| *** marst has joined #openstack-ansible | 00:02 | |
| *** MasterOfBugs has joined #openstack-ansible | 00:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/newton: Update galera running check for CentOS https://review.openstack.org/439153 | 00:16 |
|---|---|---|
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon stable/newton: Make long-running tasks use async https://review.openstack.org/439249 | 00:18 |
| *** adrian_otto has joined #openstack-ansible | 00:27 | |
| *** sdake has joined #openstack-ansible | 00:29 | |
| *** sdake_ has joined #openstack-ansible | 00:32 | |
| *** jamesdenton has joined #openstack-ansible | 00:33 | |
| *** sdake has quit IRC | 00:33 | |
| *** jamielennox is now known as jamielennox|away | 00:36 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Target specific host groups instead of 'all' https://review.openstack.org/439250 | 00:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/newton: Implementation Neutron SR-IOV https://review.openstack.org/436225 | 00:38 |
| *** jamielennox|away is now known as jamielennox | 00:44 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Use block-rescue for cache create https://review.openstack.org/439199 | 00:49 |
| *** sacharya has joined #openstack-ansible | 00:51 | |
| *** sdake has joined #openstack-ansible | 00:52 | |
| *** sdake_ has quit IRC | 00:54 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/newton: Use new ansible package module to install distro packages. https://review.openstack.org/439254 | 00:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/newton: Install packages in one step https://review.openstack.org/439255 | 00:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/ocata: Use block-rescue for cache create https://review.openstack.org/439256 | 00:59 |
| *** cuongnv has joined #openstack-ansible | 00:59 | |
| *** weezS has quit IRC | 01:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/newton: Use block-rescue for cache create https://review.openstack.org/439257 | 01:04 |
| *** svayrette has joined #openstack-ansible | 01:04 | |
| *** sdake has quit IRC | 01:06 | |
| *** sacharya has quit IRC | 01:06 | |
| *** sacharya has joined #openstack-ansible | 01:06 | |
| *** mmalik4 has quit IRC | 01:06 | |
| *** svayrette has quit IRC | 01:08 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client stable/newton: Use dynamic includes to save time https://review.openstack.org/439258 | 01:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: Implementation Neutron SR-IOV https://review.openstack.org/436227 | 01:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Install rdo package https://review.openstack.org/438990 | 01:18 |
| odyssey4me | right, that's enough from me - night all! | 01:20 |
| *** jamesdenton has quit IRC | 01:22 | |
| *** jamesdenton has joined #openstack-ansible | 01:28 | |
| *** SerenaFeng has joined #openstack-ansible | 01:28 | |
| *** jamesdenton has quit IRC | 01:30 | |
| *** phalmos_ has quit IRC | 01:31 | |
| *** phalmos has joined #openstack-ansible | 01:34 | |
| *** MasterOfBugs has quit IRC | 01:34 | |
| *** thorst has joined #openstack-ansible | 01:41 | |
| *** SerenaFeng has quit IRC | 01:45 | |
| *** thorst has quit IRC | 01:46 | |
| *** thorst has joined #openstack-ansible | 01:46 | |
| *** thorst has quit IRC | 01:50 | |
| *** jamesdenton has joined #openstack-ansible | 01:53 | |
| *** KarlKloppenborg has quit IRC | 01:54 | |
| *** jamesdenton has quit IRC | 01:57 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Only insert container hosts into lxc_hosts https://review.openstack.org/427855 | 01:58 |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Use an explicit version of urrlib3 https://review.openstack.org/439079 | 01:59 |
| *** manheim has joined #openstack-ansible | 02:05 | |
| *** agrebennikov_ has quit IRC | 02:06 | |
| *** Mahe has quit IRC | 02:08 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/ocata: Update galera running check for CentOS https://review.openstack.org/439149 | 02:08 |
| *** Mahe has joined #openstack-ansible | 02:11 | |
| *** thorst has joined #openstack-ansible | 02:13 | |
| *** woodard has joined #openstack-ansible | 02:22 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Use an explicit version of urrlib3 https://review.openstack.org/439080 | 02:26 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client stable/newton: Use dynamic includes to save time https://review.openstack.org/439258 | 02:26 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: Remove urllib3 dependency https://review.openstack.org/439084 | 02:26 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/ocata: Use an explicit version of urrlib3 https://review.openstack.org/439086 | 02:26 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/ocata: Optimize pip_install https://review.openstack.org/439242 | 02:26 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/newton: Use an explicit version of urrlib3 https://review.openstack.org/439092 | 02:26 |
| *** woodard has quit IRC | 02:27 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/newton: Add Trusty backports repo if it is not enabled https://review.openstack.org/439053 | 02:27 |
| *** ngupta has joined #openstack-ansible | 02:28 | |
| *** thorst has quit IRC | 02:33 | |
| *** thorst has joined #openstack-ansible | 02:33 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_neutron stable/newton: Install packages in one step https://review.openstack.org/439255 | 02:36 |
| mhayden | ^^ no dragonflow in stable/newton | 02:36 |
| mhayden | fixed up that backport | 02:36 |
| *** manheim has quit IRC | 02:37 | |
| *** thorst has quit IRC | 02:38 | |
| *** lucasxu has joined #openstack-ansible | 02:54 | |
| *** ngupta has quit IRC | 02:56 | |
| *** ngupta has joined #openstack-ansible | 02:56 | |
| *** lucasxu has quit IRC | 02:57 | |
| *** Jack_Iv has joined #openstack-ansible | 02:57 | |
| *** ngupta has quit IRC | 03:01 | |
| *** Jack_Iv has quit IRC | 03:02 | |
| *** thorst has joined #openstack-ansible | 03:08 | |
| *** thorst has quit IRC | 03:08 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible master: Move ceph client role to playbooks for faster run time https://review.openstack.org/438671 | 03:10 |
| *** KarlKloppenborg has joined #openstack-ansible | 03:13 | |
| *** KarlKloppenborg has quit IRC | 03:15 | |
| *** KarlKloppenborg has joined #openstack-ansible | 03:19 | |
| *** ngupta has joined #openstack-ansible | 03:26 | |
| *** galstrom_zzz is now known as galstrom | 03:34 | |
| *** thorst has joined #openstack-ansible | 03:39 | |
| *** thorst has quit IRC | 03:40 | |
| *** svayrette has joined #openstack-ansible | 03:44 | |
| *** svayrette has quit IRC | 03:49 | |
| *** thorst has joined #openstack-ansible | 03:49 | |
| *** thorst has quit IRC | 03:50 | |
| *** kamtamtun has joined #openstack-ansible | 04:00 | |
| *** shausy has joined #openstack-ansible | 04:12 | |
| *** weezS has joined #openstack-ansible | 04:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/ocata: Cap the number of worker threads https://review.openstack.org/438007 | 04:21 |
| *** KarlKloppenborg has quit IRC | 04:22 | |
| *** crushil has joined #openstack-ansible | 04:23 | |
| *** pwnall1337 is now known as zz_pwnall1337 | 04:25 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible stable/ocata: Only insert container hosts into lxc_hosts https://review.openstack.org/439378 | 04:25 |
| *** sdake has joined #openstack-ansible | 04:28 | |
| *** jrobinson is now known as jrobinson-afk | 04:29 | |
| *** Amit82 has joined #openstack-ansible | 04:29 | |
| Amit82 | Hi All, | 04:30 |
| Amit82 | I have installed Openstack using OSA 14.0.8 | 04:30 |
| Amit82 | while launching instance, I am facing this error: "No valid host was found. There are not enough hosts available." | 04:30 |
| Amit82 | systemctl status neutron-linuxbridge-agent.service is giving this: http://paste.openstack.org/show/600819/ | 04:31 |
| Amit82 | neutron logs on compute node is showing this: http://paste.openstack.org/show/600877/ | 04:32 |
| Amit82 | Any pointers that what can be the issue | 04:32 |
| *** pmannidi has quit IRC | 04:33 | |
| *** galstrom is now known as galstrom_zzz | 04:34 | |
| *** whiteveil has joined #openstack-ansible | 04:35 | |
| *** galstrom_zzz is now known as galstrom | 04:36 | |
| cloudnull | evenings | 04:37 |
| cloudnull | Amit82: if you execute the command by hand on the compute node referenced do you get anything more from the logs ? | 04:38 |
| cloudnull | Amit82: IE: /openstack/venvs/neutron-14.0.8/bin/neutron-linuxbridge-agent --config-file /etc/neutron/neutron.conf --config-file /etc/neutron/plugins/ml2/linuxbridge_agent.ini --config-file /etc/neutron/plugins/ml2/ml2_conf.ini | 04:40 |
| cloudnull | normally when you see that kinda issue its due to a network misconfiguration | 04:40 |
| cloudnull | ah, i had not seen the logs link. | 04:40 |
| cloudnull | so yes, this is a network config issue. | 04:41 |
| cloudnull | likely you need to remove the eth12 reference "flat-network" from your openstack_user_config.yml file | 04:41 |
| cloudnull | Amit82: if you copied the aio reference file | 04:42 |
| cloudnull | https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L46-L54 | 04:42 |
| *** whiteveil has quit IRC | 04:42 | |
| cloudnull | that section needs to be removed | 04:42 |
| cloudnull | then rerun the os-neutron-install.yml playbook | 04:42 |
| cloudnull | openstack-ansible os-neutron-install.yml --tags neutron-config | 04:42 |
| Amit82 | cloudnull: I am using the example configuration as provided here: https://docs.openstack.org/project-deploy-guide/openstack-ansible/newton/app-config-test.html#test-environment-config | 04:43 |
| *** adrian_otto has quit IRC | 04:44 | |
| cloudnull | yea that same section needs to be removed from the openstack_user_config file | 04:44 |
| cloudnull | the reason is: that section is trying to create a flat network for use within neutron | 04:45 |
| cloudnull | which will attempt to bind to an interface named eth12 | 04:45 |
| cloudnull | if you create that interface name the linux-bridge-agent will start | 04:45 |
| cloudnull | and everything should be ok. | 04:45 |
| cloudnull | or you can remove that section and you should be good to go | 04:46 |
| * cloudnull assuming you didn't mean to use eth12 and a flat network | 04:46 | |
| Amit82 | cloudnull: my /etc/network/interfaces/ is here: http://paste.openstack.org/show/600879/ | 04:46 |
| Amit82 | it is not containing eth12 for br-vlan | 04:46 |
| Amit82 | which you are pointing out to remove | 04:46 |
| cloudnull | yes. | 04:47 |
| Amit82 | so you are saying that even if it is not having eth12, I remove br-vlan section and rerun os-neutron-install | 04:48 |
| Amit82 | Is this correct understanding | 04:49 |
| Amit82 | ? | 04:49 |
| cloudnull | you have to remove the section for eth12 from the /etc/openstack_deploy/openstack_user_config.yml file | 04:49 |
| cloudnull | then rerun | 04:49 |
| *** pmannidi has joined #openstack-ansible | 04:49 | |
| *** thorst has joined #openstack-ansible | 04:50 | |
| Amit82 | cloudnull: ok, I will remove it from openstack_user_config.yml | 04:52 |
| Amit82 | but just for my understanding, why is this error coming up? | 04:53 |
| Amit82 | I am not running AIO, instead I am having Infra and Controller nodes on two separate machines | 04:53 |
| cloudnull | that section in the openstack_user_config.yml is instructing the playbooks to bind to an interface for neutron that does not exist | 04:53 |
| *** thorst has quit IRC | 04:55 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts master: Allow custom resolver during cache prep https://review.openstack.org/439386 | 04:56 |
| Amit82 | cloudnull: running "openstack-ansible os-neutron-install.yml --tags neutron-config" command is sufficient | 04:57 |
| Amit82 | or do I need to run any other command after that | 04:57 |
| Amit82 | ? | 04:57 |
| coolj | Amit82: see step 7 here https://docs.openstack.org/developer/openstack-ansible/mitaka/install-guide/configure-networking.html#network-configuration | 04:57 |
| *** Jack_Iv has joined #openstack-ansible | 04:58 | |
| cloudnull | Amit82: after you run that you should be all good | 04:58 |
| cloudnull | ++ what coolj mentioned | 04:59 |
| cloudnull | that covers most of what I'm referring to | 04:59 |
| *** Jack_Iv has quit IRC | 05:02 | |
| *** poopcat has quit IRC | 05:03 | |
| Amit82 | cloudnull, coolj: I am naive to this Openstack and Openstack-Ansible | 05:07 |
| Amit82 | eth12 as mentioned in br-vlan for flat nw is a virtual eth and this files talks about this interface on infra node or compute node | 05:08 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move ceph client role to playbooks for faster run time https://review.openstack.org/438671 | 05:10 |
| cloudnull | welcome Amit82 :) | 05:10 |
| openstackgerrit | Tom Jose Kalapura proposed openstack/openstack-ansible-os_horizon master: Remove TLS 1.0 and TLS 1.1 from SSLProtocol https://review.openstack.org/439390 | 05:11 |
| coolj | Amit82: just remove those lines cloudnull linked from /etc/openstack_deploy/openstack_user_config.yml and run the openstack-ansible command he gave and it should work. sorry for any confusion from the docs, i was only giving them for reference. | 05:12 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move ceph client role to playbooks for faster run time https://review.openstack.org/438671 | 05:12 |
| Amit82 | thanks coolj and cloudnull | 05:12 |
| cloudnull | ++ Amit82 that is a virtual network and is no needed. | 05:12 |
| cloudnull | if you want a flat network later you're free to add it back | 05:13 |
| Amit82 | I will try removing the section and rerun the playbook | 05:13 |
| *** ngupta has quit IRC | 05:22 | |
| *** ngupta has joined #openstack-ansible | 05:23 | |
| Amit82 | cloudnull, coolj: thanks a ton. It worked | 05:27 |
| Amit82 | Am I able to successfully launch a VM | 05:27 |
| *** ngupta has quit IRC | 05:27 | |
| *** hybridpollo has quit IRC | 05:29 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move pip_install deployment https://review.openstack.org/439397 | 05:31 |
| cloudnull | Amit82: ++ great! | 05:31 |
| *** galstrom is now known as galstrom_zzz | 05:32 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move pip_install deployment https://review.openstack.org/439397 | 05:32 |
| *** crushil has quit IRC | 05:32 | |
| *** pmannidi_ has joined #openstack-ansible | 05:38 | |
| *** sacharya has quit IRC | 05:39 | |
| *** gouthamr has quit IRC | 05:40 | |
| *** sacharya has joined #openstack-ansible | 05:40 | |
| *** pmannidi has quit IRC | 05:40 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_keystone master: Bypass LB for service setup https://review.openstack.org/438224 | 05:40 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/newton: Use block-rescue for cache create https://review.openstack.org/439257 | 05:40 |
| openstackgerrit | Merged openstack/openstack-ansible master: Updated from global requirements https://review.openstack.org/439229 | 05:41 |
| openstackgerrit | Merged openstack/openstack-ansible-security stable/newton: Fix issues from new CentOS 7 release https://review.openstack.org/438696 | 05:42 |
| *** sacharya has quit IRC | 05:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/newton: Install packages in one step https://review.openstack.org/439158 | 05:48 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create master: Allow container creation based on variant property https://review.openstack.org/438449 | 05:49 |
| *** Jack_Iv has joined #openstack-ansible | 05:55 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/newton: Fix fact set using ternary filter https://review.openstack.org/439162 | 05:56 |
| *** jrobinson-afk is now known as jrobinson | 05:59 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Provide empty default for cinder_backends https://review.openstack.org/438647 | 06:04 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move ceph client role to playbooks for faster run time https://review.openstack.org/438671 | 06:05 |
| *** rgogunskiy has joined #openstack-ansible | 06:05 | |
| *** adrian_otto has joined #openstack-ansible | 06:06 | |
| *** adrian_otto has quit IRC | 06:07 | |
| *** adrian_otto has joined #openstack-ansible | 06:09 | |
| *** lwiecek has joined #openstack-ansible | 06:09 | |
| *** kamtamtun has quit IRC | 06:10 | |
| *** adrian_otto has quit IRC | 06:13 | |
| *** pradiprwt has quit IRC | 06:13 | |
| *** pradiprwt has joined #openstack-ansible | 06:14 | |
| *** lucasxu has joined #openstack-ansible | 06:25 | |
| *** chhavi has joined #openstack-ansible | 06:31 | |
| prometheanfire | fyi, in ocata the placement service isn't optional | 06:41 |
| *** sdake_ has joined #openstack-ansible | 06:42 | |
| *** svayrette has joined #openstack-ansible | 06:44 | |
| *** sdake has quit IRC | 06:45 | |
| *** phalmos has quit IRC | 06:46 | |
| *** weezS has quit IRC | 06:47 | |
| *** svayrette has quit IRC | 06:48 | |
| prometheanfire | asettle: I think the install docs for nova don't work, ocata requires the placement api and they are not mentioned anywhere (same with the nova_cell0 db) | 06:51 |
| *** sigmavirus has quit IRC | 06:52 | |
| *** thorst has joined #openstack-ansible | 06:52 | |
| *** sigmavirus has joined #openstack-ansible | 06:53 | |
| *** sigmavirus is now known as Guest53153 | 06:54 | |
| *** lucasxu has quit IRC | 06:54 | |
| *** thorst has quit IRC | 06:56 | |
| *** Oku_OS-away is now known as Oku_OS | 07:00 | |
| *** h5t4 has joined #openstack-ansible | 07:07 | |
| *** sdake_ has quit IRC | 07:11 | |
| *** udesale has joined #openstack-ansible | 07:18 | |
| *** Jack_Iv has quit IRC | 07:20 | |
| *** pradiprwt has quit IRC | 07:32 | |
| *** pradiprwt has joined #openstack-ansible | 07:33 | |
| *** vnogin has joined #openstack-ansible | 07:47 | |
| *** jrobinson has quit IRC | 07:52 | |
| *** svayrette has joined #openstack-ansible | 08:06 | |
| *** udesale has quit IRC | 08:10 | |
| *** udesale has joined #openstack-ansible | 08:10 | |
| *** phalmos has joined #openstack-ansible | 08:10 | |
| *** svayrette has quit IRC | 08:11 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Remove ceilometer from the gate https://review.openstack.org/438960 | 08:12 |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Remove ceilometer from the gate https://review.openstack.org/438930 | 08:12 |
| *** dfflanders has quit IRC | 08:12 | |
| *** phalmos has quit IRC | 08:35 | |
| *** Andrew_jedi has joined #openstack-ansible | 08:37 | |
| Amit82 | Hi All, | 08:44 |
| Amit82 | I am new to Openstack and Openstack-Ansible | 08:44 |
| Amit82 | Could you guys provide me some links to understand how to use Openstack Ansible based scripts | 08:45 |
| Amit82 | Some basic as well advanced tutorials/videos etc. would be great | 08:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/ocata: Allow container creation based on variant property https://review.openstack.org/439457 | 08:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/newton: Allow container creation based on variant property https://review.openstack.org/439458 | 08:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Install rdo package https://review.openstack.org/439459 | 08:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: unify interface lookups https://review.openstack.org/438742 | 08:52 |
| *** vnogin has quit IRC | 08:55 | |
| *** oneswig has joined #openstack-ansible | 08:57 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Move pip_install deployment https://review.openstack.org/439113 | 09:02 |
| odyssey4me | prometheanfire I think that feedback for asettle is best given in the openstack manuals channel where the docs team resides | 09:05 |
| odyssey4me | Amit82 I'm sure that you will find plenty of information through the search engine of your choice. | 09:06 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Allow custom resolver during cache prep https://review.openstack.org/439386 | 09:06 |
| *** zerda3 has joined #openstack-ansible | 09:06 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder stable/ocata: Provide empty default for cinder_backends https://review.openstack.org/439475 | 09:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder stable/newton: Provide empty default for cinder_backends https://review.openstack.org/439476 | 09:07 |
| *** karimb has joined #openstack-ansible | 09:16 | |
| Amit82 | odyssey4me: thanks | 09:20 |
| *** T42X has quit IRC | 09:25 | |
| lwiecek | odyssey4me, im trying to install keystone using the script You gave me yesterday and I get such error: "msg": "Failed to find required executable rabbitmqctl". Where do I get & where do I have to put the rabbitmqctl executable? | 09:30 |
| *** vishwanathj has quit IRC | 09:30 | |
| lwiecek | its task [Ensure RAbbitmq vhost] | 09:30 |
| odyssey4me | lwiecek it requires the rabbit container, or a host set to run rabbit - our role will put that there | 09:30 |
| *** vishwanathj has joined #openstack-ansible | 09:31 | |
| lwiecek | btw. is there a simple way to turn off rabbitmq and such installations? I want to automate keystone deploment on physical node | 09:34 |
| odyssey4me | lwiecek sure, just set the is_metal property for the keystone container to 'True' - this can be done for all services | 09:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts stable/ocata: Remove urllib3 dependency https://review.openstack.org/439493 | 09:40 |
| odyssey4me | lwiecek which branch are you using?? | 09:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts stable/newton: Remove urllib3 dependency https://review.openstack.org/439494 | 09:41 |
| asettle | Thanks odyssey4me - prometheanfire - shoot it to the docs ML too if you can. Or raise a bug. I'll need to chase down the nova guys for that. | 09:42 |
| lwiecek | master branch | 09:42 |
| odyssey4me | lwiecek a decription of how to move a service from the host into a container is here: https://docs.openstack.org/developer/openstack-ansible-os_cinder/configure-cinder.html#configuring-cinder-to-use-ceph | 09:43 |
| odyssey4me | you can do the opposite for keystone | 09:43 |
| odyssey4me | this assumes that you're using our integrated build | 09:43 |
| odyssey4me | otherwise the role can be executed with your own custom playbook | 09:43 |
| *** karimb has quit IRC | 09:45 | |
| lwiecek | ok thank You for information | 09:46 |
| *** electrofelix has joined #openstack-ansible | 09:51 | |
| prometheanfire | asettle: thanks | 09:52 |
| prometheanfire | asettle: what's the right irc channel? | 09:53 |
| *** chhavi has quit IRC | 09:54 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-haproxy_server stable/ocata: Avoid using get_url remotely https://review.openstack.org/439508 | 09:56 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-haproxy_server stable/newton: Avoid using get_url remotely https://review.openstack.org/439509 | 09:57 |
| asettle | prometheanfire: openstack-doc :) | 10:02 |
| asettle | and the ML is openstack-docs@lists.openstack.org - might be worth attaching yourself to | 10:02 |
| asettle | Cause if it's something wrong in nova, I gotta chase *nova* down :) | 10:02 |
| odyssey4me | asettle although prometheanfire is right though, the placement API is *required* for ocata | 10:03 |
| prometheanfire | odyssey4me: it's not in any of the install docs :| | 10:03 |
| asettle | Haha I'm not disagreeing, I'm literally just stating that the nova team has to update their docs :) | 10:03 |
| odyssey4me | I know that spotz has been working on training material - not sure if she's connected into the docs community to help out. | 10:03 |
| prometheanfire | but I'll send an email | 10:03 |
| odyssey4me | prometheanfire that's not surprising - it landed in ocata milestone 3, a week before the rc | 10:04 |
| asettle | odyssey4me: she is. But again, this is a nova team thing. We can help for sure, but the *nova* team failed to let us know what was incorrect or not. It's not a complete drama, we just have to chase *nova* down | 10:04 |
| *** phalmos has joined #openstack-ansible | 10:04 | |
| prometheanfire | odyssey4me: in nova itself? | 10:04 |
| odyssey4me | asettle sure thing, now that you've auctioned them all off ;) | 10:04 |
| odyssey4me | prometheanfire yep | 10:04 |
| asettle | hahahahahah omg :( | 10:04 |
| prometheanfire | odyssey4me: that's sad :( | 10:04 |
| *** oneswig has quit IRC | 10:04 | |
| odyssey4me | prometheanfire it caused us much trouble | 10:04 |
| *** svayrette has joined #openstack-ansible | 10:04 | |
| *** karimb has joined #openstack-ansible | 10:05 | |
| odyssey4me | if you need a hand fathoming things out, take a look through our os_nova role for the changes relating to the placement API | 10:05 |
| odyssey4me | they landed quite late | 10:05 |
| prometheanfire | https://github.com/openstack/openstack-ansible-os_nova/commit/966ea269c9acb88052aca8982c37e0ba8a94a207 | 10:05 |
| prometheanfire | you mean that? | 10:05 |
| prometheanfire | I've been using it as crib notes | 10:05 |
| odyssey4me | yep that, then corrections afterwards | 10:07 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_nova/commit/2adb6500fd52620e5b415f4f2f949a98c1e611a3 | 10:07 |
| *** manheim_ has joined #openstack-ansible | 10:07 | |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_nova/commit/d2da078745c8eb2e3f579d4bbdb50bbbe12eb0db | 10:07 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_nova/commit/b9b8e08ac00ef93fca2521ba378cc16df587dad5 | 10:07 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_nova/commit/6def3da611f5029a60b3bae88eeabe35c9e37028 | 10:07 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_nova/commit/1ad0bb78cedece76064c647ebda3882cf9c7c6d0 | 10:08 |
| *** manheim_ has quit IRC | 10:08 | |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_nova/commit/6fc78d941c81a9ee8d5e291efe2c76bcf4ca3d28 | 10:08 |
| odyssey4me | that about covers it | 10:08 |
| odyssey4me | andymccr and logan- are the resident experts | 10:08 |
| odyssey4me | and of course johnthetubaguy lurks a little | 10:09 |
| *** svayrette has quit IRC | 10:09 | |
| prometheanfire | ouch | 10:09 |
| andymccr | prometheanfire: are you having issues getting it going? | 10:09 |
| prometheanfire | I run my own setup, not osa | 10:10 |
| prometheanfire | I just had a problem with this coming out of left field with no docs :P | 10:10 |
| andymccr | prometheanfire: yeah when i was setting it up hte key thing is the order in which you do your db syncs and cell setup. also the configuration for your vhost needs to be right or it doesnt forward the appropriate bits on | 10:11 |
| johnthetubaguy | asettle: yeah, we should work out how to do this better, our idea was do the docs we thought we needed, and give things a release to cook in an optional state, and turn it on if no one complains | 10:11 |
| johnthetubaguy | asettle: clearly no one chases us down till its required, which is fair enough, I know I wouldn't | 10:11 |
| asettle | johnthetubaguy: right so the nova team has not been updating their documentation for the nova-install guide? | 10:12 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Implementation Neutron SR-IOV https://review.openstack.org/436227 | 10:12 |
| asettle | aka, here: https://docs.openstack.org/ocata/install-guide-ubuntu/nova.html | 10:12 |
| asettle | Also johnthetubaguy let's move out of openstack-ansible - this is unrelated to here :) | 10:12 |
| johnthetubaguy | asettle: probably not, we always forget that exists | 10:12 |
| johnthetubaguy | asettle: yeah | 10:12 |
| prometheanfire | vhost should be good, using uwsgi/nginx atm and should be forwarding everything | 10:12 |
| *** cuongnv has quit IRC | 10:12 | |
| *** manheim_ has joined #openstack-ansible | 10:14 | |
| *** vnogin has joined #openstack-ansible | 10:15 | |
| andymccr | prometheanfire: make sure you're forwarding only /placement (depending on how you setup your endpoint) | 10:15 |
| andymccr | or copy what we have in the nginx conf for placement service in OSA | 10:15 |
| *** pcaruana has joined #openstack-ansible | 10:24 | |
| *** Jeffrey4l has quit IRC | 10:30 | |
| openstackgerrit | David Rabel proposed openstack/openstack-ansible-security stable/newton: Prepare vagrant centos/7 box for role testing https://review.openstack.org/439529 | 10:42 |
| *** Jeffrey4l has joined #openstack-ansible | 10:49 | |
| *** stuartgr has joined #openstack-ansible | 10:49 | |
| *** chhavi has joined #openstack-ansible | 10:51 | |
| *** thorst has joined #openstack-ansible | 10:53 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Implementation Neutron SR-IOV https://review.openstack.org/436224 | 10:57 |
| *** thorst has quit IRC | 10:58 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts master: Allow extra options to be passed to the cache creation https://review.openstack.org/439201 | 10:59 |
| pjm6 | morning all | 11:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts master: Allow extra options to be passed to the cache creation https://review.openstack.org/439201 | 11:01 |
| *** esberglu has joined #openstack-ansible | 11:01 | |
| *** Amit82 has quit IRC | 11:02 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts master: Allow variant and extra options to be passed to the cache creation https://review.openstack.org/439201 | 11:03 |
| *** esberglu has quit IRC | 11:06 | |
| *** Fenuks has joined #openstack-ansible | 11:08 | |
| *** Jack_Iv has joined #openstack-ansible | 11:10 | |
| nea1 | getting this error on a newly installed ubuntu 16.04 http://termbin.com/j2d1 | 11:13 |
| nea1 | with the same (systemd) error as on my other system - http://termbin.com/rg26 | 11:16 |
| andymccr | odyssey4me: on the upgrade test thats being setup are we just running "run-upgrades.sh" ? | 11:16 |
| *** udesale has quit IRC | 11:16 | |
| nea1 | could it be that one of the last commits in stable/ocata broke something with systemd / rabbitmq? | 11:16 |
| odyssey4me | andymccr yep, that's the plan | 11:16 |
| odyssey4me | andymccr well, actually no | 11:16 |
| andymccr | odyssey4me: sweet thanks, just trying to figure out those errors jamesdenton raised | 11:16 |
| odyssey4me | it'll deploy the previous stable branch, then execute run-upgrade | 11:16 |
| andymccr | ahh yeah ive done teh stable/newton deploy already :) (on 4 hosts) | 11:17 |
| andymccr | nea1: so the one rabbitmq server isnt up? | 11:18 |
| *** udesale has joined #openstack-ansible | 11:18 | |
| nea1 | yep | 11:19 |
| andymccr | nea1: can you manually start it on that failed host? or get anymore debug info on that? - also when you say newly installed do you mean you have 16.04 on all the hosts, but you just added a new one? | 11:19 |
| Andrew_jedi | andymccr: hello ,,, I am wondering whether nova-docker is supported because it appears that the role repo has not seen any updates https://github.com/openstack-ansible/ansible-role-nova-compute-docker | 11:19 |
| nea1 | nope, getting "Unexpected status from systemd "Failed to connect to bus: No such file or directory\n"" on (re)start | 11:20 |
| andymccr | Andrew_jedi: not sure - that isn't something we work on or in our namespace. | 11:20 |
| nea1 | andymccr: this is a single host setup, and ubuntu 16.04 was installed yesterday | 11:20 |
| andymccr | nea1: ahh ok - so do you have multiple rabbit containers on the same host? | 11:21 |
| nea1 | nope | 11:21 |
| prometheanfire | andymccr: glad I can use osa as a refrence :D got it working at home | 11:22 |
| Andrew_jedi | andymccr: Got it, thanks, so can I safely assume that nova-docker is not supported by OSA? | 11:22 |
| andymccr | Andrew_jedi: yeah although looking at https://github.com/openstack/nova-docker it looks to be gone anyway | 11:23 |
| andymccr | Andrew_jedi: we do have nova-lxd support if you're after a containerized approach | 11:23 |
| andymccr | but it wont be docker | 11:23 |
| Andrew_jedi | andymccr: me gusta lxd ... :D | 11:24 |
| nea1 | andymccr: my config http://termbin.com/e4bd | 11:24 |
| andymccr | Andrew_jedi: haha the Canonical folks maintain it though :) so its still working/up-to-date | 11:24 |
| prometheanfire | just have cinder to go and I'm done updating :D | 11:25 |
| andymccr | nea1: you're going to have to look into why it isn't starting. if we can figure that out we can either find a bug and fix it or decide that it isnt a bug :) but its hard to say without more info | 11:25 |
| andymccr | prometheanfire: nice. i think nova was teh main difficult one tbh the rest is reasonably ok | 11:25 |
| prometheanfire | yep | 11:26 |
| prometheanfire | nova was the only one I had to alter my configs for (I take most defaults) | 11:26 |
| Andrew_jedi | andymccr: Awesome, James Page and his team are maintaining the nova-lxd role. Cool. :) | 11:28 |
| prometheanfire | oh neat | 11:28 |
| prometheanfire | I thought that was dead :P | 11:29 |
| prometheanfire | or was it libvirt-lxc or something | 11:29 |
| nea1 | andymccr: looks like a systemd fail to me http://termbin.com/102m ... getting a lot of "Operation not permitted" | 11:29 |
| prometheanfire | been up too long, ignore me | 11:29 |
| nea1 | or maybe some perms missing on the lxc container | 11:29 |
| Andrew_jedi | nea1: Can i ask what does "rabbitmq_ctl cluster_status" say in your container ? | 11:30 |
| Andrew_jedi | sorry, "rabbitmqctl cluster_status" | 11:31 |
| nea1 | Andrew_jedi: unable to connect to epmd (port 4369) on hetzner1: address (cannot connect to host/port) | 11:31 |
| nea1 | but the port is open | 11:32 |
| Andrew_jedi | nea1: Ok, so no rabbitmwq process running, have you tried "service rabbitmq-server restart" | 11:32 |
| nea1 | yep | 11:33 |
| Andrew_jedi | nea1: Wait, could it me that the entire rabbit cluster is down, have you tried running running the rabbit playbook with "--limit <container-name> " | 11:34 |
| odyssey4me | Andrew_jedi nova_docker is not supported by anyone - it's an old dead project as far as I know | 11:39 |
| nea1 | Andrew_jedi: like this `openstack-ansible rabbitmq-install.yml --limit hetzner1_rabbit_mq_container-d17233cb` | 11:39 |
| Andrew_jedi | odyssey4me: Yes, my bad, I was not aware that the project is dead, andymccr was kind enough to update me on that. | 11:40 |
| odyssey4me | Andrew_jedi yeah, the docker philosophy is not really a good fit for nova | 11:41 |
| odyssey4me | nova is made to build VM's | 11:41 |
| odyssey4me | and nova-lxd works well with that because it builds machine containers | 11:41 |
| odyssey4me | docker does not | 11:41 |
| odyssey4me | that's why magnum was born | 11:41 |
| *** smatzek has joined #openstack-ansible | 11:41 | |
| odyssey4me | so if you want docker on your stack, then implement magnum | 11:41 |
| Andrew_jedi | nea1: If you see rabbitmq process in other containers and none in hetzner1_rabbit_mq_container-d17233cb , then yes. | 11:41 |
| odyssey4me | if you want the option of machine containers, then use nova-lxd | 11:42 |
| *** smatzek has quit IRC | 11:43 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts stable/ocata: Allow custom resolver during cache prep https://review.openstack.org/439547 | 11:43 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts stable/newton: Allow custom resolver during cache prep https://review.openstack.org/439548 | 11:43 |
| *** smatzek has joined #openstack-ansible | 11:43 | |
| *** smatzek_ has joined #openstack-ansible | 11:44 | |
| *** smatzek has quit IRC | 11:44 | |
| pjm6 | I'm doing redudancy testing, and I simulate a shutdown in all my controller nodes | 11:44 |
| nea1 | Andrew_jedi: nope, there is only a server in the rabbitmq container | 11:44 |
| pjm6 | I start controller 1 and galera DB is not running | 11:45 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts stable/newton: Allow custom resolver during cache prep https://review.openstack.org/439548 | 11:45 |
| pjm6 | from the docs, i'm reading that if we have a complete failure and seqno postiive we could restart cluster | 11:45 |
| *** phalmos has quit IRC | 11:46 | |
| pjm6 | https://docs.openstack.org/developer/openstack-ansible/newton/developer-docs/ops-galera-recovery.html | 11:46 |
| odyssey4me | pjm6 yes, if you shut them all down then you have to bootstrap the first one that comes up | 11:48 |
| odyssey4me | ie you have to tell the cluster which one is the master | 11:48 |
| *** phalmos has joined #openstack-ansible | 11:48 | |
| pjm6 | thanks odyssey4me, but to do that, i must follow the steps that are in the "Starting a cluster" right? | 11:49 |
| pjm6 | https://docs.openstack.org/developer/openstack-ansible/newton/developer-docs/ops-galera-start.html ? | 11:49 |
| Andrew_jedi | odyssey4me: gotcha! i have always preferred lxc over docker, it will be fun to try out lxd with nova :) | 11:49 |
| Andrew_jedi | nea1: I mean, you may have 3 rabbitmq containers ? right | 11:50 |
| Andrew_jedi | pjm6: If by any chance, all options are exhausted and the cluster is still down, you can try this http://paste.openstack.org/show/600923/. I always use it as my last option | 11:53 |
| pjm6 | thanks Andrew_jedi you use this configs where? | 11:54 |
| pjm6 | I'm trying all possible scenario to figure out how can I recover in each stae | 11:54 |
| Andrew_jedi | pjm6: http://stackoverflow.com/questions/23356623/mysql-galera-autostart-from-boot-wsrep-new-cluster | 11:55 |
| pjm6 | thanks Andrew_jedi :) | 11:57 |
| nea1 | Andrew_jedi: http://termbin.com/1nnd | 11:57 |
| nea1 | and no, only one | 11:57 |
| Andrew_jedi | nea1: "Your requested URL has been blocked" ... Oops | 11:58 |
| *** manheim_ has quit IRC | 11:59 | |
| Andrew_jedi | nea1: Are you using AIO ? | 11:59 |
| nea1 | no, its not a AIO | 11:59 |
| Andrew_jedi | nea1: AFAIK, you should have 3 rabbitmq containers. | 12:00 |
| *** Jack_Iv has quit IRC | 12:00 | |
| pjm6 | odyssey4me: Andrew_jedi, for instance, the state of my grastate.dat is 2 nodes with the same seqno and other with a lower value | 12:00 |
| *** Jack_Iv has joined #openstack-ansible | 12:00 | |
| pjm6 | in order to don't loose any data, I must restart the cluster in one of the higher nodes, right? | 12:01 |
| nea1 | Andrew_jedi: you mean the termbin.com link? | 12:01 |
| Andrew_jedi | nea1: If you dont have extra hosts, you can try "affinity: galera_container: 3 | 12:01 |
| nea1 | that is working for me, do you have a proxy which blocks stuff? | 12:01 |
| pjm6 | "However, because there is no guarantee that each node has an identical copy of the data, we do not recommend to restart the cluster using the --wsrep-new-cluster command on one node." -> but with this doc I shouldn't use this argument, so in order to start, i must use mysql start? | 12:02 |
| nea1 | Andrew_jedi: nope, everything should be there only once | 12:02 |
| Andrew_jedi | nea1: Yep, is it possible that you can use http://paste.openstack.org ? | 12:02 |
| *** askb has quit IRC | 12:02 | |
| nea1 | can one paste stuff there with nc? :D | 12:03 |
| Andrew_jedi | pjm6: AFAIK, yes | 12:03 |
| pjm6 | thanks, will try that | 12:03 |
| Andrew_jedi | nea1: no idea :) | 12:03 |
| odyssey4me | pjm6 you do actually need to use that argument | 12:04 |
| odyssey4me | that's what the bootstrap tag does for you | 12:04 |
| odyssey4me | and yes, choose the one with the highest transaction number IIRC | 12:04 |
| odyssey4me | https://docs.openstack.org/developer/openstack-ansible/newton/developer-docs/ops-galera-recovery.html | 12:04 |
| pjm6 | yes odyssey4me I'm seeing that docs but what confuses me is | 12:05 |
| pjm6 | in complete failure that says | 12:05 |
| evrardjp | I think at some point I'm gonna adapt the artifacting strategy to also remove ssh key generation | 12:05 |
| Andrew_jedi | odyssey4me: Is it possible to install OSA with just one rabbitmq container and one database container ? | 12:05 |
| pjm6 | "If any single node has a positive seqno value (..)" | 12:05 |
| evrardjp | Andrew_jedi: it's possible | 12:05 |
| odyssey4me | evrardjp it's already done : | 12:05 |
| odyssey4me | :p | 12:05 |
| evrardjp | we are using only one galera node in gates already | 12:05 |
| Andrew_jedi | evrardjp: TIL, thanks!! :) | 12:05 |
| evrardjp | odyssey4me: what do you mean? | 12:06 |
| pjm6 | odyssey4me: so if I use openstack-ansible galera-install.yml --tags galera-bootstrap -l container_with_higher_seqno | 12:06 |
| pjm6 | and to the others start mysql service, it should be ine? | 12:06 |
| Andrew_jedi | nea1: I guess, you should be good with "openstack-ansible rabbitmq-install.yml --limit hetzner1_rabbit_mq_container-d17233cb" | 12:06 |
| nea1 | Andrew_jedi: my config: http://paste.openstack.org/show/Rju8y8MR9JPuQTbZ7Uri/ - journal from rabbitmq container: http://paste.openstack.org/show/XxQnNflpblvHOcrjH0v9/ - output from playbook: http://paste.openstack.org/show/wd96wx3rD86sKKoaYJPe/ | 12:06 |
| evrardjp | we don't use the strategy to cleanup what's inside tasks | 12:06 |
| evrardjp | except the notifyu | 12:06 |
| evrardjp | -u | 12:06 |
| odyssey4me | pjm6 I think so. I'd advise testing in a test environment to validate, but I think that's right. cloudnull can confirm | 12:07 |
| pjm6 | odyssey4me: i'm doing in a test env. wanted to see how recover in case of failure | 12:08 |
| Andrew_jedi | pjm6: you may also want to test the complete failure scenario, it is the most challenging one | 12:08 |
| pjm6 | Andrew_jedi: idk how can I simulate to have all the cluster with seqno -1 | 12:09 |
| Andrew_jedi | pjm6: If you have all db containers in one node and you reboot that node, it can result in complete failure | 12:09 |
| Andrew_jedi | i.e. seqno -1 | 12:09 |
| pjm6 | i have in three different nodes | 12:09 |
| pjm6 | and shuted down all | 12:09 |
| pjm6 | but when I run | 12:09 |
| *** Guest53153 is now known as sigmavirus | 12:10 | |
| pjm6 | the galera-bootstrap in one container | 12:10 |
| *** sigmavirus has quit IRC | 12:10 | |
| *** sigmavirus has joined #openstack-ansible | 12:10 | |
| pjm6 | the seqno passes to -1 | 12:10 |
| Andrew_jedi | nea1: what does lxc-ls -f | grep rabbit" says, you may also rebuild the galera conatainer if you wish to | 12:13 |
| Andrew_jedi | pjm6: ohkay ,,, qnd after that, were you able to recover the cluster ? | 12:14 |
| Andrew_jedi | s/qnd/and | 12:15 |
| pjm6 | Andrew_jedi: idk, i'm doing a different approach. Manually i'm using the other node that were with higher seqno | 12:15 |
| pjm6 | attached that node | 12:15 |
| pjm6 | and doing /etc/init.d/mysql start --wserp-new-cluster | 12:16 |
| pjm6 | waiting for the status, i will post the results at the end of this | 12:16 |
| *** manheim_ has joined #openstack-ansible | 12:16 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/ocata: Install packages in one step https://review.openstack.org/437138 | 12:16 |
| Andrew_jedi | pjm6: In the mysql logs, do you see something like " unable to communicate with x, y nodes " ? | 12:17 |
| pjm6 | it's taking while starting the service, but i think this is normal | 12:17 |
| pjm6 | Andrew_jedi: in the node that failed? | 12:17 |
| Andrew_jedi | yes | 12:17 |
| pjm6 | well i get some errors yes, like | 12:18 |
| pjm6 | bind: Address already in use | 12:18 |
| pjm6 | Faiiled to open channel 'openstack_galera_cluster' | 12:18 |
| pjm6 | 1sec | 12:18 |
| pjm6 | maybe is easier with pastebin xD | 12:19 |
| Andrew_jedi | pjm6: just so i understand, mysql started succeffuly on one of the nodes where seqn was -1 | 12:19 |
| pjm6 | pastebin.com/yVLpWa7A | 12:19 |
| pjm6 | no, it was the following | 12:20 |
| *** phalmos has quit IRC | 12:21 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-repo_server master: Remove config tag for an install task https://review.openstack.org/439564 | 12:22 |
| pjm6 | 1. I shutdown all the infra hosts | 12:22 |
| pjm6 | 2. Started all and see that one node had a lower seqno (node1) and two have the same (node 2 and node 3) | 12:22 |
| pjm6 | 3. I tried to do what docs says: openstack-ansible galera-install.yml --tags galera-bootstrap -l node2 | 12:22 |
| pjm6 | 4. After that, mysql service still was down and the file grastate.dat indicates that node 2 has seqno: -1 | 12:22 |
| pjm6 | 5. Now I'm trying to recover the node 3 manually, i went to controller node 3, attach galera container and do the following: "/etc/init.d/mysql start --wsrep-new-cluster" | 12:22 |
| *** karimb has quit IRC | 12:23 | |
| Andrew_jedi | pjm6: The fun starts when all 3 have seqno: -1 ... :p | 12:25 |
| pjm6 | Andrew_jedi: well this gaves error and become with -1 to lol | 12:25 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rsyslog_server master: Fix inconsistency for tag https://review.openstack.org/439566 | 12:26 |
| pjm6 | it says that if I wantto bootstrap from that node, that i must edit the file and put: safe_to_bootstrap: 1 | 12:26 |
| pjm6 | when you have all seqno: -1 what the procedure? rebuild container? | 12:26 |
| odyssey4me | if you destroy all containers, you will destroy all your data too | 12:29 |
| Andrew_jedi | pjm6: ummm, you can try this http://stackoverflow.com/questions/23356623/mysql-galera-autostart-from-boot-wsrep-new-cluster | 12:29 |
| pjm6 | odyssey4me: yes that was I though. in the docs there's no info regarding this scenario, right? | 12:30 |
| pjm6 | will try that Andrew_jedi :) | 12:30 |
| Andrew_jedi | pjm6: this precisely, http://paste.openstack.org/show/600923/ | 12:30 |
| pjm6 | Andrew_jedi: did you maintain the other options? | 12:33 |
| Andrew_jedi | pjm6: you have to modify it according to your cluster | 12:34 |
| pjm6 | yes, my question was | 12:34 |
| pjm6 | for instance, the wsrep_sst_method is xtrabackup-v2 | 12:35 |
| pjm6 | i should maintain this or use the rsync? | 12:35 |
| nea1 | Andrew_jedi: "hetzner1_rabbit_mq_container-d17233cb RUNNING 1 onboot, openstack 10.0.3.214, 172.29.236.218 -" | 12:35 |
| lwiecek | How do I define {{ openstack_repo_url }} ? It is by default openstack_repo_url: "http://{{ internal_lb_vip_address }}:{{ repo_server_port }}". However, I defined my interal_lb as the same ip with my target vm, and because of that, in step "attempt venv download" i get the following error: | 12:36 |
| Andrew_jedi | pjm6: i am sorry, i am not sure about that. | 12:36 |
| lwiecek | "msg": "An unhandled exception occurred while running the lookup plugin 'url'. Error was a <class 'ansible.errors.AnsibleError'>, original message: Failed lookup url for http://<TARGET_VM_IP>:8181/venvs/master/centos/keystone-master-x86_64.checksum : <urlopen error timed out>" | 12:36 |
| pjm6 | Andrew_jedi: no problem, thanks for the tips :D | 12:37 |
| openstackgerrit | Ravi Gummadi proposed openstack/openstack-ansible-galera_server master: Cap the number of worker threads https://review.openstack.org/439569 | 12:38 |
| Andrew_jedi | pjm6: most welcome :) | 12:38 |
| *** retreved has joined #openstack-ansible | 12:38 | |
| *** Fenuks has quit IRC | 12:41 | |
| *** thorst has joined #openstack-ansible | 12:46 | |
| spotz | asettle: odyssey4me - spent all morning getting help on nova channel still can't create an instance in ocata:( | 12:46 |
| asettle | spotz: yeah fuck sakes. I'm going to start whacking them with a stick. Please hold. | 12:47 |
| *** phalmos has joined #openstack-ansible | 12:47 | |
| spotz | asettle: They were VERY helpful, just not sure what's up. They were looking at my logs and the compute node conf | 12:47 |
| asettle | spotz: yeah they have been good, I'm not at all saying that. It's just that prometheanfire and you mentioning this is not the first time I have heard of it being difficult (the install). "whacking them with a stick" means - encourage to update the documentation and test appropriate, and 'please hold' means - i'm going to try and implement a better doc-updating process with the base project teams | 12:49 |
| asettle | spotz: if you come across a magical elixir solution, please let me know. | 12:49 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova stable/ocata: Split upgrade testing into it's own shell script https://review.openstack.org/438002 | 12:49 |
| asettle | Nova currently does not have a docs liaison as such, only liaison for the api-ref and api-guide. So we need to encourage that first :) | 12:49 |
| pjm6 | well it seems it worked | 12:49 |
| spotz | asettle In hinsight this should have been noticed before Ocata release. It's a big hole | 12:50 |
| pjm6 | but the seqno stills -1 | 12:50 |
| asettle | spotz: completely agree. Which begs the question - why was no one reviewing that portion of the documentation? So, I will be investigating today. | 12:50 |
| spotz | I was only moving to ocata due to a library issue in newton on ubumtu which canonical has fixed for me | 12:50 |
| Andrew_jedi | pjm6: you have to restart mysql process on other nodes and wait for the sync to complete | 12:50 |
| asettle | Are you using Ubuntu, SUSE, or CentOS? | 12:50 |
| asettle | spotz: how was the upgrade path? | 12:50 |
| asettle | spotz: actually we should move this convo to docs | 12:51 |
| asettle | channel, that is | 12:51 |
| spotz | asettle I've been doing scratch installs as that's the course so no upgrading | 12:51 |
| spotz | ok moving then I need to go to the barn:) | 12:51 |
| pjm6 | Andrew_jedi: even if the wsrep_cluster is equal in all ? | 12:51 |
| Andrew_jedi | pjm6: The way i remember this that you use "--wsrep-new-cluster" one only one node and "service mysql restart" only other two nodes. | 12:53 |
| pjm6 | Yes I did that | 12:53 |
| pjm6 | everything seems to be working fine | 12:53 |
| pjm6 | but it seems that | 12:54 |
| pjm6 | seqno stills -1 | 12:54 |
| pjm6 | in that file | 12:54 |
| *** phalmos has quit IRC | 12:58 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Move ceph client role to playbooks for faster run time https://review.openstack.org/438671 | 13:04 |
| mgariepy | morning everyone | 13:11 |
| *** mgariepy has quit IRC | 13:27 | |
| pradiprwt | Hi everyone, Container are not getting ip after restarting of AIO host, Can any one please help me | 13:28 |
| *** jmckind has joined #openstack-ansible | 13:28 | |
| pradiprwt | Hi evrardjp, Is this chage is available in newton https://github.com/openstack/openstack-ansible-os_horizon/blob/stable/newton/templates/horizon_local_settings.py.j2#L728 | 13:31 |
| *** jmckind_ has joined #openstack-ansible | 13:32 | |
| *** jmckind has quit IRC | 13:32 | |
| *** karimb has joined #openstack-ansible | 13:33 | |
| *** mgariepy has joined #openstack-ansible | 13:38 | |
| odyssey4me | pradiprwt it has been backported: https://review.openstack.org/418459 | 13:39 |
| *** gouthamr has joined #openstack-ansible | 13:40 | |
| odyssey4me | pradiprwt but it will inly release with 14.1.0, so if you want it you need to use stable/newton right now until 14.1.0 is tagged | 13:40 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Move ceph client role to playbooks for faster run time https://review.openstack.org/439592 | 13:41 |
| *** klamath has joined #openstack-ansible | 13:46 | |
| *** klamath has quit IRC | 13:47 | |
| *** klamath has joined #openstack-ansible | 13:47 | |
| *** sdake has joined #openstack-ansible | 13:48 | |
| Andrew_jedi | pradiprwt: Sometimes, it is because of lxc-br0 acting strange | 13:54 |
| Andrew_jedi | s/lxc-br0/lxcbr0 | 13:54 |
| *** svayrette has joined #openstack-ansible | 13:54 | |
| Andrew_jedi | there is even a bug report for this | 13:54 |
| nea1 | anyone here with an idea why rabbitmq is failing with 'Unexpected status from systemd "Failed to connect to bus: No such file or directory\n" | 13:56 |
| nea1 | the other containers are fine | 13:56 |
| *** xdfil has quit IRC | 13:56 | |
| *** xdfil has joined #openstack-ansible | 13:56 | |
| *** lwiecek has quit IRC | 13:56 | |
| nea1 | and the only stuff I found saied that systemd may be broken, but reinstalling that doesn't help - on the other hand that was from unrelated projects | 13:57 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-ceph_client stable/ocata: Add CentOS support https://review.openstack.org/439604 | 13:59 |
| *** svayrett1 has joined #openstack-ansible | 14:01 | |
| *** svayrette has quit IRC | 14:01 | |
| *** kjw3 has joined #openstack-ansible | 14:01 | |
| pradiprwt | Andrew_jedi : any way to bring networking back with containers | 14:08 |
| pradiprwt | Here is container list http://paste.openstack.org/show/600956/ | 14:09 |
| *** jamesdenton has joined #openstack-ansible | 14:09 | |
| Andrew_jedi | pradiprwt: umm not 100% sure, but you can try removing lxc-net from /etc/network/interfaces.d/, and running setup-host playbook again | 14:09 |
| odyssey4me | palendae I just hit the bootstrap fail and realised what did it for me - I built a Trusty host and tried using OSA master | 14:10 |
| *** pjh03 has joined #openstack-ansible | 14:10 | |
| Andrew_jedi | pradiprwt: or you could unisntall lxc packages and run setup-hosts playbook again | 14:11 |
| pradiprwt | ok | 14:11 |
| *** rgogunskiy has quit IRC | 14:11 | |
| pradiprwt | let me try this | 14:11 |
| odyssey4me | pradiprwt that entirely depends on what went wrong with them - you'll need to troubleshoot why they're not coming up | 14:11 |
| odyssey4me | attach into the container and see if you can bring it up manually | 14:12 |
| Andrew_jedi | nea1: Have you tried rebuilding the container ? | 14:15 |
| Andrew_jedi | nea1: openstack-ansible lxc-containers-destroy.yml --limit <container>, openstack-ansible lxc-containers-create.yml --limit <container> and and then running rabbit playbook again | 14:17 |
| *** esberglu has joined #openstack-ansible | 14:18 | |
| *** Rodrigo_BR has joined #openstack-ansible | 14:18 | |
| *** esberglu has quit IRC | 14:21 | |
| openstackgerrit | Ravi Kumar Boyapati proposed openstack/openstack-ansible-rsyslog_client master: Logrotate not restarting rsyslog https://review.openstack.org/438193 | 14:22 |
| Rodrigo_BR | I am check the documentation about the upgrade procedure from Mitaka to Newton, And I found this "The upgrade script is still under active development and should not be run on a production environment at this time." | 14:23 |
| odyssey4me | Rodrigo_BR it was confirmed to be working at one time, but we don't have active testing on it | 14:24 |
| odyssey4me | so all that means is that you had best test it for your use case | 14:24 |
| odyssey4me | several community members have upgraded already and were happy with the results | 14:25 |
| *** shausy has quit IRC | 14:25 | |
| Rodrigo_BR | ! I will try this procedure | 14:26 |
| openstack | Rodrigo_BR: Error: "I" is not a valid command. | 14:26 |
| Rodrigo_BR | odyssey4meOk: I will try this procedure | 14:26 |
| *** FrankZhang has quit IRC | 14:26 | |
| *** karimb has quit IRC | 14:27 | |
| Rodrigo_BR | odyssey4me: wich the best way do you consider, execute run-upgrade.sh or manual upgrade ? | 14:28 |
| *** karimb has joined #openstack-ansible | 14:28 | |
| odyssey4me | Rodrigo_BR the manual procedure is meant to just be an outline of what happens in run-upgrade.sh | 14:28 |
| odyssey4me | best is usually to read run-upgrade.sh and execute each step on your own | 14:29 |
| odyssey4me | a blind major upgrade is not often a good idea, although for a small environment it's probably ok | 14:29 |
| odyssey4me | if you execute manually then you can use --limit to scope things | 14:29 |
| *** bauruine has quit IRC | 14:31 | |
| *** svayrett1 has quit IRC | 14:31 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_nova master: [WIP] Reverse placement api task order https://review.openstack.org/439157 | 14:32 |
| Rodrigo_BR | odyssey4me: Ok, At this moment I am using the version 13.3.1 I need apply the minor upgrade before or Can I upgrade to Newton direct ? | 14:32 |
| odyssey4me | Rodrigo_BR you'll have to test with your settings and implementation, as we have no testing we can provide no guarantees | 14:33 |
| pradiprwt | Andrew_jedi : Thanks, It works for me :) | 14:33 |
| *** arturb has joined #openstack-ansible | 14:34 | |
| mgariepy | I like 6.3 of this one. https://www.unix-ag.uni-kl.de/~bloch/acng/html/distinstructions.html#hints-fccore | 14:34 |
| odyssey4me | Rodrigo_BR we're working on implementing upgrade testing and will hopefully have that in place for newton->ocata and ocata->pike | 14:35 |
| *** crushil has joined #openstack-ansible | 14:36 | |
| *** bauruine has joined #openstack-ansible | 14:38 | |
| *** chris_hultin|AWA is now known as chris_hultin | 14:39 | |
| *** esberglu has joined #openstack-ansible | 14:39 | |
| *** esberglu has quit IRC | 14:39 | |
| *** esberglu has joined #openstack-ansible | 14:40 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova stable/ocata: Split upgrade testing into it's own shell script https://review.openstack.org/438002 | 14:40 |
| Rodrigo_BR | don't know how-to test this before ... Can I backup the lxc containers ? There is some way to rollback in failure case ? | 14:40 |
| Rodrigo_BR | odyssey4me: I don't know how-to test this before ... Can I backup the lxc containers ? There is some way to rollback in failure case ? | 14:41 |
| pjm6 | odyssey4me: Andrew_jedi: one question, if the mysql cluster is working fine, what it's the supposed result (seqno) of grastate.dat ? If i stop all mysql I get the same seqno | 14:43 |
| pjm6 | but if I startit, it will give me -1 | 14:43 |
| pjm6 | in that fikle | 14:43 |
| *** vnogin has quit IRC | 14:44 | |
| *** vnogin has joined #openstack-ansible | 14:45 | |
| odyssey4me | Rodrigo_BR you'll have to figure that out - you should be able to build a representative test environment using your config in /etc/openstack_deploy | 14:45 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova stable/ocata: Split upgrade testing into it's own shell script https://review.openstack.org/438002 | 14:49 |
| jamesdenton | Anyone seen the following error when booting instances? {"message": "Remote error: Error [('asn1 encoding routines', 'ASN1_get_object', 'too long'), ('asn1 encoding routines', 'ASN1_CHECK_TLEN', 'bad object header'), ('asn1 encoding routines', 'ASN1_ITEM_EX_D2I', 'nested asn1 error'), ('PEM routines', 'PEM_X509_INFO_read_b", "code": 500, "details": " File \"/openstack/venvs/nova-master/lib/python2.7/site-packages/nova/conductor/manager.py\ | 14:51 |
| jamesdenton | ", line 866, in schedule_and_build_instances | 14:51 |
| *** pjh03 has quit IRC | 14:51 | |
| nea1 | Andrew_jedi: just did so, but it didn't help | 14:53 |
| nea1 | same error again | 14:53 |
| pjm6 | anyone could test if the seqno of mysql cluster is -1 when all of them are running? | 14:56 |
| *** BjoernT has joined #openstack-ansible | 14:56 | |
| *** cjloader_ has joined #openstack-ansible | 14:57 | |
| pjm6 | i'm getting this output: pastebin.com/DZvVqnes | 14:57 |
| pjm6 | and i'm not sure if it's suppose the seqno be -1 | 14:57 |
| openstackgerrit | Merged openstack/openstack-ansible-rsyslog_server master: Fix inconsistency for tag https://review.openstack.org/439566 | 15:05 |
| *** lucasxu has joined #openstack-ansible | 15:05 | |
| *** lucasxu has quit IRC | 15:07 | |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server stable/ocata: Avoid using get_url remotely https://review.openstack.org/439508 | 15:07 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server stable/newton: Avoid using get_url remotely https://review.openstack.org/439509 | 15:07 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts stable/ocata: Remove urllib3 dependency https://review.openstack.org/439493 | 15:08 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts stable/newton: Remove urllib3 dependency https://review.openstack.org/439494 | 15:08 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/newton: Allow custom resolver during cache prep https://review.openstack.org/439548 | 15:08 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/ocata: Allow custom resolver during cache prep https://review.openstack.org/439547 | 15:10 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Allow variant and extra options to be passed to the cache creation https://review.openstack.org/439201 | 15:10 |
| *** lucasxu has joined #openstack-ansible | 15:11 | |
| *** chhavi has quit IRC | 15:11 | |
| *** TxGirlGeek has joined #openstack-ansible | 15:11 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-ceph_client master: Fix CentOS ceph repo key https://review.openstack.org/439644 | 15:12 |
| *** ngupta has joined #openstack-ansible | 15:15 | |
| *** cjloader_ has quit IRC | 15:16 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rsyslog_server stable/ocata: Fix inconsistency for tag https://review.openstack.org/439646 | 15:16 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rsyslog_server stable/newton: Fix inconsistency for tag https://review.openstack.org/439647 | 15:16 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server master: Remove config tag for an install task https://review.openstack.org/439564 | 15:16 |
| *** kamtamtun has joined #openstack-ansible | 15:17 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-ceph_client master: Fix CentOS ceph repo https://review.openstack.org/439644 | 15:19 |
| *** whiteveil has joined #openstack-ansible | 15:19 | |
| mgariepy | I did a really messy job on this one :( https://review.openstack.org/439644 | 15:20 |
| *** TxGirlGeek has quit IRC | 15:20 | |
| *** h5t4 has quit IRC | 15:21 | |
| *** whiteveil has quit IRC | 15:24 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-ceph_client stable/ocata: Add CentOS support https://review.openstack.org/439604 | 15:25 |
| andymccr | jamesdenton: i recreated that nova bug in my upgrade - also it made me realise our in role nova upgrade testing for ocata got left by the wayside. so fixing that up and should uncover that bug | 15:25 |
| *** marst has quit IRC | 15:25 | |
| jamesdenton | thanks andymccr! Which bug in particular? | 15:25 |
| andymccr | the db perms one | 15:26 |
| *** cloudtrainme has joined #openstack-ansible | 15:27 | |
| andymccr | that inventory vs ansible_hostname bug is odd, we may need a bit more info to get that one fixed up because that "should" be the right var (e.g. it should take the hostname var from the host rather than from the inventory) | 15:27 |
| *** TxGirlGeek has joined #openstack-ansible | 15:28 | |
| *** sacharya has joined #openstack-ansible | 15:28 | |
| *** The_Intern has joined #openstack-ansible | 15:28 | |
| jamesdenton | yeah i may need to recreate it. Pretty sure I set a debug var, but i could be wrong | 15:29 |
| *** karimb has quit IRC | 15:30 | |
| andymccr | hmm having a look at the nova-compute maas check we do (which does a similar thing) | 15:30 |
| jamesdenton | k | 15:30 |
| andymccr | we use "ansible_nodename" instead of ansible_hostname | 15:30 |
| *** FrankZhang has joined #openstack-ansible | 15:30 | |
| andymccr | mattt: i remember you did some work on that - whats the main diff there ^? | 15:30 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/newton: Use new ansible package module to install distro packages. https://review.openstack.org/439254 | 15:31 |
| *** cloudtrainme has quit IRC | 15:32 | |
| mattt | andymccr: bro | 15:32 |
| mattt | andymccr: i have two toddlers | 15:32 |
| mattt | andymccr: i can hardly remember stuff from last week, let alone 2 years ago | 15:32 |
| *** marst has joined #openstack-ansible | 15:32 | |
| andymccr | mattt: well if they know the answer, ask them :D | 15:32 |
| mattt | :P | 15:33 |
| mattt | andymccr: i don't remember, but i suspect it's documented in a commit message | 15:33 |
| *** karimb has joined #openstack-ansible | 15:34 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova master: Adjust ansible_hostname to ansible_nodename https://review.openstack.org/439658 | 15:35 |
| andymccr | Found some docs on it - and a bug on a diff project for a similar thing | 15:35 |
| *** TxGirlGeek has quit IRC | 15:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build master: Add non *.whl files to MANIFEST.in https://review.openstack.org/439659 | 15:37 |
| *** The_Intern has quit IRC | 15:39 | |
| mgariepy | can I have some review on this one ? https://review.openstack.org/#/c/439644/ | 15:40 |
| *** weezS has joined #openstack-ansible | 15:41 | |
| *** mocker_ has joined #openstack-ansible | 15:43 | |
| *** TxGirlGeek has joined #openstack-ansible | 15:44 | |
| *** mocker_ is now known as mocker | 15:44 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova stable/ocata: Split upgrade testing into it's own shell script https://review.openstack.org/438002 | 15:45 |
| *** agrebennikov_ has joined #openstack-ansible | 15:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/ocata: Allow variant and extra options to be passed to the cache creation https://review.openstack.org/439664 | 15:46 |
| jamesdenton | andymccr Any ideas about the nova_cell0 db access overwrite issue? | 15:46 |
| *** The_Intern has joined #openstack-ansible | 15:46 | |
| andymccr | jamesdenton: i have an idea - but hoping to recreate it in our nova upgrade gate - just fixing that up | 15:46 |
| jamesdenton | sweet | 15:46 |
| andymccr | once i can i'll involve the nova team to make sure we're upgrading properly | 15:47 |
| andymccr | i think it may be the db_migration stuff that has to happen before we can setup cells, even though the advice was to only run that at the very end | 15:47 |
| *** udesale has quit IRC | 15:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/newton: Allow variant and extra options to be passed to the cache creation https://review.openstack.org/439666 | 15:48 |
| *** cloudtrainme has joined #openstack-ansible | 15:49 | |
| *** tailDASHeF has joined #openstack-ansible | 15:50 | |
| *** TxGirlGeek has quit IRC | 15:51 | |
| *** whiteveil has joined #openstack-ansible | 15:51 | |
| *** TxGirlGeek has joined #openstack-ansible | 15:53 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move pip_install deployment https://review.openstack.org/439397 | 15:53 |
| cloudnull | mhayden: ^ alt approach to limiting pip-install runs | 15:53 |
| *** galstrom_zzz is now known as galstrom | 15:54 | |
| *** tailDASHeF has quit IRC | 15:54 | |
| *** cjloader_ has joined #openstack-ansible | 15:55 | |
| *** tailDASHeF has joined #openstack-ansible | 15:55 | |
| *** tailDASHeF has quit IRC | 15:56 | |
| *** TxGirlGeek has quit IRC | 15:56 | |
| *** whiteveil has quit IRC | 15:56 | |
| strattao | What is the recommended branch I should install from? I just tried installing from stable/newton and got an error in the nova_scheduler containers… | 15:56 |
| *** TxGirlGeek has joined #openstack-ansible | 15:56 | |
| mhayden | cloudnull: nice idea! | 15:58 |
| mocker | Does openstack ansible aio setup networking such that VMs will be accessible on the networking? Assuming simple networking of server with 192.168.1.x network. | 15:58 |
| mhayden | cloudnull: http://lists.openstack.org/pipermail/openstack-dev/2017-March/113132.html | 15:59 |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: unify interface lookups https://review.openstack.org/438742 | 16:00 |
| strattao | just trying to do another “vanilla” multi-host install similar to the production example. | 16:01 |
| strattao | or is the answer going to be “wait a week for stable/ocata”? :) | 16:02 |
| *** whiteveil has joined #openstack-ansible | 16:03 | |
| *** manheim_ has quit IRC | 16:04 | |
| *** TxGirlGeek has quit IRC | 16:04 | |
| *** Jack_Iv has quit IRC | 16:05 | |
| *** Oku_OS is now known as Oku_OS-away | 16:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/ocata: Allow variant and extra options to be passed to the cache creation https://review.openstack.org/439664 | 16:07 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/newton: Allow variant and extra options to be passed to the cache creation https://review.openstack.org/439666 | 16:07 |
| *** derekjhyang has joined #openstack-ansible | 16:07 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:08 | |
| *** whiteveil_ has joined #openstack-ansible | 16:08 | |
| *** whiteveil has quit IRC | 16:08 | |
| *** whiteveil_ is now known as whiteveil | 16:08 | |
| *** Jack_Iv has joined #openstack-ansible | 16:09 | |
| *** manheim_ has joined #openstack-ansible | 16:10 | |
| *** TxGirlGeek has quit IRC | 16:14 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova master: Adjust ansible_hostname to ansible_nodename https://review.openstack.org/439658 | 16:15 |
| *** TxGirlGeek has joined #openstack-ansible | 16:15 | |
| *** Jack_Iv has quit IRC | 16:16 | |
| openstackgerrit | Amy Marrich (spotz) proposed openstack/openstack-ansible-os_nova master: Adjust ansible_hostname to ansible_nodename https://review.openstack.org/439658 | 16:17 |
| spotz | logan-: can you redo https://review.openstack.org/439658 ?:) | 16:17 |
| logan- | done :P | 16:18 |
| spotz | :) | 16:19 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova master: Adjust ordering for nova_db_setup https://review.openstack.org/439691 | 16:19 |
| *** TxGirlGeek has quit IRC | 16:20 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:21 | |
| andymccr | jamesdenton: the above 2 patches should fix all the issues you noted | 16:22 |
| mhayden | xgerman: when you want that octavia role reviewed, just adjust the commit msg to not have [WIP] ;) | 16:22 |
| xgerman | ok, I will do the dance with the spave | 16:22 |
| andymccr | jamesdenton: i have a concern that the db_setup re-ordering may break greenfields, so we'll see if the gate passes - if not i'll talk to the novapeeps. | 16:22 |
| mhayden | woot | 16:22 |
| jamesdenton | sounds good. Thanks andymccr | 16:22 |
| xgerman | I also need some guidance on how to get other models other than AIO in my openstack-ansible piece | 16:23 |
| andymccr | np! i did a test run after it failed and that now works with those changes (although the one is just for your hostname issue) | 16:23 |
| jamesdenton | ok yeah - i will recreate that and see what happens | 16:23 |
| *** adrian_otto has joined #openstack-ansible | 16:25 | |
| *** TxGirlGeek has quit IRC | 16:26 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Octavia role hacking https://review.openstack.org/417210 | 16:27 |
| xgerman | mhayden there are two more patches - neutron and os-ansible | 16:28 |
| mhayden | ah okay | 16:28 |
| pradiprwt | odyssey4me : can you suggest me how I can make this changes with newton deployment...https://github.com/openstack/openstack-ansible-os_horizon/blob/stable/newton/templates/horizon_local_settings.py.j2#L728 | 16:28 |
| xgerman | they sort of depend on each other and one without the other will break neutron | 16:28 |
| mhayden | xgerman: so we need to wait on adding anything to openstack-ansible until we have your stuff merged into the role repo | 16:28 |
| *** TxGirlGeek has joined #openstack-ansible | 16:28 | |
| xgerman | yep | 16:28 |
| spotz | andymccr: mreidm did put a comment on the WIP doc we have going saying didn;t need db sync only the api one, but that's during the placement api install. Not sure if relevent for you | 16:28 |
| xgerman | and I probably need to add other mdoels (not only AIO) to os-ansible | 16:29 |
| andymccr | spotz: can you link me to that patch - may be useful! It's not so important in general because in a deploy we do need the db_sync overall | 16:32 |
| pjm6 | pradiprwt: not sure if is what you want, but if you want enable or disable, you just need to put in your vars "horizon_disallow_iframe_embed: True or False" | 16:32 |
| pjm6 | pradiprwt: https://github.com/openstack/openstack-ansible-os_horizon/blob/7171ff9ca0793b2d70bc630f3a08e778454d26c2/defaults/main.yml#L122 | 16:32 |
| odyssey4me | pradiprwt just add 'horizon_disallow_iframe_embed: False' into /etc/openstack_deploy/useR_variables.yml | 16:33 |
| odyssey4me | pradiprwt assuming that you're using stable/newton, not the last published tag | 16:33 |
| *** h5t4 has joined #openstack-ansible | 16:33 | |
| pjm6 | odyssey4me: do you have an running testbed ? | 16:33 |
| pjm6 | at the moment | 16:33 |
| odyssey4me | pjm6 not really, why | 16:34 |
| pjm6 | odyssey4me: well I'm trying to figure out if it's normal to have seqno: -1 in all galera nodes when they are running and working | 16:34 |
| pjm6 | if I stop them i get an positive seqno | 16:35 |
| spotz | andymccr: here you go https://review.openstack.org/#/c/438328/2 | 16:35 |
| andymccr | thanks! | 16:35 |
| pjm6 | but when are running they have -1 and I couldn't find if it's normal behavior | 16:35 |
| pjm6 | (regarding grastate.dat) | 16:35 |
| pradiprwt | odyssey4me: I am using 'horizon_disallow_iframe_embed: False' into /etc/openstack_deploy/useR_variables.yml file ... but few days befor it was not working ...evrardjp told me to they are relising update for newton.... Now, Is this change will work ? | 16:38 |
| odyssey4me | pjm6 unfortunately that would be a big context switch for me, especially since I'm terrible with databases | 16:38 |
| odyssey4me | pradiprwt as mentioned before today, it will only be released in 14.1.0 | 16:39 |
| odyssey4me | so no, it will not yet work | 16:39 |
| *** woodard has joined #openstack-ansible | 16:39 | |
| evrardjp | was that only for 14.1.0 ? | 16:39 |
| *** manheim_ has quit IRC | 16:39 | |
| odyssey4me | evrardjp yep, I checked it earlier today | 16:39 |
| evrardjp | I didn't check when was it included | 16:39 |
| evrardjp | ok | 16:39 |
| evrardjp | cool thanks for checking | 16:39 |
| pjm6 | odyssey4me: no problem, if you had an existing testbed I ask you if you could see the seqno | 16:40 |
| evrardjp | and did you follow where we are in 14.1.0 release? | 16:40 |
| evrardjp | I could have a look if you didn't got the chance | 16:40 |
| pjm6 | odyssey4me: I think I found the answer | 16:41 |
| *** weezS_ has joined #openstack-ansible | 16:41 | |
| pradiprwt | please evrardjp, I need this change ... if possible can you tell me when I can expect this change .? | 16:41 |
| pjm6 | https://www.percona.com/forums/questions-discussions/percona-xtradb-cluster/42947-grstate-dat-seqno-1-on-all-nodes | 16:42 |
| *** weezS has quit IRC | 16:42 | |
| *** weezS_ is now known as weezS | 16:42 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_neutron master: Adds Octavia to Neutron https://review.openstack.org/428979 | 16:42 |
| pjm6 | apparently: "In short, if your node is running, -1 is fine, if its not running something bad happened " | 16:42 |
| odyssey4me | pradiprwt if you checkout the head of stable/newton and run the deployment of the environment then you will get that feature | 16:42 |
| *** woodard has quit IRC | 16:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/ocata: Only insert container hosts into lxc_hosts https://review.openstack.org/439378 | 16:43 |
| *** sacharya has quit IRC | 16:45 | |
| alextricity25 | master is pike, now, correct? | 16:45 |
| *** schwicht has joined #openstack-ansible | 16:45 | |
| evrardjp | yes | 16:45 |
| alextricity25 | woot | 16:45 |
| evrardjp | well not yet | 16:46 |
| evrardjp | it's pike development cycle :) | 16:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Move pip_install deployment https://review.openstack.org/439397 | 16:46 |
| evrardjp | so master gonna become pike! | 16:46 |
| alextricity25 | oh right. that's what I meant :) | 16:46 |
| pradiprwt | thanks odyssey4me... :) | 16:46 |
| odyssey4me | master is already pike | 16:47 |
| *** The_Intern has quit IRC | 16:47 | |
| *** The_Intern has joined #openstack-ansible | 16:47 | |
| evrardjp | depends on vocabulary :p | 16:47 |
| evrardjp | but yes, master = pike ultimately | 16:47 |
| evrardjp | her | 16:47 |
| evrardjp | here now | 16:47 |
| odyssey4me | several backports waiting for review for anyone who has some time: https://review.openstack.org/#/q/project:%255Eopenstack/openstack-ansible.*+branch:stable/ocata+status:open | 16:48 |
| openstackgerrit | Merged openstack/openstack-ansible-rsyslog_server stable/ocata: Fix inconsistency for tag https://review.openstack.org/439646 | 16:48 |
| *** kstev has joined #openstack-ansible | 16:49 | |
| odyssey4me | and a bunch more here: https://review.openstack.org/#/q/project:%255Eopenstack/openstack-ansible.*+branch:stable/newton+status:open | 16:50 |
| openstackgerrit | Merged openstack/openstack-ansible-rsyslog_server stable/newton: Fix inconsistency for tag https://review.openstack.org/439647 | 16:52 |
| *** pjh03 has joined #openstack-ansible | 16:52 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova stable/ocata: Split upgrade testing into it's own shell script https://review.openstack.org/438002 | 16:52 |
| *** schwicht_at_work has joined #openstack-ansible | 16:53 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-repo_server stable/ocata: Remove config tag for an install task https://review.openstack.org/439713 | 16:54 |
| *** DanyC has joined #openstack-ansible | 16:54 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-repo_server stable/newton: Remove config tag for an install task https://review.openstack.org/439714 | 16:54 |
| alextricity25 | xgerman: Does octavia use swift? | 16:56 |
| *** schwicht has quit IRC | 16:56 | |
| *** manheim_ has joined #openstack-ansible | 16:56 | |
| xgerman | it uses glance and glance is backed by swift in AIO | 16:56 |
| *** pjh03 has quit IRC | 16:56 | |
| alextricity25 | ah. that make sense. Thanks. | 16:56 |
| xgerman | but you can make glance be backed by other stores | 16:56 |
| alextricity25 | xgerman: I was wondering why swift was built as part of the role tests, but what you said makes sense :) | 16:57 |
| *** chris_hultin is now known as chris_hultin|AWA | 16:57 | |
| *** TxGirlGeek has quit IRC | 16:58 | |
| *** tailDASHeF has joined #openstack-ansible | 16:59 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:59 | |
| evrardjp | nea1: 'Unexpected status from systemd "Failed to connect to bus: No such file or directory " seems not normal. Running ubuntu 16.04 I guess? | 17:00 |
| evrardjp | just in case :D | 17:00 |
| nea1 | yep, its a fresh ubuntu 16.04 | 17:00 |
| nea1 | installed and updated to the latest yesterday evening | 17:01 |
| nea1 | the other containers are fine too, services are beeing started there as far as i can see (can connect to mysql) | 17:02 |
| nea1 | I found some details about a simular error from some unreleated project which stated that systemd must be brocken, but reinstalling systemd didn't help and redeploying the container didn't change anything either | 17:03 |
| evrardjp | yes that's weird | 17:04 |
| evrardjp | config seems right | 17:05 |
| evrardjp | at first | 17:05 |
| evrardjp | does this happen to all your rabbitmq containers? | 17:05 |
| nea1 | i only have one | 17:05 |
| nea1 | but on my other cluster I have the same error | 17:05 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/newton: Allow container creation based on variant property https://review.openstack.org/439458 | 17:06 |
| nea1 | I'm just redeploying that | 17:06 |
| evrardjp | could you do ansible -m shell -a "ip -4 a s dev eth1" hetzner1-rabbit-mq-container-d17233cb ? | 17:06 |
| *** acormier has joined #openstack-ansible | 17:07 | |
| odyssey4me | logan- you asked for the find module, you got it: https://review.openstack.org/439659 | 17:08 |
| acormier | quick ?: For ceilometer to work, it looks like I have to set up mongodb manually? Is this correct? | 17:09 |
| evrardjp | is it me or github has changed its css? | 17:09 |
| Andrew_jedi | acormier: This is correct. | 17:09 |
| evrardjp | the colors are bright blue | 17:09 |
| evrardjp | for folders | 17:09 |
| acormier | Thanks Andrew | 17:09 |
| nea1 | " [WARNING]: No hosts matched, nothing to do" :O | 17:09 |
| nea1 | evrardjp: | 17:10 |
| logan- | yeah evrardjp, pretty blinding change | 17:10 |
| evrardjp | nea1: well that's weird with what you pasted | 17:10 |
| logan- | odyssey4me :D | 17:10 |
| odyssey4me | evrardjp yes, there were some subtle changes some time in the last few hours | 17:10 |
| evrardjp | yes but this one is bldining me | 17:10 |
| evrardjp | I hated the black ribbon on top already | 17:10 |
| evrardjp | please stop github. | 17:10 |
| evrardjp | I will have to ship my own css | 17:11 |
| odyssey4me | software as a service, you're welcome | 17:11 |
| nea1 | evrardjp: ansible -m shell -a "ip -4 a s dev eth1" hetzner1_rabbit_mq_container-d17233cb does it :) | 17:11 |
| evrardjp | oh yes, sorry | 17:12 |
| evrardjp | wrong copy paste :p | 17:12 |
| evrardjp | I was planning to ask you for rabbitmq_all | 17:12 |
| nea1 | evrardjp: http://paste.openstack.org/show/F9FMiTZxZwIdx5RqZ9Tb/ | 17:12 |
| nea1 | results in the same as the pasted output | 17:13 |
| evrardjp | well I didn't have that output before :) | 17:13 |
| *** crushil has quit IRC | 17:14 | |
| evrardjp | nea1: do you have apparmor enabled? | 17:14 |
| *** schwicht has joined #openstack-ansible | 17:14 | |
| nea1 | doesn't one of the playbooks enable it? | 17:15 |
| evrardjp | I'm thinking your problem is from container cgroups | 17:15 |
| evrardjp | permissions | 17:15 |
| *** schwicht_at_work has quit IRC | 17:16 | |
| nea1 | http://termbin.com/v58n - seams like it is enabled | 17:16 |
| *** retreved has quit IRC | 17:16 | |
| evrardjp | try disabling it, and doing a restart of rabbitmq | 17:16 |
| evrardjp | inside the container | 17:17 |
| evrardjp | disabling outside, restart inside | 17:17 |
| evrardjp | well I think you understood me :p | 17:17 |
| nea1 | sounded strange for a minute :D | 17:17 |
| *** chhavi has joined #openstack-ansible | 17:17 | |
| evrardjp | haha that's me! | 17:17 |
| *** TxGirlGeek has quit IRC | 17:18 | |
| nea1 | systemctl stop apparmor.service should do it or? - haven't really been using apparmor, if so then it didn't work | 17:19 |
| evrardjp | I only know setenforce 0 | 17:20 |
| evrardjp | :p | 17:20 |
| evrardjp | maybe it's just sudo aa-complain /etc/apparmor.d/* | 17:21 |
| evrardjp | from google search | 17:21 |
| evrardjp | :p | 17:21 |
| *** jgriffiths has joined #openstack-ansible | 17:21 | |
| *** lucasxu has quit IRC | 17:21 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: unify interface lookups https://review.openstack.org/439732 | 17:22 |
| *** retreved has joined #openstack-ansible | 17:22 | |
| *** galstrom is now known as galstrom_zzz | 17:23 | |
| nea1 | I found https://support.plesk.com/hc/en-us/articles/213909965-How-to-disable-AppArmor- | 17:24 |
| evrardjp | systemctl stop apparmor worked for me | 17:24 |
| *** galstrom_zzz is now known as galstrom | 17:24 | |
| *** sdake_ has joined #openstack-ansible | 17:25 | |
| nea1 | i restarted the rabbitmq container, didn't change anything either | 17:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/newton: unify interface lookups https://review.openstack.org/439732 | 17:25 |
| nea1 | or lets try to create it without apparmor xD | 17:25 |
| evrardjp | nea1: so after apparmor stop, a run of the playbook for rabbitmq didn't work? | 17:25 |
| *** jgriffiths has quit IRC | 17:26 | |
| evrardjp | nea1: that's weird | 17:26 |
| *** h5t4 has quit IRC | 17:26 | |
| evrardjp | there is something wrong in your host/containers, and if it's fresh ubuntu, I'm concerned :p | 17:26 |
| nea1 | oh, no i just checked it the error 'Unexpected status from systemd "Failed to connect to bus: No such file or directory\n"' had gone | 17:26 |
| *** The_Intern has quit IRC | 17:26 | |
| nea1 | but I'll just check | 17:26 |
| evrardjp | nea1: still we shouldn't have to cut apparmor | 17:27 |
| evrardjp | problem with the security role maybe ? | 17:27 |
| *** retreved has quit IRC | 17:27 | |
| nea1 | failed again | 17:27 |
| evrardjp | ok | 17:27 |
| evrardjp | the playbook or the service? | 17:27 |
| *** TxGirlGeek has joined #openstack-ansible | 17:27 | |
| nea1 | I'll redeploy it | 17:27 |
| nea1 | playbook | 17:27 |
| *** h5t4_ has joined #openstack-ansible | 17:27 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible stable/mitaka: [TEST] Testing stabke/mitaka gate https://review.openstack.org/439735 | 17:27 |
| evrardjp | ok, could you output the logs somewhere? | 17:28 |
| evrardjp | stabke? | 17:28 |
| evrardjp | mhayden: ^ | 17:28 |
| evrardjp | is that a small dutch stab? | 17:28 |
| *** sdake has quit IRC | 17:28 | |
| evrardjp | well more belgian than dutch... but that works! | 17:29 |
| nea1 | I'll redeploy the container and then create log the outputs | 17:29 |
| evrardjp | ok | 17:30 |
| evrardjp | destroy the container before, just in case | 17:30 |
| evrardjp | :p | 17:30 |
| nea1 | yep, already done | 17:30 |
| *** kamtamtun has quit IRC | 17:30 | |
| evrardjp | logan-: FYI vivaldi included sepia filter is almost fine for github. | 17:30 |
| evrardjp | :D | 17:31 |
| *** crushil has joined #openstack-ansible | 17:32 | |
| *** h5t4_ has quit IRC | 17:33 | |
| odyssey4me | mhayden andymccr what's going on with the nova role - I see two patches adjusting ordering | 17:33 |
| andymccr | odyssey4me: talking to nova now | 17:33 |
| andymccr | basically | 17:33 |
| andymccr | upgrades vs greenfields have different paths :( | 17:33 |
| odyssey4me | oh, how lovely | 17:33 |
| evrardjp | :/ | 17:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/newton: Provide empty default for cinder_backends https://review.openstack.org/439476 | 17:33 |
| evrardjp | it's time to orchestrate with ansible and do something different I guess... | 17:34 |
| odyssey4me | and do we have return codes to inform our tooling, or do we have to scrape the output of the CLI? | 17:34 |
| evrardjp | mini roles the return! | 17:34 |
| *** sdake_ is now known as sdkae | 17:34 | |
| *** sdkae is now known as sdake | 17:34 | |
| odyssey4me | evrardjp would you like to be stabbed with something sharp? :p | 17:34 |
| *** TxGirlGeek has quit IRC | 17:34 | |
| evrardjp | something like SALT? | 17:35 |
| nea1 | evrardjp: http://termbin.com/qidt | 17:35 |
| andymccr | well trying to do something around return codes | 17:35 |
| andymccr | or something | 17:35 |
| *** h5t4_ has joined #openstack-ansible | 17:35 | |
| *** TxGirlGeek has joined #openstack-ansible | 17:35 | |
| odyssey4me | andymccr if it helps, jmccrory did a great job in the keystone role around using the rc's | 17:35 |
| andymccr | odyssey4me: yeah its awesome when rcs are there :) so im trying to push for that | 17:35 |
| andymccr | but we'll see | 17:35 |
| *** lucasxu has joined #openstack-ansible | 17:36 | |
| odyssey4me | otherwise how are we supposed to detect an existing database that needs upgrading | 17:36 |
| nea1 | evrardjp: full journal log from container http://termbin.com/x9dv | 17:36 |
| evrardjp | nea1: next steps are to check if /etc/hosts and systemd logging is fine | 17:36 |
| mhayden | andymccr / odyssey4me: my patch was mainly around service setup prior to deploying the service | 17:37 |
| mhayden | since that's how we do it for other nova services | 17:37 |
| evrardjp | nea1: let me have a look if I have one 16.04 node | 17:38 |
| nea1 | evrardjp: looks fine to me http://termbin.com/2538 | 17:39 |
| nea1 | evrardjp: if you want I can give you directly access to that server | 17:39 |
| *** TxGirlGeek has quit IRC | 17:40 | |
| evrardjp | ok so, we can definitely exclude an issue with the config | 17:40 |
| evrardjp | it's a problem with lxc host/container | 17:40 |
| *** TxGirlGeek has joined #openstack-ansible | 17:40 | |
| evrardjp | that's interesting! | 17:41 |
| agrebennikov_ | hi everyone, can somebody Please vote for https://review.openstack.org/#/c/438068 please? | 17:41 |
| nea1 | I joined the the ubuntu-server channel but except of a few Tumbleweeds passing there is not much activity there | 17:42 |
| agrebennikov_ | odyssey4me, evrardjp | 17:42 |
| evrardjp | I'll have a look | 17:43 |
| evrardjp | could someone with a ubuntu16.04 check at something for nea1? | 17:44 |
| evrardjp | we'd like to see systemd logs for the rabbitmq containers | 17:44 |
| evrardjp | in the meantime I'll build a new one | 17:45 |
| odyssey4me | jmccrory hooray :) https://review.openstack.org/436999 | 17:45 |
| evrardjp | I wonder at some point when ansible will have version like 2.2.2.2.2.2.2.0-rc2 | 17:46 |
| evrardjp | agrebennikov_: wouldn't a change of haproxy be required? | 17:47 |
| *** weezS has quit IRC | 17:47 | |
| jmccrory | odyssey4me nice! | 17:47 |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Octavia role hacking https://review.openstack.org/417210 | 17:47 |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Move ceph client role to playbooks for faster run time https://review.openstack.org/439592 | 17:52 |
| *** karimb has quit IRC | 17:52 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Install rdo package https://review.openstack.org/439459 | 17:52 |
| *** manheim_ has quit IRC | 17:52 | |
| *** TxGirlGeek has quit IRC | 17:53 | |
| *** The_Intern has joined #openstack-ansible | 17:53 | |
| *** TxGirlGeek has joined #openstack-ansible | 17:53 | |
| *** sanfern has joined #openstack-ansible | 17:55 | |
| *** lucasxu has quit IRC | 17:56 | |
| *** cloudtrainme has quit IRC | 17:57 | |
| *** lucasxu has joined #openstack-ansible | 17:57 | |
| sanfern | Anyone have tested/verified ansible script to deploy octavia on multimode ? | 17:57 |
| *** TxGirlGeek has quit IRC | 17:58 | |
| andymccr | sanfern: xgerman is working on an octavia role - it's not quite merged yet and is still being developed but if you're interested https://review.openstack.org/#/c/417210/ is the PR | 17:58 |
| *** TxGirlGeek has joined #openstack-ansible | 17:58 | |
| *** mpjetta has joined #openstack-ansible | 17:58 | |
| xgerman | I haven’t done any multi-node | 17:58 |
| *** Jeffrey4l has quit IRC | 17:59 | |
| xgerman | was just asking this morning ;-) | 17:59 |
| *** agrebennikov_ has quit IRC | 17:59 | |
| sanfern | xgerman can you provide me the steps to test single node too for now | 18:01 |
| xgerman | I wrote some docs: https://review.openstack.org/#/c/417210/32/doc/source/configure-octavia.rst | 18:02 |
| xgerman | then I follow those steps: | 18:02 |
| xgerman | 1) Clone the patchec-version of openstack-ansible | 18:03 |
| xgerman | 2) Run the scripts | 18:03 |
| *** cloudtrainme has joined #openstack-ansible | 18:03 | |
| xgerman | 3) Clone patched neutron and patched octavia into roles in /etc/ansible/roles | 18:03 |
| xgerman | 4) 1. Copy aio confirmation to /etc/: cp etc/openstack_deploy/conf.d/octavia.yml.aio /etc/openstack_deploy/conf.d/octavia.yml | 18:03 |
| xgerman | 5) 2. Openstack-ansible lxc-containers-create.yml -e 'lxc_container_allow_restarts=false' --limit octavia_all | 18:04 |
| *** DimGR has joined #openstack-ansible | 18:04 | |
| *** TxGirlGeek has quit IRC | 18:04 | |
| xgerman | 6) run os-neutron playbook | 18:04 |
| xgerman | 7) create image, certs as described in docs | 18:04 |
| xgerman | 8) Run repo-build playbook | 18:05 |
| xgerman | 9) run Octavia playbook | 18:05 |
| xgerman | oh, in 7 also create the mgmt network: | 18:05 |
| xgerman | neutron net-create mgmt --provider:physical_network mgmt --provider:network_type flat | 18:05 |
| xgerman | neutron subnet-create --name submgmt mgmt --no-gateway 172.29.239.0/22 | 18:05 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client master: Tag clean up https://review.openstack.org/438723 | 18:10 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server stable/newton: Remove config tag for an install task https://review.openstack.org/439714 | 18:10 |
| sanfern | ok thanks xgerman | 18:11 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server stable/ocata: Remove config tag for an install task https://review.openstack.org/439713 | 18:11 |
| *** dimtheo_ has joined #openstack-ansible | 18:13 | |
| *** DimGR has quit IRC | 18:13 | |
| *** derekjhyang has quit IRC | 18:13 | |
| xgerman | mhayden: people are talking about multi-node :-) | 18:13 |
| *** agrebennikov_ has joined #openstack-ansible | 18:14 | |
| *** dimtheo_ has quit IRC | 18:17 | |
| *** manheim_ has joined #openstack-ansible | 18:20 | |
| *** whiteveil has quit IRC | 18:20 | |
| *** whiteveil has joined #openstack-ansible | 18:22 | |
| *** DanyC has quit IRC | 18:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/ocata: Add CentOS support https://review.openstack.org/439604 | 18:22 |
| andymccr | vars in -e @x.yml take precedence over everything else right? even vars set inside a task (e.g. import vars/ubuntu.yml) | 18:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client master: Fix CentOS ceph repo https://review.openstack.org/439644 | 18:23 |
| odyssey4me | andymccr yes, extra vars should take precedence over all else | 18:24 |
| *** whiteveil has quit IRC | 18:24 | |
| odyssey4me | unles ansible does something dumb, which is entirely plausible | 18:24 |
| odyssey4me | andymccr note that last one wins for any -e -e -e -e | 18:24 |
| andymccr | odyssey4me: it looks like it is doing something dumb :P | 18:24 |
| *** manheim_ has quit IRC | 18:24 | |
| andymccr | oh no | 18:25 |
| andymccr | its hard coded in newton | 18:25 |
| * andymccr bangs head on desk | 18:25 | |
| odyssey4me | hard coded in nova, or our stuff? | 18:26 |
| odyssey4me | can I get another vote for https://review.openstack.org/439644 please? | 18:26 |
| andymccr | odyssey4me: our stuff i guess centos support wasnt added yet :P | 18:27 |
| odyssey4me | andymccr you can use config_override if it's our stuff and in a conf file? | 18:28 |
| andymccr | odyssey4me: its the libvirt_group name :( | 18:29 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova stable/newton: Make libvirt_group a variable https://review.openstack.org/439760 | 18:29 |
| odyssey4me | oh, that thing | 18:29 |
| odyssey4me | oh of course, we have it in later versions due to the centos support | 18:30 |
| andymccr | jmccrory: odyssey4me do you have a link to where we do clever things with rc's in the keystone role? | 18:31 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_keystone/commit/d2e00019b49e6a07adb89b84536921a06c19ff0e | 18:31 |
| andymccr | ahh yeah nice | 18:32 |
| andymccr | thats perfect | 18:32 |
| odyssey4me | mgariepy mind if I update 439604 ? | 18:32 |
| odyssey4me | https://review.openstack.org/#/c/439604/ | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/ocata: Add CentOS support https://review.openstack.org/439604 | 18:35 |
| *** h5t4_ has quit IRC | 18:35 | |
| *** chhavi has quit IRC | 18:35 | |
| *** retreved has joined #openstack-ansible | 18:35 | |
| *** h5t4 has joined #openstack-ansible | 18:36 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/ocata: Add CentOS support https://review.openstack.org/439604 | 18:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/ocata: Tag clean up https://review.openstack.org/439762 | 18:37 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client master: Fix CentOS ceph repo https://review.openstack.org/439644 | 18:38 |
| *** sdake has quit IRC | 18:41 | |
| mgariepy | odyssey4me, no | 18:42 |
| mgariepy | hehe | 18:42 |
| *** Jack_Iv has joined #openstack-ansible | 18:42 | |
| *** h5t4 has quit IRC | 18:42 | |
| *** sdake has joined #openstack-ansible | 18:43 | |
| *** TxGirlGeek has joined #openstack-ansible | 18:44 | |
| *** h5t4_ has joined #openstack-ansible | 18:44 | |
| odyssey4me | mgariepy :) already done | 18:44 |
| mhayden | odyssey4me: you'll take the stabke and you'll like the stabke | 18:44 |
| odyssey4me | forgive me, I can be a little impatient at times | 18:45 |
| *** chhavi has joined #openstack-ansible | 18:45 | |
| spotz | heheh | 18:45 |
| odyssey4me | mhayden I loved the stake | 18:45 |
| odyssey4me | *stabke | 18:45 |
| odyssey4me | if only there was a *love* button in gerrit | 18:45 |
| *** electrofelix has quit IRC | 18:46 | |
| mgariepy | odyssey4me, yeah I saw, i was out for lunch | 18:46 |
| mgariepy | i don't mind people updating my patches:) | 18:47 |
| odyssey4me | mgariepy I see that the CentOS integrated build is very close to passing in time in the gate now that telemetry is off the table | 18:47 |
| mgariepy | yeah | 18:47 |
| odyssey4me | it does depend on the provider | 18:47 |
| mgariepy | ceilometer had a 5 minutes restart dealy | 18:47 |
| mgariepy | delay** | 18:47 |
| odyssey4me | but I figure that we have a ton of optimisations we can do and easily backport | 18:48 |
| mgariepy | hmm who is slow ? | 18:48 |
| *** cloudtrainme has quit IRC | 18:48 | |
| odyssey4me | speaking of which, this could do with some reviews: https://review.openstack.org/439256 | 18:49 |
| odyssey4me | and this https://review.openstack.org/439457 | 18:49 |
| odyssey4me | and I'd like some reviews for https://review.openstack.org/439659 when anyone has some time | 18:49 |
| mocker | Hmm, clean install of Ubuntu 16.04.2 and got the following error on AOI install: https://gist.github.com/ksexton/75347157f7b1051692f421d768a98b1f | 18:50 |
| mocker | Is now not the best time to try openstack-ansible because of the switch to Ocata? | 18:54 |
| *** gouthamr has quit IRC | 18:56 | |
| *** poopcat has joined #openstack-ansible | 18:57 | |
| *** vnogin has quit IRC | 18:58 | |
| *** gouthamr has joined #openstack-ansible | 19:01 | |
| *** h5t4_ has quit IRC | 19:01 | |
| *** stelouie has joined #openstack-ansible | 19:02 | |
| *** h5t4_ has joined #openstack-ansible | 19:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/ocata: Use block-rescue for cache create https://review.openstack.org/439256 | 19:08 |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Only insert container hosts into lxc_hosts https://review.openstack.org/439378 | 19:10 |
| *** whiteveil has joined #openstack-ansible | 19:19 | |
| *** jgriffiths has joined #openstack-ansible | 19:22 | |
| *** agrebennikov_ has quit IRC | 19:26 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/ocata: Allow container creation based on variant property https://review.openstack.org/439457 | 19:27 |
| stevelle | palendae: proposed some names on 418076, sorry not sorry | 19:27 |
| palendae | 123notit | 19:27 |
| palendae | stevelle: "osai"...should I just use おさい？ | 19:28 |
| palendae | Or オサイ | 19:28 |
| stevelle | kanji seems more approps. | 19:29 |
| *** stuartgr has quit IRC | 19:29 | |
| palendae | I'm not sure there's a kanji character for that | 19:30 |
| palendae | macOS didn't suggest one anyway | 19:30 |
| stevelle | :D | 19:30 |
| stevelle | seriously, rosin maybe? | 19:30 |
| *** DimGR has joined #openstack-ansible | 19:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/ocata: Add CentOS support https://review.openstack.org/439604 | 19:31 |
| palendae | https://en.wikipedia.org/wiki/Rosin ? | 19:31 |
| *** vnogin has joined #openstack-ansible | 19:31 | |
| stevelle | basically, kind of the stuff bull riders smear on a rope before they ride | 19:32 |
| palendae | Ah, so similar concept to prepping the bow on a violin, viola, or cello | 19:32 |
| palendae | Ah, there it is in the list | 19:33 |
| palendae | I'm sure asettle would appreciate skipping "erator" | 19:33 |
| stevelle | :D | 19:34 |
| palendae | I'll try to get to fixing the merge conflict for this tomorrow | 19:35 |
| *** stream10 has joined #openstack-ansible | 19:35 | |
| palendae | And while I'm thinking about it, I'll add to the agenda for tomorrow's meeting | 19:35 |
| *** vnogin has quit IRC | 19:36 | |
| *** vnogin has joined #openstack-ansible | 19:39 | |
| *** cjloader has quit IRC | 19:41 | |
| *** jwagner has quit IRC | 19:41 | |
| openstackgerrit | Satheesh Kumar Ulaganathan proposed openstack/openstack-ansible stable/newton: haproxy for ironic-api HEAD -> GET https://review.openstack.org/439808 | 19:44 |
| hamzy | Does anyone else see "E: Some packages could not be authenticated" error message with Xenial? http://paste.openstack.org/show/600994/ | 19:45 |
| odyssey4me | hamzy that typically happens when your apt source is mid-update | 19:45 |
| *** ngupta has quit IRC | 19:46 | |
| odyssey4me | basically you fetched the releases catallogue, then try to fetch the files and they're gone | 19:46 |
| odyssey4me | it's what happens when the mirror manager doesn't delay between update and clean | 19:46 |
| hamzy | hrm... like a background run of a security update? but wouldn't the apt be locked? | 19:46 |
| odyssey4me | hamzy it's the mirror, not you :) | 19:46 |
| *** ngupta has joined #openstack-ansible | 19:46 | |
| mgariepy | odyssey4me, can I recheck an already merged patch to see how far it goes for an integrated build ? | 19:46 |
| hamzy | ah... and there is not much we can do to avoid it? | 19:47 |
| odyssey4me | mgariepy a merged patch can't be rechecked, no | 19:47 |
| mgariepy | ok | 19:47 |
| odyssey4me | hamzy use a better mirror, or wait 5-15 minutes for the clean to complete, then retry | 19:47 |
| odyssey4me | mgariepy either use a test patch, or recheck any other patch that's completed its check | 19:48 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Cap the number of worker threads https://review.openstack.org/439569 | 19:48 |
| odyssey4me | using a test patch is perhaps more polite ;) | 19:48 |
| palendae | Or spin up an AIO | 19:48 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible master: DO NOT MERGE, build-test https://review.openstack.org/439816 | 19:48 |
| odyssey4me | hahaha | 19:48 |
| odyssey4me | quite right palendae | 19:48 |
| palendae | I don't like abusing the gate for that kind of thing | 19:49 |
| mgariepy | well. build gate have other stuff in them | 19:49 |
| odyssey4me | in case anyone's interested, I took a stab at adding a Vangrant file for the AIO: https://review.openstack.org/438254 | 19:49 |
| palendae | mgariepy: all their code should be in the integrated repo, no? | 19:49 |
| odyssey4me | it requires https://review.openstack.org/438252 to work because the vagrant image has no python | 19:49 |
| palendae | Granted an AIO won't check 2 OSes at once | 19:50 |
| odyssey4me | palendae mgariepy agreed, the gate images are a little weird sometimes | 19:50 |
| palendae | But also, an AIO's kind of annoying to run if you don't have a public cloud account available somewhere to spin them up | 19:51 |
| odyssey4me | I intend to add support for apt caching and a few other things in the hope that I can use vagrant on a plane for testing and development on those long, boring flights | 19:51 |
| palendae | I was thinking about using KVM for that | 19:51 |
| palendae | But I don't have a work-issued linux laptop | 19:51 |
| odyssey4me | palendae yep, vagrant/virtualbox is a little more universal | 19:51 |
| palendae | Indeed | 19:51 |
| stevelle | palendae: good call on the agenda item | 19:52 |
| *** MasterOfBugs has joined #openstack-ansible | 19:53 | |
| *** pcaruana has quit IRC | 19:58 | |
| xgerman | I am sceptical that you can run an AIO on a 16GB laptop (that’s what work gave me) | 19:59 |
| *** mmalik4 has joined #openstack-ansible | 20:00 | |
| palendae | xgerman: Gates only have 8GB RAM | 20:02 |
| palendae | It's definitely tight though | 20:02 |
| *** hybridpollo has joined #openstack-ansible | 20:02 | |
| *** adrian_otto has quit IRC | 20:02 | |
| palendae | My plan was to use a 16GB machine split in half | 20:02 |
| palendae | But on Linux. macOS seems to use a lot more | 20:02 |
| *** openstackgerrit has quit IRC | 20:03 | |
| *** TxGirlGeek has quit IRC | 20:03 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:04 | |
| *** lucasxu has quit IRC | 20:05 | |
| *** hybridpolio has joined #openstack-ansible | 20:06 | |
| xgerman | wish there would be a way to use docker on Mac OS X | 20:06 |
| palendae | Eh | 20:07 |
| *** openstackgerrit has joined #openstack-ansible | 20:07 | |
| *** chhavi has quit IRC | 20:07 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Adjust ansible_hostname to ansible_nodename https://review.openstack.org/439658 | 20:07 |
| palendae | Using a container for an AIO probably wouldn't go well | 20:07 |
| palendae | Which is why I was looking at KVM | 20:07 |
| xgerman | yep, agreed. I tried it :-( | 20:07 |
| *** hybridpollo has quit IRC | 20:07 | |
| palendae | Yeah, I don't think the cgroups are complex enough to allow lxc hosted inside of another cgroup, be it docker, lxc, or rocket | 20:08 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: Only insert container hosts into lxc_hosts https://review.openstack.org/427885 | 20:09 |
| *** nategraf has quit IRC | 20:10 | |
| *** ngupta has quit IRC | 20:10 | |
| *** ngupta has joined #openstack-ansible | 20:11 | |
| mocker | Should I be worried about error on clean install of Ubuntu 16.04.2 AOI install? https://gist.github.com/ksexton/75347157f7b1051692f421d768a98b1f | 20:11 |
| *** hybridpolio is now known as hybridpollo | 20:12 | |
| *** TxGirlGeek has quit IRC | 20:12 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:13 | |
| mgariepy | palendae, https://stgraber.org/2013/12/21/lxc-1-0-advanced-container-usage/ | 20:13 |
| *** agrebennikov_ has joined #openstack-ansible | 20:14 | |
| palendae | mgariepy: Oh, nice | 20:14 |
| mgariepy | it's getting old and i never played with it.. but it might still work. | 20:14 |
| *** TxGirlGeek has quit IRC | 20:14 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:15 | |
| *** lucasxu has joined #openstack-ansible | 20:16 | |
| mgariepy | it's used by lxd here : https://linuxcontainers.org/lxd/try-it/ | 20:16 |
| *** stelouie has quit IRC | 20:22 | |
| *** stelouie_ has joined #openstack-ansible | 20:22 | |
| *** stelouie has joined #openstack-ansible | 20:22 | |
| *** stelouie_ has quit IRC | 20:22 | |
| *** stelouie has quit IRC | 20:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install master: Use block-rescue for pip install https://review.openstack.org/439828 | 20:23 |
| palendae | Nice, didn't know lxd had a trial service | 20:24 |
| *** cloudtrainme has joined #openstack-ansible | 20:25 | |
| *** stelouie has joined #openstack-ansible | 20:26 | |
| *** stelouie_ has joined #openstack-ansible | 20:26 | |
| *** mmalik4 has quit IRC | 20:28 | |
| *** vnogin has quit IRC | 20:29 | |
| *** stelouie_ has quit IRC | 20:29 | |
| *** stelouie has quit IRC | 20:29 | |
| *** agrebennikov_ has quit IRC | 20:30 | |
| *** pradiprwt has quit IRC | 20:30 | |
| *** agrebennikov_ has joined #openstack-ansible | 20:30 | |
| *** pradiprwt has joined #openstack-ansible | 20:31 | |
| *** Jack_Iv has quit IRC | 20:33 | |
| odyssey4me | jmccrory I'd like the description of 'lovely' to be present on more patches. I think mhayden would approve. | 20:34 |
| jmccrory | haha sure thing | 20:34 |
| odyssey4me | alright, enough from me today - have a great evening y'all | 20:36 |
| *** ngupta_ has joined #openstack-ansible | 20:38 | |
| *** mmalik4 has joined #openstack-ansible | 20:39 | |
| *** galstrom is now known as galstrom_zzz | 20:40 | |
| *** vnogin has joined #openstack-ansible | 20:41 | |
| *** ngupta has quit IRC | 20:41 | |
| *** ngupta_ has quit IRC | 20:42 | |
| *** cjloader_ is now known as cjloader | 20:44 | |
| *** EmilienM has joined #openstack-ansible | 20:46 | |
| EmilienM | hello folks | 20:46 |
| EmilienM | do you have Ansible playbooks that test if OpenStack resources survive during an upgrade? A bit like Javelin was doing (back in Tempest) or what Grenade is doing now | 20:47 |
| *** adrian_otto has joined #openstack-ansible | 20:47 | |
| EmilienM | I'm asking before I create something :D | 20:47 |
| *** stream10 has quit IRC | 20:47 | |
| *** lucasxu has quit IRC | 20:48 | |
| EmilienM | cloudnull: ^ | 20:48 |
| palendae | EmilienM: Like guest resourceS? | 20:50 |
| EmilienM | palendae: yeah | 20:50 |
| *** The_Intern has quit IRC | 20:51 | |
| palendae | I think individual service roles may in their gate tests, but I haven't looked at them too closely | 20:51 |
| EmilienM | like : I've deployed OpenStack with $tool and I want to spawn VMs, volumes, users, images, networks, etc; then upgrade OpenStack; and run Ansible again to check if resources survived | 20:51 |
| MasterOfBugs | Hey All, which is the version of neutron-lib we use for newton deployment? | 20:51 |
| palendae | I don't know of anything generic | 20:51 |
| EmilienM | palendae: any pointer into the code? | 20:52 |
| asettle | palendae: very appreciated, tahnk you :P | 20:52 |
| palendae | https://github.com/openstack/openstack-ansible-os_keystone/tree/master/tests | 20:52 |
| palendae | EmilienM: ^ something like that is run for each of our role tests in the gate, and I think Keystone's the most complete for upgrades. But I don't know details | 20:53 |
| EmilienM | palendae: like https://github.com/openstack/openstack-ansible-os_keystone/blob/master/tests/test-keystone-functional.yml | 20:53 |
| EmilienM | indeed | 20:53 |
| *** pjh03 has joined #openstack-ansible | 20:53 | |
| EmilienM | I'll kick off a dicussion on openstack-dev, if we can make these playbooks more easily consummable by projects outdside OSA | 20:54 |
| palendae | Cool | 20:54 |
| cloudnull | EmilienM: not really. we basically use tempest then tempest again to make sure things still work post upgrade | 20:54 |
| cloudnull | i have used downtimer in the past | 20:54 |
| cloudnull | EmilienM: https://github.com/dstepanenko/downtimer | 20:55 |
| *** lucasxu has joined #openstack-ansible | 20:55 | |
| EmilienM | cloudnull: my use case is: I want to check if resources survived | 20:55 |
| *** sdake has quit IRC | 20:55 | |
| EmilienM | wow downtimer looks cool also | 20:55 |
| cloudnull | downtimer will check the OS api | 20:55 |
| cloudnull | then ping the instnaces using l3 networks | 20:55 |
| *** sdake has joined #openstack-ansible | 20:55 | |
| cloudnull | so when upgrading we can see if we have downtime | 20:56 |
| *** The_Intern has joined #openstack-ansible | 20:56 | |
| cloudnull | its a bit basic atm but could easily be made better. | 20:56 |
| EmilienM | cloudnull: it's cool but not exatly what I need | 20:56 |
| EmilienM | I just want to see if my cinder volume can survive an upgrade from newton to ocata for example | 20:57 |
| cloudnull | i've no tools for that atm | 20:57 |
| cloudnull | luzC: ^ | 20:57 |
| cloudnull | maybe luzC has something cool ? | 20:57 |
| EmilienM | and I thought we could use ansible | 20:57 |
| *** pjh03 has quit IRC | 20:58 | |
| EmilienM | to 1. create the resources in some tasks and 2. validate resources are still here in another task | 20:58 |
| cloudnull | EmilienM: also worth looking into https://github.com/Mirantis/Kostyor | 20:59 |
| kysse | 20:59 | |
| *** chris_hultin|AWA is now known as chris_hultin | 21:00 | |
| EmilienM | cloudnull: no doc... | 21:00 |
| EmilienM | ah | 21:00 |
| *** cloudtrainme has quit IRC | 21:00 | |
| cloudnull | https://github.com/Mirantis/Kostyor/tree/master/doc/source | 21:00 |
| EmilienM | my bad, https://github.com/Mirantis/Kostyor/tree/master/doc/source | 21:00 |
| cloudnull | yea. | 21:00 |
| cloudnull | im sure there's some rendered docs somewhere. | 21:01 |
| cloudnull | i just don't know where. | 21:01 |
| cloudnull | :) | 21:01 |
| *** rromans has quit IRC | 21:01 | |
| xgerman | mhayden/stevelle I am wondering how I should fix https://review.openstack.org/#/c/428979/6 so it passes tests… | 21:01 |
| EmilienM | cloudnull: looks very complex | 21:01 |
| *** ngupta has joined #openstack-ansible | 21:02 | |
| * cloudnull has never used it | 21:02 | |
| EmilienM | cloudnull: it reminds me how complex are upgrades :D | 21:02 |
| cloudnull | just saw it | 21:02 |
| mhayden | xgerman: i wouldn't remove the namespace driver quite yet until octavia is 100% ready to go | 21:02 |
| cloudnull | yea. idk if it works as expected but the folks at mirantis may be open to PRs or could expand on it's usage? -cc agrebennikov_ | 21:03 |
| mhayden | the test is failing because one of the variables is missing | 21:03 |
| mhayden | xgerman: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/vars/common.yml#L23-L24 | 21:03 |
| xgerman | k, so will add the var and bring namespace back (conditional) | 21:04 |
| palendae | Has anyone seen this kind of error with config_template and swift? I'm on Newton, 14.0.0 http://paste.openstack.org/show/601004/ | 21:06 |
| palendae | Tried adding debug statements to both ansible and the config_template code, but not seeing why it thinks there's a var named "0": | 21:06 |
| *** askb has joined #openstack-ansible | 21:07 | |
| *** TxGirlGeek has quit IRC | 21:07 | |
| *** TxGirlGeek has joined #openstack-ansible | 21:08 | |
| cloudnull | palendae: nope. | 21:09 |
| cloudnull | does adding -vvvvv provide any more insight in the traceback ? | 21:09 |
| *** TxGirlGeek has quit IRC | 21:09 | |
| *** TxGirlGeek has joined #openstack-ansible | 21:10 | |
| *** cloudtrainme has joined #openstack-ansible | 21:11 | |
| *** ngupta has quit IRC | 21:12 | |
| *** ngupta has joined #openstack-ansible | 21:13 | |
| *** Andrew_jedi has quit IRC | 21:20 | |
| *** FrankZhang has quit IRC | 21:20 | |
| *** jmckind_ has quit IRC | 21:23 | |
| *** rromans has joined #openstack-ansible | 21:23 | |
| *** TxGirlGeek has quit IRC | 21:25 | |
| *** TxGirlGeek has joined #openstack-ansible | 21:25 | |
| *** h5t4_ has quit IRC | 21:27 | |
| *** stream10 has joined #openstack-ansible | 21:29 | |
| *** pcaruana has joined #openstack-ansible | 21:30 | |
| *** pcaruana has quit IRC | 21:33 | |
| *** gouthamr has quit IRC | 21:33 | |
| palendae | cloudnull: Nothing that was useful to me, but added and running again | 21:35 |
| palendae | Bah, now my tmux copying is broken | 21:42 |
| jamesdenton | whats the expected drop date for v15 rc2? | 21:42 |
| palendae | jamesdenton: I wanna say this Friday or next...but I think andymccr will be double checking blockers in tomorrow's meeting | 21:43 |
| jamesdenton | cool, thanks! | 21:43 |
| palendae | There's one blocker I knew of and it got in | 21:43 |
| *** TxGirlGeek has quit IRC | 21:46 | |
| *** TxGirlGeek has joined #openstack-ansible | 21:46 | |
| *** phalmos has joined #openstack-ansible | 21:47 | |
| palendae | cloudnull: http://paste.openstack.org/show/601013/ it seems to indicate it's in config_template, but I'm not finding anything in my config files that indicates 0's being used as a key | 21:49 |
| *** ngupta has quit IRC | 21:52 | |
| *** crushil has quit IRC | 21:52 | |
| *** ngupta has joined #openstack-ansible | 21:52 | |
| *** Andrew_jedi has joined #openstack-ansible | 21:53 | |
| *** cjloader_ has joined #openstack-ansible | 21:55 | |
| *** TxGirlGeek has quit IRC | 21:56 | |
| luzC | EmilienM we have a basic persistent test resource tests in https://github.com/osic/persistent-resources-tests you can use it on your upgrade flow... | 21:56 |
| * EmilienM clicks | 21:56 | |
| *** dileepr has joined #openstack-ansible | 21:56 | |
| *** ngupta has quit IRC | 21:57 | |
| *** TxGirlGeek has joined #openstack-ansible | 21:57 | |
| palendae | Nice! | 21:57 |
| EmilienM | luzC: it's very similar to what we did with Javelin | 21:57 |
| EmilienM | if you know what I mean, | 21:57 |
| EmilienM | let me find links | 21:57 |
| luzC | yes | 21:57 |
| luzC | i remember javelin | 21:57 |
| openstackgerrit | Tom Jose Kalapura proposed openstack/openstack-ansible-os_horizon master: Use horizon_ssl_protocol variable https://review.openstack.org/439390 | 21:57 |
| EmilienM | luzC: I thought we could do that thing in ansible | 21:58 |
| lbragstad | leaving this here since it has upgrade impacts - https://review.openstack.org/#/c/438701/9 | 21:58 |
| *** cjloader has quit IRC | 21:58 | |
| spotz | lbragstad: I'm not sure that's a present.... | 21:59 |
| *** cjloader_ has quit IRC | 22:00 | |
| *** thorst has quit IRC | 22:00 | |
| lbragstad | spotz me either ;) | 22:00 |
| *** thorst has joined #openstack-ansible | 22:00 | |
| *** ngupta has joined #openstack-ansible | 22:03 | |
| *** DanyC has joined #openstack-ansible | 22:03 | |
| *** TxGirlGeek has quit IRC | 22:03 | |
| *** DanyC has quit IRC | 22:04 | |
| *** Andrew_jedi has quit IRC | 22:04 | |
| *** DanyC has joined #openstack-ansible | 22:04 | |
| *** TxGirlGeek has joined #openstack-ansible | 22:04 | |
| *** thorst has quit IRC | 22:04 | |
| luzC | EmilienM yes, actually this is a modified tempest plugin (it uses testr or ostestr test runner) and you can inject an action (create|verify|delete) resources in any flow... | 22:05 |
| openstackgerrit | Merged openstack/openstack-ansible stable/newton: haproxy for ironic-api HEAD -> GET https://review.openstack.org/439808 | 22:05 |
| luzC | EmilienM for instance we have a jenkins pipeline which installs suite, run actions (before and after upgrade)... those are groovy functions... but it can be controlled by ansible, injected on upgrade flow | 22:07 |
| strattao | what variable do I need to set to specify I want to build OSAD off of my local pypi repo? | 22:08 |
| strattao | repo_build_pip_default_index: didn’t seem to work... | 22:08 |
| *** sdake has quit IRC | 22:09 | |
| *** TxGirlGeek has quit IRC | 22:09 | |
| *** lucasxu has quit IRC | 22:09 | |
| *** sdake has joined #openstack-ansible | 22:10 | |
| *** acormier has quit IRC | 22:12 | |
| *** DanyC_ has joined #openstack-ansible | 22:12 | |
| *** acormier has joined #openstack-ansible | 22:12 | |
| luzC | EmilienM do you have an etherpad or spec so that I can have more context about your goal? is it related to osa upgrade gate jobs? | 22:12 |
| *** phalmos has quit IRC | 22:13 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_cinder stable/ocata: Use stable/ocata branch for tempest plugin repo https://review.openstack.org/439874 | 22:13 |
| *** stream10 has quit IRC | 22:14 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_cinder stable/ocata: Provide empty default for cinder_backends https://review.openstack.org/439475 | 22:14 |
| *** DanyC has quit IRC | 22:16 | |
| *** acormier has quit IRC | 22:17 | |
| *** sdake_ has joined #openstack-ansible | 22:25 | |
| *** The_Intern has quit IRC | 22:26 | |
| *** mmalik4 has quit IRC | 22:27 | |
| *** mocker has left #openstack-ansible | 22:29 | |
| *** sdake has quit IRC | 22:29 | |
| *** chris_hultin is now known as chris_hultin|AWA | 22:29 | |
| *** thorst has joined #openstack-ansible | 22:30 | |
| *** karimb has joined #openstack-ansible | 22:31 | |
| *** thorst has quit IRC | 22:35 | |
| *** sdake_ is now known as sdake | 22:37 | |
| *** vnogin has quit IRC | 22:38 | |
| openstackgerrit | Tom Jose Kalapura proposed openstack/openstack-ansible-os_horizon stable/ocata: Use horizon_ssl_protocol variable https://review.openstack.org/439885 | 22:38 |
| *** cloudtrainme has quit IRC | 22:38 | |
| *** sdake_ has joined #openstack-ansible | 22:40 | |
| *** kstev has quit IRC | 22:40 | |
| *** aludwar has quit IRC | 22:42 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_neutron master: [WIP]Adds Octavia to Neutron https://review.openstack.org/428979 | 22:42 |
| *** aludwar has joined #openstack-ansible | 22:42 | |
| *** sdake has quit IRC | 22:42 | |
| *** sdake__ has joined #openstack-ansible | 22:42 | |
| *** cloudtrainme has joined #openstack-ansible | 22:43 | |
| openstackgerrit | Tom Jose Kalapura proposed openstack/openstack-ansible-os_horizon stable/newton: Use horizon_ssl_protocol variable https://review.openstack.org/439888 | 22:44 |
| openstackgerrit | Tom Jose Kalapura proposed openstack/openstack-ansible-os_horizon stable/mitaka: Use horizon_ssl_protocol variable https://review.openstack.org/439889 | 22:44 |
| *** sdake has joined #openstack-ansible | 22:44 | |
| *** sdake_ has quit IRC | 22:44 | |
| *** cloudtrainme has quit IRC | 22:45 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible master: Additions to install OpenStack Octavia https://review.openstack.org/422062 | 22:45 |
| *** schwicht has quit IRC | 22:47 | |
| *** sdake__ has quit IRC | 22:47 | |
| *** esberglu has quit IRC | 22:48 | |
| *** smatzek_ has quit IRC | 22:49 | |
| *** hybridpollo has quit IRC | 22:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Make libvirt_group a variable https://review.openstack.org/439760 | 22:54 |
| *** mmalik4_ has joined #openstack-ansible | 22:55 | |
| *** woodard has joined #openstack-ansible | 22:55 | |
| *** gouthamr has joined #openstack-ansible | 22:58 | |
| *** sdake_ has joined #openstack-ansible | 22:59 | |
| *** sdake has quit IRC | 22:59 | |
| *** woodard has quit IRC | 23:01 | |
| *** esberglu has joined #openstack-ansible | 23:02 | |
| *** sdake_ has quit IRC | 23:03 | |
| *** sdake has joined #openstack-ansible | 23:03 | |
| *** marst has quit IRC | 23:03 | |
| *** esberglu has quit IRC | 23:06 | |
| *** hybridpollo has joined #openstack-ansible | 23:07 | |
| *** jamesdenton has quit IRC | 23:09 | |
| *** acormier has joined #openstack-ansible | 23:13 | |
| *** sdake_ has joined #openstack-ansible | 23:18 | |
| *** acormier has quit IRC | 23:18 | |
| *** sdake has quit IRC | 23:20 | |
| *** kjw3 has quit IRC | 23:21 | |
| *** sdake_ is now known as sdake | 23:23 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Use horizon_ssl_protocol variable https://review.openstack.org/439390 | 23:25 |
| *** crushil has joined #openstack-ansible | 23:27 | |
| openstackgerrit | Tom Jose Kalapura proposed openstack/openstack-ansible-haproxy_server master: Define haproxy_ssl_protocol variable. https://review.openstack.org/439912 | 23:34 |
| *** jmckind has joined #openstack-ansible | 23:36 | |
| *** DanyC_ has quit IRC | 23:42 | |
| *** schwicht has joined #openstack-ansible | 23:42 | |
| *** klamath has quit IRC | 23:46 | |
| *** whiteveil has quit IRC | 23:46 | |
| *** acormier has joined #openstack-ansible | 23:47 | |
| *** acormier has quit IRC | 23:48 | |
| *** acormier has joined #openstack-ansible | 23:48 | |
| *** dimtheo_ has joined #openstack-ansible | 23:50 | |
| *** DimGR has quit IRC | 23:51 | |
| *** crushil has quit IRC | 23:51 | |
| *** acormier has quit IRC | 23:52 | |
| *** acormier has joined #openstack-ansible | 23:53 | |
| *** acormier has quit IRC | 23:53 | |
| *** acormier has joined #openstack-ansible | 23:53 | |
| *** schwicht has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!