| *** sacharya has joined #openstack-ansible | 00:33 | |
| *** alkari has quit IRC | 01:04 | |
| *** sacharya has quit IRC | 01:13 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Use paramiko transport for gate check jobs https://review.openstack.org/248361 | 01:24 |
|---|---|---|
| *** karimb has quit IRC | 01:36 | |
| *** cfarquhar_ has quit IRC | 01:43 | |
| *** cfarquhar has joined #openstack-ansible | 01:45 | |
| *** cfarquhar has quit IRC | 01:45 | |
| *** cfarquhar has joined #openstack-ansible | 01:45 | |
| *** galstrom_zzz is now known as galstrom | 02:04 | |
| *** galstrom is now known as galstrom_zzz | 02:05 | |
| *** baker has joined #openstack-ansible | 02:16 | |
| *** baker has quit IRC | 02:41 | |
| *** sdake has joined #openstack-ansible | 02:41 | |
| *** sdake_ has joined #openstack-ansible | 02:47 | |
| *** sdake has quit IRC | 02:48 | |
| *** sdake_ has quit IRC | 03:05 | |
| *** phschwartz has quit IRC | 03:16 | |
| *** phschwartz has joined #openstack-ansible | 03:16 | |
| *** baker has joined #openstack-ansible | 03:46 | |
| *** gouthamr has quit IRC | 04:11 | |
| *** agireud has quit IRC | 04:18 | |
| *** agireud has joined #openstack-ansible | 04:20 | |
| *** sacharya has joined #openstack-ansible | 04:24 | |
| *** agireud has quit IRC | 04:25 | |
| *** galstrom_zzz is now known as galstrom | 04:33 | |
| *** agireud has joined #openstack-ansible | 04:46 | |
| *** baker has quit IRC | 04:55 | |
| *** baker has joined #openstack-ansible | 05:02 | |
| *** alkari has joined #openstack-ansible | 05:10 | |
| *** galstrom is now known as galstrom_zzz | 05:13 | |
| *** baker has quit IRC | 05:27 | |
| *** galstrom_zzz is now known as galstrom | 05:33 | |
| openstackgerrit | Matthew Oliver proposed openstack/openstack-ansible: Add statsd configuration for swift https://review.openstack.org/246182 | 05:35 |
| openstackgerrit | Matthew Oliver proposed openstack/openstack-ansible: Add container-reconciler and object-expirer to os-swift https://review.openstack.org/247290 | 05:35 |
| openstackgerrit | Merged openstack/openstack-ansible: Update neutron service handler to restart only enabled services https://review.openstack.org/245825 | 05:39 |
| *** galstrom is now known as galstrom_zzz | 05:46 | |
| *** eil397 has quit IRC | 05:51 | |
| *** mss has joined #openstack-ansible | 06:05 | |
| *** sacharya has quit IRC | 06:12 | |
| *** sacharya has joined #openstack-ansible | 06:13 | |
| *** jhesketh has quit IRC | 06:31 | |
| *** phiche has joined #openstack-ansible | 06:32 | |
| *** jhesketh has joined #openstack-ansible | 06:34 | |
| *** sacharya has quit IRC | 06:36 | |
| *** fawadkhaliq has joined #openstack-ansible | 06:41 | |
| openstackgerrit | XiaBing Yao proposed openstack/openstack-ansible: autodetect neutron service_en https://review.openstack.org/248545 | 06:46 |
| *** eil397 has joined #openstack-ansible | 06:48 | |
| *** vCloudernBeer has joined #openstack-ansible | 06:51 | |
| vCloudernBeer | Trying to setup an AIO environment | 06:52 |
| vCloudernBeer | Looking at the document, I have git cloned optnstack-ansible | 06:53 |
| vCloudernBeer | In the document, the next step is to do a "git checkout kilo" | 06:53 |
| vCloudernBeer | This it switch to a new branch? namely Kilo? | 06:54 |
| *** vCloudernBeer has quit IRC | 06:59 | |
| *** shausy has joined #openstack-ansible | 07:01 | |
| *** phiche has quit IRC | 07:02 | |
| mcarden | vCloudernBeer... ask a question then immediately disappear? Tough to get an answer to you. | 07:02 |
| *** shausy has quit IRC | 07:02 | |
| *** shausy has joined #openstack-ansible | 07:08 | |
| *** openstack has joined #openstack-ansible | 07:21 | |
| *** phiche has joined #openstack-ansible | 07:23 | |
| *** mss has quit IRC | 07:26 | |
| *** javeriak has joined #openstack-ansible | 07:31 | |
| *** javeriak has quit IRC | 07:35 | |
| *** javeriak has joined #openstack-ansible | 07:35 | |
| *** sacharya has joined #openstack-ansible | 07:37 | |
| *** sdake has joined #openstack-ansible | 07:38 | |
| *** javeriak_ has joined #openstack-ansible | 07:40 | |
| *** javeriak has quit IRC | 07:41 | |
| *** sacharya has quit IRC | 07:41 | |
| *** dolpher1 has joined #openstack-ansible | 07:48 | |
| *** karimb has joined #openstack-ansible | 07:49 | |
| *** dolpher1 has quit IRC | 07:51 | |
| *** phiche1 has joined #openstack-ansible | 07:58 | |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: Update neutron service handler to restart only enabled services https://review.openstack.org/248557 | 07:59 |
| *** phiche has quit IRC | 08:00 | |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: Update neutron service handler to restart only enabled services https://review.openstack.org/248558 | 08:01 |
| evrardjp | hello everyone | 08:04 |
| evrardjp | just to jump on the conversation of SSH connection issues, we moved to paramiko for a while | 08:04 |
| evrardjp | for openstack-ansible | 08:04 |
| evrardjp | and it's far more reliable | 08:04 |
| evrardjp | it's still not 100% reliable, and is really a little slower | 08:05 |
| evrardjp | but we felt it was better | 08:05 |
| evrardjp | we didn't move back to another connection plugin | 08:06 |
| *** fawadkhaliq has quit IRC | 08:11 | |
| *** fawadkhaliq has joined #openstack-ansible | 08:12 | |
| *** mss has joined #openstack-ansible | 08:22 | |
| iceyao | evrardjp: good job. | 08:25 |
| evrardjp | ? | 08:26 |
| openstackgerrit | XiaBing Yao proposed openstack/openstack-ansible: Update some steps in quickstart-aio.rst https://review.openstack.org/248567 | 08:26 |
| evrardjp | it's not changed upstream, so if one want to change it's connection plugin, he needs to edit his ansible.cfg | 08:27 |
| evrardjp | I'd like to summarize what has been changed for the repo-* stuff, could you tell me if I'm wrong? | 08:35 |
| evrardjp | so for master/liberty/kilo, the work to simplify repo-build has been (or will be) merged | 08:36 |
| evrardjp | this creates (for people who want it) your own repository with the software you want | 08:36 |
| *** mpavone has joined #openstack-ansible | 08:36 | |
| evrardjp | the repo-clone thingy, is not a role, only a sync of the upstream repos, and its rework has been introduced in master/liberty | 08:37 |
| evrardjp | and will be reintroduced in kilo too | 08:38 |
| evrardjp | it's purpose is only to mirror an already built mirror (with the previous repo-build playbook) | 08:38 |
| evrardjp | am I right? | 08:38 |
| *** dolpher1 has joined #openstack-ansible | 08:44 | |
| *** xek has joined #openstack-ansible | 08:46 | |
| evrardjp | I'm planning to do a cleaning of the docs | 08:48 |
| *** gparaskevas has joined #openstack-ansible | 08:49 | |
| evrardjp | is there someone from the doc team here, I'd like to know if I can edit some png | 08:50 |
| evrardjp | so I'd be happy to have the initial svgs | 08:50 |
| *** dolpher1 is now known as dolpher | 08:51 | |
| *** tiagogomes has joined #openstack-ansible | 08:54 | |
| *** fawadkhaliq has quit IRC | 09:00 | |
| *** fawadkhaliq has joined #openstack-ansible | 09:01 | |
| odyssey4me | morning all | 09:03 |
| odyssey4me | welcome iceyao :) good to see you join us in IRC! | 09:03 |
| evrardjp | welcome :) | 09:07 |
| odyssey4me | javeriak_ if you have a chance, I'd love your input on https://review.openstack.org/248271 | 09:09 |
| javeriak_ | odyssey4me sure leme see | 09:09 |
| odyssey4me | javeriak_ also https://review.openstack.org/248545 which iceyao added | 09:10 |
| javeriak_ | wow we're getting lots of new additions lately, this is nice :) | 09:11 |
| evrardjp | 248271 is nice :) | 09:14 |
| *** fawadkhaliq has quit IRC | 09:21 | |
| *** fawadkhaliq has joined #openstack-ansible | 09:22 | |
| *** mgoddard has joined #openstack-ansible | 09:23 | |
| evrardjp | 248545 is also a nice addition | 09:24 |
| odyssey4me | evrardjp if you can put feedback into the reviews, that'd be best :) | 09:28 |
| *** mpavone has quit IRC | 09:29 | |
| evrardjp | oh I have started :p | 09:33 |
| evrardjp | it was a subtle way of saaying, maybe you should check on these :p | 09:33 |
| evrardjp | -a | 09:33 |
| evrardjp | now that I have fixed my networking issue for my testing, I'll have a more flexible testing system, so I'm gonna try this | 09:35 |
| javeriak_ | odyssey4me reviewed both, evrardjp had a comment on your comment :). btw don't we have a preference of bug/bp ref for all commits? | 09:36 |
| evrardjp | what do you mean javeriak_? | 09:37 |
| evrardjp | large changes often requires blueprints | 09:37 |
| javeriak_ | evrardjp, about the comment or the bp/bug question? | 09:37 |
| evrardjp | yup | 09:38 |
| javeriak_ | oh yes that, i just thought openstack communities had a rule/preference of having every commit tied to one of these | 09:38 |
| odyssey4me | ah javeriak_ well, I'm taking a more relaxed position on the submission of reviews - blueprints/specs are for large feature changes, bugs can be registered for bugs found... but if someone finds a bug or implements a small improvement then a bug is not a hard requirement | 09:39 |
| evrardjp | bugs should be linked to launchpad | 09:39 |
| odyssey4me | I don't like people registering bugs for the sake of submitting a patch - that creates a silliness in the bug management process and results in orphaned bugs | 09:40 |
| javeriak_ | okay cool, i wont have to file new things for stuff i mean to push shortly | 09:40 |
| odyssey4me | it's more important for me to have the patches move along than to stick to a bureaucratic process | 09:41 |
| javeriak_ | yea that makes perfect sense | 09:41 |
| *** mpavone has joined #openstack-ansible | 09:41 | |
| odyssey4me | javeriak_ sure - the general position is that if the core team feels that the patch warrants a bug or spec, then they can ask for it | 09:41 |
| javeriak_ | but i remember a point in this discussion being that it would be easier to track backports for stuff other people push in one branch and then forget | 09:42 |
| odyssey4me | javeriak_ yup, but that doesn't work when people register bugs for the purpose of implementing a patch because the bug doesn't get properly triaged | 09:43 |
| evrardjp | odyssey4me: I thought that registering a bug helps tracking the issues accross branches | 09:43 |
| odyssey4me | so I'm rather facilitating proactive and almost immediate backports once patches have merged | 09:43 |
| *** alkari has quit IRC | 09:44 | |
| evrardjp | but this makes sense, more action than paperwork. I like it | 09:44 |
| odyssey4me | evrardjp that was an original intent, but it has not been my experience so I've taken a different position... if someone wants to take up a release management role for the project and define a process we should all stick to, that's fine by me... but right now it's me, so I'm working the way that makes sense for me :) | 09:44 |
| evrardjp | you're right | 09:45 |
| *** mpavone has quit IRC | 09:45 | |
| javeriak_ | so i like what https://review.openstack.org/#/c/248545/1 is doing, and i mean to add something similar that adds a condition on what core plugin is selected, ie if plumgrid then you set all these agents to false, ill push something and then you guys can tell me what you think | 09:46 |
| javeriak_ | it'll make things easier to setup on our side | 09:47 |
| evrardjp | this makes sense | 09:50 |
| evrardjp | wouldn't it be better to completely override the service_en: for each agent you want to disable? | 09:51 |
| evrardjp | my point would be: if each agent dict has a specific configuration that works for only one plugin, then it shouldn't be globally set. | 09:53 |
| evrardjp | or we should maybe introduce a way to have the service_en automatically set based on a fact | 09:54 |
| javeriak_ | evrardjp yes thats sorta what i have in mind, havent started coding yet so lets see what the optimal solution looks like, i would however like to leave these service_en's open to other conditions as well instead of assiging them all a global value dependednt on one plugin | 09:55 |
| evrardjp | but yeah, it's not that important we can keep the auxiliary variable to have an easier override | 09:55 |
| javeriak_ | yea something like that | 09:55 |
| odyssey4me | javeriak_ evrardjp agreed - I'd like to steer clear of complex jinja logic in the variables if possible too - the simpler the logic is to follow, the better | 09:56 |
| *** sacharya has joined #openstack-ansible | 09:56 | |
| odyssey4me | but yes, if there are other variables we can base decisions on to automate decisions, then that's good - like the enabling of lbaas | 09:56 |
| evrardjp | javeriak_: your input will be important, because I have no clue how we deploy this differently... | 09:57 |
| odyssey4me | it makes it much, much simpler for the deployer when less things need to be changed to implement something properly | 09:57 |
| javeriak_ | evrardjp ofcourse, thats how we're doing it in our labs all the time :) | 09:57 |
| evrardjp | odyssey4me: +1 | 09:57 |
| *** mpavone has joined #openstack-ansible | 09:58 | |
| evrardjp | for me, moving to plumgrid should be as simple as flipping a switch | 09:58 |
| evrardjp | :p | 09:58 |
| javeriak_ | btw i dont think ive expressed my appreciation at how there is someone almost all the time on this commnity to talk to and answer questions. It can get so frustrating with a few others :), so great job guys! | 09:59 |
| *** sacharya has quit IRC | 10:01 | |
| odyssey4me | javeriak_ your help evolving the neutron role into something more pluggable has been invaluable - thank you! | 10:06 |
| evrardjp | indeed | 10:06 |
| evrardjp | javeriak_: so what you're doing at plumgrid is editing these neutron_l3, neutron_metering... variables to edit the conditions, or to set them as false, right? | 10:07 |
| javeriak_ | glad to be a part of this guys :), | 10:07 |
| evrardjp | I'm wondering if the |combine filter to merge dicts would be a good way to do it | 10:08 |
| javeriak_ | evrardjp yes we just set all of them to false, as plumgrid provides the whole networking layer, see here http://docs.openstack.org/developer/openstack-ansible/install-guide/app-plumgrid.html#neutron-configurations | 10:08 |
| evrardjp | ok | 10:11 |
| evrardjp | I'll have to check more deeply in the neutron role | 10:11 |
| javeriak_ | evrardjp hmm havnt really played with the combine filter yet, so cant say | 10:11 |
| evrardjp | it's not in ansible 1 :) | 10:12 |
| evrardjp | except if you bring it yourself, which is a few lines of python | 10:12 |
| evrardjp | but ansible2 is close | 10:12 |
| javeriak_ | yea i am very anxiously waiting for thst release, theres a lot of stuff we can clean up in our roles once that arrives, and hopefully with the supporting features that we're expecting | 10:14 |
| evrardjp | did you try the rc1? | 10:26 |
| evrardjp | it's out for 4 days | 10:26 |
| javeriak_ | nope not yet, have you? | 10:26 |
| evrardjp | for small fry stuff yes | 10:26 |
| evrardjp | I didn't try with all my plugins, etc. | 10:27 |
| javeriak_ | any idea on the main release eta? | 10:27 |
| evrardjp | this month I guess | 10:27 |
| javeriak_ | i havnt really been following | 10:27 |
| javeriak_ | ok great | 10:27 |
| evrardjp | I thought they were planning on delivering it during ansible fest | 10:27 |
| evrardjp | but they delivered an RC | 10:27 |
| evrardjp | so I guess it won't take much more time anymore | 10:28 |
| evrardjp | I'll give more news when I'll now :p | 10:28 |
| evrardjp | know* | 10:30 |
| javeriak_ | great thanks evrardjp | 10:30 |
| javeriak_ | gtg meetings, catch you guys later | 10:30 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fixed haproxy backend config issue https://review.openstack.org/245883 | 10:34 |
| *** javeriak_ has quit IRC | 10:34 | |
| odyssey4me | mattt hughsaunders we need another vote for https://review.openstack.org/246392 to move it along and get the focus for new patches into the right repositories, reviews would be appreciated | 10:42 |
| odyssey4me | tiagogomes ping? | 10:48 |
| tiagogomes | odyssey4me hello | 10:49 |
| odyssey4me | tiagogomes :) did you see cloudnull's message in https://review.openstack.org/238509 ? | 10:49 |
| odyssey4me | it'd be great if your patch could be updated based on the referred review | 10:49 |
| * tiagogomes reads | 10:52 | |
| odyssey4me | evrardjp what differences are you seeing? are they due to recently landed patches? if so, we need to add reviews ASAP to get those changes into master in the new repositories | 10:54 |
| evrardjp | I'll redo a diff, but it was cosmetic changes IIRC | 10:55 |
| evrardjp | my point was that we should have a clean diff | 10:56 |
| evrardjp | and improve afterwards | 10:56 |
| evrardjp | but let me completely check and I'll give you my +1 if it's ok or not | 10:56 |
| evrardjp | it's in my pipeline | 10:56 |
| hughsaunders | odyssey4me: I'll build it and check | 10:58 |
| tiagogomes | odyssey4me so you want me to rebase my patchsets, is that right? | 10:59 |
| odyssey4me | tiagogomes as far as I saw, your patch wasn't quite functional - it never passed the gate check. cloudnull's patch was... but you did implement some changes where he didn't... so if you could compare the changes you made to those he made, implement any differences and get it fully functional, that'd be great | 11:15 |
| *** javeriak has joined #openstack-ansible | 11:17 | |
| evrardjp | odyssey4me: comment added on https://review.openstack.org/#/c/246392/. TL;DR: No concern to have | 11:18 |
| odyssey4me | evrardjp great, thanks - the .DS_Store thing should be added to .gitignore - can you prep reviews to add that to each repo? | 11:19 |
| *** javeriak has quit IRC | 11:19 | |
| odyssey4me | and whichever repo has it should have it removed also :) | 11:19 |
| evrardjp | you told me to not add review on the IRR right now, am I allowed now? | 11:20 |
| evrardjp | :p | 11:20 |
| odyssey4me | evrardjp no, reviews for the other repositories are open - there is only a freeze for those roles in the main repo | 11:20 |
| odyssey4me | ie we don't want changes in the playbook repo for any roles which are split into seperate repo's, changes should only go into the new role repositories | 11:21 |
| evrardjp | before doing these changes on these new roles, we should define how to tag them to have a long term support (branching them + tagging them), but that's a topic for Thursday :D | 11:23 |
| evrardjp | (just to keep it in mind) | 11:23 |
| evrardjp | may I also add a .gitreview file? | 11:26 |
| odyssey4me | evrardjp do the changes to the roles - for now the way we do things won't change much - when we release mitaka the role repositories will get branches, but what will be different is that each role can tag independently | 11:26 |
| odyssey4me | evrardjp yeah, good idea | 11:26 |
| evrardjp | I'll put them in their appropriate project | 11:27 |
| *** fawadkhaliq has quit IRC | 11:41 | |
| *** fawadkhaliq has joined #openstack-ansible | 11:41 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-apt_package_pinning: New git dotfiles for independant repository https://review.openstack.org/248632 | 11:50 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_container_create: New git dotfiles for independant repository https://review.openstack.org/248634 | 11:50 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_hosts: New git dotfiles for independant repository https://review.openstack.org/248635 | 11:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts: New git dotfiles for independant repository https://review.openstack.org/248636 | 11:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-pip_install: New git dotfiles for independant repository https://review.openstack.org/248637 | 11:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-py_from_git: New git dotfiles for independant repository https://review.openstack.org/248638 | 11:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rsyslog_client: New git dotfiles for independant repository https://review.openstack.org/248639 | 11:51 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_hosts: Removed undesired .DS_Store file https://review.openstack.org/248643 | 11:54 |
| *** sacharya has joined #openstack-ansible | 12:05 | |
| *** jaypipes has joined #openstack-ansible | 12:08 | |
| *** sacharya has quit IRC | 12:10 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add "httpchk /versions" for glance-api haproxy. https://review.openstack.org/248213 | 12:10 |
| iceyao | hello everyone | 12:10 |
| tiagogomes | odyssey4me I can't see why one passed and other failed in the gate. Pherpahs it would be better to abandon mine and use the other one instead | 12:16 |
| odyssey4me | o/ iceyao | 12:16 |
| odyssey4me | tiagogomes I'd prefer you to have a commit. cloudnull has enough of his own. :) | 12:16 |
| openstackgerrit | Merged openstack/openstack-ansible: Added logging for haproxy to rsyslog https://review.openstack.org/245046 | 12:21 |
| evrardjp | I dont' know anything about tox or its integration with jenkins (yet), what's best approach to learn? read tox docs first? or have someone to explain me the current status of integration of the IRR? | 12:27 |
| evrardjp | ok got it | 12:29 |
| tiagogomes | cloudnull, re https://review.openstack.org/#/c/238509/, I suggest using your patch instead :) | 12:36 |
| evrardjp | cloudnull: when you'll be back, I'd like to have a look at tox tests of the IRR with you, especially the one that effectively starts the playbook. | 12:38 |
| *** cristicalin has joined #openstack-ansible | 12:48 | |
| iceyao | tiagogomes: may be domain_id = 'default' cause jenkins failed , you should use 'Default'. | 12:51 |
| iceyao | I guess | 12:52 |
| *** woodard has joined #openstack-ansible | 12:53 | |
| tiagogomes | iceyao maybe, but I'd had thought that if I used 'Default', the openstack client would try to translate the name to the ID | 12:54 |
| odyssey4me | tiagogomes The openstack client has that functionality, but the keystone library does not. With the keystone library you have to be clear about whether you're passing a domain name or a domain ID. | 12:58 |
| tiagogomes | odyssey4me good point | 12:58 |
| mgariepy | good morning everyone ! | 13:03 |
| evrardjp | good morning mgariepy | 13:03 |
| mgariepy | how are you guys doing ? | 13:05 |
| *** baker has joined #openstack-ansible | 13:08 | |
| odyssey4me | mgariepy can you please backport https://review.openstack.org/248213 to the liberty & kilo branches? if you need some help getting that done, let me know | 13:09 |
| *** shausy has quit IRC | 13:10 | |
| mgariepy | odyssey4me, i guess I need to click the cherry-pick to button ? | 13:11 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible: Add "httpchk /versions" for glance-api haproxy. https://review.openstack.org/248688 | 13:11 |
| odyssey4me | mgariepy yep - do the cherry-pick button from the master patch for each branch | 13:12 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible: Add "httpchk /versions" for glance-api haproxy. https://review.openstack.org/248689 | 13:12 |
| mgariepy | simple enough hehe :) | 13:12 |
| odyssey4me | any volunteers to back port https://review.openstack.org/248318 ? | 13:13 |
| odyssey4me | thanks mgariepy :) | 13:13 |
| odyssey4me | evrardjp odd, if I do the tox test on a fresh cloud server it works fine | 13:14 |
| odyssey4me | although I did have to figure out how to do it | 13:14 |
| evrardjp | I have a few questions on it :) | 13:15 |
| evrardjp | first permissions, and second the envlist | 13:15 |
| evrardjp | or the other way around | 13:15 |
| evrardjp | like I said, I don't know much about it, and I'm learning | 13:16 |
| odyssey4me | evrardjp so to run a tox test I needed to clone the repo, change into the root of the dir, execute https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-aio.sh#L102-L115 (although I expect that only some of those packages are needed), install pip, then run 'tox -e ansible-functional' | 13:17 |
| evrardjp | yeah, that's what I guess | 13:18 |
| evrardjp | but is it normal that envlist doesn't mention ansible-functional ? | 13:18 |
| evrardjp | while in the same file, there is an ansible-functional environment | 13:19 |
| evrardjp | I'll try this on a server | 13:19 |
| *** baker has quit IRC | 13:20 | |
| odyssey4me | evrardjp no, that's a mistake :) | 13:24 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible: Remove duplicated Aodh host var https://review.openstack.org/248696 | 13:25 |
| evrardjp | I think this is the cause, the env is not created so the variables are incorrectly set... etc | 13:26 |
| evrardjp | it's maybe intentionnal to not create many environments, but then, we need to do some wiring I guess | 13:26 |
| evrardjp | anyway I'll test on my side first | 13:27 |
| odyssey4me | mgariepy thanks for the liberty backport on that one - the patch doesn't apply for kilo, so no more needed | 13:28 |
| mgariepy | yeah i saw that it doesn't apply to kilo ;) | 13:28 |
| *** gouthamr has joined #openstack-ansible | 13:28 | |
| *** jaypipes has quit IRC | 13:29 | |
| *** gparaskevas_ has joined #openstack-ansible | 13:30 | |
| *** gparaskevas has quit IRC | 13:32 | |
| *** jaypipes has joined #openstack-ansible | 13:32 | |
| *** dalees has quit IRC | 13:37 | |
| openstackgerrit | XiaBing Yao proposed openstack/openstack-ansible: Update some steps in quickstart-aio.rst https://review.openstack.org/248567 | 13:43 |
| *** dalees has joined #openstack-ansible | 13:49 | |
| *** antonym has joined #openstack-ansible | 13:51 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible-apt_package_pinning: Update run_tests to be more complete https://review.openstack.org/248707 | 13:51 |
| *** elgertam1 has quit IRC | 13:52 | |
| *** fawadkhaliq has quit IRC | 13:55 | |
| *** fawadkhaliq has joined #openstack-ansible | 13:55 | |
| odyssey4me | evrardjp I suspect that the actual issue is that the role presumes root access, but openstack-ci is not running the role with sudo access | 13:56 |
| evrardjp | that was my second point :p | 14:00 |
| evrardjp | we could use remote_user: root I think | 14:01 |
| evrardjp | oh you mean openstack-cli doesn't even have sudo? | 14:01 |
| evrardjp | ci* | 14:01 |
| *** greg_a has joined #openstack-ansible | 14:11 | |
| odyssey4me | evrardjp very possibly, looking into that | 14:13 |
| odyssey4me | may need to have a chat to pabelanger_ about whether we should rather not assume root, and how we should handle that | 14:13 |
| odyssey4me | FYI 12.0.1 has been tagged | 14:14 |
| odyssey4me | https://launchpad.net/openstack-ansible/liberty/12.0.1 | 14:14 |
| *** mpavone has quit IRC | 14:26 | |
| *** HimanshuGarg has joined #openstack-ansible | 14:27 | |
| *** mpavone has joined #openstack-ansible | 14:30 | |
| *** mgoddard_ has joined #openstack-ansible | 14:30 | |
| *** mgoddard has quit IRC | 14:33 | |
| *** jasondotstar_afk is now known as jasondotstar | 14:33 | |
| *** gouthamr has left #openstack-ansible | 14:36 | |
| evrardjp | nice! | 14:36 |
| evrardjp | congrats everyone | 14:36 |
| evrardjp | when I'll be able to move to work (cf. what's in the news), I'll be happy to deploy that in my pre-prod lab! | 14:37 |
| evrardjp | btw I had a weird issue with liberty AIO, galera was behaving weirdly: by default only one mysqld (1/3) was running, and service mysql start was failing. starting the mysql daemons with only 'mysql' was working perfectly fine 'though | 14:40 |
| *** alextricity_h has joined #openstack-ansible | 14:41 | |
| odyssey4me | evrardjp liberty has mariadb10, so things are a little different | 14:43 |
| odyssey4me | but that shouldn't happen, although I have seen that behaviour with both sometimes | 14:43 |
| *** spotz_zzz is now known as spotz | 14:44 | |
| evrardjp | odyssey4me: I know about mariadb change, I got the chance to follow that part | 14:44 |
| *** javeriak has joined #openstack-ansible | 14:45 | |
| *** javeriak_ has joined #openstack-ansible | 14:46 | |
| *** javeriak has quit IRC | 14:46 | |
| *** alextricity_h has quit IRC | 14:49 | |
| *** javeriak has joined #openstack-ansible | 14:50 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Disable ssh pipelining for gate check jobs https://review.openstack.org/248739 | 14:50 |
| *** javeriak_ has quit IRC | 14:52 | |
| *** KLevenstein has joined #openstack-ansible | 14:57 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Disable ssh pipelining for gate check jobs https://review.openstack.org/248739 | 14:57 |
| *** subscope has joined #openstack-ansible | 14:58 | |
| *** greg_a has quit IRC | 15:00 | |
| evrardjp | odyssey4me: I think mhayden is trying to fix that by moving to paramiko | 15:01 |
| *** openstackgerrit has quit IRC | 15:02 | |
| odyssey4me | evrardjp yep, I'm doing this as a comparison | 15:02 |
| evrardjp | ok | 15:02 |
| odyssey4me | paramiko apparently is the most reliable, but also the slowest | 15:02 |
| *** openstackgerrit has joined #openstack-ansible | 15:02 | |
| odyssey4me | so I'd just like to see the comparison | 15:02 |
| evrardjp | it's not that slow anymore | 15:02 |
| evrardjp | http://www.ansible.com/blog/2014/01/15/ssh-connection-upgrades-coming-in-ansible-1-5 | 15:04 |
| pabelanger_ | morning | 15:04 |
| *** pabelanger_ is now known as pabelanger | 15:05 | |
| evrardjp | good morning pabelanger | 15:05 |
| pabelanger | odyssey4me: just reading backsroll | 15:05 |
| pabelanger | so, some questions about gate I see :) | 15:06 |
| evrardjp | apparently | 15:06 |
| odyssey4me | o/ pabelanger short and sweet - the role functional tests are hitting http://logs.openstack.org/07/248707/1/check/gate-openstack-ansible-apt_package_pinning-tox-ansible-functional/087b326/console.html#_2015-11-23_13_53_40_532 | 15:06 |
| evrardjp | we didn't define to run the playbooks as root, atm | 15:07 |
| evrardjp | but can we? | 15:07 |
| evrardjp | or should we use sudo? | 15:07 |
| odyssey4me | my suspicion is that the role tests are running without sudo access, or we need to specify the use of sudo? | 15:07 |
| pabelanger | odyssey4me: Ah. Ya, I know the issue | 15:07 |
| evrardjp | odyssey4me: atm they are running as user | 15:07 |
| pabelanger | the tox gate drops sudo support | 15:07 |
| evrardjp | pabelanger: so, what's the solution? | 15:08 |
| evrardjp | force using sudo or, force using root user? | 15:08 |
| pabelanger | so we should port you to | 15:08 |
| odyssey4me | pabelanger hmm, so do we add sudo support back - or do we change the functional tests to run without sudo? | 15:08 |
| pabelanger | use: https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/ansible-role-jobs.yaml#L41 | 15:08 |
| pabelanger | and make voting jobs | 15:08 |
| pabelanger | I created gate-ansible-role-nodepool-dsvm-functional-centos7 gate | 15:09 |
| pabelanger | and trusty game | 15:09 |
| pabelanger | gate | 15:09 |
| pabelanger | so I could support 2 OSs | 15:09 |
| odyssey4me | pabelanger we do intend to include other OSs later on (and are looking for contributors) :) | 15:10 |
| pabelanger | evrardjp: odyssey4me: we'd update your projects to use: https://github.com/openstack-infra/project-config/blob/master/zuul/layout.yaml#L3605 | 15:10 |
| odyssey4me | but for now we obviously don't have that | 15:10 |
| pabelanger | so, what OS are you supporting ATM? | 15:10 |
| odyssey4me | pabelanger just ubuntu trusty | 15:10 |
| pabelanger | okay | 15:11 |
| pabelanger | great | 15:11 |
| evrardjp | it's just a rename of jobs then | 15:11 |
| evrardjp | or am I wrong? | 15:11 |
| pabelanger | so, we can setup gate-{name}-dsvm-functional-trusty for you | 15:11 |
| pabelanger | ya, need to add proper voting job, but I can do that if you like | 15:11 |
| evrardjp | like I said I'm like John Snow on this zuul thingy. | 15:11 |
| *** HimanshuGarg has quit IRC | 15:12 | |
| evrardjp | maybe we should also wait for cloudnull to be here | 15:13 |
| pabelanger | WFM | 15:13 |
| odyssey4me | pabelanger so simply a change like 'gate-openstack-ansible-apt_package_pinning-tox-ansible-functional' -> 'gate-openstack-ansible-apt_package_pinning-dsvm-functional-trusty' in zuul/layout.yml ? | 15:13 |
| *** HimanshuGarg has joined #openstack-ansible | 15:14 | |
| pabelanger | odyssey4me: Ya, all your modules would need to change but that should work | 15:14 |
| evrardjp | that's what I thought | 15:14 |
| odyssey4me | ok cool - let me work up a patch quickly then | 15:14 |
| pabelanger | http://logs.openstack.org/98/246998/7/check/gate-ansible-dsvm-functional-trusty-nv/bc521c0/console.html | 15:14 |
| pabelanger | for an example | 15:14 |
| pabelanger | mind you I am using connection_local, but it does access sudo | 15:15 |
| evrardjp | we are using connection_local too | 15:15 |
| pabelanger | err | 15:15 |
| pabelanger | ansible_connection=local | 15:15 |
| evrardjp | we didn't set the use_sudo: True in the playbooks | 15:15 |
| pabelanger | but I want to change that so SSH shortly | 15:15 |
| pabelanger | I have sudo: True for mine | 15:16 |
| odyssey4me | pabelanger we're hitting all sorts of pain using ssh at the moment | 15:16 |
| pabelanger | should revisit that | 15:16 |
| evrardjp | odyssey4me: do you mind sharing the commit to make everything work? this way I understand more the relationship between tox runs and this gate jobs | 15:16 |
| evrardjp | these* | 15:16 |
| *** galstrom_zzz is now known as galstrom | 15:17 | |
| odyssey4me | evrardjp sure, will do - although this isn't really tox | 15:17 |
| evrardjp | no, this is gate jobs, right? | 15:17 |
| odyssey4me | tox creates a python venv to run all the tests so that the host isn't sullied with all the python bits | 15:17 |
| odyssey4me | tox can be used locally - so, for instance, you can run it on a mac | 15:17 |
| evrardjp | yeah I understood that, but these tox runs are started from somewhere | 15:17 |
| evrardjp | that's the glue missing for me | 15:17 |
| *** gouthamr has joined #openstack-ansible | 15:18 | |
| evrardjp | I started to use it today, it doesn't seem difficult to understand | 15:18 |
| cloudnull | morning | 15:18 |
| evrardjp | CI of openstack and on what all of these jobs are running is another story... that I'd like to understand too | 15:19 |
| evrardjp | morning cloudnull | 15:19 |
| *** grumpycatt has quit IRC | 15:19 | |
| odyssey4me | pabelanger evrardjp https://review.openstack.org/248757 <- is that it? | 15:22 |
| pabelanger | odyssey4me: It is right, but will actually fail. I need to patch something first | 15:22 |
| pabelanger | right now they are only -nv jobs | 15:22 |
| odyssey4me | pabelanger ok, thanks - let me know when you have that in and I'll add the review as a dependency | 15:23 |
| pabelanger | odyssey4me: this is needed too: http://paste.openstack.org/show/479737/ | 15:24 |
| pabelanger | odyssey4me: if you want to just append it | 15:24 |
| *** mss has quit IRC | 15:24 | |
| odyssey4me | pabelanger sure, I can do that | 15:25 |
| evrardjp | I'm editing the become thingy | 15:25 |
| pabelanger | odyssey4me: what sort of SSH issues are you having? | 15:26 |
| pabelanger | curious to hear the problems | 15:26 |
| *** alkari has joined #openstack-ansible | 15:26 | |
| odyssey4me | review updated pabelanger - please check | 15:27 |
| odyssey4me | thanks evrardjp | 15:27 |
| pabelanger | odyssey4me: that should be right. Waiting for gate to confirm | 15:28 |
| odyssey4me | pabelanger as an example, look at the check results for all the jobs in https://review.openstack.org/245883 | 15:28 |
| odyssey4me | hmm, let me find a more specific one | 15:29 |
| *** mgoddard has joined #openstack-ansible | 15:29 | |
| *** mgoddard_ has quit IRC | 15:29 | |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Pass domain to some calls in the keystone library https://review.openstack.org/238509 | 15:29 |
| pabelanger | Also, forgive me if I get this wrong. Is the majority of the people in openstack-ansible at rackspace? | 15:30 |
| *** eil397 has quit IRC | 15:31 | |
| *** javeriak has quit IRC | 15:32 | |
| odyssey4me | pabelanger for the moment, yes | 15:32 |
| evrardjp | I'm not | 15:32 |
| odyssey4me | there are several contributors and even more deployers who are not, but we're still working hard at getting more development contributors | 15:34 |
| pabelanger | /me nods | 15:34 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-apt_package_pinning: [WIP] Tentative to use sudo for role testing https://review.openstack.org/248774 | 15:35 |
| evrardjp | pabelanger: does this looks right to you ^ | 15:36 |
| *** gouthamr has left #openstack-ansible | 15:36 | |
| evrardjp | I wrote using the become method instead of sudo: http://docs.ansible.com/ansible/become.html | 15:36 |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Pass domain to some calls in the keystone library https://review.openstack.org/238509 | 15:36 |
| pabelanger | evrardjp: not sure, never used that code before | 15:36 |
| pabelanger | should read up on it and test | 15:37 |
| evrardjp | we should first wait on this: https://review.openstack.org/#/c/248757/ to pass | 15:37 |
| pabelanger | Still using the old way :) | 15:37 |
| *** elgertam1 has joined #openstack-ansible | 15:38 | |
| pabelanger | evrardjp: ya, I'll try for an -infra push | 15:38 |
| evrardjp | me too, but it's time to evolve :D | 15:38 |
| *** elgertam1 has quit IRC | 15:38 | |
| evrardjp | thanks pabelanger :) | 15:38 |
| *** galstrom is now known as galstrom_zzz | 15:39 | |
| evrardjp | cloudnull: what do you think of editing the tox file and adding ansible-functional to the envlist? | 15:39 |
| evrardjp | I could do that for you if you prefer | 15:40 |
| *** galstrom_zzz is now known as galstrom | 15:42 | |
| cloudnull | evrardjp: that sounds sensible | 15:44 |
| cloudnull | evrardjp: i added the ansible-functional bits to the break our roles i've done thus far https://github.com/openstack/openstack-ansible-lxc_hosts/blob/master/tox.ini | 15:46 |
| evrardjp | yeah I've seen that | 15:48 |
| evrardjp | I'll help you on these ones | 15:48 |
| evrardjp | moreover, when I'll understand the complete process, I'll move my roles to openstack-ansible stadion | 15:48 |
| evrardjp | big tent | 15:48 |
| evrardjp | I don't know the name | 15:49 |
| evrardjp | openstack-ansible herd? | 15:49 |
| cloudnull | ansible gaggle | 15:50 |
| cloudnull | :) | 15:50 |
| evrardjp | I just google translated it. Nice term | 15:51 |
| evrardjp | "a disorderly or noisy group of people." | 15:51 |
| cloudnull | yea | 15:52 |
| *** alkari has quit IRC | 15:52 | |
| evrardjp | did some of you mentionned that you were doing openstack horizon branding internally for RPC? | 15:56 |
| evrardjp | cause I have an embryonic role that could be improved, and (maybe better) be moved upstream | 15:56 |
| evrardjp | for the branding | 15:57 |
| evrardjp | if I check at your role, maybe I can find a decent base that will be helpful for everyone | 15:57 |
| evrardjp | the only thing I see is that: https://github.com/rcbops/rpc-openstack/blob/master/rpcd/playbooks/roles/horizon_extensions/tasks/main.yml#L16 | 15:59 |
| cloudnull | evrardjp: rpc is doing branding as well as a new tab , that role https://github.com/rcbops/rpc-openstack/tree/master/rpcd/playbooks/roles/horizon_extensions is essentially all we're doing on that front | 16:00 |
| evrardjp | ok | 16:01 |
| odyssey4me | evrardjp cloudnull a clowder? | 16:01 |
| *** jmvalente has joined #openstack-ansible | 16:02 | |
| *** galstrom is now known as galstrom_zzz | 16:04 | |
| cloudnull | Im +1 for herd or gaggle :) | 16:04 |
| *** fawadkhaliq has quit IRC | 16:04 | |
| *** javeriak has joined #openstack-ansible | 16:05 | |
| *** elo has joined #openstack-ansible | 16:07 | |
| *** karimb_ has joined #openstack-ansible | 16:10 | |
| evrardjp | ansible logo is bull right? so it's maybe better to have gaggle, just for the sake of being different :p | 16:11 |
| evrardjp | or not, whatever | 16:11 |
| evrardjp | I'll check on clowder, it looks like clown and sound like cloud | 16:12 |
| *** karimb has quit IRC | 16:13 | |
| evrardjp | maybe we should put a vote in the last part of Thursday meeting if there is one minute remaining | 16:13 |
| evrardjp | :p | 16:13 |
| *** Pablo|off| has joined #openstack-ansible | 16:14 | |
| *** karimb_ has quit IRC | 16:14 | |
| *** skamithi has joined #openstack-ansible | 16:15 | |
| *** Pablo|off| is now known as pcaruana | 16:15 | |
| odyssey4me | evrardjp as I recall we last discussed that it should be simple enough to add to the horizon role... the changes are so small that it doesn't feel like it needs another role | 16:16 |
| evrardjp | it's really two tasks | 16:16 |
| odyssey4me | yup | 16:17 |
| evrardjp | I have another role, but I find it really overkill | 16:17 |
| evrardjp | I'll update horizon it's really nothing | 16:17 |
| *** woodard has quit IRC | 16:19 | |
| *** galstrom_zzz is now known as galstrom | 16:20 | |
| *** cristicalin has quit IRC | 16:24 | |
| *** daneyon_ has joined #openstack-ansible | 16:24 | |
| *** mpavone has quit IRC | 16:24 | |
| *** daneyon has quit IRC | 16:26 | |
| *** alkari has joined #openstack-ansible | 16:30 | |
| *** mss has joined #openstack-ansible | 16:31 | |
| *** javeriak_ has joined #openstack-ansible | 16:35 | |
| *** javeriak has quit IRC | 16:37 | |
| *** javeriak has joined #openstack-ansible | 16:37 | |
| *** greg_a has joined #openstack-ansible | 16:38 | |
| *** greg_a has left #openstack-ansible | 16:39 | |
| *** javeriak_ has quit IRC | 16:40 | |
| *** fawadkhaliq has joined #openstack-ansible | 16:41 | |
| *** gparaskevas_ has quit IRC | 16:43 | |
| odyssey4me | evrardjp :) | 16:44 |
| *** baker has joined #openstack-ansible | 16:45 | |
| *** subscope has quit IRC | 16:50 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Branding of OpenStack https://review.openstack.org/248819 | 16:54 |
| logan- | i have a few things regarding horizon customization too, not sure if it would be useful to anyone: https://github.com/Logan2211/openstack-ansible/commits/horizon-customization | 16:55 |
| *** sacharya has joined #openstack-ansible | 16:55 | |
| *** woodard has joined #openstack-ansible | 16:59 | |
| *** mgoddard_ has joined #openstack-ansible | 17:00 | |
| *** sacharya has quit IRC | 17:00 | |
| evrardjp | ahah copying policies! | 17:00 |
| evrardjp | I have another thing for that | 17:00 |
| evrardjp | custom module too! | 17:01 |
| evrardjp | that's what I have done in the past... | 17:01 |
| evrardjp | it proves it's needed | 17:01 |
| evrardjp | I'll add documentation on this one: https://review.openstack.org/#/c/248819 | 17:02 |
| *** mgoddard has quit IRC | 17:03 | |
| logan- | yeah that is funny, site branding was next on the list for me :P | 17:04 |
| evrardjp | now it's done | 17:04 |
| evrardjp | for the policies, I'd rather create another playbook for that | 17:05 |
| logan- | yes thanks for that. a couple pieces of these would be easier to implement for operators if we had config_template style overrides on local_settings.py in the horizon role | 17:05 |
| evrardjp | and another role | 17:05 |
| evrardjp | yup, but I was lazy | 17:05 |
| evrardjp | :p | 17:05 |
| *** phiche1 has quit IRC | 17:07 | |
| odyssey4me | evrardjp logan- it would seem that copying policies should just use the config_template? | 17:10 |
| evrardjp | policies should be handled differently | 17:10 |
| logan- | yep the policy copy does, im not sure the implementation there is the correct way | 17:10 |
| logan- | it is symlinking to all of the separate roles policy templates and using their overrides if set | 17:11 |
| evrardjp | the problem of policies: they aren't the same within the component and horizon | 17:11 |
| logan- | yup | 17:11 |
| evrardjp | what I've implemented on my side, is a way to download the policy files for each component | 17:11 |
| evrardjp | and then a way to upload on horizon, about the same way as yours | 17:12 |
| evrardjp | this way you have time to adjust the mapping | 17:12 |
| logan- | ah | 17:12 |
| evrardjp | it's maybe not required for you | 17:13 |
| *** galstrom is now known as galstrom_zzz | 17:13 | |
| evrardjp | but with config_template is maybe possible to take the override dict from each component, and fill it with other values to override the override, and then send it to horizon | 17:14 |
| logan- | just curious.. what is the reason you would have differing policies on horizon vs. the service container? | 17:15 |
| evrardjp | because sometimes horizon just doesn't map with the same name (different cycles of deprecations) | 17:16 |
| logan- | ah ok | 17:16 |
| evrardjp | the other way would be to just use config_template for each component, and let the deployer edit its own dicts for that | 17:17 |
| evrardjp | and not care about the automatic way | 17:17 |
| evrardjp | it's maybe the best solution, because we could (later) combine override from a component into these variables | 17:18 |
| *** baker has quit IRC | 17:20 | |
| *** phiche has joined #openstack-ansible | 17:20 | |
| *** galstrom_zzz is now known as galstrom | 17:33 | |
| *** baker has joined #openstack-ansible | 17:34 | |
| *** rebase has joined #openstack-ansible | 17:35 | |
| *** tiagogomes has quit IRC | 17:43 | |
| *** baker has quit IRC | 17:50 | |
| *** egonzalez has joined #openstack-ansible | 17:54 | |
| odyssey4me | stevelle sigmavirus24_awa here's an odd one for you - if I set the glance {api,registry} workers to n, the number of processes running always seems to end up being n+1 | 17:55 |
| *** egonzalez is now known as egonzalez90 | 17:55 | |
| *** baker has joined #openstack-ansible | 17:55 | |
| palendae | odyssey4me: I think they're both out all week, fwiw | 17:55 |
| *** egonzalez90 has quit IRC | 17:55 | |
| *** galstrom is now known as galstrom_zzz | 17:56 | |
| jmccrory | odyssey4me: think the +1 is the parent, workers sets the number of child processes | 17:58 |
| odyssey4me | jmccrory oh? nova works differently though | 17:59 |
| odyssey4me | jmccrory but you're right - one is a parent of the other | 17:59 |
| jmccrory | possibly, remember running into the n+1 before with something but can't remember exactly which service it was | 17:59 |
| *** mgoddard_ has quit IRC | 18:01 | |
| *** mgoddard has joined #openstack-ansible | 18:01 | |
| logan- | http://paste.gentoolinux.info/oricidojax.hs | 18:01 |
| logan- | is what im seeing for nova api | 18:01 |
| *** egonzalez has joined #openstack-ansible | 18:02 | |
| *** daneyon has joined #openstack-ansible | 18:03 | |
| odyssey4me | logan- jmccrory quite right - I see now that nova has the same behaviour :) | 18:05 |
| odyssey4me | mhayden cloudnull fyi - I'm working on some tweaks to gate configs to reduce resource usage again... it seems that a lot of the gate fails relate to low resource availability again | 18:07 |
| *** daneyon_ has quit IRC | 18:07 | |
| *** baker has quit IRC | 18:10 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Allow Horizon Custom Branding https://review.openstack.org/248819 | 18:10 |
| *** sacharya has joined #openstack-ansible | 18:13 | |
| *** egonzalez has quit IRC | 18:16 | |
| *** eil397 has joined #openstack-ansible | 18:16 | |
| evrardjp | is it possible to merge stuff even with Jenkins giving a -1? | 18:17 |
| evrardjp | All these "New git dotfiles for independant repository" should be merged as soon as possible IMO | 18:17 |
| eil397 | as i know it possible | 18:17 |
| odyssey4me | evrardjp yes, but only if we merge something into the gate checks to change the checks done | 18:18 |
| evrardjp | I see :p | 18:19 |
| odyssey4me | evrardjp the fix for the gate is waiting for one more +2 | 18:19 |
| evrardjp | yeah sorry | 18:19 |
| evrardjp | anyway I shouldn't continue to work for today :p | 18:19 |
| *** rromans_ is now known as rromans | 18:22 | |
| *** fawadkhaliq has quit IRC | 18:24 | |
| *** sacharya has quit IRC | 18:24 | |
| *** fawadkhaliq has joined #openstack-ansible | 18:24 | |
| cloudnull | have a good one evrardjp | 18:25 |
| evrardjp | I'll follow these gate changes, this way I can adapt the (I hope soon) working solution for all the independant roles | 18:26 |
| evrardjp | bye | 18:26 |
| cloudnull | odyssey4me: do we have an OSA dashboard setup for the new repos ? | 18:27 |
| odyssey4me | cloudnull see the link list on https://launchpad.net/openstack-ansible | 18:27 |
| odyssey4me | the bottom two links | 18:27 |
| *** fawadkhaliq has quit IRC | 18:28 | |
| *** fawadkhaliq has joined #openstack-ansible | 18:29 | |
| *** fawadkhaliq has quit IRC | 18:29 | |
| kmARC | Hi all again! I have problems with the repo container deployment. The error message says that one or two containers are unreachable. | 18:31 |
| kmARC | The problem stems from - I guess - that there are for example DUP answers from ping | 18:32 |
| kmARC | now I use different networks for mgmt, vxlan and storage | 18:32 |
| kmARC | on a /23 network | 18:32 |
| odyssey4me | kmARC oh? we're hitting the same problem fairly frequently in the gate checks so we're trying to figure out why | 18:32 |
| kmARC | and blacklisted the hosts | 18:32 |
| kmARC | Oh thank God | 18:33 |
| kmARC | Soooo.... any workarounds? :) | 18:33 |
| odyssey4me | kmARC well, retry is one - but that doesn't work for the gate :/ | 18:33 |
| odyssey4me | kmARC another attempt is to use 'transport = paramiko' in ansible.cfg | 18:33 |
| odyssey4me | that switches how connections are setup and torn down, and seems to be reliable | 18:34 |
| cloudnull | kmARC odyssey4me: https://review.openstack.org/#/c/248361/ | 18:34 |
| odyssey4me | yep, I'm trying this too: https://review.openstack.org/248739 | 18:34 |
| odyssey4me | but hitting resource issues causing failures - so busy working on that now | 18:34 |
| cloudnull | just as a note, the nv job seems to be passing with more regularity than dsvm | 18:35 |
| * cloudnull lunching | 18:35 | |
| *** _hanhart has joined #openstack-ansible | 18:36 | |
| odyssey4me | cloudnull oh nice... if that's the case then there may be a package diff... although it could also be something to do with the hosts they run on :/ | 18:36 |
| kmARC | so which section should I define the paramiko transport? | 18:36 |
| kmARC | [ssh_connection]? (I guess) | 18:36 |
| odyssey4me | kmARC in defaults | 18:37 |
| kmARC | OK now I see in the git review | 18:44 |
| kmARC | I teared down, now let's see | 18:44 |
| *** _hanhart has quit IRC | 18:49 | |
| *** alkari has quit IRC | 18:52 | |
| *** cemmason1 has joined #openstack-ansible | 18:53 | |
| *** baker has joined #openstack-ansible | 19:00 | |
| mgariepy | quick question for git experts ;), i'm on tag 11.2.3 which i checkedout with : git checkout -b kilo-11.2.3 11.2.3, on top of that i pull some patches (cherry-pick), is there a simple way to rebase to update to tag 11.2.5 ? | 19:14 |
| *** baker has quit IRC | 19:15 | |
| mgariepy | without loosing my patches | 19:15 |
| *** ashishb has joined #openstack-ansible | 19:23 | |
| *** gouthamr has joined #openstack-ansible | 19:35 | |
| *** Bjoern_ has joined #openstack-ansible | 19:35 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Optimised the workers used for all services for the AIO https://review.openstack.org/248887 | 19:36 |
| *** jmvalente has quit IRC | 19:36 | |
| *** Bjoern_ is now known as BjoernT | 19:36 | |
| cloudnull | mgariepy: you should be able to do a `git rebase -i gerrit/11.2.5` which will take you step by step through the rebase. | 19:38 |
| *** baker has joined #openstack-ansible | 19:38 | |
| cloudnull | I'd recommend commiting your patches and making a backup | 19:38 |
| cloudnull | also if your patches have been picked clean you can rebase the kilo branch and then apply your patch branch on top | 19:39 |
| *** skamithi has quit IRC | 19:39 | |
| *** baker has quit IRC | 19:43 | |
| mgariepy | cloudnull, my patches are all commited and most of them are cherry-pick from master or kilo branches. | 19:45 |
| cloudnull | so it should cleanly apply | 19:46 |
| cloudnull | you may have a conflict but rebasing your feature branch with the tag you want to go to should be all thats needed. | 19:47 |
| kmARC | odyssey4me: I still get the DUP pings with transport = paramiko | 19:47 |
| cloudnull | kmARC: is it always to the first container in the build process ? | 19:48 |
| cloudnull | and does a rerun make it happy ? | 19:48 |
| *** greg_a has joined #openstack-ansible | 19:48 | |
| cloudnull | or is it intermitant ? | 19:48 |
| cloudnull | kmARC: sorry if im asking questions that have already been aswered . | 19:48 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Keystone domain fix https://review.openstack.org/238509 | 19:52 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Optimise the workers used for all services for the AIO https://review.openstack.org/248887 | 19:53 |
| *** neillc_ is now known as neillc | 19:56 | |
| *** mss_ has joined #openstack-ansible | 20:04 | |
| *** mss has quit IRC | 20:07 | |
| kmARC | oh sorry was out | 20:11 |
| kmARC | yes, the first dies immediately | 20:11 |
| kmARC | in setup-infrastructure | 20:11 |
| kmARC | if I retry with the suggested --limit, I get the same. Authentication Failed | 20:12 |
| kmARC | and still DUP ping | 20:12 |
| cloudnull | thats an ssh "authentication failed"? | 20:15 |
| *** egonzalez has joined #openstack-ansible | 20:17 | |
| *** ashishb has quit IRC | 20:18 | |
| kmARC | yes | 20:18 |
| kmARC | I guess | 20:18 |
| *** egonzalez has quit IRC | 20:18 | |
| kmARC | yes it is | 20:18 |
| odyssey4me | kmARC and can you ssh to the container? | 20:19 |
| kmARC | of course :) | 20:19 |
| odyssey4me | kmARC are you running as a non root user? | 20:19 |
| kmARC | nope, everything is root | 20:19 |
| kmARC | this worked when I wasn't using paramiko, tho | 20:20 |
| *** tlian has joined #openstack-ansible | 20:21 | |
| *** tlian has quit IRC | 20:23 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Update Liberty SHA's - 23.11.2015 https://review.openstack.org/245964 | 20:25 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Update Liberty SHA's - 23.11.2015 https://review.openstack.org/245964 | 20:26 |
| odyssey4me | kmARC try removing the paramico line, but setting ssh_pipelining to False? | 20:27 |
| kmARC | ok, now it didn't stop after the first container | 20:29 |
| kmARC | but I still have the DUPs | 20:30 |
| kmARC | should I tear down first? | 20:30 |
| odyssey4me | I'm not sure if the DUPs are bad or what may cause them. Perhaps rackertom or Apsu can help you figure that out? | 20:31 |
| *** tlian has joined #openstack-ansible | 20:31 | |
| * Apsu perks up | 20:32 | |
| Apsu | What's up? | 20:32 |
| rackertom | Hello | 20:32 |
| rackertom | DUPE ping responses? | 20:32 |
| Apsu | Network Rangers, deploy | 20:32 |
| *** greg_a has quit IRC | 20:32 | |
| odyssey4me | yeah, kmARC appears to be getting duplicate ping responses - from containers, I believe | 20:33 |
| Apsu | kmARC: If you're getting duplicate ping responses when pinging a container IP, you probably have (wait for it)... duplicate IPs. :D | 20:33 |
| Apsu | Easy way to tell is run a tcpdump and check the MAC addresses sending the responses. | 20:33 |
| Apsu | tcpdump -i br-mgmt (or whatever) -npevl icmp | 20:33 |
| odyssey4me | argh, with all the teardowns, etc that might actually be possible | 20:34 |
| Apsu | Background that, do your ping, and look at the MACs in the ICMP Echo Reply | 20:34 |
| Apsu | Then, if you're feeling enterprising... you can figure out which host veth the offending MACs are coming in on with brctl showmacs br-mgmt | grep $MAC | 20:35 |
| Apsu | It'll show the port | 20:35 |
| Apsu | If you're using recent enough OSA that your veths are named after the containers they're hooked to... yay | 20:35 |
| Apsu | If not, life gets harder. | 20:35 |
| rackertom | Apsu: Could possibly see the dupe entries in `ip neigh show`, no? | 20:35 |
| *** KLevenstein has quit IRC | 20:35 | |
| Apsu | rackertom: It would rotate, after each ARP | 20:36 |
| Apsu | Flapping like crazy, another good sign you've got duplicate IPs | 20:36 |
| rackertom | Yeah, I meant as you were performing a ping, you should see the MAC change in the output. Not as positive a detection as tcpdump, but people may have varying skill levels and comfort with tcpdump :) | 20:37 |
| Apsu | Sure | 20:37 |
| *** greg_a has joined #openstack-ansible | 20:38 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Fixes playbook runtime issues with ldap https://review.openstack.org/248900 | 20:40 |
| kmARC | Thanks Apsu, I really wanted not to tuch tcpdump :) But my problem is that I now have a /23 space. All the IP's assigned by me are blacklisted in the user.yaml. And after teardown and removing ansible_facts my networks are fine. Running setup-hosts screws it | 20:40 |
| Apsu | kmARC: After you tear everything down can you still ping the IP(s) in question? :P | 20:40 |
| kmARC | whatever happens, the playbooks should NOT set up any containters with clashing IPs | 20:40 |
| kmARC | Apsu, no, of cors not | 20:41 |
| Apsu | Well that's a good sign. Means the duplicate is not somewhere else on the network. | 20:41 |
| Apsu | Ok... next question... | 20:41 |
| kmARC | The fact is, I have a perfectly set up network. With big enough address space. Then I run the ansible scripts, and my network is screwed | 20:41 |
| SamYaple | bonding with an old kernel causes DUPs | 20:41 |
| Apsu | You setup your own bridges on your host(s), and are any of those bridges set to promiscuous mode? | 20:42 |
| Apsu | ip -o l | grep PROMISC | 20:42 |
| Apsu | If your bridges are in promiscuous mode, you can get duplicate ping responses without duplicate IPs | 20:42 |
| kmARC | no outpu | 20:42 |
| kmARC | I gues that means no they arent | 20:42 |
| Apsu | Indeed. Also a good sign. | 20:42 |
| Apsu | To be really clear, when you say "of course not", did you actually try pinging the IP to verify? :P | 20:43 |
| kmARC | ... yes? :) | 20:43 |
| Apsu | Just making sure. Because if the IP existed elsewhere on the network then no troubleshooting of containers would ever find it | 20:43 |
| Apsu | So we have to confirm that we can't ping it before deploying the containers, is all. | 20:44 |
| kmARC | no it didn't. Actually I configured a separate vlan and tried it too | 20:44 |
| Apsu | Cool. | 20:44 |
| Apsu | So definitely hiding in containers somehow. | 20:44 |
| Apsu | Either through broken bridging creating a loop or actual duplicate IPs. | 20:44 |
| kmARC | I guess so. Is there a convenient way to squeeze this info out from one of the lxc-* commands? | 20:44 |
| odyssey4me | ok, that's me out for the night - catch y'all tomorrow | 20:45 |
| Apsu | Well the correct way to assess the situation is to tcpdump in the same location you're pinging from, to see if you get different MAC addresses on replies. | 20:45 |
| Apsu | It's really not that scary. Just use the flags and parameters I gave above | 20:45 |
| kmARC | ah okay. let's learn how to do this then :) | 20:45 |
| kmARC | okay | 20:45 |
| Apsu | -i for interface, which is the bridge for the subnet you're pinging on. I assume br-mgmt | 20:46 |
| Apsu | -npevl for the other flags, and lastly 'icmp' for the protocol to listen for | 20:46 |
| kmARC | so basically when I ping from the infra-2-memcahced the deploy machine (and vice versa) I have the DUP syndrome | 20:46 |
| kmARC | okay | 20:46 |
| Apsu | Are you pinging on the management subnet? br-mgmt bridge IPs? | 20:47 |
| kmARC | yes | 20:47 |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Creating nova-secret with safe file permissions https://review.openstack.org/248904 | 20:47 |
| Apsu | Ok. Well, lxc-ls -f will show all the IPs assigned to all the containers | 20:47 |
| Apsu | So you can grep the output for the IP in question and see if it shows up twice | 20:48 |
| *** gouthamr has quit IRC | 20:49 | |
| Apsu | But as for tcpdump, use those flags, do your ping, and look at the ICMP Echo Reply output | 20:49 |
| Apsu | Near the beginning of each line will be the MAC addresses for source and destination | 20:49 |
| Apsu | In the replies, we care about the source MAC | 20:49 |
| Apsu | (the first one) | 20:50 |
| rackertom | Apsu: Can confirm, finding the error with ip neigh would be very difficult | 20:50 |
| Apsu | rackertom: Seems legit, but yeah, it's rough to pin down without logging of state changes | 20:50 |
| *** gouthamr has joined #openstack-ansible | 20:50 | |
| *** elo has quit IRC | 20:51 | |
| kmARC | hm. without pinging I can see this message in tcpdump's output: br-mgmt: received packet on eth1 with own add | 20:51 |
| kmARC | ress as source address | 20:51 |
| Apsu | Interesting | 20:51 |
| Apsu | Run: dmesg | tail | 20:51 |
| openstackgerrit | Miguel Alex Cantu proposed openstack/openstack-ansible: Add documentation for HA ceilometer https://review.openstack.org/248905 | 20:51 |
| Apsu | See if the last few lines from the kernel log are complaining loudly about network loops | 20:51 |
| kmARC | yeah, full of it | 20:51 |
| *** skamithi has joined #openstack-ansible | 20:52 | |
| Apsu | So you've got a loop! Neat. Now to figure out if it's software or hardware. | 20:52 |
| Apsu | First should find out what interfaces are in the br-mgmt bridge | 20:52 |
| *** javeriak has quit IRC | 20:52 | |
| rackertom | kmARC Apsu I think we found an issue a few months ago where if the address space was small enough, then we would get duplicate addresses anyway | 20:52 |
| rackertom | I think cloudnull saw it, too? | 20:52 |
| Apsu | You can either run ip -o | grep br-mgmt, or brctl show br-mgmt | 20:53 |
| kmARC | rackertom: I'm using a /23 is that small? | 20:53 |
| Apsu | Nah, that's fine | 20:53 |
| Apsu | Also that's not what's happening here | 20:53 |
| Apsu | tcpdump/the kernel complaining about "same address" isn't IP address | 20:53 |
| Apsu | It's MAC | 20:53 |
| Apsu | And that's a network loop :) | 20:53 |
| rackertom | a /23 should be just over 500 addresses | 20:53 |
| kmARC | well, 600 :) | 20:54 |
| Apsu | 510 | 20:54 |
| kmARC | Oh, sure | 20:54 |
| rackertom | yeah, 512-2 | 20:54 |
| Apsu | 2^n-2 | 20:54 |
| Apsu | kmARC: So, aside from a bunch of veth pairs, what else is in br-mgmt? | 20:54 |
| kmARC | nope | 20:54 |
| Apsu | Any physical interfaces? | 20:54 |
| kmARC | yes | 20:54 |
| kmARC | et1 | 20:54 |
| kmARC | eth1 | 20:54 |
| Apsu | Lets take eth1 out of br-mgmt for a moment and see if the dupes go away | 20:55 |
| Apsu | You can easily do so with ip link | 20:55 |
| Apsu | ip link set eth1 nomaster | 20:55 |
| kmARC | okay, I pullet that out | 20:55 |
| Apsu | Ping some of the containers local to this box | 20:56 |
| kmARC | yeah this works fine | 20:56 |
| Apsu | Cool. So the loop is external somehow | 20:56 |
| kmARC | oh wait | 20:56 |
| kmARC | I just looked at dmesg | 20:57 |
| kmARC | it says "dev eth1 left promisc mode" | 20:57 |
| kmARC | is this okay? | 20:57 |
| Apsu | Yes, that's ok | 20:57 |
| kmARC | okay that makes sense yes. It actually needs to be in promisc if my guess is right | 20:57 |
| *** KLevenstein has joined #openstack-ansible | 20:58 | |
| Apsu | Yeah it's fine. Links in bridges behave promiscuously | 20:58 |
| Apsu | So they can pass traffic to MACs not their own | 20:58 |
| Apsu | You just don't usually want the bridge Itself to behave promiscuously | 20:59 |
| Apsu | So, with eth1 out of br-mgmt, no more dupes? | 20:59 |
| kmARC | well, now I can't ping the deploy machine | 20:59 |
| kmARC | so can't tell. However in dmesg I cant see any suspicous messages either | 20:59 |
| Apsu | Ok | 20:59 |
| Apsu | Well we can put eth1 back in and take a look on the deploy machine | 21:00 |
| Apsu | Maybe the issue is there | 21:00 |
| Apsu | ip link set eth1 master br-mgmt | 21:00 |
| *** gouthamr has quit IRC | 21:00 | |
| Apsu | Is there a similar bridge on the deploy machine? | 21:00 |
| kmARC | yes | 21:00 |
| Apsu | I assume no containers there? | 21:00 |
| kmARC | nope | 21:00 |
| Apsu | Cool. Mind pasting a link to the output of "ip -o l"? | 21:01 |
| kmARC | so now I don't have the DUPs | 21:01 |
| Apsu | It won't include IPs so shouldn't have to sanitize | 21:01 |
| Apsu | Er... ok | 21:01 |
| kmARC | oh wait | 21:01 |
| kmARC | I was pinging from outside the container | 21:01 |
| kmARC | cool | 21:01 |
| kmARC | now I tried from inside the container | 21:02 |
| kmARC | now _ALL_ three "memcahced" containers have DUPs | 21:02 |
| Apsu | Let's try something | 21:02 |
| Apsu | Take eth1 back out, and ping one container from another | 21:02 |
| Apsu | That are on the same box | 21:02 |
| Apsu | Make sure that works without dupes | 21:02 |
| kmARC | ok | 21:03 |
| Apsu | Which will tend to rule out the bridge/veth/container setup | 21:03 |
| Apsu | I suspect it's on the deploy box or the switching inbetween | 21:03 |
| kmARC | So now I'm pinging from the memcahced container the utility container, and don't have dupes. eth1 is out | 21:04 |
| Apsu | Cool. | 21:04 |
| Apsu | Let's see the deploy host links :D | 21:07 |
| *** sdake_ has joined #openstack-ansible | 21:08 | |
| kmARC | http://paste.openstack.org/show/VCNGaPBsZnqxbxJ9ksoZ/ | 21:08 |
| *** sdake has quit IRC | 21:09 | |
| Apsu | No :) | 21:09 |
| Apsu | That's ifconfig. I need to see "ip -o link" | 21:09 |
| kmARC | http://paste.openstack.org/show/B3hh89S4PEHSJpI9Vtza/ | 21:09 |
| Apsu | There it is! | 21:10 |
| Apsu | Ok... so... good news is it looks good. | 21:10 |
| Apsu | Bad news is... it looks good. | 21:10 |
| Apsu | Might be a switch loop. Do you see errors in dmesg on this box too? | 21:11 |
| kmARC | br-mgmt: received packet on eth1 with own address as source address | 21:11 |
| kmARC | full of these | 21:11 |
| Apsu | Yep | 21:11 |
| Apsu | Pretty sure it's external to the boxes | 21:12 |
| *** jasondotstar has quit IRC | 21:12 | |
| kmARC | So these boxes are vm's on a vmware single hypervisor. No distributed switch, no nothing, only a vlan | 21:12 |
| kmARC | but now I'm thinking of putting these guys on a completely different, new vlan created by me | 21:13 |
| Apsu | That sounds like a good plan | 21:13 |
| Apsu | The VMware virtual switch the VMs are on can definitely have a loop created | 21:13 |
| Apsu | Assuming there's not a physical switch somewhere with a patch cable no one noticed ;P | 21:13 |
| kmARC | the problem is that I already tried putting the management interfaces on different vlan | 21:13 |
| *** rromans has quit IRC | 21:14 | |
| *** rromans has joined #openstack-ansible | 21:15 | |
| kmARC | Yeah, so now I double checked we have only one other machine on this hypervisor, and it's IP is not the dupe :) | 21:16 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Use paramiko transport for gate check jobs https://review.openstack.org/248361 | 21:16 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: AIO bootstrap in Ansible https://review.openstack.org/239525 | 21:18 |
| *** rromans has quit IRC | 21:18 | |
| *** jasondotstar has joined #openstack-ansible | 21:18 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Use Ansible AIO playbook for run-aio-build.sh https://review.openstack.org/244708 | 21:19 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Updating AIO docs for Ansible playbook https://review.openstack.org/244720 | 21:19 |
| Apsu | kmARC: Another good sign! | 21:19 |
| Apsu | But I don't think it's duplicate IPs | 21:20 |
| Apsu | I think it's duplicate MACs :) | 21:20 |
| kmARC | oh shit | 21:20 |
| mhayden | Apsu: just don't use a Mac | 21:21 |
| *** elgertam1 has joined #openstack-ansible | 21:21 | |
| Apsu | mhayden: GENIUS! | 21:22 |
| mhayden | Apsu: had to come in on my day off to heckle you at least once | 21:22 |
| Apsu | haha | 21:22 |
| mhayden | and it can't be a monday without one additional patch set for the AIO bootstrap review | 21:22 |
| Apsu | Day = complete | 21:22 |
| kmARC | hm. So for a second I thought when I cloned the machine's then the MAC's weren't replaced but this is not the case | 21:23 |
| Apsu | kmARC: Maybe the vmware MAC assignment overlaps our LXC container MAC assignment | 21:23 |
| *** skamithi has quit IRC | 21:23 | |
| kmARC | chances are...? :-) | 21:23 |
| Apsu | Hehe, that would explain a lot! | 21:23 |
| kmARC | I mean I retried like 5 times, with different settings, always removing the containers. | 21:24 |
| spotz | mhayden don't change it again I reviewed it:) | 21:24 |
| *** sdake_ has quit IRC | 21:42 | |
| *** sdake has joined #openstack-ansible | 21:42 | |
| *** tlian has quit IRC | 21:46 | |
| *** gouthamr has joined #openstack-ansible | 21:48 | |
| *** elo has joined #openstack-ansible | 21:48 | |
| *** woodard_ has joined #openstack-ansible | 21:51 | |
| *** elgertam1 has quit IRC | 21:51 | |
| *** alkari has joined #openstack-ansible | 21:52 | |
| *** elgertam1 has joined #openstack-ansible | 21:52 | |
| *** elgertam1 has quit IRC | 21:52 | |
| *** woodard has quit IRC | 21:55 | |
| *** jhesketh has quit IRC | 21:55 | |
| *** jhesketh has joined #openstack-ansible | 21:55 | |
| *** woodard_ has quit IRC | 21:56 | |
| *** rromans has joined #openstack-ansible | 21:57 | |
| *** rromans has quit IRC | 21:57 | |
| *** rromans has joined #openstack-ansible | 21:57 | |
| *** harlowja_ has joined #openstack-ansible | 21:59 | |
| *** harlowja has quit IRC | 21:59 | |
| *** cemmason1 has quit IRC | 22:00 | |
| *** cemmason has joined #openstack-ansible | 22:00 | |
| *** greg_a has quit IRC | 22:29 | |
| *** daneyon_ has joined #openstack-ansible | 22:36 | |
| *** phiche has quit IRC | 22:36 | |
| *** elo has quit IRC | 22:37 | |
| *** sdake has quit IRC | 22:37 | |
| *** daneyon has quit IRC | 22:39 | |
| *** sdake has joined #openstack-ansible | 22:39 | |
| *** rromans has quit IRC | 22:56 | |
| *** rromans has joined #openstack-ansible | 22:56 | |
| *** BjoernT has quit IRC | 22:57 | |
| *** HimanshuGarg has quit IRC | 23:07 | |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Creating nova-secret with safe file permissions https://review.openstack.org/248904 | 23:15 |
| *** jwagner is now known as jwagner_away | 23:16 | |
| *** galstrom_zzz is now known as galstrom | 23:27 | |
| *** alkari has quit IRC | 23:28 | |
| *** spotz is now known as spotz_zzz | 23:30 | |
| *** jhesketh has quit IRC | 23:31 | |
| *** josh6627 has joined #openstack-ansible | 23:31 | |
| *** cemmason has quit IRC | 23:31 | |
| *** skamithi13 has joined #openstack-ansible | 23:35 | |
| *** skamithi13 has quit IRC | 23:35 | |
| *** josh6627 has quit IRC | 23:35 | |
| *** skamithi13 has joined #openstack-ansible | 23:36 | |
| *** KLevenstein has quit IRC | 23:38 | |
| *** neillc is now known as neillc_away | 23:51 | |
| *** sacharya has joined #openstack-ansible | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!