*** TxGirlGe_ has quit IRC | 00:02 | |
*** thorst has joined #openstack-ansible | 00:07 | |
*** vnogin has quit IRC | 00:10 | |
*** thorst has quit IRC | 00:11 | |
*** phalmos has quit IRC | 00:17 | |
*** phalmos has joined #openstack-ansible | 00:18 | |
*** threestrands has quit IRC | 00:21 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 00:21 |
---|---|---|
cloudnull | bndzor: I'm glad it wasn't something terrible going on within the host but that's frustrating. | 00:22 |
*** thorst has joined #openstack-ansible | 00:25 | |
*** yifei has joined #openstack-ansible | 00:29 | |
cloudnull | any cores around that might want to give this a push https://review.openstack.org/#/c/523525/ | 00:31 |
cloudnull | seems we keep hitting http://logs.openstack.org/50/523850/5/check/openstack-ansible-functional-opensuse-423/2ce7655/job-output.txt.gz#_2017-11-29_19_55_12_653934 | 00:32 |
*** chyka has quit IRC | 00:32 | |
*** phalmos has quit IRC | 00:38 | |
*** pedja has quit IRC | 01:03 | |
*** thorst has quit IRC | 01:05 | |
*** rpittau_ has joined #openstack-ansible | 01:05 | |
*** rpittau has quit IRC | 01:08 | |
*** rromans has joined #openstack-ansible | 01:17 | |
*** rromans has joined #openstack-ansible | 01:18 | |
*** smatzek has joined #openstack-ansible | 01:19 | |
*** smatzek has quit IRC | 01:24 | |
*** rpittau__ has joined #openstack-ansible | 01:25 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 01:26 |
*** rpittau_ has quit IRC | 01:29 | |
*** thorst has joined #openstack-ansible | 01:34 | |
*** thorst has quit IRC | 01:34 | |
*** jafeha has quit IRC | 01:36 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-memcached_server master: Clean up the role and further isolate the service https://review.openstack.org/524036 | 01:48 |
*** savvas_ has joined #openstack-ansible | 01:50 | |
*** rromans has quit IRC | 01:54 | |
*** rromans has joined #openstack-ansible | 01:55 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 02:00 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-memcached_server master: Clean up the role and further isolate the service https://review.openstack.org/524036 | 02:04 |
*** savvas_ has quit IRC | 02:07 | |
*** savvas_ has joined #openstack-ansible | 02:07 | |
*** savvas_ has quit IRC | 02:08 | |
*** savvas_ has joined #openstack-ansible | 02:08 | |
*** savvas_ has quit IRC | 02:08 | |
*** savvas_ has joined #openstack-ansible | 02:09 | |
*** savvas_ has quit IRC | 02:09 | |
*** savvas_ has joined #openstack-ansible | 02:10 | |
*** savvas_ has quit IRC | 02:10 | |
*** dave-mccowan has quit IRC | 02:18 | |
openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Fix galera_monitoring_allowed_source https://review.openstack.org/523850 | 02:26 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Update the basic systemd extra configs for isolation and consistency https://review.openstack.org/524043 | 02:31 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-memcached_server master: Clean up the role and further isolate the service https://review.openstack.org/524036 | 02:33 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-memcached_server master: Clean up the role and further isolate the service https://review.openstack.org/524036 | 02:38 |
*** rromans has quit IRC | 02:39 | |
*** savvas_ has joined #openstack-ansible | 02:41 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 02:45 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Change the galera health check for better cluster health https://review.openstack.org/520673 | 02:45 |
*** savvas_ has quit IRC | 02:45 | |
*** dave-mccowan has joined #openstack-ansible | 02:46 | |
*** rromans has joined #openstack-ansible | 02:51 | |
*** savvas_ has joined #openstack-ansible | 02:51 | |
*** threestrands has joined #openstack-ansible | 02:55 | |
*** thorst has joined #openstack-ansible | 02:59 | |
*** masber has joined #openstack-ansible | 03:03 | |
*** masuberu has quit IRC | 03:07 | |
*** savvas_ has quit IRC | 03:26 | |
*** masber has quit IRC | 03:30 | |
*** masber has joined #openstack-ansible | 03:34 | |
*** Taseer has quit IRC | 03:36 | |
*** Taseer has joined #openstack-ansible | 03:36 | |
*** thorst has quit IRC | 03:46 | |
*** savvas_ has joined #openstack-ansible | 03:52 | |
*** savvas_ has quit IRC | 03:56 | |
*** SmearedBeard has quit IRC | 03:58 | |
*** Smeared_Beard has joined #openstack-ansible | 03:59 | |
*** savvas_ has joined #openstack-ansible | 04:00 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_client master: Fix SSL handling for galera clusters https://review.openstack.org/524059 | 04:04 |
*** dave-mccowan has quit IRC | 04:07 | |
*** thorst has joined #openstack-ansible | 04:16 | |
*** udesale has joined #openstack-ansible | 04:21 | |
*** thorst has quit IRC | 04:22 | |
*** thorst has joined #openstack-ansible | 04:24 | |
*** thorst has quit IRC | 04:29 | |
*** Jeffrey4l has quit IRC | 04:42 | |
*** nshetty has joined #openstack-ansible | 04:46 | |
openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Update Percona XtraDB Backup version https://review.openstack.org/523074 | 04:51 |
*** savvas__ has joined #openstack-ansible | 04:54 | |
*** thorst has joined #openstack-ansible | 04:55 | |
*** savvas_ has quit IRC | 04:55 | |
*** savvas__ has quit IRC | 04:55 | |
*** savvas has quit IRC | 04:55 | |
*** savvas has joined #openstack-ansible | 04:56 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 04:57 |
*** thorst has quit IRC | 04:57 | |
*** vnogin has joined #openstack-ansible | 04:59 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 05:02 |
*** poopcat has quit IRC | 05:02 | |
*** vnogin has quit IRC | 05:03 | |
*** Jeffrey4l has joined #openstack-ansible | 05:04 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Ensure the role works when run in serial https://review.openstack.org/524065 | 05:06 |
*** armaan has quit IRC | 05:08 | |
*** armaan has joined #openstack-ansible | 05:09 | |
*** cshen_ has joined #openstack-ansible | 05:15 | |
*** germs has quit IRC | 05:15 | |
*** cshen_ has quit IRC | 05:20 | |
*** savvas_ has joined #openstack-ansible | 05:21 | |
*** savvas_ has quit IRC | 05:26 | |
*** hybridpollo has quit IRC | 05:26 | |
*** chyka has joined #openstack-ansible | 05:27 | |
*** thorst has joined #openstack-ansible | 05:31 | |
*** chyka has quit IRC | 05:32 | |
*** savvas_ has joined #openstack-ansible | 05:34 | |
*** thorst has quit IRC | 05:35 | |
*** taseer1 has joined #openstack-ansible | 05:39 | |
*** taseer2 has joined #openstack-ansible | 05:40 | |
*** Taseer has quit IRC | 05:40 | |
*** taseer1 has quit IRC | 05:44 | |
*** savvas__ has joined #openstack-ansible | 05:44 | |
*** chhavi has joined #openstack-ansible | 05:46 | |
*** savvas has quit IRC | 05:47 | |
*** taseer3 has joined #openstack-ansible | 05:47 | |
*** taseer2 has quit IRC | 05:48 | |
*** taseer4 has joined #openstack-ansible | 05:48 | |
*** cshen_ has joined #openstack-ansible | 05:50 | |
*** cshen_ has quit IRC | 05:52 | |
*** taseer3 has quit IRC | 05:52 | |
*** armaan has quit IRC | 05:56 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 05:56 |
*** armaan has joined #openstack-ansible | 05:57 | |
*** udesale__ has joined #openstack-ansible | 06:01 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_client master: Fix SSL handling for galera clusters https://review.openstack.org/524059 | 06:03 |
*** armaan has quit IRC | 06:03 | |
*** udesale has quit IRC | 06:04 | |
*** pcaruana has joined #openstack-ansible | 06:05 | |
*** thorst has joined #openstack-ansible | 06:07 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Remove the AIO scenario & add new scenarios to maintain coverage https://review.openstack.org/516002 | 06:09 |
*** thorst has quit IRC | 06:12 | |
*** jbadiapa has quit IRC | 06:15 | |
*** pcaruana has quit IRC | 06:16 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 06:25 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_client master: Fix SSL handling for galera clusters https://review.openstack.org/524059 | 06:31 |
*** savvas_ has quit IRC | 06:34 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Ensure the role works when run in serial https://review.openstack.org/524065 | 06:35 |
*** mcarden has quit IRC | 06:38 | |
*** masuberu has joined #openstack-ansible | 06:42 | |
*** masber has quit IRC | 06:46 | |
*** thorst has joined #openstack-ansible | 06:46 | |
*** threestrands has quit IRC | 06:50 | |
*** thorst has quit IRC | 06:51 | |
*** savvas has joined #openstack-ansible | 07:02 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Fix the key lookup names for self-signed SSL https://review.openstack.org/524081 | 07:04 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_client master: Fix SSL handling for galera clusters https://review.openstack.org/524059 | 07:05 |
*** hw_wutianwei has joined #openstack-ansible | 07:06 | |
*** savvas has quit IRC | 07:07 | |
*** sxc731 has joined #openstack-ansible | 07:08 | |
*** savvas has joined #openstack-ansible | 07:10 | |
*** sxc731_ has joined #openstack-ansible | 07:14 | |
*** sxc731_ has quit IRC | 07:17 | |
*** Oku_OS-away is now known as Oku_OS | 07:18 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Tune-up the rabbitmq role for efficiency https://review.openstack.org/524028 | 07:18 |
*** pmannidi has quit IRC | 07:20 | |
*** thorst has joined #openstack-ansible | 07:20 | |
*** gouthamr has quit IRC | 07:20 | |
*** michelv has joined #openstack-ansible | 07:20 | |
*** gkadam_ has quit IRC | 07:24 | |
*** thorst has quit IRC | 07:25 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Further simplify the upgrade check https://review.openstack.org/524086 | 07:26 |
*** zkynet has joined #openstack-ansible | 07:28 | |
*** rgogunskiy has joined #openstack-ansible | 07:32 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-plugins master: Add basic plugin support for nspawn type containers https://review.openstack.org/490689 | 07:33 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Add nspawn container driver https://review.openstack.org/477017 | 07:33 |
*** sxc731 has quit IRC | 07:37 | |
*** markvoelker has quit IRC | 07:37 | |
*** udesale has joined #openstack-ansible | 07:38 | |
*** jbadiapa has joined #openstack-ansible | 07:39 | |
evrardjp | morning | 07:39 |
evrardjp | cloudnull: is on a row! | 07:39 |
*** udesale__ has quit IRC | 07:39 | |
*** armaan has joined #openstack-ansible | 07:48 | |
*** mbuil has joined #openstack-ansible | 07:48 | |
*** thorst has joined #openstack-ansible | 07:52 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/pike: Fix in-tree version https://review.openstack.org/522853 | 07:53 |
*** pcaruana has joined #openstack-ansible | 07:55 | |
*** thorst has quit IRC | 07:57 | |
*** nshetty is now known as nshetty|lunch | 07:58 | |
*** zkynet has quit IRC | 08:00 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Use Ansible 2.4 https://review.openstack.org/522778 | 08:00 |
jmccrory | it's out https://pypi.python.org/pypi/ansible/2.4.2.0 | 08:02 |
*** zkynet has joined #openstack-ansible | 08:10 | |
*** savvas has quit IRC | 08:14 | |
*** holmsten has joined #openstack-ansible | 08:23 | |
openstackgerrit | Dimitrios Markou proposed openstack/openstack-ansible-os_neutron master: Add OvS-NSH support https://review.openstack.org/517259 | 08:24 |
*** epalper has joined #openstack-ansible | 08:25 | |
*** thorst has joined #openstack-ansible | 08:27 | |
*** jafeha has joined #openstack-ansible | 08:28 | |
jafeha | good morning | 08:28 |
*** thorst has quit IRC | 08:32 | |
evrardjp | good morning jafeha | 08:35 |
evrardjp | how are things | 08:35 |
evrardjp | ? | 08:35 |
*** vnogin has joined #openstack-ansible | 08:35 | |
jafeha | evrardjp: we've talked about the telemetry stack, mbe we can help a bit after we've successfully deployed. currently we're debugging ovs. again. :/ | 08:37 |
mardim | evrardjp, Godd morning :) | 08:38 |
*** armaan has quit IRC | 08:38 | |
*** markvoelker has joined #openstack-ansible | 08:38 | |
mardim | good* | 08:38 |
mardim | evrardjp, I noticed yesterday | 08:38 |
mardim | evrardjp, that all the gates were failing | 08:38 |
*** armaan has joined #openstack-ansible | 08:39 | |
mardim | evrardjp, Because of rabbit_mq tasks | 08:39 |
mardim | evrardjp, Are you aware of this ? | 08:39 |
*** vnogin has quit IRC | 08:39 | |
*** savvas has joined #openstack-ansible | 08:41 | |
*** zkynet has quit IRC | 08:41 | |
*** nyloc has quit IRC | 08:43 | |
*** savvas has quit IRC | 08:46 | |
mardim | hwoarang, odyssey4me ^^ | 08:46 |
*** nyloc has joined #openstack-ansible | 08:46 | |
*** zkynet has joined #openstack-ansible | 08:46 | |
*** gkadam_ has joined #openstack-ansible | 08:46 | |
*** savvas has joined #openstack-ansible | 08:49 | |
evrardjp | jafeha: you can help ovs story too :) | 08:56 |
evrardjp | jafeha: I am surprised ppl chose ovs tbh | 08:56 |
jafeha | evrardjp: sure, and ceph story :) | 08:56 |
evrardjp | mardim: I know gates were failing yesterday but I saw a few patches in this night | 08:57 |
evrardjp | do you still have the issue? | 08:57 |
jafeha | evrardjp: yeah, we've had some "all features on" days when we've decided to switch to ovs. it is damn cool, but still very hard to get it running | 08:58 |
*** armaan has quit IRC | 08:58 | |
evrardjp | jafeha: do you do flow manipulation? | 08:59 |
evrardjp | or do you have a sdn controller? | 08:59 |
jafeha | we'd love to use sdn controllers, yes. | 09:00 |
*** thorst has joined #openstack-ansible | 09:00 | |
mardim | evrardjp, I upload a patch just now I am waiting for the gates to run | 09:01 |
mardim | will see ;) | 09:01 |
evrardjp | mardim: ok | 09:01 |
evrardjp | if you see a gate problem, ping me anytime | 09:01 |
mardim | evrardjp, sure thanks | 09:01 |
evrardjp | if you can't reach me, please file a bug | 09:01 |
mardim | Yes ok | 09:02 |
mardim | aslo I want to ask you something more | 09:02 |
evrardjp | wow! | 09:02 |
mardim | I want to ask you where I can Document the variables which the user should include | 09:02 |
evrardjp | how dare you! | 09:02 |
evrardjp | :p | 09:02 |
evrardjp | depends | 09:02 |
mardim | in the user_variables.yaml | 09:02 |
evrardjp | on which role? | 09:02 |
mardim | neutron role | 09:02 |
evrardjp | then in neutron docs | 09:03 |
mardim | hahahah sorry master :P | 09:03 |
evrardjp | I mean os_neutron role | 09:03 |
*** pbandark has joined #openstack-ansible | 09:03 | |
evrardjp | yeah master branch is fine | 09:03 |
mardim | evrardjp, ok thanks | 09:03 |
evrardjp | mardim: https://docs.openstack.org/openstack-ansible-os_neutron/latest/ | 09:03 |
mardim | no master branch I mean you are the MAster :P | 09:03 |
*** chyka has joined #openstack-ansible | 09:03 | |
*** sxc731 has joined #openstack-ansible | 09:04 | |
*** vnogin has joined #openstack-ansible | 09:04 | |
odyssey4me | evrardjp posed a question in https://review.openstack.org/#/c/523854/2/group_vars/galera_all.yml which has already merged. | 09:05 |
*** thorst has quit IRC | 09:05 | |
*** nshetty|lunch is now known as nshetty | 09:05 | |
*** sxc731 has quit IRC | 09:07 | |
evrardjp | odyssey4me: that's a terrible practice, you're right | 09:08 |
*** sxc731 has joined #openstack-ansible | 09:08 | |
*** chyka has quit IRC | 09:08 | |
evrardjp | odyssey4me: what we should ideally do is use an arbitrary variable for hosts in the role, then override it properly in group vars. | 09:09 |
odyssey4me | evrardjp there is already that variable in the role | 09:09 |
evrardjp | odyssey4me: so I guess here, what we can do is reuse galera_cluster_members, and properly define it in group vars | 09:09 |
mardim | evrardjp, odyssey4me The gates are still failing today check this out http://logs.openstack.org/59/517259/23/check/openstack-ansible-functional-ubuntu-xenial/b14cb05/job-output.txt.gz | 09:09 |
odyssey4me | yes, that would be a good solution | 09:09 |
evrardjp | in the role itself, we should instead get rid of the inventory link by using anything else | 09:09 |
evrardjp | this way it applies for other ppl than us | 09:10 |
odyssey4me | ah, a galera failure: http://logs.openstack.org/59/517259/23/check/openstack-ansible-functional-ubuntu-xenial/b14cb05/job-output.txt.gz#_2017-11-30_08_45_02_381152 | 09:10 |
evrardjp | but I don't think we're ever gonna get there | 09:10 |
odyssey4me | I think the tests repo may need a similar fix | 09:10 |
odyssey4me | lemme push that up | 09:10 |
evrardjp | that is true. We've hardened things ! | 09:11 |
evrardjp | mardim: odyssey4me it's always fun times around milestone 2 | 09:11 |
odyssey4me | hmm, wait a minute - these tests use a single galera - why are they failing? | 09:12 |
*** sxc731 has left #openstack-ansible | 09:12 | |
mardim | evrardjp, the fun is unlimited :) | 09:12 |
*** sboyron has joined #openstack-ansible | 09:13 | |
odyssey4me | I'll fire up a test environment to try and figure out what's going on here. The galera patch should not be affecting a single node environment. | 09:14 |
evrardjp | odyssey4me: there are more than one patch that merged in galera_server | 09:15 |
evrardjp | I think there was one from kevin that changed permissions | 09:15 |
evrardjp | let me check | 09:15 |
odyssey4me | ok, patch(es) | 09:15 |
evrardjp | https://github.com/openstack/openstack-ansible-galera_server/commit/ca054bd8f74ac9f7e4d2a41589faf12806863553 has a permission change | 09:16 |
evrardjp | but I don't think it would be a problem it's more permissions | 09:17 |
evrardjp | odyssey4me: so fun stuff, we are using galera_cluster_members as a variable, and it points to galera_all group nodes | 09:18 |
*** m|y|k has joined #openstack-ansible | 09:18 | |
evrardjp | is there a reason we want to override this? | 09:18 |
evrardjp | I don't see any | 09:18 |
odyssey4me | yes, I'm aware of that | 09:18 |
odyssey4me | the point is that the var is being used in group_vars, with no value set in group_vars | 09:18 |
evrardjp | yeah I agree | 09:19 |
evrardjp | but I am thinking ahead now | 09:19 |
odyssey4me | so I would suggest closing the loop and having the group_vars set the same var and value | 09:19 |
evrardjp | there is NO reason why this variable should exist in the first place | 09:19 |
odyssey4me | well, that var exists because we wanted to abstract the role from the inventory | 09:19 |
odyssey4me | this was a step in that direction | 09:19 |
evrardjp | odyssey4me: well not at all | 09:20 |
evrardjp | because it is directly in the defaults/ groups['galera_all'] that's pure link :p | 09:20 |
evrardjp | The only thing I see, is that it's used to determine the first node | 09:20 |
evrardjp | so it's more reliable than ansible_play_hosts | 09:20 |
odyssey4me | yes, but that was a necessary evil to transition - the transition was never completed | 09:20 |
evrardjp | I'd be enclined to remove this variable, this way we don't have to care about it :p | 09:21 |
odyssey4me | ok, but then replace it with what? | 09:21 |
evrardjp | replace with nothing as it's not used | 09:21 |
openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible-specs master: blueprint spec for opendaylight bgpvpn support https://review.openstack.org/523171 | 09:21 |
evrardjp | or almost not used | 09:21 |
odyssey4me | it is used - it's used whenever there's a need to loop through the cluster members | 09:22 |
evrardjp | every iteration could be replaced with something else, like ansible_play_hosts | 09:22 |
evrardjp | which is live and dynamic and works for everyone | 09:22 |
evrardjp | even ppl outside openstack-ansible | 09:23 |
*** loudgefly has joined #openstack-ansible | 09:23 | |
evrardjp | and if we need to, we override in group vars | 09:23 |
odyssey4me | sure, do whatever you think is best - I didn't mean to create a debate - just to call it out | 09:23 |
odyssey4me | those patches are all intended for backporting, so be aware and consider the implications | 09:24 |
evrardjp | I think we need to go through all our variables at one point, and skim them. Start with group vars, host vars, then role defaults | 09:24 |
evrardjp | and we take the decision if a role is opiniated or can be used everywhere | 09:24 |
odyssey4me | by 'those patches' I mean the ones implementing xinetd and the script to take the node out of the LB pool while syncing | 09:24 |
evrardjp | odyssey4me: I wasn't aware they were meant for backporting. They were not built for that: almost no reno, barely manually tested, no improved automatic test coverage | 09:25 |
odyssey4me | well, then let's get that properly sorted out before they get ported back | 09:27 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Define galera_cluster_members https://review.openstack.org/524107 | 09:28 |
*** udesale has quit IRC | 09:31 | |
*** udesale has joined #openstack-ansible | 09:31 | |
*** thorst has joined #openstack-ansible | 09:32 | |
evrardjp | yeah | 09:34 |
evrardjp | so I have a machine | 09:34 |
evrardjp | GRANT ALL PRIVILEGES ON *.* TO 'root'@'localhost' IDENTIFIED BY PASSWORD | 09:34 |
*** taseer4 is now known as Taseer | 09:34 | |
evrardjp | so it looks ok | 09:34 |
evrardjp | but still my issue is: failed: [infra1 -> 10.1.0.2] (item=localhost) => {"failed": true, "item": "localhost", "msg": "(1044, \"Access denied for user 'root'@'localhost' to database 'keystone'\")"} | 09:35 |
evrardjp | database keystone exists and has inherited the permissions | 09:35 |
evrardjp | there is something dubious there | 09:35 |
*** pester has joined #openstack-ansible | 09:36 | |
*** thorst has quit IRC | 09:37 | |
*** asettle_ has joined #openstack-ansible | 09:38 | |
*** asettle has quit IRC | 09:39 | |
*** EmilienM has quit IRC | 09:39 | |
*** fxpester has quit IRC | 09:39 | |
*** andymccr has quit IRC | 09:39 | |
*** pcaruana has quit IRC | 09:39 | |
*** pcaruana has joined #openstack-ansible | 09:40 | |
*** andymccr has joined #openstack-ansible | 09:40 | |
*** EmilienM has joined #openstack-ansible | 09:41 | |
*** john51 has quit IRC | 09:41 | |
*** strattao has quit IRC | 09:41 | |
*** strattao has joined #openstack-ansible | 09:42 | |
*** john51 has joined #openstack-ansible | 09:42 | |
*** armaan has joined #openstack-ansible | 09:43 | |
*** armaan_ has joined #openstack-ansible | 09:45 | |
*** chason has quit IRC | 09:45 | |
*** chason has joined #openstack-ansible | 09:45 | |
*** armaan has quit IRC | 09:48 | |
*** flaviosr has quit IRC | 09:49 | |
evrardjp | odyssey4me: are you debugging it? | 09:51 |
*** flaviosr has joined #openstack-ansible | 09:51 | |
*** wspthr has quit IRC | 09:54 | |
loudgefly | Hello everyone, I'm getting an error while using glance in a test infrastructure, "glance.common.wsgi ConnectionError: Unable to establish connection: HTTPSConnectionPool(host='10.1.16.61', port=8776): Max retries exceeded with u | 09:55 |
loudgefly | rl: /v2/2ac0be436ef544c3b233a37db25430a7/volumes (Caused by SSLError(SSLError("bad handshake: Error([('SSL routines', 'ssl3_get_server_certificate', 'certificate verify failed')] | 09:55 |
loudgefly | It'clearly a self-signed certificate related issue, | 09:55 |
evrardjp | yeah | 09:55 |
evrardjp | it looks like it | 09:55 |
evrardjp | --insecure? | 09:56 |
evrardjp | maybe we should have let's encrypt for everyone. | 09:56 |
evrardjp | but with all those offline installations, meh. | 09:57 |
loudgefly | I can't find a way to use ss certificates and skip validity check | 09:57 |
evrardjp | you can use your own ssl certificates | 09:57 |
evrardjp | for the validity check, I don't know, I never got your issue, what triggered that issue? Was that an API call to the external api? | 09:57 |
loudgefly | Yes to the ha | 09:58 |
*** wspthr has joined #openstack-ansible | 09:58 | |
loudgefly | external haproxy vip | 09:58 |
*** electrofelix has joined #openstack-ansible | 10:00 | |
*** hw_wutianwei has quit IRC | 10:00 | |
*** armaan has joined #openstack-ansible | 10:01 | |
*** savvas has quit IRC | 10:01 | |
*** yifei has quit IRC | 10:02 | |
*** armaan_ has quit IRC | 10:05 | |
evrardjp | so can't you use --insecure in the call? | 10:06 |
evrardjp | if it's for tests, can you, in the meantime, use the internal API from the utility container? | 10:07 |
*** MasterofJOKers has quit IRC | 10:09 | |
*** MasterofJOKers has joined #openstack-ansible | 10:09 | |
*** flaviosr has quit IRC | 10:09 | |
*** epalper has quit IRC | 10:09 | |
*** xgerman_ has quit IRC | 10:09 | |
*** kklimonda has quit IRC | 10:09 | |
*** berendt has quit IRC | 10:09 | |
*** dgonzalez has quit IRC | 10:09 | |
loudgefly | Oh yes, 'I'm using --insecure, the call is "openstack --insecure image create --container-format bare --disk-format qcow2 --file debian-9.2.2-20171105-openstack-amd64.qcow2 debian-9-openstack-amd64" | 10:10 |
*** flaviosr has joined #openstack-ansible | 10:13 | |
*** epalper has joined #openstack-ansible | 10:15 | |
*** xgerman_ has joined #openstack-ansible | 10:15 | |
*** kklimonda has joined #openstack-ansible | 10:15 | |
*** berendt has joined #openstack-ansible | 10:15 | |
*** dgonzalez has joined #openstack-ansible | 10:15 | |
*** armaan has quit IRC | 10:15 | |
*** udesale has quit IRC | 10:16 | |
odyssey4me | evrardjp - no, busy debugging some other stuff right now | 10:16 |
*** MasterofJOKers has quit IRC | 10:17 | |
jrosser | letsencrypt can fail for internet installs too if you're unlucky enough to be on a high profile letsencrypt blocked domain | 10:18 |
*** MasterofJOKers has joined #openstack-ansible | 10:20 | |
evrardjp | yeah I think it's not a great idea. We should just say to deployers to user proper certificates. | 10:23 |
evrardjp | or they generate them beforehand with whatever they want | 10:23 |
evrardjp | odyssey4me: ok I will have a look | 10:24 |
*** pbandark has quit IRC | 10:25 | |
mardim | evrardjp, odyssey4me If you guys find a solution for the gates problem please ping me to run rechecks :) | 10:26 |
odyssey4me | wow, that's a fun time jrosser | 10:27 |
*** savvas has joined #openstack-ansible | 10:27 | |
jrosser | odyssey4me: yes it turns out letsencrypt have a gigantic blacklist they wont issue for | 10:30 |
*** savvas has quit IRC | 10:32 | |
*** savvas has joined #openstack-ansible | 10:36 | |
*** sboyron has quit IRC | 10:37 | |
*** thorst has joined #openstack-ansible | 10:45 | |
*** masuberu has quit IRC | 10:49 | |
*** thorst has quit IRC | 10:50 | |
*** pedja has joined #openstack-ansible | 10:56 | |
*** stuartgr has joined #openstack-ansible | 11:03 | |
*** chhavi has quit IRC | 11:13 | |
*** evrardjp[m] has quit IRC | 11:13 | |
*** csmart has quit IRC | 11:13 | |
*** zkynet has quit IRC | 11:14 | |
*** thorst has joined #openstack-ansible | 11:24 | |
*** thorst has quit IRC | 11:28 | |
*** sboyron has joined #openstack-ansible | 11:42 | |
*** udesale has joined #openstack-ansible | 11:47 | |
*** thorst has joined #openstack-ansible | 11:55 | |
*** savvas has quit IRC | 11:56 | |
*** sboyron has quit IRC | 11:58 | |
*** thorst has quit IRC | 12:01 | |
*** michelv_ has joined #openstack-ansible | 12:02 | |
*** michelv has quit IRC | 12:02 | |
*** goldenfri has quit IRC | 12:03 | |
*** dave-mccowan has joined #openstack-ansible | 12:06 | |
*** savvas__ has quit IRC | 12:10 | |
*** nshetty is now known as nshetty|afk | 12:11 | |
*** savvas has joined #openstack-ansible | 12:11 | |
*** ThomasS has joined #openstack-ansible | 12:20 | |
ThomasS | hi | 12:21 |
ThomasS | i created ssl certs + key + ca certs and placed them on the deploymnet host | 12:21 |
ThomasS | the i ran | 12:22 |
ThomasS | openstack-ansible haproxy-install.yml --tags haproxy-config | 12:22 |
ThomasS | after adding the 3 entries to user_varibales.yml | 12:22 |
ThomasS | but the certs is not picked up | 12:22 |
ThomasS | what is wrong? | 12:22 |
*** savvas_ has joined #openstack-ansible | 12:23 | |
ThomasS | are they not set by this playbook? | 12:23 |
evrardjp | that doesn't sound right ThomasS | 12:24 |
evrardjp | it should work | 12:24 |
*** evrardjp[m] has joined #openstack-ansible | 12:24 | |
evrardjp | could you paste your user_variables changes somewhere? | 12:25 |
ThomasS | sure | 12:26 |
ThomasS | works! | 12:26 |
ThomasS | it was the --tags haproxy-config | 12:26 |
ThomasS | without the tag it worked :-) | 12:26 |
*** savvas_ has quit IRC | 12:27 | |
ThomasS | what i have seen in hatop | 12:30 |
ThomasS | all 3 adoh container son teh controllers are shown as down | 12:30 |
ThomasS | they are all pingable and up | 12:30 |
*** nshetty|afk is now known as nshetty | 12:30 | |
evrardjp | ThomasS: could you file a bug? | 12:31 |
ThomasS | suere | 12:31 |
evrardjp | Because what you did sounded ok: This certificate copy is not at build time, more at configuration time | 12:31 |
evrardjp | what you did was not wrong :p | 12:31 |
ThomasS | ok but without --tags it worked | 12:32 |
ThomasS | btw where are all the tags listed? | 12:32 |
*** chyka has joined #openstack-ansible | 12:34 | |
*** savvas_ has joined #openstack-ansible | 12:37 | |
*** chyka has quit IRC | 12:38 | |
*** pbandark has joined #openstack-ansible | 12:39 | |
*** zkynet has joined #openstack-ansible | 12:43 | |
ThomasS | is the adoh bug know? | 12:44 |
ThomasS | all containers are up and pingable and haproxy shows them as down | 12:44 |
*** zkynet has quit IRC | 12:45 | |
openstackgerrit | Merged openstack/openstack-ansible master: Allow pip_install to run in repo-server playbook https://review.openstack.org/522022 | 12:45 |
openstackgerrit | Merged openstack/openstack-ansible master: Disable offloading in test by default https://review.openstack.org/523549 | 12:45 |
ThomasS | the connect 8042 is reachable from the haproxy | 12:46 |
*** sboyron has joined #openstack-ansible | 12:46 | |
*** armaan has joined #openstack-ansible | 12:46 | |
mardim | evrardjp, Any news with the problem at the gates ? | 12:48 |
*** pbandark has quit IRC | 12:48 | |
*** zkynet has joined #openstack-ansible | 12:49 | |
Bico_Fino | Morning! | 12:52 |
*** csmart has joined #openstack-ansible | 12:52 | |
openstackgerrit | Merged openstack/openstack-ansible master: Lighter ceph gates https://review.openstack.org/522766 | 12:55 |
openstackgerrit | Merged openstack/openstack-ansible master: Remove jinja2 delimiters in when https://review.openstack.org/523936 | 12:55 |
*** smatzek has joined #openstack-ansible | 12:56 | |
*** m|y|k_ has joined #openstack-ansible | 12:59 | |
*** m|y|k has quit IRC | 13:01 | |
gokhan | ThomasS, which openstack version you installed | 13:03 |
*** zkynet has quit IRC | 13:04 | |
*** m|y|k has joined #openstack-ansible | 13:04 | |
*** thorst has joined #openstack-ansible | 13:05 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest master: Fix tempest plugins installation https://review.openstack.org/524166 | 13:06 |
gokhan | ThomasS, on pike, ocata I think there is a bug on aodh haproxy config. if you remove "http expect status 401" under the aodh_backend in haproxy.cfg, then restart haproxy service I think it will work | 13:06 |
*** m|y|k_ has quit IRC | 13:06 | |
evrardjp | mardim: didn't got the chance to work on it. | 13:07 |
mardim | evrardjp, ok thanks for the update :) | 13:07 |
evrardjp | ThomasS: aodh doesn't get some love, as do the rest of the telemetry stack. If there is an issue, you should probably submit a patch :) | 13:07 |
evrardjp | it should be simple enough if that's haproxy healthchecks | 13:07 |
mardim | evrardjp, I think you should take a look on this patches when you find some time https://review.openstack.org/#/c/523907/ | 13:08 |
mardim | evrardjp, https://review.openstack.org/#/c/523907/ | 13:08 |
evrardjp | ThomasS: https://github.com/openstack/openstack-ansible/blob/master/group_vars/haproxy_all/haproxy.yml#L239 | 13:08 |
mardim | I think we should discuss in some point the refactoring of neutron role | 13:08 |
*** zkynet has joined #openstack-ansible | 13:08 | |
evrardjp | mardim: I am fine with that | 13:08 |
mardim | evrardjp, with what ? | 13:08 |
evrardjp | with a refactor of the role if you need to | 13:09 |
evrardjp | but I think you should write a spec with your ideas | 13:09 |
mardim | evrardjp, I was talking more for a discussion at the ansible meeting so we can see waht we can do | 13:09 |
mardim | evrardjp, Yes that could work too | 13:09 |
evrardjp | mardim: you gave twice the same link | 13:09 |
mardim | op sorry | 13:09 |
evrardjp | mardim: yeah conversations without anything written generally don't work | 13:10 |
mardim | evrardjp, https://review.openstack.org/#/c/523171/ https://review.openstack.org/#/c/523907/ | 13:10 |
mardim | evrardjp, When I will find some time I will do that | 13:10 |
mardim | thanks | 13:10 |
*** sboyron has quit IRC | 13:11 | |
evrardjp | mardim: but to be honest it's not that bad | 13:11 |
evrardjp | I would be happy to hear your concerns first if you don't want to go through the spec | 13:12 |
evrardjp | gokhan: ThomasS patches welcome on this :) | 13:12 |
evrardjp | Morning Bico_Fino | 13:12 |
openstackgerrit | Merged openstack/openstack-ansible stable/pike: Fix in-tree version https://review.openstack.org/522853 | 13:13 |
Bico_Fino | So, additional services like designate, magnum, etc can I setup in the initial deploy or just after setup-openstack playbook? | 13:13 |
*** mcarden has joined #openstack-ansible | 13:14 | |
evrardjp | Bico_Fino: if you're confident you can do it on the first run! | 13:15 |
evrardjp | I tend to say I'd do it iteratively | 13:15 |
evrardjp | then do a last full redeploy when ready | 13:15 |
*** vnogin has quit IRC | 13:15 | |
evrardjp | but technically magnum, designate can be setup in the initial deploy | 13:16 |
evrardjp | it's just that those are complex services, so you may want to test the others before :) | 13:16 |
Bico_Fino | before setup-openstack it complains that keystone isn't running. keystoneauth1.exceptions.http.ServiceUnavailable: Service Unavailable (HTTP 503) | 13:16 |
Bico_Fino | Going try run after setup-openstack | 13:17 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Disable offloading in test by default https://review.openstack.org/524173 | 13:19 |
odyssey4me | evrardjp cloudnull need to add that newline back: https://review.openstack.org/#/c/523541/8 | 13:21 |
*** markvoelker has quit IRC | 13:23 | |
*** markvoelker has joined #openstack-ansible | 13:24 | |
jrosser | does anyone have an example/list of things to do in order to make OSA ceph clients pick up a pre-existing ceph cluster? | 13:24 |
gun1x | anybody did OSA install with proxy ? | 13:26 |
gun1x | http proxy | 13:26 |
evrardjp | jrosser: the old ceph documentation probably has it, so you can probably take that info | 13:27 |
evrardjp | but it's maybe worth sharing in the ceph docs | 13:28 |
evrardjp | our ceph docs* | 13:28 |
*** savvas_ has quit IRC | 13:28 | |
odyssey4me | we should ideally have an example for building with a cluster, and building and attaching to acluster | 13:28 |
evrardjp | yeah | 13:28 |
evrardjp | agreed | 13:28 |
gun1x | i get stuck here: https://bpaste.net/show/36e6f27cceff | 13:28 |
evrardjp | two different user stories | 13:28 |
jrosser | yes the example that builds the cluster makes a lot of things opaque | 13:29 |
evrardjp | gun1x: you should try the offline mode I guess | 13:29 |
jrosser | like where the responsibility for creating pools etc lies | 13:29 |
evrardjp | jrosser: yes I'd like to improve our documentation a little further | 13:29 |
gun1x | evrardjp: what is offline mode ? | 13:29 |
evrardjp | I will speak with spotz tomorrow about it | 13:29 |
evrardjp | gun1x: did you look at https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-limited-connectivity.html ? | 13:30 |
*** ThomasS has quit IRC | 13:30 | |
gun1x | evrardjp: yea, i added env variables | 13:30 |
gun1x | and also the removed checks for certificates | 13:31 |
odyssey4me | and so I discover another subtle difference for Ansible 2.3+ which can cause headaches with the synchronize module: https://gist.github.com/odyssey4me/5817160155cc9ff7bd46aaef87fc2c07 | 13:31 |
gun1x | as described in that link | 13:31 |
evrardjp | we don't speak about pip offline mode there | 13:31 |
evrardjp | darn | 13:31 |
odyssey4me | that was a fun two days | 13:31 |
jrosser | gun1x: that could be no_proxy not listing the lb internal addr? | 13:31 |
gun1x | jrosser: i am using osa-aio | 13:31 |
*** yifei has joined #openstack-ansible | 13:31 | |
evrardjp | gun1x: pip_offline_install: false | 13:31 |
gun1x | jrosser: afaik this gets set by default with aio | 13:31 |
evrardjp | can be modified to pip_offline_install: true | 13:31 |
evrardjp | see https://docs.openstack.org/openstack-ansible-pip_install/latest/ | 13:31 |
gun1x | evrardjp: ok i will try that | 13:32 |
odyssey4me | evrardjp that doesn't help with a proxy | 13:32 |
odyssey4me | that just changes where the download happens | 13:32 |
jrosser | with http proxies it is well worth double checking that the contents of no_proxy is correct | 13:32 |
evrardjp | I guess so | 13:32 |
gun1x | /etc/openstack_deploy/user_variables.yml: no_proxy: "localhost,127.0.0.1" | 13:33 |
gun1x | it's the same for containers | 13:33 |
odyssey4me | gun1x looks like you didn't apply all of https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/user_variables.yml#L120-L138 | 13:34 |
odyssey4me | well, either of the two options there | 13:34 |
jrosser | gun1x: i don't think your no_proxy includes enough things | 13:35 |
evrardjp | yeah it looks like something is missing :) | 13:35 |
evrardjp | how do you access your load balancer internally through your external proxy? Not saying it's impossible... | 13:35 |
*** yifei has quit IRC | 13:36 | |
gun1x | odyssey4me: proxy_env_url: http://username:pa$$w0rd@10.10.10.9:9000/ | 13:36 |
gun1x | this is missing | 13:36 |
gun1x | i have the rest | 13:36 |
evrardjp | gun1x: so the no_proxy address is complete with all the nodes? | 13:37 |
gun1x | evrardjp: no, that has only localhost and 127.0.0.1 | 13:38 |
evrardjp | gun1x: that's your issue | 13:39 |
gun1x | i have to add all containers to no_proxy, even for AIO ? | 13:39 |
odyssey4me | yes, you're adjusting the environment to pass all traffic through a proxy | 13:39 |
*** zkynet has quit IRC | 13:39 | |
evrardjp | as I said how do you access from within your container your internal vip if you're passing all its traffic through the proxy? | 13:39 |
odyssey4me | that will affect everything | 13:39 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Disable offloading in test by default https://review.openstack.org/523541 | 13:39 |
gun1x | can't i just add the management subnet from br-mgmt, where all proxies exist ? | 13:40 |
*** zkynet has joined #openstack-ansible | 13:40 | |
gun1x | *where all containers exist ? | 13:40 |
gun1x | instead of adding all containers ... just saying ... | 13:40 |
jrosser | sadly no | 13:40 |
odyssey4me | gun1x just add the jinja expression as noted in user_variables | 13:40 |
odyssey4me | https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/user_variables.yml#L125 | 13:40 |
odyssey4me | exactly as shown there | 13:41 |
odyssey4me | just without the # prefix | 13:41 |
jrosser | no_proxy does not have the same semantics as http_proxy, and you cannot reliably use wildcards or network ranges | 13:41 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/pike: Update all SHAs for 16.0.5 https://review.openstack.org/524179 | 13:41 |
evrardjp | this time I followed the procedure! | 13:42 |
*** zkynet has quit IRC | 13:42 | |
gun1x | odyssey4me: ok, tyring that, thank you! | 13:42 |
evrardjp | yw :p | 13:43 |
*** zkynet has joined #openstack-ansible | 13:43 | |
evrardjp | odyssey4me: could you have a look at this: https://review.openstack.org/#/c/524166/ it blocks dragonflow team | 13:43 |
evrardjp | also this would be nice: https://review.openstack.org/#/c/523920/ | 13:44 |
evrardjp | and this: https://review.openstack.org/#/c/523468/ to fix translations | 13:44 |
evrardjp | so now I will fix the rest. | 13:44 |
jrosser | odyssey4me: that proxy setting could do with another note to the effect of it all blowing up when the env var length exceeds 1024 chars, which it does with only a modest deployment | 13:49 |
openstackgerrit | Merged openstack/openstack-ansible-repo_server master: Implement pypiserver and pypi proxy cache https://review.openstack.org/521754 | 13:50 |
jrosser | or even fail if the resulting no_proxy_env is too long | 13:50 |
*** woodard has joined #openstack-ansible | 13:51 | |
*** woodard has quit IRC | 13:53 | |
*** woodard has joined #openstack-ansible | 13:53 | |
mgariepy | morning everyone | 13:54 |
*** savvas_ has joined #openstack-ansible | 13:55 | |
*** armaan has quit IRC | 13:56 | |
gun1x | odyssey4me: now it got stuck on TASK [Update apt when proxy is added] | 13:56 |
gokhan | evrardjp, yep ThomasS patches would be good :) | 13:57 |
*** jbadiapa has quit IRC | 13:58 | |
mardim | evrardjp, Sure I will collect my thoughts and I will ping you to discuss this in later time :0 | 13:58 |
mardim | :) | 13:58 |
*** savvas_ has quit IRC | 13:59 | |
jrosser | gun1x: this is one of the things you have to set up yourself before running the deploy | 14:01 |
jrosser | the host mush be generally sorted for use behind a proxy as a pre-requisite | 14:01 |
jrosser | *must | 14:02 |
*** yifei has joined #openstack-ansible | 14:02 | |
gun1x | jrosser: the proxy is working fine. i did an apt update and apt dist-upgrade just before running osa scripts | 14:04 |
jrosser | ok maybe i misunderstand then - pastebin a bit more that just the error log for context would help | 14:05 |
*** jbadiapa has joined #openstack-ansible | 14:05 | |
*** jwitko has quit IRC | 14:05 | |
*** savvas_ has joined #openstack-ansible | 14:07 | |
*** m|y|k has quit IRC | 14:10 | |
*** m|y|k has joined #openstack-ansible | 14:10 | |
*** m|y|k has quit IRC | 14:10 | |
gun1x | jrosser: if it fails again i am pasting the config | 14:12 |
*** ThomasS has joined #openstack-ansible | 14:15 | |
gun1x | yea, it failed | 14:17 |
ThomasS | OSA with proxy does not work | 14:17 |
ThomasS | i also gave up | 14:17 |
gun1x | this time it failed on Install distro packages | 14:17 |
gun1x | ThomasS: yea well i tried multiple ways to configure it and it keeps failing somewhere | 14:18 |
*** jbadiapa has quit IRC | 14:21 | |
jrosser | we've dont a reasonable number of deploys behind a proxy before with ocata, both AIO, mnaio and on-metal | 14:21 |
jrosser | and that did work | 14:21 |
odyssey4me | evrardjp have you come to the bottom of the mysql user denied issue yet? | 14:21 |
evrardjp | not yet, I was sidetracked | 14:21 |
evrardjp | I will work on it now | 14:22 |
evrardjp | I just know which commit causes the issue for now | 14:22 |
evrardjp | according to git bisect at least | 14:22 |
evrardjp | https://github.com/openstack/openstack-ansible-galera_server/commit/ca054bd8f74ac9f7e4d2a41589faf12806863553 | 14:22 |
evrardjp | so i have to double check first | 14:23 |
odyssey4me | yeah, but given that's pretty much a rewrite it's not surprising | 14:23 |
* hwoarang looks around | 14:25 | |
bndzor | cloudnull: yeah, really wierd that deb 8 is broken by default.. but with kernel 4.9 not a single issue and no more nagging about serial etc in dmesg.. all clean + responsiveness is much better | 14:25 |
evrardjp | you had that on role test, right? | 14:26 |
gun1x | bndzor: you run OSA on deb? is that possible ? | 14:26 |
bndzor | was a guest vm in nova | 14:27 |
bndzor | but i cant see how it would be a problem to run openstack in deb | 14:27 |
*** jwitko has joined #openstack-ansible | 14:29 | |
evrardjp | bndzor: that would be possible if there was some ppl willing to work on it. It's probably just a few vars, packages and package repos changes :) | 14:29 |
bndzor | ah OSA.. im blind.. well, that would be possible with some modification if setup with run on metal | 14:29 |
evrardjp | bndzor: you still need to include the proper vars etc. | 14:29 |
evrardjp | so not possible for now | 14:29 |
bndzor | like what vars ? | 14:29 |
evrardjp | but I guess the work to be done is likely small | 14:29 |
evrardjp | like all of them distro based? | 14:30 |
evrardjp | example: https://github.com/openstack/openstack-ansible-os_neutron/tree/master/vars | 14:30 |
bndzor | ye, thats why i said "some modification" :p | 14:30 |
*** esberglu has joined #openstack-ansible | 14:33 | |
*** nshetty has quit IRC | 14:36 | |
*** loudgefly has quit IRC | 14:39 | |
*** pbandark has joined #openstack-ansible | 14:47 | |
*** savvas_ has quit IRC | 14:47 | |
*** rgogunskiy has quit IRC | 14:54 | |
*** pbandark has quit IRC | 14:55 | |
*** sxc731 has joined #openstack-ansible | 14:56 | |
jmccrory | evrardjp the 'Create galera users' task probably needs append_privs, without it these items overwrite each other https://github.com/openstack/openstack-ansible-galera_server/blob/master/tasks/galera_setup.yml#L26 | 14:57 |
evrardjp | haha that's the cause? | 14:57 |
jmccrory | haven't tested, but seen it happen before | 14:57 |
evrardjp | I am trying a revert right now, because it seems something was missing. | 14:57 |
jmccrory | https://review.openstack.org/#/c/483176/ | 14:58 |
evrardjp | I will propose a revert and then I'll do proper testing on the initial patch causing issue + include any fixes. | 14:58 |
evrardjp | jmccrory: so you think it's just db_append_privs: "yes" that's missing from tests? | 14:59 |
jmccrory | probably, these first two were added with the latest change and the last one use to be absent https://github.com/openstack/openstack-ansible-galera_server/blob/master/tasks/galera_setup.yml#L32-L46 | 15:00 |
evrardjp | or you mean in the role itself? I am confused in waht you said | 15:00 |
evrardjp | yes I've seen that | 15:00 |
jmccrory | just that one task, 'Create galera users' | 15:00 |
*** yifei has quit IRC | 15:01 | |
evrardjp | I mean we changed this, so it's risky, and we need to think about it properly | 15:01 |
evrardjp | so the revert sounded like the easier solution until we merge a properly tested version of the first patch | 15:01 |
jmccrory | mysql_user module needs append_privs: true, but not sure why those items were even added since a '%' host would already cover them http://docs.ansible.com/ansible/latest/mysql_user_module.html#options | 15:01 |
evrardjp | I want to do a real investigation of this, to see what are the issue with the "tune-up galera" patchset is. But in the meantime, a revert sounds fast and better. | 15:02 |
*** chhavi has joined #openstack-ansible | 15:02 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_server master: Revert "Tune-up the galera role for efficiency" https://review.openstack.org/524205 | 15:03 |
jmccrory | ok, think upgrades might be broken with that patch too anyway so revert/more testing sounds good. commented here on that https://review.openstack.org/#/c/524086/ | 15:03 |
evrardjp | oh great. Nice catch jmccrory | 15:06 |
evrardjp | maybe we should check versions and after, and see if they have changed. | 15:06 |
evrardjp | fail upgrade test if not | 15:06 |
odyssey4me | also see https://review.openstack.org/521229 - I think perhaps this is going a little too far, or needs a lot more scrutiny | 15:07 |
jmccrory | only issue with that is the same packages are being installed in both releases, both 10.1 and minor | 15:07 |
mhayden | what's our current status on the galera gate shenanigans? | 15:11 |
jmccrory | odyssey4me : yeah, if all nodes get stopped a same time with that it'll cause a lot of issues too and takes some manual steps to get galera back up | 15:12 |
*** udesale has quit IRC | 15:13 | |
odyssey4me | perhaps that efficiency patch should get picked apart and added bit by bit if that's even possible | 15:13 |
odyssey4me | anyway, heading out for a bit | 15:14 |
*** udesale has joined #openstack-ansible | 15:14 | |
*** udesale has quit IRC | 15:14 | |
*** savvas_ has joined #openstack-ansible | 15:14 | |
evrardjp | odyssey4me: I've -W this patch then. | 15:14 |
evrardjp | it went in silently that's not good. | 15:15 |
mhayden | evrardjp: is the revert (https://review.openstack.org/524205) happening because of gate breakage? | 15:15 |
* mhayden assumes so | 15:15 | |
evrardjp | mhayden: we are reverting the patches causing the issue, we'll improve tests to avoid this happy merging from happening | 15:15 |
mhayden | okay | 15:15 |
evrardjp | yes | 15:15 |
evrardjp | I'd like to leave it a few days with what's already IN it | 15:16 |
evrardjp | (we have a wsrep patch, this is only the tune up revert) | 15:16 |
evrardjp | I'd like to leave it a few days before merging something again that breaks, and we should have proper testing of those things | 15:16 |
mhayden | could someone take a look at the ansible-hardening contrib patch? https://review.openstack.org/514385 (besides odyssey4me) ;) | 15:17 |
evrardjp | sure | 15:17 |
*** marst has joined #openstack-ansible | 15:17 | |
*** savvas_ has quit IRC | 15:19 | |
openstackgerrit | Lihi Wishnitzer proposed openstack/openstack-ansible-os_neutron master: Update Dragonflow configuration https://review.openstack.org/523870 | 15:21 |
*** savvas_ has joined #openstack-ansible | 15:22 | |
evrardjp | jmccrory: could you vote on https://review.openstack.org/#/c/524205/ ? | 15:22 |
evrardjp | when you're satisfied it fixes gates ofc | 15:23 |
*** rodolof has joined #openstack-ansible | 15:23 | |
openstackgerrit | Adrien Cunin proposed openstack/openstack-ansible-ops master: Added ability to cleanup venv tgz https://review.openstack.org/523827 | 15:25 |
openstackgerrit | Lihi Wishnitzer proposed openstack/openstack-ansible-os_neutron master: Update Dragonflow configuration https://review.openstack.org/523870 | 15:26 |
*** gouthamr has joined #openstack-ansible | 15:27 | |
evrardjp | Adri2000: you will hate me :p | 15:27 |
evrardjp | but it's already better! | 15:27 |
Adri2000 | evrardjp: ahaha, no tell me what I can improve it I'll happily (try to) do it :) | 15:29 |
evrardjp | :) | 15:29 |
evrardjp | well I like this one better, because it's easier on idempotency and lint tests | 15:30 |
evrardjp | but I'll try something quick when I have time | 15:30 |
Adri2000 | evrardjp: I definetely tried to exclude the current {{ venv_tag }} version directly with find, but couldn't find (erm) a way to do that... | 15:30 |
Adri2000 | agreed it'd be nicer | 15:30 |
*** SmearedBeard has joined #openstack-ansible | 15:31 | |
*** Smeared_Beard has quit IRC | 15:31 | |
*** weezS has joined #openstack-ansible | 15:36 | |
-openstackstatus- NOTICE: if you receieved a result of "RETRY_LIMIT" after 14:15 UTC, it was likely due to an error since corrected. please "recheck" | 15:36 | |
cloudnull | evrardjp: I was testing your patch https://review.openstack.org/#/c/524107/ with https://review.openstack.org/#/c/520673/ - looks like you need to modify to use the ip, not the hostname. | 15:38 |
cloudnull | I commented in your patch | 15:38 |
cloudnull | otherwise it results in a hostname lookup which may or may not based on the inventory hostname | 15:39 |
*** jwitko has quit IRC | 15:40 | |
cloudnull | as it stands it results in http://paste.openstack.org/show/627885 | 15:41 |
evrardjp | cloudnull: the inventory name is in /etc/hosts but I can fix it | 15:41 |
*** sxc731 has quit IRC | 15:42 | |
evrardjp | cloudnull: is that the issue though? | 15:42 |
cloudnull | as soon as I changed hostname to IP, keystone worked | 15:42 |
evrardjp | because these IPs are only used for monitoring, right? | 15:42 |
evrardjp | wow. | 15:43 |
cloudnull | the resulting http://paste.openstack.org/show/627886/ xinetd config is functional in my mnaio | 15:43 |
evrardjp | weird because I just reverted something else and my galera started to work, so I am not sure we are talking about the same issue | 15:44 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Define galera_cluster_members https://review.openstack.org/524107 | 15:46 |
evrardjp | I need more time to demystify all this :) | 15:46 |
cloudnull | I have this in my galera role http://paste.openstack.org/show/627888/ | 15:46 |
cloudnull | and those two prs in osa | 15:47 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Define galera_cluster_members https://review.openstack.org/524107 | 15:47 |
cloudnull | evrardjp: ^ no quotes around 127.0.0.1 | 15:47 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Define galera_cluster_members https://review.openstack.org/524107 | 15:47 |
cloudnull | haha | 15:47 |
evrardjp | yeah I was blind | 15:48 |
*** jbadiapa has joined #openstack-ansible | 15:48 | |
evrardjp | and dumb | 15:48 |
evrardjp | cloudnull: the tune-up is what causing me the issues. | 15:48 |
evrardjp | the rest is fine | 15:48 |
evrardjp | or at least I think it's fine | 15:48 |
cloudnull | what is going on with that change? | 15:49 |
evrardjp | breaking all the role tests | 15:50 |
evrardjp | except itself | 15:50 |
cloudnull | with login issues maybe related to the wsrep changes? | 15:50 |
*** savvas_ has quit IRC | 15:50 | |
cloudnull | https://review.openstack.org/#/c/510909/ | 15:51 |
cloudnull | that test seems to work | 15:51 |
evrardjp | It is indeed connection issues | 15:51 |
cloudnull | besides http://logs.openstack.org/09/510909/6/check/openstack-ansible-functional-opensuse-423/2590a43/job-output.txt.gz#_2017-10-31_18_39_37_235754 which I think can be bandained with "https://review.openstack.org/#/c/523525/" | 15:52 |
evrardjp | let me rephrase something: | 15:52 |
evrardjp | where is this mysqlchk xinetd service used | 15:53 |
*** zkynet has quit IRC | 15:53 | |
evrardjp | I thought it was only used when you reach it | 15:53 |
evrardjp | for monitoring* | 15:53 |
cloudnull | via haproxy | 15:53 |
evrardjp | so I thought the usual galera process are working fine without it | 15:53 |
evrardjp | ok | 15:54 |
evrardjp | so I am not mistaken | 15:54 |
evrardjp | let's not confuse things :) | 15:55 |
odyssey4me | cloudnull evrardjp currently all role tests are busted with http://logs.openstack.org/59/517259/23/check/openstack-ansible-functional-ubuntu-xenial/b14cb05/job-output.txt.gz#_2017-11-30_08_45_02_381152 which appears to be since the tune-up patch landed | 15:55 |
odyssey4me | then in the chat history jmccrory raised some issues too which would eb good to address | 15:55 |
evrardjp | odyssey4me: I did a git bisect, and it has proven me the tune-up patch is the cause | 15:55 |
evrardjp | I think it's worth reverting, and improving our test coverage , that's it | 15:56 |
mgariepy | anyone setup saml2 on pike ? I cannot get it to work and i'm wondering why, I'm testing on AIO inside a vm but keystone doesn't followup :5000/Shibboleth.sso/Login , always return a 404. | 15:56 |
hwoarang | cloudnull: actually the opensuse log you posted is a bit weird. let me investigate | 15:57 |
evrardjp | mgariepy: shib2? | 15:57 |
hwoarang | ah that's a very old log | 15:58 |
mgariepy | evrardjp, https://docs.openstack.org/openstack-ansible-os_keystone/pike/configure-federation-sp.html | 15:58 |
evrardjp | mgariepy: that's not helping me | 16:00 |
evrardjp | I am just wondering if you have shibboleth2 installed by default on ubuntu16. | 16:00 |
mgariepy | yes | 16:01 |
evrardjp | mmm. | 16:01 |
evrardjp | ok | 16:01 |
mgariepy | i want to configure the deployment as an SP. | 16:01 |
evrardjp | yeah I got that part :) | 16:01 |
evrardjp | Did you check the apache config? | 16:01 |
mgariepy | i'm wondeing if the natted stuff is causing me issues. | 16:01 |
evrardjp | to see if the mod_shib is enabled, and well configured? | 16:01 |
evrardjp | mgariepy: oh yeah probably | 16:02 |
evrardjp | well | 16:02 |
evrardjp | depends on what you mean | 16:02 |
mgariepy | mod_shib is enabled and configured. | 16:02 |
evrardjp | but it will not help you later, as the DS (if any, else the IdP) needs direct https access to the SP | 16:02 |
mgariepy | well. enabled at least. | 16:02 |
evrardjp | good start | 16:02 |
evrardjp | and the endpoints? | 16:03 |
evrardjp | It's been a while I didn't touch those technologies | 16:03 |
mgariepy | me too lol | 16:04 |
mgariepy | well it was working on kilo. | 16:04 |
evrardjp | haha I know the feeling :p | 16:05 |
mgariepy | my endpoint is set to v3. | 16:06 |
mgariepy | i wanted to help a guys i work with.. .. thought it would be quite easy and fast to deploy.. haha | 16:06 |
mgariepy | well i was mistaken | 16:06 |
evrardjp | :) | 16:07 |
evrardjp | shibboleth has received no love in the community. 0. nada :) | 16:07 |
evrardjp | on top of that, keystone changed I guess | 16:07 |
evrardjp | mgariepy: did you check latest videos of the summit? | 16:08 |
evrardjp | I think there was a few ones about federation | 16:08 |
mgariepy | kk | 16:08 |
mgariepy | i'm starting to hate what is not automatically tested on every commit ;) | 16:08 |
mgariepy | it's often broken. | 16:09 |
*** sxc731 has joined #openstack-ansible | 16:09 | |
evrardjp | yeah | 16:10 |
evrardjp | if there is no test, we can't say it's working :p | 16:10 |
evrardjp | that's why I'd like to see this kinda of new scenarios | 16:10 |
evrardjp | I'd be happy to have one with the federation | 16:10 |
mgariepy | i would prefer to have one with LDAP first haha | 16:11 |
evrardjp | or one with OVS :) | 16:11 |
evrardjp | haha true | 16:11 |
evrardjp | who needs LDAP though, when you can have sql. | 16:11 |
evrardjp | :D | 16:11 |
mgariepy | well. i just don't want to be responsible for the accounts. | 16:12 |
mgariepy | so ldap is great for that.. | 16:12 |
evrardjp | RBAC to someone? :p | 16:12 |
evrardjp | policy are great :) | 16:12 |
evrardjp | but yes I got your point | 16:15 |
evrardjp | I am trolling because I don't really like ldap myself :p | 16:15 |
*** jwitko has joined #openstack-ansible | 16:16 | |
*** savvas_ has joined #openstack-ansible | 16:17 | |
mgariepy | I don't like ldap, but i dont want to have to manage the 30k accounts with password reset and so on in sql. | 16:17 |
mgariepy | :P | 16:17 |
evrardjp | well funny story, I had more issues with federation log-offs than I could ever have password resets failures in SQL :D | 16:21 |
evrardjp | but i 100% agree | 16:21 |
evrardjp | :) | 16:21 |
*** holmsten has quit IRC | 16:21 | |
evrardjp | who doesn't like a good troll while gates are busy? | 16:21 |
*** savvas_ has quit IRC | 16:22 | |
odyssey4me | mgariepy I did some work recently on that stuff and can confirm it works for newton. I expect that perhaps some of the newton+ changes might have broken some things. | 16:22 |
odyssey4me | And yes, I would love to have an on-commit test for it. | 16:22 |
mgariepy | odyssey4me, were you doing it on a natted env ? | 16:23 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron master: [DNM] Gate testing https://review.openstack.org/524239 | 16:23 |
odyssey4me | mgariepy hmm, that shouldn't matter if your config and host headers are right | 16:24 |
evrardjp | odyssey4me: do you know if it's tested in keystone gates? | 16:24 |
odyssey4me | evrardjp some tests were going in - k2k only as far as I recall | 16:25 |
odyssey4me | mgariepy this is my old reference for making sure the public endpoint is usable: https://gist.github.com/odyssey4me/4af6a759b7ce1a4df9b36df412f57f0a | 16:25 |
odyssey4me | it's important for the ssl cert to use the name for the keystone endpoint, and for horizon to use that as an endpoint if you're using websso | 16:26 |
*** woodard has quit IRC | 16:26 | |
odyssey4me | this stuff should still be mostly good: https://docs.openstack.org/openstack-ansible-os_keystone/latest/configure-federation-sp.html | 16:27 |
SamYaple | https://github.com/rabbitmq/rabbitmq-server/releases/tag/v3.7.0 | 16:28 |
SamYaple | w00t | 16:28 |
evrardjp | SamYaple: OH WHAT? | 16:29 |
evrardjp | I just bumped! | 16:29 |
evrardjp | already need of rebumping? :p | 16:29 |
evrardjp | what's the changelog? | 16:29 |
SamYaple | idk. change log is 3.7.0 is released? | 16:31 |
SamYaple | unless you were running the 3.7.0 rc before | 16:32 |
evrardjp | :) | 16:32 |
evrardjp | the question was more: it's a minor version bump, so it's probably a good interesting change | 16:32 |
*** savvas_ has joined #openstack-ansible | 16:32 | |
evrardjp | (loading changelog) | 16:32 |
evrardjp | rabbitmqctl and rabbitmq plugins rewritten from the ground up! | 16:33 |
SamYaple | 3.7.0 is a major bump. they dont follow semver | 16:33 |
evrardjp | helpful error messages! | 16:33 |
evrardjp | SamYaple: yeah that's what I meant with wrong words, once again :) | 16:34 |
evrardjp | oh no I am misreading again? | 16:35 |
*** rodolof has quit IRC | 16:35 | |
*** rodolof has joined #openstack-ansible | 16:36 | |
*** vnogin has joined #openstack-ansible | 16:36 | |
*** chyka has joined #openstack-ansible | 16:39 | |
*** vnogin has quit IRC | 16:42 | |
*** thorst has quit IRC | 16:42 | |
*** thorst has joined #openstack-ansible | 16:42 | |
*** thorst has quit IRC | 16:44 | |
SamYaple | evrardjp: haha i dont know what youre saying | 16:48 |
evrardjp | me neither. | 16:49 |
*** germs has joined #openstack-ansible | 16:49 | |
SamYaple | well 3.7.0 is something ive been waiting for for like a year. so happy its finally here. lots of cool stuff | 16:50 |
*** germs1 has joined #openstack-ansible | 16:51 | |
evrardjp | oh so you know what's in it :) | 16:53 |
evrardjp | I don't see much on the changelog (I guess they need to update it) | 16:53 |
*** germs has quit IRC | 16:54 | |
*** pbandark has joined #openstack-ansible | 16:54 | |
cloudnull | looks like our role tests are failing here: https://github.com/openstack/openstack-ansible-tests/blob/master/create-grant-db.yml#L24-L39 but the db is getting created in the previous task. | 16:58 |
cloudnull | ive got to run to a dr appt but will be back in a few | 16:58 |
*** woodard has joined #openstack-ansible | 17:01 | |
*** pbandark has quit IRC | 17:02 | |
*** TxGirlGeek has joined #openstack-ansible | 17:02 | |
*** woodard has quit IRC | 17:05 | |
*** woodard has joined #openstack-ansible | 17:06 | |
*** savvas_ has quit IRC | 17:08 | |
*** pcaruana has quit IRC | 17:08 | |
spotz | bye cloudnull | 17:08 |
Bico_Fino | I'm getting this warning on galera container -> [Warning] Could not increase number of max_open_files to more than 65535 (request: 80491) | 17:12 |
*** rpittau__ has quit IRC | 17:14 | |
*** gkadam_ has quit IRC | 17:22 | |
evrardjp | cloudnull: yeah it's probably fixable but right now I'd say let's rollback and take our time to fix all the issues we've seen. | 17:23 |
*** RandomTech has joined #openstack-ansible | 17:28 | |
*** Oku_OS is now known as Oku_OS-away | 17:29 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron master: [DNM] Gate testing https://review.openstack.org/524239 | 17:30 |
RandomTech | Hey , quick question. My second playbook failed because it said galera's root password wasnt set, however, i cant find anywhere that says where to configure it in the steps. Anyone know what all passwords need configured and how? | 17:33 |
RandomTech | wait nevermind found it | 17:33 |
Bico_Fino | RandomTech https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/configure.html | 17:34 |
*** savvas_ has joined #openstack-ansible | 17:34 | |
Bico_Fino | You need run python pw-token-gen.py | 17:34 |
evrardjp | RandomTech: :) | 17:35 |
RandomTech | ya we missed the secrets file | 17:36 |
*** savvas_ has quit IRC | 17:38 | |
*** weezS has quit IRC | 17:46 | |
*** savvas_ has joined #openstack-ansible | 17:49 | |
*** michelv_ has quit IRC | 17:52 | |
evrardjp | Bico_Fino: large cluster? | 17:53 |
Bico_Fino | evrardjp 4 nodes | 17:53 |
Bico_Fino | was getting too many connections on mysql | 17:53 |
evrardjp | Generally it shows you have another issue somewhere | 17:53 |
evrardjp | or some configuration need tweaking | 17:54 |
evrardjp | can't really help, sorry | 17:54 |
Bico_Fino | np. I'm increased the max_connections and fixed it | 17:54 |
RandomTech | evrardjp: you there? | 18:02 |
evrardjp | I am still for a few minutes | 18:03 |
evrardjp | how can I help? | 18:03 |
RandomTech | we ran into a an error with galera while running: http://paste.openstack.org/show/627907/ | 18:04 |
RandomTech | im not seeing anything in the user config that mentions galera besides shared infra hosts | 18:05 |
evrardjp | RandomTech: you are running master sir! | 18:07 |
evrardjp | yeah our gates are broken, I am sorry | 18:07 |
evrardjp | for you these should fix it: | 18:07 |
evrardjp | https://review.openstack.org/#/c/524107/4/group_vars/galera_all.yml | 18:07 |
evrardjp | and this: https://review.openstack.org/#/c/524205/ | 18:08 |
evrardjp | so they aren't merged | 18:08 |
evrardjp | either you apply them manually to your environment, or you wait a little :) | 18:08 |
*** thorst has joined #openstack-ansible | 18:08 | |
Bico_Fino | or use a stable branch. :P | 18:08 |
evrardjp | I am sorry for that, that happens sometimes on master, but we should catch most of those in our testing :) | 18:08 |
evrardjp | milestone 2 is where we shove many things, so this and next week will be fun ! | 18:09 |
evrardjp | yeah a stable branch is more reliable :) | 18:09 |
RandomTech | will it hurt things if we go ahead and pull the last tagged and rerun the playbooks | 18:09 |
evrardjp | but I like living on the edge :) | 18:09 |
evrardjp | RandomTech: depends | 18:10 |
evrardjp | getting backwards is generally a bad idea. | 18:10 |
evrardjp | But far backwards is terribad | 18:10 |
evrardjp | and for master, we only have one tag: milestone 1 | 18:11 |
*** chhavi has quit IRC | 18:11 | |
evrardjp | so I'd just apply the patches | 18:11 |
evrardjp | or wait a little | 18:11 |
RandomTech | i believe we got this just the other day | 18:11 |
*** sxc731 has quit IRC | 18:12 | |
*** armaan has joined #openstack-ansible | 18:13 | |
*** epalper has quit IRC | 18:13 | |
evrardjp | I am sorry it's taking some time to fix it | 18:13 |
evrardjp | I do my best | 18:14 |
RandomTech | how do you apply patches? | 18:15 |
evrardjp | your roles are in /etc/ansible/roles | 18:15 |
evrardjp | so you can git clone the role that needs to patch there (pay attention to the proper name) | 18:16 |
evrardjp | then you can go to the links I gate you | 18:16 |
evrardjp | gave* | 18:16 |
evrardjp | there is a download on the top right | 18:16 |
RandomTech | do you think if we hacnt run all three playbooks yet if running 'git clone -b 17.0.0.0b1 https://git.openstack.org/openstack/openstack-ansible \ /opt/openstack-ansible' and the bootstrap will allow us to rerun the first 2 playbooks okay? | 18:19 |
RandomTech | we plan on tearing it down and reinstalling everything after doing it successfully so that we have a nice clean install | 18:21 |
*** mbuil has quit IRC | 18:21 | |
evrardjp | beta 1 is far ago | 18:21 |
evrardjp | I'd destroy the containers | 18:21 |
evrardjp | and start again | 18:21 |
RandomTech | oh, that was the one on the https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/deploymenthost.html | 18:22 |
evrardjp | there is a playbook for that | 18:22 |
evrardjp | you need queens or Pike is enough? | 18:22 |
RandomTech | i honestly dont know the difference | 18:23 |
evrardjp | go for stable | 18:23 |
RandomTech | how would i get the stable one then? the command i posted above? | 18:23 |
evrardjp | if you aren't ready for developing and being on the edge, then for a stable | 18:23 |
evrardjp | no | 18:23 |
evrardjp | 17.0.0.0b1 is the first beta of queens | 18:23 |
evrardjp | you should first destroy your existing containers | 18:24 |
evrardjp | by running the destroy playbook | 18:24 |
evrardjp | did you run galera and things? | 18:24 |
RandomTech | we ran the first 2 playbooks but the second one errored | 18:24 |
evrardjp | which first 2? | 18:25 |
evrardjp | setup-hosts setup-infra? | 18:25 |
RandomTech | setup-hosts and setup-infastructure | 18:25 |
evrardjp | ok | 18:25 |
RandomTech | like it says on the website | 18:25 |
evrardjp | please follow the stable branch documentation instead of the pre-release one: https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/ | 18:25 |
evrardjp | you'll be in a more stable world if you're not ready to develop :) | 18:25 |
RandomTech | ya were not developing openstack | 18:26 |
evrardjp | ok | 18:26 |
evrardjp | so stable/pike it is! | 18:26 |
evrardjp | cleanup your nodes first | 18:26 |
evrardjp | you don't want to have too recent galera or things like that | 18:27 |
evrardjp | if you destroy your containers and destroy everything in /openstack you should be good. | 18:27 |
evrardjp | For safety you can reroll your nodes. | 18:27 |
evrardjp | good luck! | 18:27 |
RandomTech | will our user config file still work or will that need redun? | 18:27 |
evrardjp | they are probably still valid, same for the openstack_user_config | 18:28 |
RandomTech | thats what i meant | 18:28 |
evrardjp | I thought you meant user_* variables files | 18:28 |
evrardjp | but anyway, they should still apply :) | 18:28 |
RandomTech | so we run the destruction playbook, then we pull pike, finally we rerun? | 18:28 |
evrardjp | yeah, run the destruction playbook, remove everything in /openstack on your nodes, then pull pike and rerun everything | 18:29 |
*** weezS has joined #openstack-ansible | 18:29 | |
RandomTech | do you mean /opt/openstack? or did it make a new root directory? | 18:29 |
RandomTech | oh i found it | 18:30 |
RandomTech | is the destruction playbook just in playbooks? | 18:30 |
odyssey4me | lxc-containers-destroy ;) | 18:31 |
odyssey4me | when you run it, be sure to have this pic in your mind https://media.giphy.com/media/JIUYwgWQ5INSU/giphy.gif | 18:31 |
*** savvas_ has quit IRC | 18:34 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Fix galera server local db permission access https://review.openstack.org/524311 | 18:36 |
cloudnull | odyssey4me: evrardjp: jmccrory: ^ that puts the user grants back to what we had before. | 18:36 |
*** stuartgr has quit IRC | 18:37 | |
*** weezS_ has joined #openstack-ansible | 18:42 | |
*** weezS has quit IRC | 18:43 | |
*** weezS_ is now known as weezS | 18:43 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests master: Gather more logs when running tests https://review.openstack.org/523647 | 18:43 |
cloudnull | and that pr should tell us if it's fixed, else I'll remove my -2 and we can revert the tune-up bits. | 18:44 |
odyssey4me | thanks cloudnull - will monitor those | 18:46 |
cloudnull | sorry for the issues. | 18:46 |
odyssey4me | it is better to move forward, but sometimes you have to revert to regroup | 18:46 |
cloudnull | ++ I just didn't want to let it go without taking a crack at getting it fixed. | 18:46 |
odyssey4me | it shows more clearly that our tests are missing something | 18:47 |
odyssey4me | so perhaps we need to figure out what sort of test to add to the galera role to pick this up next time | 18:47 |
cloudnull | I think we can do more cross project work to fill those gaps | 18:47 |
odyssey4me | we could perhaps add a keystone non-voting cross-project test to the galera repo? | 18:48 |
cloudnull | Example https://review.openstack.org/#/c/524059/ - i added the galera server + ssl test to the client role | 18:48 |
*** openstackgerrit has quit IRC | 18:48 | |
*** armaan has quit IRC | 18:48 | |
cloudnull | should help protect against breakage | 18:48 |
RandomTech | Should i delete everything in /openstack or delete /openstack itself? | 18:48 |
cloudnull | we should be able to do the same same elsewhere. | 18:48 |
odyssey4me | nice, although I think cross-project should always be non-voting to help prevent wedges | 18:48 |
odyssey4me | especially on dependent roles | 18:49 |
cloudnull | odyssey4me: ++ that makes sense. I'll update. | 18:49 |
cloudnull | RandomTech: if you're trying to clean up, yes. | 18:49 |
cloudnull | the lxc-container-destroy playbook should cleanup things in that dir | 18:49 |
odyssey4me | RandomTech usually better to re-kick the hosts - it's a better recipe for success in a lab | 18:50 |
odyssey4me | but sure, delete the dir is fine | 18:50 |
RandomTech | re-kick? | 18:50 |
odyssey4me | re-install the OS | 18:50 |
RandomTech | ah | 18:50 |
*** openstackgerrit has joined #openstack-ansible | 18:50 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_client master: Fix SSL handling for galera clusters https://review.openstack.org/524059 | 18:50 |
RandomTech | we will the second time we set it up. we just want to make sure we can configure everything correctly and get something running first | 18:50 |
*** pcaruana has joined #openstack-ansible | 18:50 | |
RandomTech | we then tear down and try again | 18:50 |
odyssey4me | RandomTech for a lab env, best to get yourself geared to re-kick early and often | 18:51 |
odyssey4me | so many issues arise when you're trying to re-use a existing host | 18:51 |
odyssey4me | it all depends on what you're trying to work out, but networking issues especiallly linger | 18:52 |
cloudnull | ^ especially when there's been a lot of hacking | 18:52 |
RandomTech | ya were going to rekick multiple times difinately | 18:52 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests master: Gather more logs when running tests https://review.openstack.org/523647 | 18:55 |
cloudnull | ^ fixed stupid bashate failure. | 18:56 |
evrardjp | is the upgrade concern fixed too? | 18:57 |
cloudnull | w/ the perms ? ? | 18:58 |
cloudnull | what's the upgrade concern ? | 18:58 |
*** gkadam has joined #openstack-ansible | 18:58 | |
evrardjp | https://review.openstack.org/#/c/524205/1/tasks/main.yml@a59 | 18:58 |
evrardjp | Ok I am done for today, I will not check anymore, I just trust you guys :) | 19:00 |
*** savvas_ has joined #openstack-ansible | 19:00 | |
*** poopcat has joined #openstack-ansible | 19:00 | |
cloudnull | hahaha. | 19:00 |
*** pbandark has joined #openstack-ansible | 19:00 | |
cloudnull | that's not addressed here. | 19:01 |
cloudnull | however I think that's something we can add in the galera_server playbook | 19:01 |
evrardjp | I still believe it's better to rollback | 19:01 |
cloudnull | for upgrades | 19:01 |
evrardjp | then* | 19:01 |
cloudnull | i disagree. | 19:01 |
evrardjp | because we are keeping a failure hidden for now, and when the time will come and everybody will have forgotten, it will bite us back | 19:01 |
evrardjp | I know | 19:02 |
evrardjp | It's fine, we've agreed to disagreed! | 19:02 |
evrardjp | for once we are in disagreement | 19:02 |
* evrardjp is marking the day to the calendar | 19:03 | |
evrardjp | :p | 19:03 |
evrardjp | Anyway, I trust you to do the right thing for upgrades if you are taking the fix route | 19:03 |
evrardjp | I am gone for today, someone is waiting for me here :) | 19:04 |
evrardjp | sorry! | 19:04 |
*** savvas_ has quit IRC | 19:04 | |
*** poopcat1 has joined #openstack-ansible | 19:05 | |
cloudnull | https://review.openstack.org/#/c/524086/ | 19:06 |
cloudnull | I need to make one change there. | 19:06 |
*** poopcat has quit IRC | 19:06 | |
cloudnull | but I think that will address those upgrade issues | 19:06 |
*** pbandark has quit IRC | 19:07 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Further simplify the upgrade check https://review.openstack.org/524086 | 19:08 |
*** savvas_ has joined #openstack-ansible | 19:09 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Further simplify the upgrade check https://review.openstack.org/524086 | 19:12 |
cloudnull | ^ added a comment. | 19:12 |
SamYaple | cloudnull: since you are ok with additional services (xinetd) we can avoid the need for etcd3 i believe. we just need *someway* for containers to get status from all other containers | 19:14 |
cloudnull | I quite liked the approach you were working on with etcd3 | 19:15 |
cloudnull | it sounded nice at least !) | 19:15 |
SamYaple | cloudnull: are you good with pulling down an etcd3 binary? because they do roll standalone binaries | 19:16 |
odyssey4me | I have to say, using etcd seems a far better approach than the stick-tape we're doing now. | 19:16 |
SamYaple | odyssey4me: actually with the haproxy healthchecking cloudnull is far superior to my stuff, luckily they dont conflict, they compliement eachother | 19:16 |
odyssey4me | etc3 may not be required, depending on what's intended | 19:16 |
cloudnull | I'm good with a static binary | 19:17 |
SamYaple | odyssey4me: we just need any kind of distributed locking | 19:17 |
odyssey4me | well, if a local script stored database state in etcd - and haproxy grabbed that state from its local proxy, that's better than using xinetd | 19:17 |
odyssey4me | although I suppose using the current method is more LB agnostic | 19:18 |
odyssey4me | whereas my suggested alternative would be more specifically a tie-in to haproxy | 19:19 |
SamYaple | odyssey4me: fair point. haproxy could be setup so the backends are dynamic based on the stats in etcd (which expire after a TTL) | 19:20 |
SamYaple | so it oculd know when a backend cant recieve data, and it will remove old backend automagically | 19:20 |
cloudnull | ^ that'd be quite cool | 19:20 |
SamYaple | coolaboration! it makes the world go round | 19:21 |
odyssey4me | yes, instead of an over-the-wire check, it just does a local etcd check | 19:21 |
SamYaple | ill start working on that piece actually. because i want that | 19:21 |
cloudnull | the xinetd things are there and implementing the percona spec but if we can devise a MOAR better way, I say do it! | 19:21 |
SamYaple | cloudnull: yea thats why im totally cool with them. re-use is always good. | 19:22 |
SamYaple | good chance whatever we finalize they will accept as a helper script/service too | 19:22 |
odyssey4me | it'd probably be better for us to retain both methods, at least until people get used to the new etcd service around | 19:22 |
*** xingchao has joined #openstack-ansible | 19:29 | |
RandomTech | Hey anyone know why 'TASK [ansible-hardening : Ensure debsums is installed]' would fail on only my controllers? | 19:30 |
RandomTech | this is the error: http://paste.openstack.org/show/627919/ | 19:31 |
cloudnull | anyone around mind giving https://review.openstack.org/#/c/524081/ a tap? -cc hwoarang | 19:32 |
hwoarang | cloudnull: do you need to change the when conditionals? | 19:37 |
*** electrofelix has quit IRC | 19:38 | |
cloudnull | those variables are noted in the mains defaults | 19:38 |
cloudnull | https://github.com/openstack/openstack-ansible-galera_server/blob/master/defaults/main.yml#L181-L183 | 19:38 |
cloudnull | but I can pull those conditions out if we think it best. | 19:39 |
hwoarang | Yeah but I mean now that you do stuff with the server certificate does it make sense to check the user ones in the when statement? | 19:39 |
cloudnull | I guess I could just move the task https://review.openstack.org/#/c/524081/1/tasks/galera_ssl.yml@32 | 19:39 |
cloudnull | to the galera_ssl_key_create task file. | 19:40 |
cloudnull | and drop those extra conditions. | 19:40 |
*** xingchao has quit IRC | 19:47 | |
cloudnull | hwoarang: I could do something like this http://paste.openstack.org/show/627920 | 19:47 |
cloudnull | opps http://paste.openstack.org/show/627921/ | 19:48 |
cloudnull | was missing the add | 19:48 |
hwoarang | yeah ok that looks ok too | 19:48 |
cloudnull | odyssey4me: https://review.openstack.org/#/c/523647/ | 19:49 |
cloudnull | seems https://review.openstack.org/#/c/524311/ makes things happier | 19:49 |
cloudnull | if folks can give that a look ^ that unblocks the role gates. | 19:51 |
*** savvas_ has quit IRC | 19:51 | |
RandomTech | hey, im getting a strange error when running setup-hosts.yml. anyone able to help me take a look? | 19:52 |
cloudnull | RandomTech: what are you seeing? | 19:53 |
RandomTech | http://paste.openstack.org/show/627919/ | 19:53 |
RandomTech | TASK [ansible-hardening : Ensure debsums is installed] is failing on my controller(Infrastructure) nodes | 19:54 |
RandomTech | producing a strange 503 | 19:54 |
odyssey4me | cloudnull I wonder if jmccrory could perhaps give it a looksee | 19:54 |
mhayden | RandomTech: looking | 19:55 |
mhayden | RandomTech: interesting -- perhaps some shenanigans on ubuntu's mirror there? | 19:55 |
cloudnull | RandomTech: if this is a rebuild check for the file /etc/apt/apt.conf.d/00apt-cacher-proxy | 19:56 |
cloudnull | and remove | 19:56 |
cloudnull | then apt-get update | 19:57 |
cloudnull | ansible -m shell -a 'rm /etc/apt/apt.conf.d/00apt-cacher-proxy' hosts | 19:57 |
RandomTech | on the erroring nodes? | 19:57 |
cloudnull | ansible -m shell -a 'apt-get update' hosts. | 19:57 |
cloudnull | yup | 19:57 |
RandomTech | should i replace hosts with something? | 19:57 |
cloudnull | when the cluster is online it uses an apt proxy to install packages which runs through the repo server. | 19:57 |
cloudnull | if those containers are gone | 19:58 |
cloudnull | then the proxy is too . | 19:58 |
RandomTech | should i replace hosts in that command? | 19:58 |
cloudnull | it's same to remove /etc/apt/apt.conf.d/00apt-cacher-proxy | 19:58 |
cloudnull | we put that file in place. | 19:58 |
cloudnull | hosts is the group within inventory | 19:59 |
cloudnull | so this is what you'd run: `ansible -m shell -a 'rm /etc/apt/apt.conf.d/00apt-cacher-proxy; apt-get update' hosts` | 19:59 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Fix the key lookup names for self-signed SSL https://review.openstack.org/524081 | 20:00 |
cloudnull | hwoarang: ^ | 20:00 |
*** smatzek has quit IRC | 20:00 | |
*** smatzek has joined #openstack-ansible | 20:00 | |
*** smatzek has quit IRC | 20:01 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Further simplify the upgrade check https://review.openstack.org/524086 | 20:04 |
cloudnull | ok, stepping out for a min . | 20:04 |
RandomTech | it seems to be running better cloudnull thanks | 20:05 |
*** sxc731 has joined #openstack-ansible | 20:08 | |
odyssey4me | cloudnull heh, jmccrory noted some good feedback in https://review.openstack.org/524311 - maybe you can claw some security back ;) | 20:13 |
cloudnull | ah cool! - i'll follow on to make things more restrictive once again but with a cross project test. | 20:15 |
cloudnull | thanks odyssey4me evrardjp jmccrory mhayden hwoarang - I appreciate all of the reviews. | 20:15 |
cloudnull | odyssey4me: mind nudging this through https://review.openstack.org/#/c/523647/ | 20:16 |
cloudnull | given the other is now approved | 20:16 |
cloudnull | more logs! | 20:17 |
*** savvas_ has joined #openstack-ansible | 20:18 | |
*** savvas_ has quit IRC | 20:22 | |
*** vnogin has joined #openstack-ansible | 20:24 | |
*** vnogin has quit IRC | 20:25 | |
*** savvas_ has joined #openstack-ansible | 20:25 | |
*** TxGirlGeek has quit IRC | 20:26 | |
*** sboyron has joined #openstack-ansible | 20:35 | |
odyssey4me | cloudnull done! | 20:37 |
cloudnull | thanks odyssey4me | 20:39 |
*** poopcat1 is now known as poopcat | 20:40 | |
*** xingchao has joined #openstack-ansible | 20:48 | |
*** xingchao has quit IRC | 20:48 | |
*** pcaruana has quit IRC | 20:55 | |
SamYaple | so rabbitmq 3.7.0 has a file store *per* vhost | 20:56 |
*** mhayden has quit IRC | 20:56 | |
SamYaple | that removes any advantage to exploding out a rabbitmq cluster per service in my mind | 20:56 |
*** mhayden has joined #openstack-ansible | 20:57 | |
*** gkadam has quit IRC | 21:01 | |
*** savvas_ has quit IRC | 21:09 | |
RandomTech | cloudnull: do you think any rabbit mq files could have been left over during the teardown? | 21:11 |
*** threestrands has joined #openstack-ansible | 21:12 | |
RandomTech | or does anyone else know? | 21:15 |
cloudnull | RandomTech: not normally. | 21:15 |
cloudnull | SamYaple: that, kinda, sucks. | 21:15 |
cloudnull | i wonder what the rational for rthat was? | 21:16 |
RandomTech | we ran the second playbook and it had an error on TASK [rabbitmq_server : Read rabbit cookie] and the join | 21:16 |
cloudnull | hwoarang: http://logs.openstack.org/11/524311/1/gate/openstack-ansible-functional-opensuse-423/bcaef3c/job-output.txt.gz#_2017-11-30_20_13_51_822006 - did you by chance have time to look into that? | 21:17 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server master: Fix the key lookup names for self-signed SSL https://review.openstack.org/524081 | 21:21 |
openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Add more testing and increase build timeout for suse/cent https://review.openstack.org/523525 | 21:21 |
*** pcaruana has joined #openstack-ansible | 21:24 | |
RandomTech | anyone have an idea why it could fail on [rabbitmq_server : Read rabbit cookie] | 21:26 |
*** woodard has quit IRC | 21:31 | |
*** dave-mccowan has quit IRC | 21:35 | |
*** savvas_ has joined #openstack-ansible | 21:36 | |
*** pcaruana has quit IRC | 21:39 | |
RandomTech | okay just rerunning it fixed that error | 21:40 |
*** savvas_ has quit IRC | 21:40 | |
jrosser | RandomTech: https://github.com/openstack/openstack-ansible-rabbitmq_server/commit/3779124eec4386a2a9db88e585e85ded43d85489 | 21:42 |
RandomTech | oh thanks | 21:42 |
RandomTech | still got one more error to fix though | 21:42 |
*** esberglu has quit IRC | 21:43 | |
*** savvas_ has joined #openstack-ansible | 21:44 | |
RandomTech | something about applying rabbitMQ policies | 21:44 |
RandomTech | jrosser: is that going to be commited to 16.0.2 soon? | 21:45 |
jrosser | i suspect it is in 16.0.4? | 21:46 |
RandomTech | i noticed the documentation is labled as 16.0.2 and tells you to install 16.0.1 | 21:48 |
RandomTech | is the documentation out of date? | 21:54 |
*** esberglu has joined #openstack-ansible | 21:55 | |
*** TxGirlGeek has joined #openstack-ansible | 21:56 | |
*** rodolof has quit IRC | 21:56 | |
*** pbandark has joined #openstack-ansible | 22:00 | |
*** RandomTech has quit IRC | 22:03 | |
*** vnogin has joined #openstack-ansible | 22:05 | |
*** gokhan has quit IRC | 22:06 | |
*** gokhan has joined #openstack-ansible | 22:07 | |
*** TxGirlGeek has quit IRC | 22:07 | |
*** pbandark has quit IRC | 22:09 | |
*** vnogin has quit IRC | 22:09 | |
*** TxGirlGeek has joined #openstack-ansible | 22:14 | |
*** marst has quit IRC | 22:19 | |
*** savvas_ has quit IRC | 22:20 | |
*** TxGirlGeek has quit IRC | 22:23 | |
*** jwitko_ has joined #openstack-ansible | 22:25 | |
*** TxGirlGeek has joined #openstack-ansible | 22:27 | |
*** vnogin has joined #openstack-ansible | 22:28 | |
*** savvas_ has joined #openstack-ansible | 22:28 | |
*** jwitko has quit IRC | 22:29 | |
*** savvas has quit IRC | 22:31 | |
*** TxGirlGeek has quit IRC | 22:31 | |
*** TxGirlGeek has joined #openstack-ansible | 22:31 | |
*** jwitko_ has quit IRC | 22:32 | |
*** savvas_ has quit IRC | 22:35 | |
*** savvas has joined #openstack-ansible | 22:36 | |
*** TxGirlGeek has quit IRC | 22:39 | |
*** TxGirlGeek has joined #openstack-ansible | 22:39 | |
*** woodard has joined #openstack-ansible | 22:40 | |
*** TxGirlGeek has quit IRC | 22:43 | |
*** TxGirlGeek has joined #openstack-ansible | 22:43 | |
*** savvas_ has joined #openstack-ansible | 22:47 | |
cloudnull | randomtech: there was an issue which evrardjp fixed. | 22:47 |
cloudnull | the release versions were out of sync | 22:48 |
cloudnull | I think that's in ? | 22:48 |
*** savvas_ has quit IRC | 22:51 | |
xgerman_ | this errro confuses me: http://logs.openstack.org/95/521795/2/check/openstack-ansible-functional-ubuntu-xenial/6aac84a/job-output.txt.gz | 22:53 |
*** woodard has quit IRC | 22:53 | |
*** savvas_ has joined #openstack-ansible | 22:54 | |
*** TxGirlGeek has quit IRC | 22:56 | |
*** TxGirlGeek has joined #openstack-ansible | 22:57 | |
*** weezS has quit IRC | 22:59 | |
cloudnull | hwoarang: https://review.openstack.org/#/c/516002/ -- look at all that opensuse passing the gate :) | 23:02 |
cloudnull | xgerman_: the error http://logs.openstack.org/95/521795/2/check/openstack-ansible-functional-ubuntu-xenial/6aac84a/job-output.txt.gz#_2017-11-30_22_49_34_188686 | 23:02 |
*** TxGirlGeek has quit IRC | 23:02 | |
cloudnull | is being fixed here - https://review.openstack.org/#/c/524311/ | 23:02 |
cloudnull | as soon as the cent and container repos stop being shitty | 23:03 |
xgerman_ | awesome - thanks! | 23:03 |
*** TxGirlGeek has joined #openstack-ansible | 23:04 | |
*** jwitko has joined #openstack-ansible | 23:04 | |
*** vnogin has quit IRC | 23:07 | |
*** TxGirlGeek has quit IRC | 23:08 | |
*** TxGirlGeek has joined #openstack-ansible | 23:08 | |
*** sxc731 has quit IRC | 23:13 | |
*** TxGirlGeek has quit IRC | 23:14 | |
*** TxGirlGeek has joined #openstack-ansible | 23:15 | |
*** vnogin has joined #openstack-ansible | 23:15 | |
*** TxGirlGeek has quit IRC | 23:17 | |
*** TxGirlGeek has joined #openstack-ansible | 23:18 | |
*** weezS has joined #openstack-ansible | 23:19 | |
SamYaple | cloudnull: i wasnt saying the rabbitmq as a bad thing. im saying like the advantages i would hope to see are now part of normal vhost seperation in rabbitmq | 23:23 |
*** savvas_ has quit IRC | 23:24 | |
*** sboyron has quit IRC | 23:37 | |
*** thorst has quit IRC | 23:41 | |
*** savvas_ has joined #openstack-ansible | 23:51 | |
*** weezS has quit IRC | 23:52 | |
*** savvas_ has quit IRC | 23:56 | |
*** vnogin has quit IRC | 23:58 | |
*** chyka_ has joined #openstack-ansible | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!