes through *all* the things | 10:37 |
| asettle | You'll find most of the stuff there, or, here: https://docs.openstack.org/contributor-guide/index.html | 10:38 |
| asettle | Both are equally helpful in different ways | 10:38 |
| asettle | So, feel free to ping if you ever have questions, but you'll also most likely find it there somewhere | 10:38 |
| mpranjic | thnx | 10:39 |
| mpranjic | will dig into it | 10:39 |
| mpranjic | it is also good to see some things on example. thank you for sharing one | 10:39 |
| asettle | You're welcome :) | 10:44 |
| *** SerenaFeng has quit IRC | 10:51 | |
| *** chas has joined #openstack-ansible | 11:04 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-galera_client master: Conditional pip dependency https://review.openstack.org/454674 | 11:06 |
| mpranjic | asettle: can we go though one example when you have some time pls? https://review.openstack.org/#/c/453557/ | 11:07 |
| asettle | mpranjic: shoot :) | 11:07 |
| asettle | Ah, that's an interesting patch you've chosen :P hahahaha | 11:07 |
| mpranjic | I can pick the right ones :) | 11:08 |
| mpranjic | so ... | 11:08 |
| mpranjic | I see 1 file to be added/changed and whole bunch of them to be deleted. correct? | 11:08 |
| *** chas has quit IRC | 11:09 | |
| asettle | mpranjic: well, not added, it's not new. If the file was to be added, there would be an A next to the file, similar to how there is a D next to the ones being deleted. | 11:09 |
| mpranjic | true | 11:09 |
| mpranjic | changed | 11:09 |
| asettle | I edited that file, and moved that text over. That's why it's all bright green. | 11:09 |
| asettle | The rest is to be deleted, correct :) | 11:10 |
| mpranjic | okido.. looking at that file now ... and patches... how can I be sure all comments are resolved? | 11:10 |
| asettle | mpranjic: what exactly do you mean? | 11:12 |
| asettle | Give me an example | 11:12 |
| mpranjic | asettle: https://review.openstack.org/#/c/453557/5/doc/config-reference/source/object-storage.rst | 11:13 |
| *** cNilesh has joined #openstack-ansible | 11:13 | |
| asettle | Ah | 11:14 |
| asettle | mpranjic: so you can use that trick I showed you earlier where you can view two different patchsets | 11:14 |
| asettle | So, you can see here: https://review.openstack.org/#/c/453557/5..6/doc/config-reference/source/object-storage.rst | 11:14 |
| asettle | THat I have removed teh toctree as per Andreas' comments | 11:14 |
| *** markvoelker has joined #openstack-ansible | 11:14 | |
| mpranjic | oh | 11:15 |
| mpranjic | i see | 11:15 |
| mpranjic | I presumed you wpuld mark that commend as OK/fixed | 11:15 |
| mpranjic | *comment | 11:15 |
| mpranjic | so I don't see any new comments | 11:16 |
| mpranjic | who accepts the review and when in this case? | 11:16 |
| *** openstackgerrit has quit IRC | 11:18 | |
| asettle | mpranjic: well, normally people do :P I just am a bit lazy in that respect. | 11:19 |
| *** markvoelker has quit IRC | 11:19 | |
| asettle | It's usually helpful to add a 'fixed' or 'done' to the comment | 11:19 |
| asettle | mpranjic: who accepts the review? What do you mean by that exactly? Like, who approves it? | 11:19 |
| *** shardy is now known as shardy_lunch | 11:21 | |
| *** pcaruana has quit IRC | 11:22 | |
| mpranjic | asettle: yes. the point in time when it is aproved to be merged | 11:23 |
| mpranjic | I presume the one who had a comment on it becomes on of the reviewers and his aproval is needed | 11:24 |
| asettle | Yes, so, each project team within openstack has a 'core' team. The patch you're currently looking at is for the openstack-manuals repo. | 11:24 |
| asettle | The list of core individuals are here: https://review.openstack.org/#/admin/groups/30,members | 11:24 |
| mpranjic | for example, in phabricator all reviewers have to accept the review before we land it | 11:24 |
| mpranjic | so I am trying to understand the proces based on this example | 11:24 |
| asettle | So, everyone and anyone can review a patch. But we need at least one +1, a +2 (from a core, only cores can give +2's) and a +2 with an 'approval' for another core. ANy core can give approval | 11:25 |
| asettle | Here is the list of openstack-ansible cores: https://review.openstack.org/#/admin/groups/490,members | 11:25 |
| asettle | Documentation changes within openstack-ansible are slightly different. There are 3 doc cores working for openstack-ansible. These are rromans spotz and myself. If there is a doc change, it is required that one of us gives the patch a +2 before it is merged. | 11:26 |
| asettle | This is to ensure the patch has been reviewed for spelling/grammar | 11:26 |
| mpranjic | understood | 11:26 |
| mpranjic | so, currently this review has no open issues if I see it correctly and waiting for someone from the list to aprove it. correct? | 11:27 |
| asettle | mpranjic: that's right :) | 11:28 |
| asettle | And since it is my own patch, although I am a core, I cannot approve it. So I have to wait for others. | 11:28 |
| mpranjic | of course :) | 11:29 |
| mpranjic | that would be silly | 11:30 |
| asettle | Hahhaha quite quite :P | 11:30 |
| asettle | I mean, I have done it before, especially if it's just me adding a line space, or fixing a very small spelling error in the openstack-ansible repo | 11:30 |
| asettle | But that is not good practice. That's just me being impatient :p | 11:30 |
| *** hw_wutianwei has quit IRC | 11:33 | |
| *** vnogin has joined #openstack-ansible | 11:36 | |
| *** dixiaoli has quit IRC | 11:37 | |
| mpranjic | is there a rule how review should wait before aproval even if there are no objections? | 11:38 |
| spotz | mpranjic: If no one has looked at it post it on channel we probably just missed it | 11:39 |
| asettle | mpranjic: basically what spotz said. If you need it urgently merged, message people that can help. | 11:40 |
| asettle | But otherwise, people *will* get there. | 11:40 |
| asettle | It's just not everyone has the time 100% to look always :) (and some people really don't like being prodded for reviews) | 11:40 |
| mpranjic | understandable :) | 11:41 |
| mpranjic | I have one more: Jenkins test: what does it mean in this case? | 11:42 |
| asettle | mpranjic: so, Jenkins is our gating system. When we push patches, it tests the code or documentation against a series of pre-defined criteria. | 11:43 |
| asettle | So, when you push a docs patch - it looks for a 'niceness' check. THis ensures that you don't have any whitespace, or wrongly formatted RST. | 11:43 |
| asettle | For code patches, it's a little more complicated. | 11:43 |
| mpranjic | I use jenkins at work too. it can be helpful :) | 11:43 |
| *** vnogin has quit IRC | 11:43 | |
| asettle | mpranjic: it is good :) | 11:44 |
| asettle | So, different projects 'gate' against different things | 11:44 |
| mpranjic | so jenkins will do basic syntax checks. nice. | 11:44 |
| asettle | Yep! | 11:44 |
| mpranjic | okido. | 11:44 |
| mpranjic | I also noticed: | 11:44 |
| mpranjic | Code-Review | 11:44 |
| mpranjic | +2 Andreas Jaeger | 11:44 |
| mpranjic | this means review is aproved? | 11:44 |
| asettle | It means that the first core has agreed that it's the right/correct change. | 11:45 |
| asettle | I will still need another core to give +2, and +a (approval) | 11:45 |
| asettle | And then it will be checked by Jenkins again, before passing through and being 'merged' into the repository | 11:45 |
| mpranjic | so 2 cores are needed | 11:45 |
| mpranjic | you need to trigger jenkins job or it will be run automatically? | 11:45 |
| mpranjic | after 2nd +2 | 11:46 |
| spotz | It will run when it gets a +1 workflow | 11:46 |
| mpranjic | thnx. that explains :) | 11:47 |
| spotz | Keep in mind other people besides cores can +1/-1 a patch | 11:47 |
| *** thorst has joined #openstack-ansible | 11:47 | |
| spotz | Gives more then 2 sets of eyes, plus if there's not a whole lot of cores if a bunch of known contributers have +1'd a project might +1 it to let it go | 11:47 |
| spotz | last +1 was for workflow, still waking up:) | 11:48 |
| boxrick1 | If Python-pip is installed before the AIO the deploy ansible script fails.. | 11:48 |
| andymccr | boxrick1: hmm what is the failure, is it something we can account for/fix up? i guess we want a consistent version of pip | 11:49 |
| asettle | spotz: good point! | 11:50 |
| spotz | Ok pony riding time, I'll be back in a bit:) | 11:51 |
| asettle | mpranjic: does that sort of explain the process? Those links I gave you earlier will have some more context and explanations within them | 11:51 |
| asettle | spotz: enjoy! | 11:51 |
| boxrick1 | I am just looking into it now, was just confirming if it was python-pip full stop or simply version 9.01 | 11:51 |
| mpranjic | asettle: pretty much, yes :) | 12:00 |
| mpranjic | thanks a lot | 12:00 |
| mpranjic | will continue my reading now ... :) | 12:00 |
| asettle | mpranjic: no problem :) | 12:02 |
| mgariepy | morning everyone | 12:03 |
| admin0 | what name are we voting :D ? | 12:11 |
| admin0 | root ? | 12:11 |
| admin0 | got the R release name | 12:11 |
| mgariepy | revolution | 12:13 |
| admin0 | openstack ramble release - intentionally made complicated :) | 12:13 |
| evrardjp | morning | 12:13 |
| mgariepy | openstack revolution release ! haha | 12:13 |
| mgariepy | much better. | 12:13 |
| mgariepy | morning evrardjp | 12:13 |
| evrardjp | I was auto pinged by revolution. | 12:13 |
| evrardjp | :p | 12:13 |
| admin0 | there is no revolution in the choices | 12:13 |
| evrardjp | resolution? | 12:14 |
| mgariepy | recheck ? | 12:14 |
| mgariepy | haha | 12:14 |
| evrardjp | oh you are talking about release names | 12:14 |
| admin0 | root rock rex rocky rambles radium razorback :) | 12:14 |
| admin0 | yeah | 12:14 |
| asettle | Morning guys | 12:14 |
| evrardjp | it's already done, no? | 12:14 |
| evrardjp | "release" would be simple | 12:14 |
| *** markvoelker has joined #openstack-ansible | 12:14 | |
| admin0 | i got the voting link 30 mins back | 12:15 |
| admin0 | i think they sent it out today | 12:15 |
| evrardjp | it's a place right | 12:15 |
| evrardjp | it's gonna be 17.x good enough for me. | 12:15 |
| evrardjp | I don't care about the naming :p | 12:16 |
| evrardjp | When openstack gonna turn 18, that's when we get major issues. | 12:16 |
| evrardjp | #terriblejoke | 12:16 |
| *** gouthamr has joined #openstack-ansible | 12:18 | |
| admin0 | :) | 12:19 |
| *** cNilesh has quit IRC | 12:19 | |
| *** woodard has joined #openstack-ansible | 12:20 | |
| *** Jack_Iv_ has quit IRC | 12:23 | |
| *** Jack_Iv has joined #openstack-ansible | 12:23 | |
| *** Jack_Iv has quit IRC | 12:23 | |
| *** Jack_Iv has joined #openstack-ansible | 12:24 | |
| *** Jack_Iv has quit IRC | 12:24 | |
| *** Jack_Iv has joined #openstack-ansible | 12:25 | |
| *** Jack_Iv has quit IRC | 12:26 | |
| *** Jack_Iv has joined #openstack-ansible | 12:26 | |
| cloudnull | mornings | 12:29 |
| admin0 | ¯\_(ツ)_/¯ | 12:30 |
| *** shardy_lunch is now known as shardy | 12:32 | |
| *** acormier has joined #openstack-ansible | 12:36 | |
| *** acormier has quit IRC | 12:38 | |
| logan- | o/ cloudnull | 12:41 |
| *** esberglu has joined #openstack-ansible | 12:43 | |
| cloudnull | o/ logan- | 12:43 |
| cloudnull | hows it today | 12:43 |
| *** vnogin has joined #openstack-ansible | 12:44 | |
| *** vnogin has quit IRC | 12:45 | |
| logan- | goin alright | 12:45 |
| logan- | i have a question on something with the linear strategy plugin | 12:46 |
| logan- | http://cdn.pasteraw.com/oz8iyy4by39kfiewghqtd4fepjbyvbr -- it seems like it is using inventory_hostname instead of ansible_host when handling delegated tasks | 12:47 |
| *** acormier has joined #openstack-ansible | 12:47 | |
| cloudnull | yes that may be. | 12:47 |
| * cloudnull goes to look | 12:47 | |
| cloudnull | logan-: https://github.com/openstack/openstack-ansible-plugins/blob/master/strategy/linear.py#L133-L143 | 12:48 |
| *** vnogin has joined #openstack-ansible | 12:48 | |
| cloudnull | it's using physical_host | 12:48 |
| cloudnull | then it pulls the ansible_host | 12:49 |
| logan- | right but with a delegated task I'm inside https://github.com/openstack/openstack-ansible-plugins/blob/master/strategy/linear.py#L106-L131 | 12:49 |
| cloudnull | https://github.com/openstack/openstack-ansible-plugins/blob/master/strategy/linear.py#L140 | 12:49 |
| *** acormier has quit IRC | 12:49 | |
| logan- | it works fine if the task isn't delegated | 12:49 |
| logan- | i think we need to rework this so it does the same lookups for the delegated_to hostname | 12:51 |
| cloudnull | we could pull out the else | 12:51 |
| cloudnull | https://github.com/openstack/openstack-ansible-plugins/blob/master/strategy/linear.py#L132 | 12:51 |
| cloudnull | and just have it tweak a delegated task and do all the normal things | 12:51 |
| logan- | testing that out | 12:53 |
| mpranjic | asettle: I have one more if I may :) | 12:53 |
| asettle | mpranjic: please go on :) | 12:54 |
| mpranjic | asettle: | 12:55 |
| mpranjic | asettle: I would like to see how the proposed version will look | 12:55 |
| mpranjic | for example: https://review.openstack.org/#/c/410561/5/doc/networking-guide/source/config-dhcp-ha.rst@328 | 12:55 |
| *** Jack_Iv_ has joined #openstack-ansible | 12:55 | |
| mpranjic | is that possible? | 12:55 |
| *** manheim has quit IRC | 12:55 | |
| logan- | cloudnull: do we need to perform another lookup to get the vars for the delegated host though? won't all of the _vars.get for the container detection incorrectly lookup the playbook target, not the delegated_to host vars? | 12:56 |
| asettle | mpranjic: sure is! | 12:56 |
| *** manheim has joined #openstack-ansible | 12:56 | |
| asettle | So you can do it one of two ways, pull the PR down and manually build it yourself. | 12:56 |
| asettle | Instructions are here: https://docs.openstack.org/contributor-guide/docs-builds.html | 12:56 |
| asettle | OR | 12:56 |
| asettle | You can go here: https://review.openstack.org/#/c/410561/5 click on "gate-openstack-manuals-tox-doc-publish-checkbuild" | 12:56 |
| asettle | And you're looking at the networking guide, so click "networking-guide" | 12:57 |
| asettle | And you can see the build here: http://docs-draft.openstack.org/61/410561/8/check/gate-openstack-manuals-tox-doc-publish-checkbuild/8c59298//publish-docs/draft/networking-guide/index.html | 12:57 |
| cloudnull | logan-: hum | 12:58 |
| *** Jack_Iv has quit IRC | 12:58 | |
| cloudnull | yea we may need to try and pull the delegated_host and set it as the _physical_host object. | 12:58 |
| mpranjic | asettle: ok. that gives me start page for the entire guide. I would prefer to see changed page in that review (without to navigate from the beginning of the guide) | 13:00 |
| mpranjic | dunno if that can be done | 13:00 |
| asettle | mpranjic: nope, both options build the entire guide. So you have to navigate to it :) | 13:00 |
| logan- | also need to keep in mind delegated_to may be a container also, so we should do the same physical_host lookup work on it and set the container awareness | 13:00 |
| logan- | in this case its just the normal OSA common-task for setting up a mysql user/db where i'm seeing this | 13:01 |
| jcookeman | hey channel, I'm using os_server to create instances with boot_from_volume which creates a new volume if it doesn't exist. What do you recommend is the most practical way to name that volume? | 13:01 |
| mpranjic | asettle: oh. okido. will stash more beer close by to help me through it :) | 13:02 |
| asettle | Hahhaa mpranjic smart move :p | 13:02 |
| asettle | mpranjic: if you're interested in working on some more openstack-manuals stuff, jump in our chan #openstack-doc | 13:02 |
| *** chas has joined #openstack-ansible | 13:05 | |
| cloudnull | logan-: so our approach so far to task delegation has been to +/- leave things alone when its used. | 13:07 |
| cloudnull | are you seeing the strategy plugin break that | 13:08 |
| cloudnull | or would this be more of an optimization ? | 13:08 |
| logan- | yeah. in the case where the playbook is running against a container, and the task is delegated to a different container, it doesn't seem to work | 13:08 |
| cloudnull | huh, it should just disable pipelining and switch to paraminko for the delegated task | 13:09 |
| cloudnull | and then carry on normally | 13:09 |
| *** smatzek has joined #openstack-ansible | 13:10 | |
| *** chas has quit IRC | 13:10 | |
| logan- | http://cdn.pasteraw.com/mwdkn0su8q21f1jfk1ffq22xn3giarx a little more context. on the ssh plugin I see the ansible_host inside the brackets, on the paraimiko stuff with the delegated_task it has the inventory_hostname in the brackets, and the failure seems to be caused by it trying to connect to the inventory_hostname instead of the ansible_host there | 13:10 |
| cloudnull | we might be able to remove https://github.com/openstack/openstack-ansible-plugins/blob/master/strategy/linear.py#L106-L132 | 13:10 |
| cloudnull | and just change it to if not task.delegate_to: | 13:11 |
| cloudnull | that stuff was there due to issues in task delegation in 2.1 | 13:11 |
| cloudnull | which may no longer be a problem | 13:11 |
| *** smatzek has quit IRC | 13:12 | |
| *** smatzek has joined #openstack-ansible | 13:12 | |
| logan- | i removed it and it doesn't seem to perform delegation without it | 13:16 |
| *** cathrichardson has joined #openstack-ansible | 13:17 | |
| logan- | it just ran the delegated task on the non-delegated container heh | 13:17 |
| cloudnull | maybe ansible has broken delegation once again? | 13:17 |
| *** winggundamth has quit IRC | 13:18 | |
| cloudnull | just looking through the linear and ssh plugins it doesn't look like were doing much else to munge that | 13:18 |
| *** cathrich_ has quit IRC | 13:19 | |
| cloudnull | logan-: if you pull https://github.com/openstack/openstack-ansible-plugins/blob/master/connection/ssh.py out of the way does it work ? | 13:19 |
| logan- | looking. but also this is what i saw when ripping out the paramiko stuff http://cdn.pasteraw.com/igwsh6j4cw2didih629to1vy4kk6fi7 | 13:22 |
| logan- | no nothing works if i move the ssh connection plugin out of the way | 13:24 |
| logan- | because these containers don't have ssh, they need to have the lxc-attach wrapper | 13:24 |
| logan- | same test.yml with a normal ssh connection/strategy plugin: http://cdn.pasteraw.com/ogiewfdkifhhfwxwfjhlqgc1qjdlo7a ... see what i'm saying about the delegated connection having the inventory hostname instead of the IP address? | 13:27 |
| logan- | brb coffee | 13:28 |
| logan- | i guess this isn't going to work with paramiko anyway. | 13:34 |
| logan- | since using paramiko as the connection plugin will not lxc-attach wrap the command | 13:34 |
| logan- | i think we just don't have support built for container -> container delegation currently | 13:35 |
| *** adreznec has quit IRC | 13:39 | |
| *** Jack_Iv_ has quit IRC | 13:43 | |
| *** Jack_Iv has joined #openstack-ansible | 13:46 | |
| *** openstackgerrit has joined #openstack-ansible | 13:47 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Adds iptables rules to protect octavia server container https://review.openstack.org/447151 | 13:47 |
| cloudnull | logan- getting coffee myself. | 13:52 |
| logan- | im gonna hack on it some and see what happens | 13:52 |
| boxrick1 | andymccr: I mentioned this morning about me thinking it was python-pip, it actually turns out the problem is somehow related to my ARA role. I thought it may have been caused by the env vars I set, but I already add in the same paths you use in ansible.cfg so that won't cause an issue. Perhaps you can have a quick glance over my role and error and may be | 13:53 |
| boxrick1 | able to provide a bit of insight? https://gist.github.com/boxrick/bdf492848f02c1f3d5ea4af4b7933729 | 13:53 |
| *** cjloader_ has joined #openstack-ansible | 13:54 | |
| boxrick1 | The bootstrap-ansible.sh all runs sucessfully without error | 13:54 |
| *** udesale has joined #openstack-ansible | 13:55 | |
| andymccr | hmm boxrick1 that looks like the openstack-ansible bin isn't working properly - or is pointing at the wrong location for ansible-playbook | 13:59 |
| andymccr | try a which openstack-ansible and then check that file out? (its just a wrapper that automatically adds -e @filename for vars files in /etc/openstack_deploy" | 13:59 |
| andymccr | lemme try a bootstrap quickly on a vm and see | 14:00 |
| *** adreznec has joined #openstack-ansible | 14:00 | |
| strattao | Hi everyone. I have someone on my team that thinks deploying with OSA in production is nice for small deployments, but does not work at scale for larger ones. Is there any ammo you can give on some large-scale deployments using OSA to set his mind at ease? | 14:04 |
| andymccr | boxrick1: hmm i wonder why /opt/ansible-runtime/bin/ansible-playbook doesnt exist then... | 14:04 |
| andymccr | strattao: cloudnull can probably give you a rundown - the OSIC cloud runs using OSA, additionally the infra assigned to OpenStack gating from OSIC is running OSA | 14:04 |
| andymccr | strattao: additionally, its super flexible - e.g. you can run it on metal or in containers and with x number of containers per host or any number of archs so its really as scalable as you make it. | 14:05 |
| cloudnull | strattao: what are you looking for ? | 14:05 |
| boxrick1 | '/opt/ansible-runtime/bin' is missing all the ansible binaries which are ultimately there on a working install | 14:06 |
| andymccr | boxrick1: that is weird. | 14:06 |
| cloudnull | strattao: http://grafana.openstack.org/dashboard/db/nodepool-osic | 14:07 |
| cloudnull | https://cloud1.osic.org/grafana/dashboard/db/openstack-all-compute-aggregates | 14:07 |
| boxrick1 | If I symbolic link it from where they actually are it will all work | 14:07 |
| cloudnull | idk if those metrics will help your case | 14:07 |
| cloudnull | but gives you decent insight into what we're doing at large scale under constant workload | 14:08 |
| andymccr | boxrick1: https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L128-L142 i think thats what that is meant to do | 14:08 |
| boxrick1 | Yea I was looking at that bit and trying to see why it didnt run properly | 14:08 |
| cloudnull | also plumgrid did a scale test using OSA talk here: https://www.openstack.org/videos/video/sdn-scale-out-testing-at-openstack-innovation-center-osic blog here: https://osic.org/blogs/plumgrid-proves-sdn-scalability-osic -- sadly the whitepaper which was on plumgrid.com is gone now that vmware owns them | 14:11 |
| strattao | cloudnull: Thanks for taking a moment. I’m looking for a couple data points like the largest number of nodes in a single cluster have been deployed, why vxlan was required for deployment. What an ideal large scale deployment looks like in terms of number of servers. | 14:11 |
| *** cjloader_ has quit IRC | 14:12 | |
| strattao | andymccr: are there any specs on the physical size of OSIC somewhere? How many servers, how many users, etc? | 14:13 |
| *** thorst has quit IRC | 14:15 | |
| cloudnull | strattao: at the moment the osic is 242 compute nodes, 100 swift/cinder, with a 5 node control plane in region 1. its 250 ironic compute nodes and a 3 node control plane in region 2 | 14:15 |
| andymccr | ^ there we go :) | 14:15 |
| strattao | thank you | 14:15 |
| cloudnull | you can kinda see that here https://cloud1.osic.org/grafana/dashboard/db/openstack-all-compute-aggregates | 14:15 |
| strattao | this is all very helpful! | 14:15 |
| cloudnull | compute node count at the top | 14:16 |
| cloudnull | the system metrics dashboard will give you all the data and things throughout the cloud https://cloud1.osic.org/grafana/dashboard/db/openstack-system-metrics | 14:16 |
| cloudnull | over the last 90 days | 14:16 |
| andymccr | strattao: i think one of the key things though, is that we don't force an architecture - we aim to be flexible around what a deployer wants, so if OpenStack can scale, OSA should be able to scale. | 14:16 |
| andymccr | for example I know some testing happened around swift and found it performed far better in a large cluster, running swift-proxy on physical hosts /w dedicated memcache - you can set that up with OSA if you want. | 14:17 |
| *** agrebennikov has joined #openstack-ansible | 14:17 | |
| *** thorst has joined #openstack-ansible | 14:18 | |
| strattao | one of the big concerns is with network traffic. He is VERY leery about using VXLAN in production. For bigger deployments, any insights as to what has worked well at scale? | 14:19 |
| cloudnull | strattao: i guess its also worth mentioning that the rackspace private cloud uses osa for the basis of their product (shameless plug, as i work for rackspace) and we've huge clouds under our management though I'm not at liberty to share customer specific data. | 14:20 |
| cloudnull | strattao: if your nics support vxlan offloading, vxlan works great! | 14:20 |
| andymccr | cloudnull: do we force vxlan? | 14:20 |
| strattao | cloudnull: yeah, fair enough. I think that was one of his arguments is that OSA is very Rackspacy | 14:20 |
| cloudnull | intel x710, melenox connect x3pro+ | 14:20 |
| cloudnull | I mean OSA has lots of rackers working on it. but when I was PTL I kicked all of the rackspacey things out | 14:21 |
| andymccr | ^ also the % of commits in OSA from Rackspace has dropped by about 5-10% each cycle, it was just over 50% in Ocata | 14:22 |
| cloudnull | few other nics have vxlan offloading so there are options out their | 14:22 |
| cloudnull | but if you're networking gear does not support vxlan offloading its not a great experience | 14:22 |
| cloudnull | that said, we do not force you to have vxlan | 14:22 |
| cloudnull | or even use it | 14:23 |
| cloudnull | if you want a pure l2 nework topology, that works great | 14:23 |
| cloudnull | also if you want to use OVS you can have GRE for L3 networks | 14:23 |
| *** adreznec has quit IRC | 14:23 | |
| cloudnull | vxlan is the default because its what the linuxbridge agent supports | 14:23 |
| cloudnull | and that is the default because it'll run everywhere. | 14:24 |
| cloudnull | ... and that is the default (linuxbridge) because ... | 14:24 |
| *** shashank_t_ has joined #openstack-ansible | 14:25 | |
| *** david-lyle has joined #openstack-ansible | 14:25 | |
| cloudnull | many of our rackspace customers just use l2 w/ vlans | 14:25 |
| cloudnull | and just about all commodity network gear has support for vlan offloading. | 14:26 |
| strattao | so, if you had your choice for the simplest to manage large scale deployment, would you choose VXLAN and ensure you had NICs with VXLAN offloading, or would you have network gear that supports a large VLAN-based deployment? | 14:27 |
| strattao | or something else? | 14:27 |
| cloudnull | for simplest to manage, and cheaper for that matter, I'd go all L2 and not use any vxlan for tenant networks. | 14:28 |
| cloudnull | its super performant, easy to understand. | 14:28 |
| agrebennikov | my 2 cents - it was found by one of the large customers that currently vxlan performance is terrible when used with lxbr approach | 14:29 |
| agrebennikov | that is was very quickly decided to completely comment out vxlans and just go with vlans | 14:29 |
| agrebennikov | also there are so many prod clouds where tenants only use corp IP addresses straight, so basically there are only 2-3 shared networks in the whole cloud (like /22 subnets or so) | 14:30 |
| agrebennikov | and if you want to expose the instances to outside many of them already use just ipv6 | 14:31 |
| cloudnull | ++ the osic uses a dual stack ipv6 + 10/8 tenant network which has been amazing | 14:31 |
| cloudnull | all of the infra gates on the osic are using that | 14:32 |
| admin0 | strattao: but if you are to do a public cloud, or some networy heavy deployment, vlans will limit you to around 4000 networks only | 14:32 |
| strattao | cool | 14:32 |
| admin0 | per region | 14:32 |
| logan- | strattao: if you want direct routing to vms with no overlay, don't have a need for tenant networks, and don't require the additional fancy features like lbaas, calico is worth a look too. | 14:33 |
| strattao | yeah, I think one item that was tripping us up is in the deployment guide it says that br-vxlan is a required bridge: https://docs.openstack.org/project-deploy-guide/openstack-ansible/newton/overview-network-arch.html | 14:33 |
| cloudnull | we can get near linespeed w/ linuxbridge + vxlan, but only when everything in-path supports vxlan offloading (+acceleration) | 14:33 |
| admin0 | strattao: public or private cloud ? | 14:33 |
| strattao | private cloud | 14:34 |
| *** Dinesh_Bhor has quit IRC | 14:34 | |
| admin0 | will it have more than 4000 networks per region ? | 14:34 |
| cloudnull | ohh yea good point logan- calico is quite amazing ! | 14:34 |
| admin0 | if answer is no, go vlans :) | 14:34 |
| cloudnull | o/ admin0 | 14:35 |
| admin0 | hey | 14:35 |
| evrardjp | I need a pip wheel builder expert, cloudnull? :D | 14:35 |
| * cloudnull runs | 14:35 | |
| evrardjp | haha! | 14:35 |
| cloudnull | what's up ? | 14:36 |
| admin0 | “pip wheel builder expert” | 14:36 |
| *** lucasxu has joined #openstack-ansible | 14:36 | |
| evrardjp | so behind my repo-build there is this command that runs | 14:36 |
| evrardjp | http://paste.openstack.org/show/605762/ | 14:36 |
| admin0 | whoever comes forward, i will definately endore in linkedin for this :) | 14:36 |
| evrardjp | haha | 14:37 |
| evrardjp | odyssey4me: and cloudnull would do the trick there I think :p | 14:37 |
| cloudnull | evrardjp: and then ? | 14:37 |
| evrardjp | let me copy paste! | 14:37 |
| admin0 | for recruiters “its a hardcore openstack expertise “ only a few have this .. | 14:37 |
| cloudnull | AND THEN! | 14:37 |
| cloudnull | :) | 14:37 |
| *** smatzek has quit IRC | 14:37 | |
| evrardjp | http://paste.openstack.org/show/605763/ | 14:37 |
| evrardjp | and indeed the pools/ is empty | 14:38 |
| cloudnull | https://www.youtube.com/watch?v=CkdyU_eUm1U | 14:38 |
| evrardjp | loading video | 14:38 |
| *** thorst has quit IRC | 14:39 | |
| *** udesale has quit IRC | 14:39 | |
| cloudnull | evrardjp: is /var/www/repo/pools/elasticsearch/elasticsearch-2.3.0-py2.py3-none-any.whl missing ? | 14:39 |
| evrardjp | /var/www/repo/pools/ is empty | 14:39 |
| cloudnull | looks like the os-release has a link to it | 14:39 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-plugins master: Support delegation in strategy plugin https://review.openstack.org/454742 | 14:40 |
| logan- | cloudnull ^ | 14:40 |
| spotz | Poke for reviews:) | 14:40 |
| evrardjp | cloudnull: I didn't see that movie | 14:40 |
| cloudnull | evrardjp: you should fix that | 14:41 |
| cloudnull | both the pool thing and the move | 14:41 |
| cloudnull | **movie | 14:41 |
| cloudnull | :) | 14:41 |
| evrardjp | haha | 14:41 |
| cloudnull | logan-: so does that address the delegated commands? | 14:41 |
| evrardjp | That I already know | 14:41 |
| evrardjp | :p | 14:41 |
| evrardjp | logan-: thanks for this commit :) I will have a look! | 14:41 |
| cloudnull | and its working as expected in your env ? | 14:41 |
| * cloudnull pulling into a test node now | 14:42 | |
| *** udesale has joined #openstack-ansible | 14:42 | |
| *** adreznec has joined #openstack-ansible | 14:42 | |
| logan- | yes it now sets the physical_host and container_name stuff relative to task.delegate_to | 14:42 |
| evrardjp | cloudnull: isn't the pools building part of the repo_build process? | 14:42 |
| cloudnull | it is | 14:42 |
| cloudnull | however is this an existing deployment | 14:43 |
| logan- | so we get container awareness parity to non-delegated tasks | 14:43 |
| evrardjp | cloudnull: new | 14:43 |
| *** thorst has joined #openstack-ansible | 14:43 | |
| *** thorst has quit IRC | 14:43 | |
| evrardjp | python artifact building so kinda really empty. | 14:43 |
| *** shashank_t_ has quit IRC | 14:43 | |
| *** thorst has joined #openstack-ansible | 14:44 | |
| evrardjp | odyssey4me: did the previous work on this, I just follow up during his holidays :) | 14:44 |
| *** shashank_t_ has joined #openstack-ansible | 14:44 | |
| strattao | thanks everyone for the insight - this was really helpful and appreciated! | 14:45 |
| *** shashank_t_ has quit IRC | 14:46 | |
| *** shashank_t_ has joined #openstack-ansible | 14:46 | |
| evrardjp | cloudnull: if you check the first paste, you see the find-links is the only reference to pools | 14:47 |
| *** shashank_t_ has quit IRC | 14:49 | |
| *** shashank_t_ has joined #openstack-ansible | 14:52 | |
| *** adreznec has quit IRC | 14:52 | |
| *** adreznec has joined #openstack-ansible | 14:53 | |
| evrardjp | cloudnull: what's the reason the pip wheel doesn't check extra-index-url ? | 14:53 |
| *** chris_hultin|AWA is now known as chris_hultin | 14:54 | |
| evrardjp | ok I'll replace man 5 cloudnull by apt-get install man && man pip | 14:55 |
| *** shashank_t_ has quit IRC | 14:56 | |
| *** markus_z has quit IRC | 14:57 | |
| *** winggundamth has joined #openstack-ansible | 14:59 | |
| *** vnogin has quit IRC | 15:01 | |
| *** markus_z has joined #openstack-ansible | 15:01 | |
| *** markus_z has quit IRC | 15:01 | |
| *** vnogin has joined #openstack-ansible | 15:02 | |
| cloudnull | evrardjp: I suspect your pip.conf file has no-index=true | 15:02 |
| cloudnull | and --extra-index-url https://rpc-repo.rackspace.com/pools wont work. | 15:02 |
| cloudnull | "https://rpc-repo.rackspace.com/pools " is not an idex | 15:03 |
| *** shashank_t_ has joined #openstack-ansible | 15:03 | |
| cloudnull | index | 15:03 |
| evrardjp | I don't have no-index=true | 15:03 |
| *** shashank_t_ has quit IRC | 15:03 | |
| evrardjp | But agreed on not an index. | 15:03 |
| *** chris_hultin is now known as chris_hultin|AWA | 15:04 | |
| *** shashank_t_ has joined #openstack-ansible | 15:04 | |
| evrardjp | I thought it was looking into the html for links if used | 15:04 |
| *** winggundamth has quit IRC | 15:04 | |
| evrardjp | but I didn't modify this. | 15:04 |
| cloudnull | if rpc-repo needs to be a pypi index, it could use something like https://pypi.python.org/pypi/pypiserver#using-a-different-wsgi-server | 15:05 |
| cloudnull | and then it'd work | 15:05 |
| *** chas has joined #openstack-ansible | 15:06 | |
| *** vnogin has quit IRC | 15:07 | |
| admin0 | anyone tried, heard of OVN ? | 15:07 |
| *** shashank_t_ has quit IRC | 15:08 | |
| *** Oku_OS is now known as Oku_OS-away | 15:09 | |
| cloudnull | not i | 15:10 |
| *** jamesdenton has joined #openstack-ansible | 15:10 | |
| cloudnull | but the IBM folks were (are) running it | 15:10 |
| *** foutatoro has quit IRC | 15:10 | |
| *** chas has quit IRC | 15:11 | |
| *** shashank_t_ has joined #openstack-ansible | 15:11 | |
| *** shashank_t_ has quit IRC | 15:11 | |
| *** Pavo has joined #openstack-ansible | 15:14 | |
| *** Pavo has left #openstack-ansible | 15:15 | |
| *** udesale has quit IRC | 15:16 | |
| *** manheim has quit IRC | 15:21 | |
| evrardjp | open virtual networking? | 15:24 |
| evrardjp | we have dragonflow FYI | 15:25 |
| evrardjp | work in progress in OSA, patches welcome. | 15:25 |
| *** BjoernT has joined #openstack-ansible | 15:34 | |
| *** chyka has joined #openstack-ansible | 15:41 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_horizon master: Use central testing repo Horizon plays https://review.openstack.org/454765 | 15:42 |
| *** rmelero has joined #openstack-ansible | 15:48 | |
| *** admin0 has quit IRC | 15:54 | |
| *** cjloader_ has joined #openstack-ansible | 15:59 | |
| *** jcookeman has quit IRC | 15:59 | |
| *** cjloader_ has quit IRC | 16:01 | |
| *** lostRhino has joined #openstack-ansible | 16:02 | |
| *** admin0 has joined #openstack-ansible | 16:02 | |
| *** cmart has joined #openstack-ansible | 16:04 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-galera_client master: Conditional pip dependency https://review.openstack.org/454674 | 16:04 |
| lostRhino | We’ve been getting the following error: [DEPRECATION WARNING]: Skipping task due to undefined Error, in the future this\n will be a fatal error.: 'ssh_key_url' is undefined. all the research I’ve done points to an AIO install but the install is not an AIO (I’m working on building a disconnected build w/repos and the such)- If anyone could point me to a variable or something more than this | 16:05 |
| lostRhino | (https://github.com/openstack/openstack-ansible-openstack_hosts/blob/master/tasks/openstack_authorized_keys.yml) file I’d be glad … I’m probably missing something and its probably small but any and all help would be greatly appreciated | 16:05 |
| lostRhino | it fails at the following step “TASK [install-osad : OSA | Setup hosts]” | 16:06 |
| *** johnmilton has joined #openstack-ansible | 16:07 | |
| *** johnmilton has quit IRC | 16:09 | |
| *** Jack_Iv has quit IRC | 16:12 | |
| andymccr | lostRhino: where is that task/role? i assume the install-osad role is diff? i think that error you linked is unimportant - its a deprecation warning but its not going to be the reason it isnt installing | 16:18 |
| lostRhino | under setup-hosts.yml | 16:18 |
| lostRhino | wait… hold on I think I see something | 16:19 |
| lostRhino | andymccr: let me get back to you | 16:19 |
| lostRhino | thanks for your assistance | 16:19 |
| andymccr | lostRhino: no problems! | 16:20 |
| lostRhino | a little frustrated with myself - forgot to do a commit after a change.. | 16:21 |
| lostRhino | :-/ | 16:21 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_horizon master: Use central testing repo Horizon plays https://review.openstack.org/454765 | 16:23 |
| andymccr | haha welcome to the team :P | 16:23 |
| spotz | lostRhino welcome!:) | 16:24 |
| *** lucasxu has quit IRC | 16:24 | |
| *** cmart has quit IRC | 16:26 | |
| *** teng has joined #openstack-ansible | 16:30 | |
| *** acormier has joined #openstack-ansible | 16:32 | |
| *** cmart has joined #openstack-ansible | 16:33 | |
| *** winggundamth has joined #openstack-ansible | 16:35 | |
| *** jbadiapa has quit IRC | 16:36 | |
| boxrick1 | Would anyone be kind enough to try and assist me with my problem? Specifically this line: https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L125 info: https://gist.github.com/boxrick/bdf492848f02c1f3d5ea4af4b7933729 | 16:37 |
| boxrick1 | When I use the included role, the ansible files in /opt/ansible-runtime/bin don't get populated | 16:38 |
| *** stuartgr has left #openstack-ansible | 16:38 | |
| boxrick1 | I have narrowed it down to that specific line. | 16:38 |
| boxrick1 | Almost like it gets skipped over because something is present | 16:39 |
| *** rpittau has quit IRC | 16:41 | |
| *** teng has quit IRC | 16:42 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/ocata: nova user can read kernel for libguestfs on Ubuntu https://review.openstack.org/454269 | 16:43 |
| *** Jack_Iv has joined #openstack-ansible | 16:43 | |
| andymccr | boxrick1: the first (broken) one it skipped over installing ansible (+ reqs like jinja2 etc) because they were already there (just in the wrong location) https://gist.github.com/boxrick/bdf492848f02c1f3d5ea4af4b7933729#file-bootstrap-ansible-sh-broken-L77-L83 | 16:43 |
| andymccr | there must be a way to force that. | 16:43 |
| *** acormier has quit IRC | 16:44 | |
| andymccr | im guessing because they are in a diff location and it didnt install them the linking never happens | 16:44 |
| andymccr | i wonder if we could patch it to be smarter - e.g. either reinstall (less smart) or check where ansible lives and link from that location instead of a flat expected location | 16:45 |
| boxrick1 | Oddly Ansible is not present at all on that system. | 16:45 |
| boxrick1 | Oh no, ignore that looking at wrong thing. Too many tabs | 16:45 |
| *** winggundamth has quit IRC | 16:47 | |
| *** Jack_Iv has quit IRC | 16:48 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Ensure proxy check is short circuit https://review.openstack.org/454604 | 16:48 |
| *** vnogin has joined #openstack-ansible | 16:49 | |
| boxrick1 | Seems that ARA installs through pip: setuptools, Frozen-Flask, Flask-SQLAlchemy, junit-xml, Flask, ansible, pygments, decorator, Flask-Migrate, cliff, debtcollector, Flask-Script, pbr | 16:52 |
| *** vnogin has quit IRC | 16:54 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Converge neutron agents onto Baremetal https://review.openstack.org/454450 | 16:54 |
| *** lostRhino has left #openstack-ansible | 16:55 | |
| *** acormier has joined #openstack-ansible | 16:55 | |
| andymccr | you could set PIP_OPTS += --force-reinstall | 16:57 |
| andymccr | perhaps? | 16:57 |
| andymccr | that'd force those packages to reinstall | 16:57 |
| andymccr | ideally we should probably use a venv for this, so OSA uses its own version of everything | 16:58 |
| andymccr | avoids a situation where ara has dependencies that osa doesnt match or vice versa | 16:59 |
| boxrick1 | Would that not conflict with the already present "--upgrade"? | 17:00 |
| *** japestinho_ has joined #openstack-ansible | 17:01 | |
| andymccr | i dont think it would. i mean its still a hacky way to get around that - either checking where those binaries are before linking, or using a venv for osa so that reqs dont overlap is a far better way to go i think | 17:03 |
| *** jpstnh has quit IRC | 17:04 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] reduce container counts for cinder, ironic and trove https://review.openstack.org/454802 | 17:07 |
| *** chas has joined #openstack-ansible | 17:07 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] reduce container counts for cinder, ironic and trove https://review.openstack.org/454802 | 17:09 |
| boxrick1 | Is it worth raising a bug, or is this something that will naturally be dealt with? I can use a workaround for now in my environment. | 17:11 |
| openstackgerrit | Merged openstack/openstack-ansible master: Set a sane default haproxy_bind_on_non_local https://review.openstack.org/454606 | 17:11 |
| *** chas has quit IRC | 17:12 | |
| *** 5EXAATI3T has joined #openstack-ansible | 17:12 | |
| andymccr | boxrick1: id call that a bug and im not sure it'll get dealt with otherwise - would probably be a good idea to raise a bug about that | 17:12 |
| *** pbandark has quit IRC | 17:13 | |
| *** japestinho_ has quit IRC | 17:16 | |
| *** acormier has quit IRC | 17:22 | |
| andymccr | have a great weekend all! im out for today :) | 17:22 |
| *** lucasxu has joined #openstack-ansible | 17:24 | |
| *** acormier has joined #openstack-ansible | 17:26 | |
| spotz | seeya andymccr | 17:27 |
| openstackgerrit | Merged openstack/openstack-ansible master: Configurable haproxy network whitelists https://review.openstack.org/454219 | 17:28 |
| *** acormier has quit IRC | 17:31 | |
| *** shardy has quit IRC | 17:35 | |
| *** BjoernT has quit IRC | 17:44 | |
| *** teng has joined #openstack-ansible | 17:52 | |
| *** mattt has quit IRC | 17:57 | |
| *** mattt has joined #openstack-ansible | 17:57 | |
| teng | I'm using OpenStack Ansible 15.1.0 to deploy OpenStack 3 nodes, but I found neutron-L3-agent outside of neutron agent lxc. In this case is normally, isn't it. | 18:01 |
| *** poopcat has joined #openstack-ansible | 18:04 | |
| *** albertcard1 has joined #openstack-ansible | 18:11 | |
| *** poopcat has quit IRC | 18:12 | |
| *** albertcard1 is now known as poopcat | 18:13 | |
| xgerman | if somebody wants to take a look: https://review.openstack.org/#/c/447151/ | 18:20 |
| *** smatzek has joined #openstack-ansible | 18:21 | |
| *** teng has quit IRC | 18:28 | |
| *** smatzek has quit IRC | 18:33 | |
| *** electrofelix has quit IRC | 18:33 | |
| *** MasterOfBugs has joined #openstack-ansible | 18:39 | |
| *** thorst has quit IRC | 19:03 | |
| *** chas has joined #openstack-ansible | 19:08 | |
| *** vnogin has joined #openstack-ansible | 19:11 | |
| *** vnogin has quit IRC | 19:12 | |
| *** chas has quit IRC | 19:12 | |
| *** vnogin has joined #openstack-ansible | 19:13 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Converge neutron agents onto Baremetal https://review.openstack.org/454450 | 19:15 |
| cloudnull | teng where was it deployed ? | 19:16 |
| cloudnull | xgerman: looking | 19:16 |
| xgerman | thanks | 19:16 |
| *** openstackgerrit has quit IRC | 19:19 | |
| *** teng has joined #openstack-ansible | 19:21 | |
| *** teng has quit IRC | 19:21 | |
| *** pbandark has joined #openstack-ansible | 19:26 | |
| *** thorst has joined #openstack-ansible | 19:32 | |
| *** cjloader_ has joined #openstack-ansible | 19:35 | |
| *** vnogin has quit IRC | 19:35 | |
| *** pbandark has quit IRC | 19:38 | |
| *** schwicht has joined #openstack-ansible | 19:38 | |
| *** cjloader_ has quit IRC | 19:40 | |
| *** pbandark has joined #openstack-ansible | 19:40 | |
| *** openstackgerrit has joined #openstack-ansible | 20:05 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] reduce container counts for cinder, ironic and trove https://review.openstack.org/454802 | 20:05 |
| *** Jack_Iv has joined #openstack-ansible | 20:07 | |
| *** MasterOfBugs has quit IRC | 20:10 | |
| *** Jack_Iv has quit IRC | 20:16 | |
| *** lucasxu has quit IRC | 20:17 | |
| *** mcrafty has quit IRC | 20:19 | |
| *** vnogin has joined #openstack-ansible | 20:22 | |
| *** manheim has joined #openstack-ansible | 20:24 | |
| *** gouthamr has quit IRC | 20:26 | |
| *** esberglu has quit IRC | 20:27 | |
| *** manheim has quit IRC | 20:29 | |
| *** esberglu has joined #openstack-ansible | 20:30 | |
| *** foutatoro has joined #openstack-ansible | 20:33 | |
| *** agrebennikov has quit IRC | 20:38 | |
| foutatoro | hi all, I would like to consume openstack api with curl from remote server. | 20:48 |
| foutatoro | but I get "[Errno 1] _ssl.c:510: error:14090086:SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed" | 20:48 |
| foutatoro | which certs should be exported with OS vars ? | 20:48 |
| *** deadnull has joined #openstack-ansible | 20:52 | |
| *** gouthamr has joined #openstack-ansible | 20:53 | |
| *** mamitchl has joined #openstack-ansible | 20:55 | |
| *** vnogin has quit IRC | 20:56 | |
| *** vnogin has joined #openstack-ansible | 20:59 | |
| foutatoro | any suggestion ? | 21:02 |
| *** jamesdenton has quit IRC | 21:02 | |
| foutatoro | cloudnull | 21:02 |
| *** chas has joined #openstack-ansible | 21:09 | |
| *** chas has quit IRC | 21:14 | |
| *** esberglu has quit IRC | 21:19 | |
| *** thorst has quit IRC | 21:22 | |
| *** gouthamr has quit IRC | 21:28 | |
| *** mamitchl has quit IRC | 21:40 | |
| *** SamYaple has quit IRC | 22:01 | |
| *** SamYaple has joined #openstack-ansible | 22:01 | |
| *** pbandark has quit IRC | 22:05 | |
| *** SamYaple has quit IRC | 22:06 | |
| *** julian1 has joined #openstack-ansible | 22:06 | |
| *** SamYaple has joined #openstack-ansible | 22:06 | |
| *** esberglu has joined #openstack-ansible | 22:07 | |
| *** strattao has quit IRC | 22:08 | |
| *** strattao has joined #openstack-ansible | 22:09 | |
| cloudnull | foutatoro: is this using a self signed cert ? | 22:10 |
| cloudnull | maybe you need curl -k ... | 22:11 |
| *** esberglu has quit IRC | 22:12 | |
| foutatoro | cloudnull: I can exported OS_CACERT var but I don't now where to find the according *.pem | 22:14 |
| cloudnull | for the lb that should be under /etc/ssl/private/haproxy.pem | 22:15 |
| foutatoro | cloudnull: you mean in horizon container ? | 22:21 |
| cloudnull | that should be wherever haproxy is running | 22:22 |
| *** thorst has joined #openstack-ansible | 22:22 | |
| cloudnull | I assume you're curl'ing to keystone via the external vip address ? | 22:22 |
| foutatoro | cloud: yes | 22:23 |
| foutatoro | there is no haproxy container. I'm running AIO | 22:23 |
| *** schwicht has quit IRC | 22:24 | |
| *** manheim has joined #openstack-ansible | 22:26 | |
| *** esberglu has joined #openstack-ansible | 22:28 | |
| *** vnogin has quit IRC | 22:30 | |
| cloudnull | haproxy will be on the aio physical host | 22:30 |
| *** manheim has quit IRC | 22:30 | |
| *** manheim has joined #openstack-ansible | 22:31 | |
| *** julian1 has quit IRC | 22:32 | |
| *** foutatoro has quit IRC | 22:39 | |
| *** thorst has quit IRC | 22:41 | |
| *** rmelero has quit IRC | 22:46 | |
| *** Hosam has joined #openstack-ansible | 22:59 | |
| *** mamitchl has joined #openstack-ansible | 23:02 | |
| *** manheim has quit IRC | 23:03 | |
| *** esberglu has quit IRC | 23:03 | |
| *** chas has joined #openstack-ansible | 23:09 | |
| DimGR | hey | 23:12 |
| *** chas has quit IRC | 23:14 | |
| *** Hosam has quit IRC | 23:17 | |
| *** Hosam has joined #openstack-ansible | 23:29 | |
| *** chyka has quit IRC | 23:29 | |
| *** chyka has joined #openstack-ansible | 23:30 | |
| *** Hosam has quit IRC | 23:32 | |
| *** Hosam has joined #openstack-ansible | 23:33 | |
| *** esberglu has joined #openstack-ansible | 23:34 | |
| *** chyka has quit IRC | 23:34 | |
| *** esberglu has quit IRC | 23:38 | |
| *** thorst has joined #openstack-ansible | 23:39 | |
| *** thorst has quit IRC | 23:43 | |
| *** thorst has joined #openstack-ansible | 23:56 | |
| *** esberglu has joined #openstack-ansible | 23:56 | |
| *** thorst has quit IRC | 23:57 | |
| *** askb has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!