*** weezS has joined #openstack-ansible | 00:06 | |
*** maeker has quit IRC | 00:07 | |
*** maeker has joined #openstack-ansible | 00:09 | |
*** maeker has quit IRC | 00:11 | |
*** maeker has joined #openstack-ansible | 00:12 | |
*** marst has quit IRC | 00:20 | |
*** Jeffrey4l_ is now known as Jeffrey4l | 00:28 | |
*** weezS has quit IRC | 00:28 | |
*** gouthamr has joined #openstack-ansible | 00:30 | |
*** maeker has quit IRC | 00:46 | |
*** whiteveil_ has joined #openstack-ansible | 00:51 | |
*** whiteveil has quit IRC | 00:51 | |
*** whiteveil_ is now known as whiteveil | 00:51 | |
*** jamielennox is now known as jamielennox|away | 00:57 | |
*** May-meimei has quit IRC | 00:59 | |
*** sdake_ has quit IRC | 01:00 | |
*** jamielennox|away is now known as jamielennox | 01:01 | |
*** asettle has joined #openstack-ansible | 01:08 | |
*** woodard has quit IRC | 01:10 | |
*** vnogin has quit IRC | 01:10 | |
*** woodard has joined #openstack-ansible | 01:10 | |
*** vnogin has joined #openstack-ansible | 01:11 | |
*** woodard has quit IRC | 01:11 | |
*** vnogin has quit IRC | 01:11 | |
*** woodard has joined #openstack-ansible | 01:11 | |
*** asettle has quit IRC | 01:13 | |
*** markvoelker has joined #openstack-ansible | 01:14 | |
*** sdake has joined #openstack-ansible | 01:14 | |
*** markvoelker has quit IRC | 01:19 | |
*** May-meimei has joined #openstack-ansible | 01:22 | |
*** messy has quit IRC | 01:30 | |
*** sdake has quit IRC | 01:35 | |
*** sdake has joined #openstack-ansible | 01:39 | |
*** jamielennox is now known as jamielennox|away | 01:41 | |
*** hw_wutianwei has joined #openstack-ansible | 01:44 | |
*** rmelero has quit IRC | 01:44 | |
*** rmelero has joined #openstack-ansible | 01:44 | |
*** jamielennox|away is now known as jamielennox | 01:47 | |
*** rmelero has quit IRC | 01:48 | |
*** v1k0d3n has quit IRC | 01:50 | |
*** sdake has quit IRC | 02:00 | |
*** Mahe has quit IRC | 02:04 | |
*** Mahe has joined #openstack-ansible | 02:05 | |
*** asettle has joined #openstack-ansible | 02:09 | |
*** Mudpuppy has quit IRC | 02:12 | |
*** vnogin has joined #openstack-ansible | 02:12 | |
*** Mudpuppy has joined #openstack-ansible | 02:12 | |
*** klamath has quit IRC | 02:13 | |
*** asettle has quit IRC | 02:13 | |
*** messy has joined #openstack-ansible | 02:16 | |
*** Mudpuppy has quit IRC | 02:17 | |
*** vnogin has quit IRC | 02:17 | |
*** sdake has joined #openstack-ansible | 02:22 | |
*** Jeffrey4l has quit IRC | 02:31 | |
*** Jeffrey4l has joined #openstack-ansible | 02:31 | |
*** markvoelker has joined #openstack-ansible | 02:31 | |
*** Jeffrey4l has quit IRC | 02:33 | |
*** Jeffrey4l has joined #openstack-ansible | 02:33 | |
*** markvoelker has quit IRC | 02:36 | |
*** sdake has quit IRC | 02:42 | |
*** sdake has joined #openstack-ansible | 02:43 | |
*** Mahe has quit IRC | 03:01 | |
*** sdake has quit IRC | 03:03 | |
*** Mahe has joined #openstack-ansible | 03:04 | |
*** ngupta has joined #openstack-ansible | 03:09 | |
*** ngupta has joined #openstack-ansible | 03:09 | |
*** gouthamr has quit IRC | 03:10 | |
*** asettle has joined #openstack-ansible | 03:10 | |
*** asettle has quit IRC | 03:14 | |
*** whiteveil has quit IRC | 03:15 | |
*** sdake has joined #openstack-ansible | 03:16 | |
*** sdake has quit IRC | 03:35 | |
*** sdake has joined #openstack-ansible | 03:36 | |
openstackgerrit | Kyle L. Henderson proposed openstack/openstack-ansible-repo_server: Fix apt-cacher-ng file owners during rsync https://review.openstack.org/410916 | 04:04 |
---|---|---|
*** dxiri has joined #openstack-ansible | 04:05 | |
*** dxiri has quit IRC | 04:09 | |
*** asettle has joined #openstack-ansible | 04:10 | |
*** maeker has joined #openstack-ansible | 04:12 | |
*** vnogin has joined #openstack-ansible | 04:13 | |
*** asettle has quit IRC | 04:15 | |
*** v1k0d3n has joined #openstack-ansible | 04:15 | |
*** vnogin has quit IRC | 04:18 | |
*** thorst has joined #openstack-ansible | 04:18 | |
*** ngupta has quit IRC | 04:34 | |
*** maeker has quit IRC | 04:34 | |
*** thorst has quit IRC | 04:45 | |
*** chhavi has joined #openstack-ansible | 04:46 | |
*** thorst has joined #openstack-ansible | 04:49 | |
*** thorst has quit IRC | 04:57 | |
*** qiliang27 has quit IRC | 04:57 | |
*** qiliang27 has joined #openstack-ansible | 04:58 | |
*** shausy has joined #openstack-ansible | 05:10 | |
*** asettle has joined #openstack-ansible | 05:11 | |
*** asettle has quit IRC | 05:15 | |
*** sdake_ has joined #openstack-ansible | 05:24 | |
*** sdake has quit IRC | 05:25 | |
*** v1k0d3n has quit IRC | 05:35 | |
*** ngupta has joined #openstack-ansible | 05:42 | |
*** thorst has joined #openstack-ansible | 05:44 | |
*** thorst has quit IRC | 05:52 | |
*** vnogin has joined #openstack-ansible | 06:15 | |
*** vnogin has quit IRC | 06:19 | |
*** jimbaker has quit IRC | 06:22 | |
*** jimbaker has joined #openstack-ansible | 06:26 | |
*** ngupta has quit IRC | 06:27 | |
*** ngupta has joined #openstack-ansible | 06:27 | |
*** Jack_Iv has joined #openstack-ansible | 06:29 | |
*** whiteveil has joined #openstack-ansible | 06:30 | |
*** ngupta has quit IRC | 06:31 | |
*** sdake_ has quit IRC | 06:32 | |
*** sdake has joined #openstack-ansible | 06:34 | |
*** thorst has joined #openstack-ansible | 06:39 | |
*** thorst has quit IRC | 06:46 | |
*** askb has quit IRC | 06:53 | |
*** Oku_OS-away is now known as Oku_OS | 06:55 | |
*** Mudpuppy has joined #openstack-ansible | 07:01 | |
*** fxpester has joined #openstack-ansible | 07:02 | |
*** sdake has quit IRC | 07:03 | |
*** Mudpuppy has quit IRC | 07:06 | |
*** asettle has joined #openstack-ansible | 07:12 | |
*** asettle has quit IRC | 07:17 | |
*** adreznec has quit IRC | 07:21 | |
*** karimb has joined #openstack-ansible | 07:36 | |
*** thorst has joined #openstack-ansible | 07:38 | |
*** thorst has quit IRC | 07:44 | |
*** karimb has quit IRC | 07:55 | |
*** pcaruana has joined #openstack-ansible | 07:56 | |
*** sacharya has quit IRC | 08:06 | |
*** jamielennox is now known as jamielennox|away | 08:07 | |
*** sacharya has joined #openstack-ansible | 08:08 | |
*** adreznec has joined #openstack-ansible | 08:09 | |
*** sacharya has quit IRC | 08:12 | |
*** jamielennox|away is now known as jamielennox | 08:14 | |
*** karimb has joined #openstack-ansible | 08:32 | |
*** karimb has quit IRC | 08:32 | |
*** asettle has joined #openstack-ansible | 08:34 | |
*** chhavi has quit IRC | 08:37 | |
*** asettle has quit IRC | 08:39 | |
*** thorst has joined #openstack-ansible | 08:39 | |
*** chhavi has joined #openstack-ansible | 08:43 | |
*** thorst has quit IRC | 08:44 | |
*** whiteveil has quit IRC | 08:58 | |
*** sacharya has joined #openstack-ansible | 09:09 | |
*** sacharya has quit IRC | 09:13 | |
andymccr | mornin | 09:25 |
winggundamth | morning | 09:29 |
*** ngupta has joined #openstack-ansible | 09:30 | |
*** asettle has joined #openstack-ansible | 09:33 | |
*** ngupta has quit IRC | 09:34 | |
*** qiliang27 has quit IRC | 09:35 | |
*** thorst has joined #openstack-ansible | 09:39 | |
*** thorst has quit IRC | 09:44 | |
*** jamielennox is now known as jamielennox|away | 09:48 | |
*** vnogin has joined #openstack-ansible | 09:49 | |
*** fxpester has quit IRC | 09:54 | |
*** jamielennox|away is now known as jamielennox | 09:55 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_searchlight: [WIP] Import initial os_searchlight role. https://review.openstack.org/404419 | 10:03 |
*** karimb has joined #openstack-ansible | 10:04 | |
*** kylek3h has quit IRC | 10:05 | |
*** chhavi has quit IRC | 10:09 | |
*** Jack_iv_ has joined #openstack-ansible | 10:12 | |
*** openstackgerrit has quit IRC | 10:18 | |
*** vnogin has quit IRC | 10:24 | |
*** vnogin has joined #openstack-ansible | 10:24 | |
*** openstackgerrit has joined #openstack-ansible | 10:26 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-tests: Remove Trusty support from tests repository https://review.openstack.org/411744 | 10:26 |
*** vnogin has quit IRC | 10:27 | |
*** ngupta has joined #openstack-ansible | 10:31 | |
*** uthng has joined #openstack-ansible | 10:32 | |
uthng | hi all | 10:32 |
uthng | anyone can tell me if I can install magnum on AIO ? maybe odyssey4me ? because when I tried to run os-magnum-install.yml, it told me that no host matched | 10:33 |
*** thorst has joined #openstack-ansible | 10:33 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_tempest: Fix tempest for os_horizon role https://review.openstack.org/411748 | 10:35 |
*** ngupta has quit IRC | 10:35 | |
*** v1k0d3n has joined #openstack-ansible | 10:36 | |
*** sliver has joined #openstack-ansible | 10:38 | |
*** stuartgr has joined #openstack-ansible | 10:39 | |
*** mcrafty has quit IRC | 10:40 | |
andymccr | uthng: you should have a magnum.yml file inside /etc/openstack_deploy/conf.d/magnum.yml and if so you should be able to run it | 10:41 |
*** thorst has quit IRC | 10:41 | |
*** v1k0d3n has quit IRC | 10:42 | |
uthng | andymccr: arff I dont have ! How is it possible ? | 10:43 |
uthng | I have only aodh.yml ceilometer.yml swift.yml ! | 10:43 |
uthng | after upgrade ! | 10:44 |
uthng | i have magnum.yml in env.d | 10:45 |
andymccr | uthng: if you did an upgrade perhaps you upgraded from a release that didnt have magnum - id recommend reading the docs on how to configure, but if its just an AIO you can copy over the AIO sample file from the repository (etc/openstack_deploy/conf.d/magnum.yml.aio) to your local etc directory (/etc/openstack_deploy/conf.d/magnum.yml) - also note to drop the .aio when you copy it over | 10:45 |
andymccr | and by didn't have magnum, i mean in the aio build | 10:45 |
*** kylek3h has joined #openstack-ansible | 10:47 | |
uthng | andymccr: thats right that I am doing upgrade from Mitaka. But I donot understand that why other component *.yml like cinder, glance etc. are not in conf.d ? They do not need ? | 10:48 |
*** philippe_ has joined #openstack-ansible | 10:49 | |
*** philippe_ is now known as lemouchon2 | 10:49 | |
andymccr | uthng: the way we setup the AIO was slightly different in mitaka - so all of those are most likely in a single /etc/openstack_deploy/openstack_user_config.yml | 10:53 |
andymccr | which will still work fine in newton/master etc | 10:54 |
uthng | andymccr: ah ok. after copying magnum.yml in conf.d, I have to rerun setup-hosts and lxc-containers-create.yml ? | 10:55 |
andymccr | uthng: if its an aio probably just lxc-containers-create.yml since it should be setup | 10:56 |
uthng | and do I have to rerun other install playbook for other components ? Becasue I already upgraded all other existing components such as cinder, leystone, glance etc. | 10:56 |
uthng | andymccr: ok, I will try it. Thanks for your helps | 10:57 |
andymccr | no problem! | 10:57 |
uthng | andymccr: ah It seems that the script user-secrets-adjustment.yml does not do correctly the work | 10:58 |
uthng | I think it save the existing user_secrets.yml and add new secrets for magnum or sarah etc. in it | 10:58 |
uthng | But finally it overrides existing secrets with the new generated ones ! | 10:59 |
andymccr | uthng: it should just rename some secrets that have changed, add in new secrets that didnt exist, and then generate secrets for the new ones | 11:00 |
uthng | I donot know if I misunderstood anything but I had to report all old passwords found in the components that are not upgraded yet | 11:00 |
*** shausy has quit IRC | 11:02 | |
uthng | thats why when I upgraded glance and cinder, they cannot authenticate anymore at Keystone | 11:03 |
*** bpetit has quit IRC | 11:04 | |
uthng | I had to use curl api, as odyssey4me told me, to set existing cinder or glance passwords in keystone to the new ones generated in user_secrets.yml | 11:04 |
*** bpetit has joined #openstack-ansible | 11:04 | |
andymccr | uthng: ok - im not 100% sure, i've not run into that error before - what odyssey4me told you would fix it so you can continue, but that isnt an expected behaviour, if its consistently happening and you could file a bug for it that'd be really good :) | 11:06 |
*** sacharya has joined #openstack-ansible | 11:06 | |
uthng | yes, just to talk to you before if you knew anything about that. I will try to run this script againts on another server. If it happens again, I will open a bug then. | 11:09 |
*** sacharya has quit IRC | 11:11 | |
*** hw_wutianwei has quit IRC | 11:13 | |
*** fxpester has joined #openstack-ansible | 11:17 | |
*** dgonzalez has quit IRC | 11:22 | |
*** dgonzalez has joined #openstack-ansible | 11:24 | |
*** thorst has joined #openstack-ansible | 11:29 | |
*** thorst has quit IRC | 11:36 | |
*** chhavi has joined #openstack-ansible | 11:40 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ceilometer: Add required packages for Ceilometer https://review.openstack.org/411775 | 12:01 |
openstackgerrit | Logan V proposed openstack/openstack-ansible: WIP: ceph-ansible integration https://review.openstack.org/409407 | 12:02 |
odyssey4me | uthng FYI for future purposes you may wish to check into using the openstack CLI instead of curl: http://docs.openstack.org/developer/python-openstackclient/command-objects/user.html#user-set | 12:06 |
odyssey4me | but the end result is the same :) | 12:06 |
*** Andrew_jedi has joined #openstack-ansible | 12:09 | |
Andrew_jedi | Hello Folks, Is there a command to find out all the overrides available? | 12:09 |
odyssey4me | Andrew_jedi each role's defaults/main.yml files are the documented overrides | 12:10 |
odyssey4me | some of which are used in group_vars in the integrated repo | 12:10 |
*** deadnull has joined #openstack-ansible | 12:13 | |
Andrew_jedi | odyssey4me: Ohh, ok. Could you please also let me know whether its possible to override this :https://github.com/openstack/openstack-ansible-os_heat/blob/master/tasks/heat_domain_setup.yml#L72 | 12:14 |
odyssey4me | Andrew_jedi It's not using a variable, so not at this time. | 12:15 |
odyssey4me | That does seem like an oversight. | 12:15 |
odyssey4me | although, the 'admin' role name is special - it's the super admin that allows changes to anything and everything | 12:16 |
odyssey4me | why would you want to set that to something different? | 12:16 |
odyssey4me | so it looks like the keystone role has 'keystone_role_name' as a var to set which defines the role name for the admin | 12:18 |
andymccr | odyssey4me: Andrew_jedi's moving from another platform that has admin as like "Admin" or something similar - so yeah we should just use that var | 12:18 |
odyssey4me | aha, makes sense | 12:18 |
andymccr | i said the same last time hahah :D | 12:18 |
odyssey4me | so the var 'keystone_role_name' was created quite recently - Newton I think | 12:18 |
andymccr | thats a bug we can backport that fix too | 12:18 |
odyssey4me | the heat role doesn't get much love - so I expect that we may need to survey the other roles and implement the use of that var if it's set | 12:19 |
odyssey4me | I can do a sample if you like? | 12:19 |
Andrew_jedi | odyssey4me andymccr : Yes, i am trying to migrate Juju db to OSA. And juju uses "Admin" role instead of "admin". | 12:19 |
odyssey4me | Andrew_jedi I hope you're keeping notes on your experience and will share them :) | 12:20 |
odyssey4me | ok, I'll put a quick patch together for os_heat | 12:20 |
Andrew_jedi | odyssey4me: Notes written in blood. | 12:20 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Allow admin role name to be changed https://review.openstack.org/411780 | 12:23 |
odyssey4me | andymccr ^ seems the simplest mechanism | 12:23 |
odyssey4me | we haven't really gone through all the roles to namespace all var usage, and to some degree I think that may be overcomplicating things | 12:23 |
Andrew_jedi | odyssey4me: Please correct me if i am wrong but if i want to change this https://github.com/openstack/openstack-ansible-os_heat/blob/master/defaults/main.yml#L100 then i have to use "heat_default_yaml_overrides" ? | 12:23 |
odyssey4me | Andrew_jedi The 'heat_service_role_name' variable? Nope - just add it to your user_variables.yml file with the value you want | 12:24 |
*** thorst has joined #openstack-ansible | 12:25 | |
Andrew_jedi | odyssey4me: oh ... so you mean any variable which is in roles/default/main.yam can be overridden by just adding it to user_variables.yaml ? | 12:26 |
Andrew_jedi | odyssey4me: Any chance i can use this patch in Liberty :P https://review.openstack.org/411780 | 12:27 |
odyssey4me | Andrew_jedi yes, that's correct | 12:27 |
odyssey4me | Andrew_jedi perhaps, it's a safe patch which just uses the var if it exists | 12:28 |
odyssey4me | Our liberty branch is gone though as it's EOL, so you'll have to fork and patch yourself. | 12:28 |
Andrew_jedi | odyssey4me: Gotcha! Thanks! | 12:29 |
odyssey4me | ok, I don't see any other roles which have a similar pattern | 12:30 |
odyssey4me | so it looks like that'll do it | 12:30 |
odyssey4me | if you find any other you need, ping us :) | 12:30 |
Andrew_jedi | odyssey4me: roger that :) | 12:31 |
openstackgerrit | Merged openstack/openstack-ansible-pip_install: Remove Trusty support from pip_install role https://review.openstack.org/411367 | 12:31 |
andymccr | odyssey4me: agree with that | 12:32 |
*** ngupta has joined #openstack-ansible | 12:32 | |
odyssey4me | andymccr I think we should actually work through the roles and implement similar defaulting for all those keystone vars and any other vars outside the role | 12:32 |
odyssey4me | hmm, let me register some bugs for that | 12:32 |
odyssey4me | it's a nice low-hanging-fruit item | 12:33 |
andymccr | odyssey4me: yeah i thinkso - but do the others not have it? | 12:33 |
andymccr | if Andrew_jedi only ran into this issue on heat im guessing the other major projects have it atleast | 12:33 |
*** thorst has quit IRC | 12:33 | |
odyssey4me | well, there is the use of keystone_* vars in almost all the service roles... those aren't present in the defaults and aren't defaulted either | 12:34 |
odyssey4me | it's not *necessary*, but it is a bit of tidying up that would be nice | 12:34 |
openstackgerrit | Merged openstack/openstack-ansible-plugins: Remove Trusty support from plugins role https://review.openstack.org/411376 | 12:36 |
*** ngupta has quit IRC | 12:36 | |
openstackgerrit | Merged openstack/openstack-ansible-rsyslog_client: Remove Trusty support from rsyslog_client role https://review.openstack.org/411386 | 12:37 |
openstackgerrit | Merged openstack/openstack-ansible-rsyslog_server: Remove Trusty support from rsyslog_server role https://review.openstack.org/411390 | 12:40 |
openstackgerrit | Merged openstack/openstack-ansible-repo_server: Remove Trusty Support from repo_server role https://review.openstack.org/411385 | 12:40 |
openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Remove Trusty support from os_magnum role https://review.openstack.org/411278 | 12:44 |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-repo_build: Add libxml2-dev package required for lxml install https://review.openstack.org/411787 | 12:44 |
*** markvoelker has joined #openstack-ansible | 12:45 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ceilometer: Add required packages for Ceilometer https://review.openstack.org/411775 | 12:45 |
odyssey4me | andymccr I see you're in the thick of things there. Need any help with anything? | 12:46 |
*** gouthamr has joined #openstack-ansible | 12:46 | |
andymccr | odyssey4me: think im done now! found out why the 3 are failing, and once the appropriate fixes merge it should be good to go | 12:47 |
andymccr | im tempted to set ceilometer to build in the integrated build, ignoring requirements | 12:47 |
andymccr | for now | 12:47 |
*** jamesdenton has joined #openstack-ansible | 12:47 | |
andymccr | so we can ensure it stays up to date and catch issues sooner | 12:47 |
andymccr | and then before actual release we remove that | 12:47 |
odyssey4me | andymccr sure, the trick will be to remember to do that! | 12:49 |
andymccr | haha yeah i know... | 12:49 |
andymccr | maybe i'll leave it a bit longer | 12:49 |
odyssey4me | you could create a blueprint for the Ocata release and add a note to it | 12:49 |
andymccr | there is working to fix that stuff | 12:49 |
*** thorst has joined #openstack-ansible | 12:49 | |
andymccr | there is work happening, rather :P | 12:49 |
odyssey4me | it's very likely that more and more projects will start doing this | 12:49 |
andymccr | ignoring reqs? | 12:49 |
odyssey4me | yeah, the whole co-installable thing is not something everyone subscribes to | 12:50 |
*** jamesden_ has joined #openstack-ansible | 12:50 | |
odyssey4me | but yeah, that is a current community contract which ceilometer really should be working within the community to resolve | 12:50 |
odyssey4me | I mean if monasca is the hold back, then perhaps monasca should solve it or step out of the co-installable promise | 12:51 |
*** jamesdenton has quit IRC | 12:52 | |
andymccr | odyssey4me: yeah it looks like oslo is gonna update and ceilometer will consume oslo rather than its own thing | 12:53 |
andymccr | but yeah | 12:53 |
openstackgerrit | Merged openstack/openstack-ansible-os_trove: Remove Trusty support from os_trove role https://review.openstack.org/411366 | 12:53 |
odyssey4me | ah ok nice | 12:54 |
odyssey4me | will they be able to achieve that in Ocata? | 12:54 |
openstackgerrit | Merged openstack/openstack-ansible-repo_build: Remove Trusty support from repo_build role https://review.openstack.org/411380 | 12:55 |
*** thorst has quit IRC | 12:56 | |
*** thorst has joined #openstack-ansible | 12:57 | |
andymccr | odyssey4me: i thinkso - we arent the only ones who will have issues | 12:58 |
andymccr | i imagine packagers will have bigger issues | 12:58 |
odyssey4me | yeah, much | 12:58 |
odyssey4me | that's exactly who the co-installability promise is designed for | 12:58 |
andymccr | we can get around it (at a cost) | 13:00 |
andymccr | but then i might request upper constraints in their reqs file | 13:00 |
andymccr | to ensure consistency | 13:01 |
*** Mudpuppy has joined #openstack-ansible | 13:04 | |
*** thorst has quit IRC | 13:05 | |
*** johnmilton has joined #openstack-ansible | 13:06 | |
odyssey4me | yes, if they're going to ignore the requirements process, then they need to publish their own properly tested upper bounds in their own repo | 13:07 |
*** sacharya has joined #openstack-ansible | 13:07 | |
odyssey4me | that then gives us something to work with as a consistent set which guarantees that what was tested is what is deployed | 13:07 |
*** Mudpuppy has quit IRC | 13:08 | |
*** v1k0d3n has joined #openstack-ansible | 13:10 | |
*** woodard has quit IRC | 13:11 | |
*** woodard has joined #openstack-ansible | 13:11 | |
*** sacharya has quit IRC | 13:11 | |
*** jamesden_ has quit IRC | 13:12 | |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Remove Trusty support from os_tempest role https://review.openstack.org/411365 | 13:16 |
mgariepy | good morning everyone | 13:19 |
openstackgerrit | Kyle L. Henderson proposed openstack/openstack-ansible: Add retries to apt update in cache proxy task https://review.openstack.org/411805 | 13:35 |
openstackgerrit | Merged openstack/openstack-ansible: Revert role SHA pin for Ocata-3 development https://review.openstack.org/411188 | 13:35 |
mgariepy | anyone would have an idea why http://logs.openstack.org/58/411358/1/check/gate-openstack-ansible-os_nova-ansible-func-centos-7-nv/c7a6100/console.html#_2016-12-15_20_39_01_452233 fails and this one doesn't http://logs.openstack.org/58/411358/1/check/gate-openstack-ansible-os_nova-ansible-func-centos-7-nv/c7a6100/console.html#_2016-12-15_20_11_28_816156 | 13:37 |
*** retreved has joined #openstack-ansible | 13:45 | |
kylek3h | mgariepy: what version of ansible are these runs using? perhaps 1.9.x? | 13:46 |
mgariepy | 2.2.something | 13:47 |
kylek3h | hmm....intersting. we saw that error alot with 1.9.x | 13:47 |
mgariepy | when I run the test on my vm it passes correctly. | 13:48 |
kylek3h | there were no internal reties with apt in 1.9.x we saw that about 10% of the time. we patched ansible 1.9 to add retries and it went away. | 13:48 |
mgariepy | this is centos ;) | 13:49 |
kylek3h | ah... | 13:49 |
kylek3h | well...maybe similar issue? | 13:49 |
mgariepy | there is retry but still the cert doesn't validate. | 13:50 |
mgariepy | and it tries on github and bootstrap.pypa.io | 13:50 |
kylek3h | right...it was really flaky with apt....and a simple retry worked. | 13:50 |
odyssey4me | mgariepy hmm, that sounds like the usual SNI issues | 13:53 |
odyssey4me | we had that badly with trusty | 13:53 |
odyssey4me | any python version < 2.7.9 does not support SNI SSL certs natively | 13:53 |
mgariepy | why does it work the first time ? | 13:53 |
odyssey4me | we have patched in a bunch of extra packages and python bits to make that work, so it should work... but yeah | 13:54 |
mgariepy | and it's consistent on the gate job :) haha | 13:54 |
*** fxpester has quit IRC | 13:54 | |
odyssey4me | that said, yes - I have seen sometimes it works on the host but not in containers | 13:54 |
odyssey4me | in that case it's better to look for networking shenanigans | 13:55 |
odyssey4me | race conditions and such | 13:55 |
mgariepy | when i run it on my openstack installation it works fine. | 13:55 |
*** ngupta has joined #openstack-ansible | 13:55 | |
odyssey4me | hmm, I wonder if there may not be something gate specific involved then | 13:55 |
mgariepy | can we stat get-pip.py to see if it's there an skip the download if it's not older than a day ? | 13:56 |
mgariepy | might speed up the gate check a tiny bit also. | 13:56 |
odyssey4me | well, so that was the point of implementing the fetch of get-pip to the repo server, then using it from there for all other hosts/containers | 13:57 |
odyssey4me | it means it would skip automatically and not have to reach out to the internet | 13:57 |
odyssey4me | but obviously you're looking at a role test here where there is no repo server | 13:57 |
mgariepy | i saw errors in the past with github certs but those where linked to mtu issue (1500 in container but 1450 on the hosts) | 13:58 |
mgariepy | i don't think this kind of issue would come up in the gate job. | 13:59 |
odyssey4me | it could? it all depends on how we're setting things up | 13:59 |
odyssey4me | but there is the fact that it did work earlier in the sequence, then doesn't work later | 13:59 |
mgariepy | it wouldn't work the first time for the container ;) | 13:59 |
odyssey4me | so if it is that, then something's changing the setup | 14:00 |
mgariepy | hmm. | 14:00 |
*** shausy has joined #openstack-ansible | 14:01 | |
mgariepy | dmsimard, how can i enable the ARA logging on that patch ? | 14:01 |
dmsimard | mgariepy: which one | 14:01 |
mgariepy | https://review.openstack.org/#/c/411358 | 14:01 |
odyssey4me | I wonder if it would work if we add 'creates: /opt/get-pip.py' to that task, but make it default to omit that... and only activate that for the role tests | 14:01 |
odyssey4me | lemme work up a patch to try that out | 14:02 |
dmsimard | mgariepy: I'm not sure how the gate is set up. Maybe a Depends-On https://review.openstack.org/#/c/396324/ ? | 14:02 |
dmsimard | mgariepy: or https://review.openstack.org/#/c/396565/ | 14:03 |
dmsimard | I'm not sure which one you need | 14:03 |
mgariepy | depends don't pull patches AFAIK for openstack-ansible. | 14:04 |
odyssey4me | no, not yet - we still need to work out how to do it | 14:06 |
mgariepy | odyssey4me, could we have a check-experimental-with-ara ? | 14:07 |
odyssey4me | mgariepy what are you hoping to achieve? | 14:07 |
mgariepy | having a easy way to see what's happening on the server | 14:07 |
dmsimard | FWIW I feel https://review.openstack.org/#/c/396565/ is good to merge if you guys are satisfied with it | 14:07 |
odyssey4me | a way to do it would be to change https://review.openstack.org/#/c/396565/ so that it's opt-in rather than enabled by default | 14:08 |
dmsimard | However, https://review.openstack.org/#/c/396324/ should definitely not merge until ARA is optimized for that amount of content | 14:08 |
odyssey4me | then in any patch you want to use it you could just active it | 14:08 |
odyssey4me | *activate it | 14:08 |
mgariepy | just like check experimental is ? | 14:08 |
*** marc_ab has joined #openstack-ansible | 14:09 | |
odyssey4me | mgariepy check experimental would require a lot more moving parts | 14:10 |
odyssey4me | andymccr thoughts on setting up ARA as an opt-in item via an env var? | 14:10 |
uthng | odyssey4me: yeah thanks for tuto. It is simpler :D but I did not find it out hehe :) | 14:10 |
odyssey4me | that way it would be easy to adjust the env var temporarily in a patch to use it, but not always have it enabled by default | 14:10 |
*** shausy has quit IRC | 14:11 | |
*** mennie has quit IRC | 14:11 | |
*** shausy has joined #openstack-ansible | 14:11 | |
dmsimard | odyssey4me, mgariepy, andymccr: I added comments to the ARA reviews: https://review.openstack.org/#/q/topic:osa-ara | 14:13 |
*** mennie has joined #openstack-ansible | 14:18 | |
*** whiteveil has joined #openstack-ansible | 14:21 | |
*** fguillot has joined #openstack-ansible | 14:22 | |
*** woodard has quit IRC | 14:23 | |
*** woodard has joined #openstack-ansible | 14:23 | |
*** shausy has quit IRC | 14:26 | |
*** jamesdenton has joined #openstack-ansible | 14:28 | |
*** shausy has joined #openstack-ansible | 14:32 | |
odyssey4me | mgariepy so, interestingly the get_url module is meant to actually ignore the task if the file is already there - or at least the docs say so | 14:33 |
*** Andrew_jedi has quit IRC | 14:33 | |
odyssey4me | ah, we have 'force: yes' | 14:33 |
mgariepy | we might want it to get updated once in a while ? | 14:35 |
mgariepy | if we remove the force, we might loose updating it on upgrades | 14:37 |
*** klamath has joined #openstack-ansible | 14:38 | |
*** ngupta has quit IRC | 14:38 | |
*** klamath has quit IRC | 14:39 | |
*** ngupta has joined #openstack-ansible | 14:39 | |
*** klamath has joined #openstack-ansible | 14:40 | |
*** thorst_ has joined #openstack-ansible | 14:41 | |
*** marst has joined #openstack-ansible | 14:41 | |
*** shausy has quit IRC | 14:41 | |
*** shausy has joined #openstack-ansible | 14:42 | |
*** TxGirlGeek has joined #openstack-ansible | 14:43 | |
*** thorst_ has quit IRC | 14:46 | |
*** shausy has quit IRC | 14:49 | |
*** shausy has joined #openstack-ansible | 14:50 | |
*** thorst_ has joined #openstack-ansible | 14:51 | |
*** shausy has quit IRC | 14:55 | |
*** chhavi has quit IRC | 14:58 | |
*** dxiri has joined #openstack-ansible | 14:58 | |
*** Andrew_jedi has joined #openstack-ansible | 15:00 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install: Provide toggle for get-pip.py get_url task force option https://review.openstack.org/411854 | 15:04 |
odyssey4me | mgariepy ^ | 15:04 |
*** whiteveil has quit IRC | 15:08 | |
*** sacharya has joined #openstack-ansible | 15:08 | |
mgariepy | odyssey4me, thanks a lot :) | 15:08 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install: Provide toggle for get-pip.py get_url task force option https://review.openstack.org/411854 | 15:08 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Prevent repeated get-pip.py fetches in role tests https://review.openstack.org/411855 | 15:09 |
odyssey4me | mgariepy ^ that combination should improve the role test results | 15:09 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install: Provide toggle for get-pip.py get_url task force option https://review.openstack.org/411854 | 15:11 |
*** sacharya has quit IRC | 15:13 | |
*** thorst_ has quit IRC | 15:14 | |
*** dxiri has quit IRC | 15:18 | |
*** dxiri has joined #openstack-ansible | 15:19 | |
*** Disova has quit IRC | 15:23 | |
*** deadnull has quit IRC | 15:23 | |
mgariepy | thanks a lot for this odyssey4me. | 15:24 |
odyssey4me | mgariepy well, andymccr may correct me here but the original patch to force it to fetch all the time was because of Ansible 1.9 | 15:25 |
odyssey4me | I think that patch implements enough of a safeguard, but will really help improve the role test execution. | 15:26 |
* mgariepy hope that os_nova centos will not fail a bit later after this :D | 15:27 | |
odyssey4me | I don't know if any cores are around to vote that one through. | 15:28 |
odyssey4me | but I expect that it will go through at some point | 15:29 |
*** thorst_ has joined #openstack-ansible | 15:35 | |
openstackgerrit | Merged openstack/openstack-ansible-pip_install: Provide toggle for get-pip.py get_url task force option https://review.openstack.org/411854 | 15:37 |
dmsimard | I'm going to ask a crazy question | 15:45 |
dmsimard | Will there ever be binary deployments in OSAD, either through UCA or RDO ? | 15:46 |
*** thorst_ has quit IRC | 15:46 | |
dmsimard | If someone interested enough contributes it | 15:46 |
*** jmckind has joined #openstack-ansible | 15:49 | |
*** whiteveil has joined #openstack-ansible | 15:51 | |
*** maeker has joined #openstack-ansible | 15:55 | |
*** thorst_ has joined #openstack-ansible | 15:57 | |
*** weezS has joined #openstack-ansible | 15:58 | |
odyssey4me | dmsimard that's the key thing right there | 15:59 |
odyssey4me | if someone is prepared to contribute it and maintain it, then all it takes is a spec | 15:59 |
dmsimard | odyssey4me: okay -- because I haven't looked at the scope of this but I am assuming there are a lot of assumptions that everything runs off of source and it'd probably be a huge undertaking | 16:00 |
odyssey4me | yep, I expect it wouldn't be too hard - but it would be a fair body of work | 16:01 |
*** thorst_ has quit IRC | 16:01 | |
odyssey4me | and the trouble we've had in the past when working on chef-based deployments (where we did use packages only) was that each distro made its own assumptions | 16:01 |
odyssey4me | so we'd have to skirt the line between those - perhaps we could help normalise them a bit | 16:02 |
dmsimard | meh, like different paths and so on ? | 16:02 |
odyssey4me | different paths are fine - I think we have that covered now | 16:02 |
odyssey4me | but different conf files was what I remember | 16:02 |
odyssey4me | some do multiple, some do a single file | 16:02 |
odyssey4me | not terrible to overcome - I think we have a fair framework in place to cater for distro differences | 16:03 |
*** sdake has joined #openstack-ansible | 16:03 | |
odyssey4me | a good start for anyone wanting to do that would be for them to get involved in what we already have | 16:03 |
dmsimard | yeah | 16:03 |
*** marc_ab has quit IRC | 16:03 | |
dmsimard | hey, whatever happened to that config templating module in Ansible ? Did that get merged ? | 16:03 |
odyssey4me | gain community confidence, understand where every current stakeholder is coming from | 16:04 |
odyssey4me | nah, Ansible doesn't want to merge it - they feel that using filters you can achieve the same thing | 16:04 |
odyssey4me | regardless of performance differences | 16:04 |
odyssey4me | so we know that the ceph-ansible crew vendor it | 16:04 |
odyssey4me | and we'll continue to use it | 16:04 |
odyssey4me | and extend it as we go | 16:04 |
odyssey4me | here's a new extension https://review.openstack.org/382109 | 16:05 |
*** chhavi has joined #openstack-ansible | 16:06 | |
*** chhavi has quit IRC | 16:06 | |
odyssey4me | andymccr any chance for a vote on https://review.openstack.org/411855? | 16:06 |
*** chhavi has joined #openstack-ansible | 16:06 | |
*** rmelero has joined #openstack-ansible | 16:08 | |
odyssey4me | thx | 16:08 |
*** maeker has quit IRC | 16:16 | |
*** maeker has joined #openstack-ansible | 16:16 | |
*** uthng has quit IRC | 16:19 | |
*** marc_ab has joined #openstack-ansible | 16:20 | |
*** pcaruana has quit IRC | 16:25 | |
openstackgerrit | Merged openstack/openstack-ansible-tests: Prevent repeated get-pip.py fetches in role tests https://review.openstack.org/411855 | 16:28 |
*** adrian_otto has joined #openstack-ansible | 16:32 | |
*** whiteveil has quit IRC | 16:32 | |
*** Oku_OS is now known as Oku_OS-away | 16:32 | |
*** whiteveil has joined #openstack-ansible | 16:33 | |
*** whiteveil has quit IRC | 16:34 | |
*** karimb has quit IRC | 16:39 | |
*** karimb has joined #openstack-ansible | 16:40 | |
*** dxiri has quit IRC | 16:41 | |
*** kstev has joined #openstack-ansible | 16:44 | |
*** dxiri has joined #openstack-ansible | 16:46 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Setup ARA reporting for central tests repository https://review.openstack.org/396565 | 16:48 |
*** maeker has quit IRC | 16:55 | |
*** woodard_ has joined #openstack-ansible | 17:01 | |
*** woodard has quit IRC | 17:01 | |
*** whiteveil has joined #openstack-ansible | 17:01 | |
*** ngupta has quit IRC | 17:03 | |
*** dxiri has quit IRC | 17:04 | |
*** dxiri has joined #openstack-ansible | 17:05 | |
*** adrian_otto has quit IRC | 17:08 | |
*** sacharya has joined #openstack-ansible | 17:09 | |
*** chhavi has quit IRC | 17:14 | |
*** maeker has joined #openstack-ansible | 17:19 | |
*** lemouchon2 has quit IRC | 17:20 | |
*** lemouchon2 has joined #openstack-ansible | 17:20 | |
*** maeker has quit IRC | 17:23 | |
*** maeker_ has joined #openstack-ansible | 17:24 | |
andymccr | ok i think im done for today! Have a great weekend everybody | 17:26 |
*** sdake has quit IRC | 17:28 | |
*** maeker_ has quit IRC | 17:29 | |
*** maeker has joined #openstack-ansible | 17:30 | |
*** maeker has quit IRC | 17:33 | |
*** chhavi has joined #openstack-ansible | 17:34 | |
*** adrian_otto has joined #openstack-ansible | 17:38 | |
*** Jack_Iv__ has joined #openstack-ansible | 17:42 | |
*** maeker has joined #openstack-ansible | 17:45 | |
*** maeker has quit IRC | 17:48 | |
*** ngupta has joined #openstack-ansible | 17:53 | |
palendae | You too andymccr | 17:54 |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Fix tempest for os_horizon role https://review.openstack.org/411748 | 17:55 |
*** wadeholler has joined #openstack-ansible | 17:57 | |
*** maeker has joined #openstack-ansible | 17:59 | |
*** Jack_Iv__ has quit IRC | 18:00 | |
*** asettle has quit IRC | 18:01 | |
*** Jack_Iv__ has joined #openstack-ansible | 18:01 | |
*** asettle has joined #openstack-ansible | 18:01 | |
*** whiteveil has quit IRC | 18:03 | |
*** Jack_Iv__ has quit IRC | 18:05 | |
*** asettle has quit IRC | 18:06 | |
*** ngupta has quit IRC | 18:11 | |
*** ngupta has joined #openstack-ansible | 18:11 | |
*** ngupta has quit IRC | 18:13 | |
*** ngupta has joined #openstack-ansible | 18:13 | |
*** Jack_Iv__ has joined #openstack-ansible | 18:15 | |
*** Andrew_jedi has quit IRC | 18:16 | |
*** Jack_Iv__ has quit IRC | 18:18 | |
*** Mudpuppy has joined #openstack-ansible | 18:18 | |
*** Jack_Iv__ has joined #openstack-ansible | 18:19 | |
*** Jack_Iv__ has quit IRC | 18:24 | |
palendae | logan-, cloudnull: So I dug a bit deeper on potential group_var overlaps; in these two tests, I couldn't find conflicts. https://gist.github.com/nrb/179053241c3035f57bd9a8d89b0f8410 | 18:24 |
*** weezS has quit IRC | 18:24 | |
*** kelv has joined #openstack-ansible | 18:26 | |
*** weezS has joined #openstack-ansible | 18:29 | |
*** ianychoi has quit IRC | 18:30 | |
*** adrian_otto has quit IRC | 18:30 | |
*** Mudpuppy has quit IRC | 18:30 | |
*** Mudpuppy has joined #openstack-ansible | 18:31 | |
*** asettle has joined #openstack-ansible | 18:31 | |
*** asettle has quit IRC | 18:32 | |
*** sdake has joined #openstack-ansible | 18:33 | |
*** Mudpuppy has quit IRC | 18:35 | |
logan- | agreed palendae. that has been my experience with ansible's base inventory outside of OSA. | 18:36 |
palendae | https://github.com/ansible/ansible/issues/17559; https://github.com/ansible/ansible/issues/17243 | 18:36 |
palendae | I did not try the scenario where a host was directly in 2 groups, or trying --limit | 18:36 |
palendae | Will try it after lunch | 18:36 |
palendae | My 2 parents thing was a *group* in 2 different groups | 18:37 |
palendae | Hmmmmmmmmm https://github.com/ansible/ansible/pull/15942 | 18:37 |
palendae | ^ that may be why it's working actually | 18:37 |
palendae | Oh, that links back to 17243, nvm | 18:38 |
palendae | Kind of spelunking so I can compare OSA's approach, and at least write down some whys | 18:39 |
palendae | Even if they're historical and we can move | 18:39 |
*** galstrom_zzz is now known as galstrom | 18:39 | |
logan- | seems like that PR would make overlaps a lot safer | 18:40 |
palendae | Yeah, the claim is that https://github.com/ansible/ansible/issues/17243 fixed it | 18:40 |
palendae | Which might have gotten into 2.2, which my test was using | 18:40 |
*** Jack_Iv__ has joined #openstack-ansible | 18:41 | |
logan- | right. it looks like it is inthe stable-2.2 branch but not any tags yet if im parsing github correctly. | 18:41 |
palendae | I think the first issue https://github.com/ansible/ansible/issues/17559, where --limit behaves unexpectedly, may still be a problem though | 18:41 |
logan- | yeah | 18:42 |
palendae | --limit is *just* for deciding where to execute tasks, not for processing variable information | 18:42 |
palendae | So, more experiments to do this afternnon | 18:42 |
palendae | But I'm hungry :) | 18:42 |
logan- | i think that is actually the correct way to do it. --limiting execution shouldn't change how inventory is compiled imo | 18:43 |
wadeholler | hi all I have a 16.04 upgrade issue that I need little help with: doing first node of 3 node infra group; re-pxe'd / reinstalled host, ran setup-hosts,setup-infrastructure,setup-openstack which relative to the problem im looking at ran clean. however, new services don't start with a pip problem. | 18:43 |
wadeholler | http://pastebin.com/E6DvWdHy | 18:43 |
wadeholler | I bet it's something simple but I'm new. Can someone point me in a direction ? | 18:44 |
palendae | logan-: Yeah; as long as that's understood, I'm fine, but I'm still trying to dig out the why of our inventory structure so I can write that down :) | 18:44 |
logan- | wadeholler: this is going to be interesting to dig into. I take it you had a trusty env, upgraded the trusty env to newton, and now you are working on upgrading your newton+trusty install to newton+xenial one controller at a time? | 18:46 |
wadeholler | logan: you got it | 18:47 |
*** Jack_Iv__ has quit IRC | 18:48 | |
*** maeker has quit IRC | 18:48 | |
*** Jack_Iv__ has joined #openstack-ansible | 18:48 | |
logan- | ok first question... can you check which host the repo-build.yml playbook is running the builds against? i'm guessing your wheels/venvs are getting built on a trusty controller, but can you confirm? | 18:49 |
*** stuartgr has left #openstack-ansible | 18:50 | |
*** maeker has joined #openstack-ansible | 18:50 | |
wadeholler | will do if you tel me how ot do that ..... ? | 18:51 |
*** maeker has quit IRC | 18:52 | |
*** maeker_ has joined #openstack-ansible | 18:52 | |
openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Remove Trusty support from os_horizon role https://review.openstack.org/411270 | 18:52 |
*** Jack_Iv__ has quit IRC | 18:53 | |
wadeholler | login: p.s. "pip list" on the container before I run /openstack/venv/glance...../bin/activate runs OK | 18:53 |
*** sdake has quit IRC | 18:54 | |
logan- | if you run openstack-ansible repo-build.yml, just curious which release the host running tasks like "Create OpenStack-Ansible requirement wheels" has | 18:54 |
wadeholler | logan: ok - will run that and report back. | 18:55 |
logan- | also did you remove the ansible facts cache on your deployment host after reinstalling the cpa00003 server? | 18:55 |
*** maeker_ has quit IRC | 18:55 | |
wadeholler | logan: no..... | 18:55 |
logan- | it should be in /etc/openstack_deploy/ansible_facts | 18:55 |
logan- | ok | 18:56 |
*** maeker has joined #openstack-ansible | 18:56 | |
*** sdake has joined #openstack-ansible | 18:56 | |
logan- | well lets check on the repo_build, then i'm curious if we delete the glance container you're having problems with, delete the facts cache, and re-build the container if it still has the same failure. | 18:57 |
logan- | i'm guessing it will but its worth checking | 18:57 |
*** Jack_Iv__ has joined #openstack-ansible | 18:59 | |
*** BjoernT has joined #openstack-ansible | 19:00 | |
*** jmckind_ has joined #openstack-ansible | 19:00 | |
*** jmckind has quit IRC | 19:01 | |
*** poopcat has joined #openstack-ansible | 19:02 | |
wadeholler | yes "Create requirement wheels" ran on a trusty backed host - at least so far in the repo-build.xml execution | 19:04 |
wadeholler | logan: regardless of where "repo-build" built the wheels - go ahead and delete container, facts cache, rebuild ...? | 19:05 |
wadeholler | logan: as long as I don't dork with openstack_inventory the new container should get the same IP right? | 19:06 |
logan- | ok. thanks. if you could lxc-stop and lxc-destroy that glance container, delete /etc/openstack_deploy/ansible_facts, and then run setup-hosts and setup-openstack again that would be the next thing i'd test | 19:06 |
wadeholler | will do. | 19:06 |
logan- | yes the facts will be re-gathered when you run ansible again. the /etc/openstack_deploy/ansible_facts is just a cache that is refreshed. I think it is a 1 day cache time by default | 19:07 |
logan- | yep, it'll get built with the same IP as long as the inventory does not change | 19:07 |
wadeholler | ok. thank you | 19:07 |
logan- | I think we're running into the issue here https://bugs.launchpad.net/openstack-ansible/+bug/1641131 where wheels built in the trusty repo containers aren't compatible with xenial hosts, and vice versa if your repo_build was occurring on a xenial container | 19:09 |
openstack | Launchpad bug 1641131 in openstack-ansible "In mixed OS environments, the libvirt-python wheel does not work" [High,Confirmed] | 19:09 |
wadeholler | logan: delete ansible_facts or just files in that directory for glance container ? | 19:09 |
logan- | it should be safe to delete the entire facts directory, and I would recommend that since you rebuilt the entire host and it probably has old facts cached for all of the containers on that host | 19:10 |
*** poopcat has quit IRC | 19:10 | |
wadeholler | ok | 19:10 |
*** poopcat has joined #openstack-ansible | 19:11 | |
*** phalmos has joined #openstack-ansible | 19:11 | |
*** whiteveil has joined #openstack-ansible | 19:13 | |
*** sdake has quit IRC | 19:16 | |
*** weezS has quit IRC | 19:17 | |
*** sdake has joined #openstack-ansible | 19:17 | |
*** whiteveil has quit IRC | 19:20 | |
*** thorst_ has joined #openstack-ansible | 19:20 | |
*** Jack_I___ has joined #openstack-ansible | 19:21 | |
*** whiteveil has joined #openstack-ansible | 19:23 | |
*** Jack_Iv__ has quit IRC | 19:23 | |
*** Jack_iv_ has quit IRC | 19:24 | |
*** Jack_Iv has quit IRC | 19:24 | |
*** phalmos has quit IRC | 19:25 | |
*** Jack_Iv has joined #openstack-ansible | 19:25 | |
*** chhavi has quit IRC | 19:25 | |
*** thorst_ has quit IRC | 19:27 | |
*** asettle has joined #openstack-ansible | 19:33 | |
*** asettle has quit IRC | 19:36 | |
*** asettle has joined #openstack-ansible | 19:36 | |
*** electrofelix has quit IRC | 19:37 | |
*** sdake has quit IRC | 19:37 | |
*** asettle has quit IRC | 19:39 | |
*** sdake has joined #openstack-ansible | 19:42 | |
*** jmckind_ has quit IRC | 19:51 | |
*** jamesdenton has quit IRC | 19:52 | |
*** jamesdenton has joined #openstack-ansible | 19:53 | |
*** ngupta has quit IRC | 19:54 | |
*** ngupta has joined #openstack-ansible | 19:54 | |
*** jamesdenton has quit IRC | 19:58 | |
*** ngupta has quit IRC | 19:59 | |
*** jmckind has joined #openstack-ansible | 20:00 | |
*** galstrom is now known as galstrom_zzz | 20:00 | |
*** sdake has quit IRC | 20:02 | |
*** Jack_I___ has quit IRC | 20:03 | |
*** Jack_Iv has quit IRC | 20:03 | |
*** Jack_Iv has joined #openstack-ansible | 20:04 | |
*** sdake has joined #openstack-ansible | 20:04 | |
*** Jack_Iv_ has joined #openstack-ansible | 20:06 | |
*** galstrom_zzz is now known as galstrom | 20:06 | |
*** Jack_Iv has quit IRC | 20:09 | |
*** Jack_Iv has joined #openstack-ansible | 20:11 | |
*** galstrom is now known as galstrom_zzz | 20:12 | |
*** johnmilton has quit IRC | 20:20 | |
wadeholler | logan: ran procedure you advised - delete glance container, delete ansible_facts, setup-hosts, setup-openstack | 20:22 |
wadeholler | logan: same error | 20:22 |
*** maeker has quit IRC | 20:22 | |
*** sdake has quit IRC | 20:24 | |
wadeholler | logan: other thoughts - delete repo containers - force to xenial based physical only, repo-build ? | 20:25 |
wadeholler | seems heavy handed and troublesome to other nodes not yet upgraded to xenial | 20:26 |
*** sdake has joined #openstack-ansible | 20:26 | |
*** galstrom_zzz is now known as galstrom | 20:27 | |
*** maeker_ has joined #openstack-ansible | 20:27 | |
*** maeker_ has quit IRC | 20:32 | |
*** Andrew_jedi has joined #openstack-ansible | 20:32 | |
*** gouthamr has quit IRC | 20:44 | |
*** sdake has quit IRC | 20:46 | |
*** gouthamr has joined #openstack-ansible | 20:48 | |
*** sdake has joined #openstack-ansible | 20:49 | |
*** asettle has joined #openstack-ansible | 20:59 | |
*** asettle has quit IRC | 21:00 | |
*** galstrom is now known as galstrom_zzz | 21:03 | |
*** askb has joined #openstack-ansible | 21:05 | |
*** sdake has quit IRC | 21:09 | |
*** sdake has joined #openstack-ansible | 21:10 | |
*** Andrew_jedi has quit IRC | 21:11 | |
*** BjoernT has quit IRC | 21:12 | |
*** jwitko has joined #openstack-ansible | 21:14 | |
*** Jack_Iv_ has quit IRC | 21:16 | |
*** Jack_Iv has quit IRC | 21:17 | |
*** ngupta has joined #openstack-ansible | 21:25 | |
*** gouthamr has quit IRC | 21:25 | |
*** wadeholler has quit IRC | 21:25 | |
*** gouthamr has joined #openstack-ansible | 21:26 | |
*** thorst_ has joined #openstack-ansible | 21:26 | |
*** gouthamr has quit IRC | 21:27 | |
*** Mudpuppy has joined #openstack-ansible | 21:29 | |
*** uthng has joined #openstack-ansible | 21:30 | |
uthng | hello all | 21:30 |
*** sdake has quit IRC | 21:30 | |
uthng | while running os-magnum-install.yml, I got this error : systemctl status rsyslog.service rsyslog.service Loaded: error (Reason: No such file or directory) Active: inactive (dead) | 21:31 |
uthng | anyone can tell me why please ? | 21:31 |
*** Mudpuppy has quit IRC | 21:32 | |
*** Mudpuppy has joined #openstack-ansible | 21:32 | |
*** sdake has joined #openstack-ansible | 21:32 | |
*** Jeffrey4l has quit IRC | 21:35 | |
*** whiteveil has quit IRC | 21:35 | |
*** retreved has quit IRC | 21:35 | |
*** thorst_ has quit IRC | 21:35 | |
*** Jeffrey4l has joined #openstack-ansible | 21:36 | |
*** thorst_ has joined #openstack-ansible | 21:36 | |
*** fguillot has quit IRC | 21:41 | |
*** adrian_otto has joined #openstack-ansible | 21:42 | |
*** adrian_otto has quit IRC | 21:49 | |
*** Mudpuppy has quit IRC | 21:53 | |
*** marc_ab has quit IRC | 21:56 | |
*** jproulx has quit IRC | 22:04 | |
*** aludwar has quit IRC | 22:06 | |
*** aludwar has joined #openstack-ansible | 22:07 | |
*** adrian_otto has joined #openstack-ansible | 22:12 | |
*** Jack_Iv has joined #openstack-ansible | 22:13 | |
*** smatzek has joined #openstack-ansible | 22:17 | |
*** adrian_otto has quit IRC | 22:31 | |
*** jamesdenton has joined #openstack-ansible | 22:34 | |
*** jamesden_ has joined #openstack-ansible | 22:37 | |
*** jamesdenton has quit IRC | 22:38 | |
*** jmckind has quit IRC | 22:42 | |
*** aludwar has quit IRC | 22:42 | |
*** aludwar has joined #openstack-ansible | 22:43 | |
*** Jack_Iv has quit IRC | 22:45 | |
*** smatzek has quit IRC | 22:48 | |
*** aludwar has quit IRC | 22:52 | |
uthng | noone here please ? | 22:55 |
*** Jack_Iv has joined #openstack-ansible | 22:56 | |
*** chris_hultin|AWA is now known as chris_hultin | 22:56 | |
*** dxiri has quit IRC | 22:56 | |
*** dxiri has joined #openstack-ansible | 22:57 | |
*** aludwar has joined #openstack-ansible | 23:01 | |
*** Jack_Iv has quit IRC | 23:02 | |
*** Jack_Iv has joined #openstack-ansible | 23:02 | |
*** Jack_Iv_ has joined #openstack-ansible | 23:03 | |
*** Jack_Iv_ has quit IRC | 23:03 | |
*** Jack_Iv_ has joined #openstack-ansible | 23:04 | |
*** Jack_Iv__ has joined #openstack-ansible | 23:04 | |
*** Jack_Iv_ has quit IRC | 23:08 | |
*** TxGirlGeek has quit IRC | 23:12 | |
*** adrian_otto has joined #openstack-ansible | 23:12 | |
*** maeker has joined #openstack-ansible | 23:15 | |
*** marst has quit IRC | 23:21 | |
*** Mudpuppy has joined #openstack-ansible | 23:22 | |
*** maeker has quit IRC | 23:26 | |
*** ngupta has quit IRC | 23:27 | |
*** ngupta has joined #openstack-ansible | 23:28 | |
*** maeker has joined #openstack-ansible | 23:28 | |
*** chris_hultin is now known as chris_hultin|AWA | 23:29 | |
*** ngupta has quit IRC | 23:30 | |
uthng | hmm magnum & sahara container lxc use systemd | 23:30 |
*** ngupta has joined #openstack-ansible | 23:30 | |
uthng | other containers still use upstart. Thats why they work | 23:30 |
uthng | why ansible_service_mgr on magnum and sahara detects systemd ? anyone has an idea about it ? | 23:31 |
*** retreved has joined #openstack-ansible | 23:31 | |
*** retreved has quit IRC | 23:32 | |
*** dxiri_ has joined #openstack-ansible | 23:34 | |
*** dxiri has quit IRC | 23:34 | |
uthng | nobody awaked here ? odyssey4me ? andymccr ? :( | 23:35 |
*** maeker has quit IRC | 23:37 | |
*** maeker has joined #openstack-ansible | 23:40 | |
*** dalees has quit IRC | 23:40 | |
*** dalees has joined #openstack-ansible | 23:43 | |
*** Jack_Iv__ has quit IRC | 23:45 | |
jmccrory | uthng : trusty with OSA 14.0.3? you're probably hitting this problem https://review.openstack.org/#/c/409312/ | 23:46 |
*** Jack_Iv_ has joined #openstack-ansible | 23:46 | |
uthng | jmccrory: for other components such as glance, cinder etc. I have trusty 14.04.4. | 23:47 |
uthng | But new components : magnum and sahara have 14.04.5 | 23:48 |
jmccrory | right, and you're deploying newton, or the 14.0.3 tag of openstack-ansible? | 23:48 |
uthng | newton | 23:48 |
*** dalees has quit IRC | 23:48 | |
jmccrory | ok, try installing the systemd-shim package in your containers | 23:49 |
*** dalees has joined #openstack-ansible | 23:50 | |
*** Jack_Iv_ has quit IRC | 23:51 | |
uthng | jmccrory: the problem you sais is due to /run/systemd/system. But when I check ansible/module_utils/facts.py, I saw that it does not check /run/systemd/system but check the path to systemctl | 23:54 |
*** Jack_Iv has quit IRC | 23:54 | |
uthng | I dont know if by default, trusty 14.04.05 installs systemd or not. But there are both systemd (all related packages installed) and initctl | 23:56 |
uthng | I wonder if I can uninstall systemd in magnum and sahara containers | 23:56 |
*** maeker has quit IRC | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!