| admin0 | root@c3:~# lxc-attach -n c3_utility_container-d12ce6d9 => root@c3_utility_container-d12ce6d9:~# | 00:02 |
|---|---|---|
| admin0 | after root@c3_ -- underscore | 00:02 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Remove the "max_fail_percentage" option https://review.openstack.org/547218 | 00:02 |
| admin0 | when i bypass 1 failed VM and rerun all root@c3:~# lxc-attach -n c3_utility_container-d12ce6d9 => root@c3-utility-container-d12ce6d9:~# | 00:02 |
| admin0 | now - | 00:02 |
| admin0 | c1_utility_container-e351625c : ok=34 changed=8 -- first glance .. changed=8 = 8 things skipped due to this stuff .. that were all green and OK before | 00:03 |
| cloudnull | that change should make the entire playbook fail when a transient issue happens, making green == grean | 00:03 |
| cloudnull | **green == green :) | 00:04 |
| admin0 | so straight calculation .. in my setup 23x3 + 1 containers = 69 containers on infra and 1 on ironic node was created .. 1 on ironic node showed red, but the rest 69 showed OK .. all green .. however it did not change the hostname . when i rerun it again, without the 1 green, all 69 = 8 more changes | 00:06 |
| admin0 | thats a lot of changes skipped 69 x 8 = 552 | 00:07 |
| admin0 | due to 1 failed container somewhere else in the platform | 00:07 |
| admin0 | ok .. i am off .. finally i have all the answers :D | 00:08 |
| admin0 | setup-openstack will be only tomorrow .. hopefully will go without any edge cases | 00:09 |
| cloudnull | I'm curious if we could make the above change in your environment and test the same failure scenario | 00:09 |
| cloudnull | it should now pull the one host out of a failed run which will fail the entire playbook | 00:10 |
| cloudnull | which should notify you that it needs to be rerun | 00:10 |
| cloudnull | instead of producing this silent failure | 00:10 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [Docs] Guide users more https://review.openstack.org/547219 | 00:13 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [Docs] Guide users more https://review.openstack.org/547219 | 00:14 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Drop unnecessary dependencies from role https://review.openstack.org/547139 | 00:23 |
| *** lbragstad has quit IRC | 00:23 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_panko stable/queens: Zuul: Remove project name https://review.openstack.org/546776 | 00:25 |
| *** abelur_ has joined #openstack-ansible | 00:28 | |
| *** admin0 has quit IRC | 00:30 | |
| *** abelur_ has quit IRC | 00:30 | |
| *** admin0 has joined #openstack-ansible | 00:32 | |
| *** chyka_ has joined #openstack-ansible | 00:33 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_panko master: Zuul: Remove project name https://review.openstack.org/546773 | 00:34 |
| *** kstev has quit IRC | 00:36 | |
| *** chyka has quit IRC | 00:36 | |
| *** chyka_ has quit IRC | 00:37 | |
| *** threestrands has joined #openstack-ansible | 00:44 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Fix cache update after initial apt_repository fail https://review.openstack.org/547004 | 00:52 |
| *** admin0 has quit IRC | 00:57 | |
| *** admin0 has joined #openstack-ansible | 01:18 | |
| admin0 | cloudnull, another issue: https://gist.github.com/a1git/5f6a7591ea206683b2fd397cda73d8a0 | 01:19 |
| admin0 | rabbitmq somehow gets /etc/hosts mapping of hostnames without - in them . and the script fails | 01:20 |
| *** woodard has quit IRC | 01:42 | |
| *** woodard has joined #openstack-ansible | 01:42 | |
| *** weezS has joined #openstack-ansible | 01:47 | |
| *** weezS has quit IRC | 01:50 | |
| *** vedin has quit IRC | 01:57 | |
| *** bhujay has joined #openstack-ansible | 02:24 | |
| *** admin0 has quit IRC | 02:26 | |
| *** acormier has quit IRC | 02:26 | |
| *** acormier has joined #openstack-ansible | 02:32 | |
| *** mattoliverau_ is now known as mattoliverau | 02:35 | |
| *** acormier has quit IRC | 02:40 | |
| *** acormier has joined #openstack-ansible | 02:40 | |
| *** acormier has quit IRC | 02:48 | |
| *** acormier has joined #openstack-ansible | 02:56 | |
| *** tinwood has quit IRC | 03:05 | |
| *** tinwood has joined #openstack-ansible | 03:06 | |
| *** acormier has quit IRC | 03:09 | |
| *** acormier has joined #openstack-ansible | 03:10 | |
| *** bhujay has quit IRC | 03:10 | |
| *** Faster-Fanboi_ has quit IRC | 03:10 | |
| *** Faster-Fanboi has joined #openstack-ansible | 03:12 | |
| *** woodard has quit IRC | 03:21 | |
| *** woodard has joined #openstack-ansible | 03:22 | |
| *** woodard has quit IRC | 03:26 | |
| *** acormier has quit IRC | 03:43 | |
| *** woodard has joined #openstack-ansible | 03:52 | |
| *** woodard has quit IRC | 03:57 | |
| *** udesale has joined #openstack-ansible | 04:25 | |
| *** Faster-Fanboi has quit IRC | 04:46 | |
| *** woodard has joined #openstack-ansible | 04:47 | |
| *** woodard has quit IRC | 04:51 | |
| *** Faster-Fanboi has joined #openstack-ansible | 04:52 | |
| *** poopcat has quit IRC | 05:04 | |
| *** Faster-Fanboi has quit IRC | 05:14 | |
| *** Faster-Fanboi has joined #openstack-ansible | 05:20 | |
| *** chyka has joined #openstack-ansible | 05:21 | |
| *** chyka has quit IRC | 05:26 | |
| *** dariko has joined #openstack-ansible | 05:40 | |
| *** woodard has joined #openstack-ansible | 05:45 | |
| nikm | if I want to update the inventory with updated openstack_user_config.yml, what are the steps I need to follow ? | 05:48 |
| *** woodard has quit IRC | 05:49 | |
| *** udesale_ has joined #openstack-ansible | 05:58 | |
| *** aruns has joined #openstack-ansible | 05:58 | |
| *** vishwana_ has joined #openstack-ansible | 05:59 | |
| *** udesale has quit IRC | 06:00 | |
| *** idlemind has quit IRC | 06:02 | |
| *** vishwanathj has quit IRC | 06:03 | |
| cloudnull | nikm: for most things in the openstack_user_config the inventory will be updated on the next playbook run | 06:07 |
| *** threestrands has quit IRC | 06:13 | |
| nikm | cloudnull : it is not updating, we commented ceph-mon_hosts in /etc/openstack_deploy/openstack_user_config.yml and rerun openstack-ansible setup-infrastructure.yml | 06:16 |
| nikm | but ceph-mon_hosts entries are still inside inventory | 06:17 |
| cloudnull | ah, yes. if you want to remove hosts from the inventory you need to use the inventory-manage.py script | 06:17 |
| nikm | cloudnull: we want to remove group | 06:17 |
| nikm | ceph-mon_hosts: infra1: ip: 192.168.20.11 infra2: ip: 192.168.20.12 infra3: ip: 192.168.20.13 | 06:18 |
| *** udesale_ is now known as udesale | 06:18 | |
| nikm | so we want to remove ceph-mon_hosts group from the inventory | 06:18 |
| nikm | what is the command | 06:18 |
| cloudnull | hosts/groups remain in inventory as a fail-safe for accidental removal. I think there's a group remove switch, it's been a while since i've messed with that | 06:19 |
| cloudnull | let me go look | 06:19 |
| cloudnull | the script is `scripts/inventory-manage.py` | 06:19 |
| cloudnull | looking at the code, it doesn't seem to support the removal of a specific group | 06:21 |
| cloudnull | but leaving the hosts intact | 06:21 |
| cloudnull | so in this case if all you wanted to do is remove the group, you's edit the inventory.json file and remove the group entry | 06:22 |
| cloudnull | it should be a list | 06:22 |
| cloudnull | just to be clear, you should worry to much about editing the json file. in the openstack_inventory.tar archive there are backups of the inventory | 06:26 |
| cloudnull | so if you edit it, make a mess of it and need to revert there's are many restore point | 06:27 |
| nikm | cloudnull : thanks | 06:37 |
| *** acormier has joined #openstack-ansible | 06:44 | |
| *** acormier has quit IRC | 06:48 | |
| *** woodard has joined #openstack-ansible | 06:48 | |
| *** woodard has quit IRC | 06:53 | |
| *** pcaruana has joined #openstack-ansible | 06:56 | |
| niraj_singh | i am not able to access nova list inopenstack ansible. Getting error like "ERROR (CommandError): You must provide a username or user ID via --os-username, --os-user-id, env[OS_USERNAME] or env[OS_USER_ID]" | 07:00 |
| niraj_singh | Also not able to find openrc file. Any help???? | 07:00 |
| *** pcaruana has quit IRC | 07:02 | |
| *** pcaruana has joined #openstack-ansible | 07:02 | |
| *** aruns__ has joined #openstack-ansible | 07:11 | |
| *** udesale_ has joined #openstack-ansible | 07:12 | |
| niraj_singh | evrardjp: any help? | 07:13 |
| *** aruns has quit IRC | 07:13 | |
| *** udesale has quit IRC | 07:15 | |
| PTO | Im trying to deploy an installation, but the setup-infrastructure.yml failes becasue the repo_container cannot apt-update. Its a result of a missing dns resolver in the interface configuration. Where is this part defined? | 07:19 |
| *** berendt has quit IRC | 07:30 | |
| *** Miouge has quit IRC | 07:30 | |
| *** Taseer has quit IRC | 07:30 | |
| *** xdfil has quit IRC | 07:30 | |
| *** kysse has quit IRC | 07:30 | |
| *** hwoarang has quit IRC | 07:30 | |
| *** tobberydberg has quit IRC | 07:30 | |
| *** coolj has quit IRC | 07:31 | |
| *** electrocucaracha has quit IRC | 07:31 | |
| *** dgonzalez has quit IRC | 07:31 | |
| jrosser | niraj_singh: you should find an openrc file in the utility container | 07:34 |
| evrardjp[m] | niraj_singh: use your clouds.yaml | 07:35 |
| evrardjp[m] | --os-cloud=default | 07:35 |
| PTO | im totally lost... should i append --os-cloud=default to the openstack-ansible setup-infrastructure.yml? (Sorry - im still a newbie here) | 07:38 |
| evrardjp[m] | nope | 07:39 |
| evrardjp[m] | pto your host resolving is fine ? | 07:39 |
| PTO | yes it is. I did several checks on all my hosts to verify connectivity to storage, vxlan, mgmt - bridges and internet connectivity | 07:41 |
| PTO | Is it required to deploy unbount resolvers in the 16.0.7 release? | 07:42 |
| PTO | Or can i use upstream DNS? | 07:42 |
| *** woodard has joined #openstack-ansible | 07:43 | |
| *** sxc731 has joined #openstack-ansible | 07:45 | |
| evrardjp[m] | could you print your lxc-ls -f ? | 07:46 |
| *** woodard has quit IRC | 07:48 | |
| *** hamza21 has joined #openstack-ansible | 07:50 | |
| *** hwoarang has joined #openstack-ansible | 07:52 | |
| *** coolj has joined #openstack-ansible | 07:53 | |
| *** electrocucaracha has joined #openstack-ansible | 07:53 | |
| *** dgonzalez has joined #openstack-ansible | 07:53 | |
| PTO | evrardjp[m]: Im working on a 24 node deployment. openstack_user_config.yml: http://paste.openstack.org/show/682887/, user_variables.yml: http://paste.openstack.org/show/682886/, output from lxc-ls -f from the infrastrucrure node with repo: http://paste.openstack.org/show/682885/ | 07:55 |
| *** berendt has joined #openstack-ansible | 07:57 | |
| *** Miouge has joined #openstack-ansible | 07:57 | |
| *** Taseer has joined #openstack-ansible | 07:57 | |
| *** xdfil has joined #openstack-ansible | 07:57 | |
| *** kysse has joined #openstack-ansible | 07:57 | |
| *** tobberydberg has joined #openstack-ansible | 07:57 | |
| *** csmart has quit IRC | 08:00 | |
| *** evrardjp[m] has quit IRC | 08:00 | |
| evrardjp | PTO will have a look | 08:04 |
| evrardjp | you have twice used_ips: | 08:04 |
| evrardjp | PTO the openstack_user_config look ok except that but it's the morning for me :) | 08:06 |
| evrardjp | PTO your haproxy_keepalived_external_vip_cidr: is wrong | 08:08 |
| evrardjp | it needs a vip IP and cidr | 08:08 |
| evrardjp | external being a hostname you need to resolve it | 08:08 |
| evrardjp | (you can use a lookup if you want) | 08:10 |
| evrardjp | http://docs.ansible.com/ansible/latest/playbooks_lookups.html#the-dns-lookup-dig | 08:10 |
| *** berendt has quit IRC | 08:11 | |
| *** Miouge has quit IRC | 08:11 | |
| *** Taseer has quit IRC | 08:11 | |
| *** xdfil has quit IRC | 08:11 | |
| *** kysse has quit IRC | 08:11 | |
| *** tobberydberg has quit IRC | 08:11 | |
| *** coolj has quit IRC | 08:11 | |
| *** electrocucaracha has quit IRC | 08:11 | |
| *** dgonzalez has quit IRC | 08:11 | |
| PTO | i have changed the haproxy_keepalived_external_vip_cidr to the ip 10.207.134.250 - as i had in my "dry run" test on 3 nodes. I changed to the fqdn, because it was mentioned in the configuration examples. Maybe a note in them stating it should be the IP would be nice | 08:12 |
| evrardjp | it's better to use the fqdn in the external_lb_vip_address | 08:12 |
| PTO | evrardjp: Im running a redeploy now which include unbound on the infrasturcture hosts. Is this part required? | 08:13 |
| evrardjp | yes | 08:13 |
| evrardjp | all the playbooks are useful :) | 08:13 |
| evrardjp | sometimes they are skipped though | 08:13 |
| evrardjp | PTO it seems our doc need improvement in terms of keepalived configuration -- But it's very disruptive. Let me find an easy solution to that. | 08:14 |
| *** aruns has joined #openstack-ansible | 08:14 | |
| PTO | evrardjp: I guess the error is identified then. Last time unbound was not part of the deployment | 08:14 |
| evrardjp | I think it will be skipped | 08:14 |
| evrardjp | let me double check | 08:14 |
| *** fusmu has joined #openstack-ansible | 08:14 | |
| evrardjp | PTO: https://github.com/openstack/openstack-ansible/blob/d057e9522f37fbbe446aa79cbff5f6c23150d944/inventory/host_vars/localhost/unbound.yml#L16 | 08:16 |
| evrardjp | unbound will not be used | 08:16 |
| evrardjp | by default in your environment | 08:16 |
| evrardjp | but re-running the infrastructure with your change of keepalived would change your haproxy/keepalived configuration which is good | 08:17 |
| evrardjp | there is just something else we want to check first | 08:17 |
| *** aruns__ has quit IRC | 08:17 | |
| evrardjp | can you attach to one of your containers and ping from there ? | 08:17 |
| evrardjp | you said dns resolution is broken, but I am checking the rest. | 08:17 |
| evrardjp | also please allow me to explain how it works: | 08:18 |
| evrardjp | the external network connectivity is provided by default on the 10.0.3 range, on the bridge lxcbr0. There is dnsmasq running to provide dns resolution/host dhcp | 08:19 |
| *** acormier has joined #openstack-ansible | 08:19 | |
| *** armaan has joined #openstack-ansible | 08:19 | |
| evrardjp | that interface is natted, so if your host has a well configured dns configuration, it should be fine | 08:19 |
| *** acormier has quit IRC | 08:20 | |
| PTO | evrardjp: currently running setup-hosts.yml again to check that everything is complete | 08:20 |
| evrardjp | PTO please also note you can edit your network configuration | 08:20 |
| evrardjp | for your containers | 08:20 |
| evrardjp | *including* the dns resolution | 08:20 |
| *** acormier has joined #openstack-ansible | 08:20 | |
| evrardjp | my concern in your case would be a flaky dnsmasq | 08:20 |
| PTO | evrardjp: is that required? In the initial "dry run" test, i recall the resolvers was setup correct | 08:20 |
| *** dgonzalez has joined #openstack-ansible | 08:22 | |
| *** electrocucaracha has joined #openstack-ansible | 08:22 | |
| evrardjp | PTO it's not required and should work out of the box. | 08:22 |
| *** coolj has joined #openstack-ansible | 08:22 | |
| evrardjp | like I said -- dnsmasq should provide the container dns resolution. | 08:23 |
| evrardjp | I think it's time for me to add a playbook healthcheck-hosts.yml | 08:23 |
| PTO | evrardjp: That was also my impression. I think it a nice clean way you deploy with the dnsmasq running. Strange why all /etc/resolv.conf files were empty (except the default comments header). | 08:23 |
| evrardjp | PTO isn't that standard ubuntu behavior to use their resolvconf script to feed /etc/resolv.conf ? | 08:25 |
| *** hachi has joined #openstack-ansible | 08:25 | |
| evrardjp | PTO: I think we should focus on two things now, when you have run your playbooks | 08:26 |
| gokhan | hi evrardjp, ı tried to add a new compute node, but I am getting errors on ceph client. it says 'ceph_mon_host' is undefined. error logs are in http://paste.openstack.org/show/682894/ . How can I solve this problem ? | 08:26 |
| PTO | evrardjp: The dns should be configured in /etc/network/interfaces (dns-nameservers variable), which will populate the /etc/resolv.conf. Both were missing | 08:26 |
| evrardjp | 1) is your host behaving correctly 2) is your container behaving properly | 08:26 |
| evrardjp | PTO let me check | 08:26 |
| evrardjp | gokhan: just a moment please | 08:27 |
| evrardjp | gokhan: oh yeah, you're using ceph for the first time? I think it's a group | 08:27 |
| evrardjp | ceph-mon_hosts | 08:28 |
| evrardjp | gokhan: oh we might have a typo somewhere! | 08:28 |
| evrardjp | gokhan: could you point to where you got that -- from which playbook was the ceph_client called? | 08:29 |
| gokhan | evrardjp, I have defined it ceph-mon_host in my openstack-user-config.yml | 08:29 |
| evrardjp | gokhan: that's good | 08:29 |
| gokhan | evrardjp, nova playbook calls it | 08:29 |
| evrardjp | what did you run ? | 08:29 |
| evrardjp | openstack-ansible os-nova-install.yml ? | 08:30 |
| gokhan | evrardjp, I run openstack-ansible setup-openstack.yml --tags nova-key --limit compute_hosts | 08:30 |
| *** hamza21 has quit IRC | 08:30 | |
| *** kysse has joined #openstack-ansible | 08:33 | |
| *** berendt has joined #openstack-ansible | 08:33 | |
| *** tobberydberg has joined #openstack-ansible | 08:33 | |
| *** Taseer has joined #openstack-ansible | 08:33 | |
| *** Miouge has joined #openstack-ansible | 08:33 | |
| *** xdfil has joined #openstack-ansible | 08:33 | |
| evrardjp | gokhan: ok that's the reason | 08:34 |
| evrardjp | https://github.com/openstack/openstack-ansible-ceph_client/blob/159d8164f6080db68d641f0ff5a54b0fd92449d8/tasks/ceph_get_mon_host.yml#L17-L35 | 08:34 |
| evrardjp | didn't run | 08:34 |
| evrardjp | becayse --tags nova-key | 08:34 |
| evrardjp | wrong docs? | 08:34 |
| evrardjp | gokhan: just run openstack-ansible os-nova-install.yml | 08:35 |
| evrardjp | wait | 08:35 |
| evrardjp | no | 08:35 |
| evrardjp | new compute? | 08:35 |
| evrardjp | openstack-ansible setup-openstack.yml | 08:35 |
| evrardjp | take your time :) | 08:35 |
| evrardjp | or you can do --limit localhost,compute_hosts | 08:35 |
| gokhan | evrardjp, yes new compute node | 08:36 |
| evrardjp | openstack-ansible setup-openstack.yml (nicer) | 08:36 |
| evrardjp | or openstack-ansible setup-openstack.yml --limit localhost,compute_hosts | 08:36 |
| *** hachi has quit IRC | 08:36 | |
| gokhan | evrardjp, ok firstly I will try openstack-ansible setup-openstack.yml --limit localhost,compute_hosts. thanks :9 | 08:37 |
| evrardjp | PTO I checked | 08:37 |
| evrardjp | on my machines the eth0 network is configured as dhcp -- no need to set dns-nameservers. | 08:38 |
| *** sxc731 has quit IRC | 08:38 | |
| gokhan | evrardjp, for these reasons I sometimes run 'ansible _infrastructure_hosts -m setup'. maybe I need to run this command also compute hosts | 08:38 |
| evrardjp | the resolv.conf get filled by resolvconf(5) | 08:38 |
| evrardjp | gokhan: refreshing your fact cache is always useful. You can indeed do it. | 08:39 |
| evrardjp | always | 08:39 |
| evrardjp | .... | 08:39 |
| evrardjp | Not always .. Often! | 08:39 |
| PTO | evrardjp: Im trying to run the playbooks again. Then ill get back with the results | 08:40 |
| evrardjp | PTO ok | 08:43 |
| evrardjp | so the long story short -- it doesn't need manual configuration by default and the dhcp should give you the ip and fill the /etc/resolv.conf automatically. So dnsmasq on the host or the dhcp client on the container seem busted. | 08:44 |
| *** mbuil has joined #openstack-ansible | 08:45 | |
| ThomasS | hi | 08:47 |
| ThomasS | i opened this bug about the vpnaas dashboard not beeing copied | 08:49 |
| ThomasS | https://bugs.launchpad.net/openstack-ansible/+bug/1750789 | 08:49 |
| openstack | Launchpad bug 1750789 in openstack-ansible "VPNaaS dashboard not copied to horizon container" [Undecided,New] | 08:49 |
| gokhan | evrardjp, same error . how can ı ignore ceph client role . or run command with --step and pass this step. is it sensible ? | 08:49 |
| ThomasS | can someone tell me which files to copy to make it work manually? | 08:49 |
| *** woodard has joined #openstack-ansible | 08:51 | |
| *** acormier has quit IRC | 08:53 | |
| *** udesale_ is now known as udesale | 08:53 | |
| *** woodard has quit IRC | 08:55 | |
| *** csmart has joined #openstack-ansible | 08:59 | |
| evrardjp | ThomasS: manually is not a good idea. It would be better to probably adapt the role to make it happen | 09:05 |
| evrardjp | ThomasS: isn't this documented in the vpnaas dashboard documentation? | 09:05 |
| evrardjp | gokhan: no it's not | 09:06 |
| evrardjp | it should work | 09:06 |
| evrardjp | could you print your log to show that the task has run? | 09:06 |
| evrardjp | the one I expressed above | 09:06 |
| evrardjp | maybe it's worth adding ceph_all into the limit | 09:07 |
| evrardjp | (please show the command you used and its full output) | 09:07 |
| *** Sha000000 has joined #openstack-ansible | 09:08 | |
| evrardjp | PTO don't hesitate to ask questions. | 09:08 |
| evrardjp | we are here to help, as much as we can. | 09:08 |
| ThomasS | i will dig into the docs of vpnaas | 09:09 |
| evrardjp | ThomasS: thanks | 09:09 |
| evrardjp | that's what we'll do anyway :) | 09:09 |
| evrardjp | so you have a head start there | 09:10 |
| evrardjp | worth sharing in the bug | 09:10 |
| evrardjp | ThomasS: just for your information, it's a process - we triage bugs on Tuesday, except the last tuesday of the month (that's a community meeting instead of bug triage) | 09:10 |
| evrardjp | Then depending on the severity ppl tackle the bug directly or we just have them hanging. | 09:11 |
| evrardjp | We are organizing bug smash days | 09:11 |
| evrardjp | but generally it all depends on the community -- it's a community work here, it's not a commercial offering. | 09:11 |
| evrardjp | we can only affort "best effort" | 09:12 |
| evrardjp | afford* | 09:12 |
| *** sxc731 has joined #openstack-ansible | 09:12 | |
| PTO | evrardjp: Thanks. I sure will as questions and give feedback. | 09:14 |
| evrardjp | :) | 09:14 |
| evrardjp | thank you | 09:14 |
| *** flaviosr_ has quit IRC | 09:15 | |
| *** vedin has joined #openstack-ansible | 09:16 | |
| *** flaviosr_ has joined #openstack-ansible | 09:17 | |
| *** pbandark has joined #openstack-ansible | 09:18 | |
| *** flaviosr_ has quit IRC | 09:22 | |
| vedin | we are facing mariadb issue inside galera container http://paste.openstack.org/show/682913/ | 09:22 |
| vedin | openstack-ansible setup-infrastructure.yml script completed successfully but ansible galera_container -m shell \ -a "mysql -h localhost -e 'show status like \"%wsrep_cluster_%\";'" | 09:23 |
| vedin | command showed errors for controller2 and controller 3 mariadb | 09:23 |
| sar | I also had that. It was a mariadb bug | 09:24 |
| vedin | mariadb service is unable to start on controller2 and controller3 | 09:24 |
| vedin | sar: any luck | 09:24 |
| vedin | i am using centos7 | 09:25 |
| vedin | and ocata | 09:25 |
| sar | Which ocata version? I think they fixed the bug in 15.1.14 | 09:25 |
| vedin | 15.1.13 | 09:27 |
| vedin | can i cherry pick | 09:27 |
| vedin | that fix | 09:27 |
| vedin | what is fix | 09:27 |
| sar | Well, i changed the sst method to rsync. Not optimal, but it worked | 09:28 |
| vedin | sar : where and how did u change it, can u please tell | 09:28 |
| vedin | can u also give reference of bug fix | 09:29 |
| *** epalper has joined #openstack-ansible | 09:29 | |
| *** evrardjp[m] has joined #openstack-ansible | 09:29 | |
| sar | I put this in user_variables.yml: galera_wsrep_sst_method: rsync | 09:29 |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/openstack-ansible-ops master: Remove empty file https://review.openstack.org/547338 | 09:31 |
| *** gameon has joined #openstack-ansible | 09:31 | |
| sar | You can see the bug referenced here: https://docs.openstack.org/releasenotes/openstack-ansible/ocata.html | 09:31 |
| sar | "The latest version, 10.1.31, has presented problems with the state transfer for multi-node environments when the variable galera_wsrep_sst_method is set to xtrabackup-v2 (the default value). This causes a new cluster to fail, or an existing cluster to be unable to transfer state when a node is rebooted." | 09:31 |
| gameon | Hi all - I have destroyed my keystone containers, and I cannot recreate them using "openstack-ansible lxc-containers-create.yml --limit "*keystone*"" | 09:31 |
| gameon | lxc_container_create : Gather variables for each operating system: fatal: [infra1_keystone_container-abcdefg]: FAILED! => {"failed": true, "msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"} | 09:32 |
| gameon | Does anyone have any advice please | 09:33 |
| vedin | sar: shall i use galera_repo_url: https://downloads.mariadb.com/MariaDB/mariadb-10.1.30/yum/centos7-amd64/ galera_client_repo_url: "{{ galera_repo_url }}" | 09:36 |
| vedin | and try | 09:36 |
| sar | I guess that's what the page says, but i haven't tried it yet | 09:38 |
| sar | You are welcome to try, then report back ;D | 09:38 |
| *** admin0 has joined #openstack-ansible | 09:43 | |
| *** prometheanfire has quit IRC | 09:45 | |
| admin0 | morning | 09:45 |
| evrardjp | sar: vedin if you're using old code, you'll have an issue if the var wasn't set before hand. | 09:48 |
| evrardjp | beforehand | 09:48 |
| evrardjp | because you'll have two repos defined. One with 10.1.30 and one with latest 10.1, which is currently 10.1.31 | 09:48 |
| evrardjp | the fix is included in 10.1.32 | 09:48 |
| evrardjp | in the meantime, you should cleanup your sources, removing references of 10.1.31 | 09:49 |
| gameon | evrardjp: Do you suggest using 10.1.32 if I'm performing an upgrade? | 09:49 |
| evrardjp | gameon: 10.1.32 is not released yet | 09:49 |
| gameon | Ah OK - I will leave at 10.1.30 as per the docs then | 09:49 |
| evrardjp | yeah. | 09:49 |
| evrardjp | 10.1.30 is fine | 09:50 |
| gameon | Thanks - might you have any ideas on my issue where I deleted keystone containers & the data but cannot recreate them | 09:50 |
| vedin | evrardjp: which version I should go for .. | 09:50 |
| *** woodard has joined #openstack-ansible | 09:50 | |
| evrardjp | 10.1.30 | 09:50 |
| evrardjp | 10.2 has other issues with openstack (on top of the clustering issue), and 10.1.31 has the clustering issues. | 09:51 |
| vedin | shall we use this user variables : | 09:51 |
| vedin | # Use these overrides for CentOS/RHEL: galera_repo_url: https://downloads.mariadb.com/MariaDB/mariadb-10.1.30/yum/centos7-amd64/ galera_client_repo_url: "{{ galera_repo_url }}" | 09:51 |
| evrardjp | vedin: what's documented indeed. | 09:51 |
| evrardjp | did you run something with 10.1.31? | 09:52 |
| evrardjp | if you did an apt or yum upgrade, and updated your packages to 10.1.31 you're in a bad state. | 09:52 |
| evrardjp | you have to fix it. | 09:52 |
| vedin | I am deleting containers and re creating it | 09:52 |
| evrardjp | that sounds an easy way to fix it :) | 09:53 |
| evrardjp | That's what I'd do. | 09:53 |
| *** woodard has quit IRC | 09:55 | |
| *** Smeared_Beard has quit IRC | 09:57 | |
| *** masber has quit IRC | 10:01 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [Docs] Add explicit warnings on common mistake https://review.openstack.org/547353 | 10:14 |
| *** MikeW has quit IRC | 10:17 | |
| *** SmearedBeard has joined #openstack-ansible | 10:22 | |
| *** udesale_ has joined #openstack-ansible | 10:28 | |
| *** udesale has quit IRC | 10:29 | |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: fix the url https://review.openstack.org/547363 | 10:34 |
| *** PTO has quit IRC | 10:37 | |
| *** aruns__ has joined #openstack-ansible | 10:38 | |
| *** lvdombrkr has joined #openstack-ansible | 10:39 | |
| *** aruns has quit IRC | 10:40 | |
| *** woodard has joined #openstack-ansible | 10:42 | |
| *** chyka has joined #openstack-ansible | 10:45 | |
| *** woodard has quit IRC | 10:48 | |
| *** chyka has quit IRC | 10:50 | |
| *** udesale__ has joined #openstack-ansible | 11:15 | |
| *** udesale_ has quit IRC | 11:18 | |
| *** hamza21 has joined #openstack-ansible | 11:25 | |
| *** armaan has quit IRC | 11:26 | |
| *** armaan has joined #openstack-ansible | 11:27 | |
| *** pbandark has quit IRC | 11:32 | |
| Miouge | Is it OK/should I +1 a change that I co-authored? (#517856) | 11:33 |
| *** aruns has joined #openstack-ansible | 11:35 | |
| gokhan | evrardjp, same error . all logs are in http://paste.openstack.org/show/682936/ | 11:36 |
| *** udesale__ has quit IRC | 11:36 | |
| *** aruns__ has quit IRC | 11:37 | |
| *** Sha000000 has quit IRC | 11:37 | |
| *** woodard has joined #openstack-ansible | 11:39 | |
| *** stuartgr has joined #openstack-ansible | 11:40 | |
| *** pbandark has joined #openstack-ansible | 11:41 | |
| *** PTO has joined #openstack-ansible | 11:43 | |
| *** woodard has quit IRC | 11:43 | |
| evrardjp | gokhan: don't use --tags nova-key | 11:45 |
| evrardjp | I didn't write it in my command above :) | 11:46 |
| evrardjp | it skips the tasks that are required | 11:46 |
| evrardjp | is the docs wrong? | 11:46 |
| armaan | evrardjp: Hello, how are you today? | 11:47 |
| gokhan | evrardjpi ohh sorry, yep doc is wrong I think | 11:47 |
| gokhan | evrardjp, https://docs.openstack.org/openstack-ansible/pike/admin/maintenance-tasks/scale-environment.html | 11:47 |
| armaan | evrardjp: fyi, my colleague upgrades Ocata to Pike last night. Worked very well, no errors. Thanks a lot for your help there :D | 11:48 |
| armaan | s/upgrades/upgraded | 11:48 |
| PTO | evrardjp: the setup-hosts.yml completed successful. I tested all containers by pinging google, and ~30% of them fails due to no nameservers in /etc/resovl.conf. Should i try and remove all containers (lxc-containers-destroy.yml) and try again? | 11:48 |
| *** taseer1 has joined #openstack-ansible | 11:49 | |
| *** Taseer has quit IRC | 11:49 | |
| *** taseer2 has joined #openstack-ansible | 11:49 | |
| *** fusmu has quit IRC | 11:52 | |
| *** fusmu_ has joined #openstack-ansible | 11:52 | |
| *** taseer1 has quit IRC | 11:53 | |
| *** gameon has quit IRC | 11:54 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547393 | 11:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547394 | 11:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547395 | 12:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547396 | 12:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547397 | 12:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547398 | 12:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547399 | 12:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547400 | 12:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547401 | 12:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-openstack_hosts stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547402 | 12:04 |
| evrardjp | PTO yeah that's weird. are those containers on the same host? | 12:19 |
| evrardjp | armaan: good to hear | 12:19 |
| evrardjp | armaan: upgrading openstack is not a small task | 12:19 |
| armaan | evrardjp: What do you think about upgrading Mitaka to Ocata? ;) | 12:20 |
| PTO | evrardjp: I did some digging, and it seems like some of the containers networking are not wokring (on the same host) | 12:20 |
| evrardjp | armaan: haha. | 12:21 |
| armaan | :P | 12:21 |
| evrardjp | M to N is fine then good luck on the ubuntu upgrade I guess :) | 12:21 |
| evrardjp | PTO: interesting | 12:21 |
| evrardjp | PTO so the ips are listed in the lxc-ls -f but the connectivity itself is wrong? | 12:21 |
| evrardjp | because of dns | 12:21 |
| evrardjp | correct? | 12:21 |
| evrardjp | I am working on a healthchecker since this morning. I will have that ready today | 12:22 |
| gokhan | evrardjp, it is worked. damn nova-keys tag :) | 12:23 |
| *** Sha000000 has joined #openstack-ansible | 12:23 | |
| evrardjp | gokhan: :) | 12:23 |
| evrardjp | gokhan: could you fix it or log a bug for it? | 12:23 |
| evrardjp | this way I don't forget | 12:24 |
| *** taseer2 is now known as Taseer | 12:32 | |
| gokhan | evrardjp , ok solution is only remove nova-key tag from http://paste.openstack.org/show/682942/ . is it right ? | 12:34 |
| PTO | evrardjp: i have restarted the lxc.service and lxc-dnsmasq.service on the host. lxc-stop && lxc-start all containers, and not it seems to work | 12:38 |
| *** masber has joined #openstack-ansible | 12:38 | |
| evrardjp | gokhan: I'd not use the tags at all for filtering | 12:39 |
| evrardjp | PTO mmm | 12:39 |
| evrardjp | I think you're hitting an edge case that mattt talked about a few days ago | 12:39 |
| evrardjp | He told me rebooting fixed it. | 12:40 |
| evrardjp | But if you could dig deeper into why that would be great | 12:40 |
| mhayden | morning | 12:41 |
| PTO | evrardjp: think the restart of the lxc-dnsmasq.service resolved the problem. Is this done in the playbooks? | 12:42 |
| sar | Hi evrardjp, I finally got the volume migration to work, also on attached volumes. Un-attached volumes worked all along since the nova client wasn't involved. (I was the guy who got "Could not determine a suitable URL" in cinder-volume.log) | 12:47 |
| sar | I got a step further by specifying os_privileged_user_auth_url, os_privileged_user_name, os_privileged_user_password and os_privileged_user_tenant in the DEFAULT section in cinder.conf, though a new error message appeared: "Expecting to find domain in project. The server could not comply with the request since it is either malformed or otherwise incorrect." | 12:47 |
| gokhan | evrardjp, ok I will fix this tomorrow. today I am a little busy. also I lost in rabbitmq clusters :( It gives timeout error on nova-compute. it is about updating source available periodic task. but on new added compute node(I created instances on it also) I don't see any timeout errors. it is weird :( | 12:47 |
| sar | I have multidomain-support enabled by using "horizon_keystone_multidomain_support: True" in user_variables.yml. | 12:47 |
| sar | I couldn't figure out how to set the domain using config so i was looking at the source code for stable/ocata: https://github.com/openstack/cinder/blob/stable/ocata/cinder/compute/nova.py | 12:48 |
| *** hamza21 has quit IRC | 12:48 | |
| sar | Specifically this code: | 12:48 |
| sar | auth = identity.Password(auth_url=url, | 12:48 |
| sar | username=context.user_id, | 12:48 |
| sar | password=context.auth_token, | 12:48 |
| sar | project_name=context.project_name, | 12:48 |
| sar | project_domain_id=context.project_domain, user_domain_id=context.user_domain) | 12:48 |
| sar | If i'm not mistaken, it's not possible to specify project_domain or user_domain. The context.project_domain and context.user_domain variables are not set so they will always be None, causing the error I saw in the logs. By hard coding them to 'default' i got it to work. | 12:48 |
| *** woodard has joined #openstack-ansible | 12:49 | |
| *** aruns__ has joined #openstack-ansible | 12:52 | |
| sar | In ocata the context object is created like this: | 12:53 |
| sar | context = ctx.RequestContext( | 12:53 |
| sar | CONF.os_privileged_user_name, None, | 12:53 |
| sar | auth_token=CONF.os_privileged_user_password, | 12:53 |
| sar | project_name=CONF.os_privileged_user_tenant, | 12:53 |
| sar | service_catalog=context.service_catalog) | 12:53 |
| *** woodard has quit IRC | 12:54 | |
| sar | in pike it's not created like that but instead is passed in through the context parameter in the novaclient constructor | 12:54 |
| *** aruns has quit IRC | 12:54 | |
| sar | The context object code is here: https://github.com/openstack/cinder/blob/stable/ocata/cinder/context.py | 12:55 |
| sar | Can't see those variables being set there either, except when using the class method from_dict | 12:56 |
| sar | Doesn't this mean volume migration won't work in ocata with multidomain support enabled? | 12:57 |
| sar | Am i missing something? | 12:57 |
| *** aruns__ has quit IRC | 12:57 | |
| *** aruns__ has joined #openstack-ansible | 13:01 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Add base kit scenario https://review.openstack.org/543534 | 13:02 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: CentOS 7 integrated gate optimization https://review.openstack.org/545455 | 13:02 |
| *** Sha000000 has quit IRC | 13:05 | |
| *** fusmu_ has quit IRC | 13:06 | |
| *** fusmu has joined #openstack-ansible | 13:08 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Install vanilla Python 2.7 on CentOS 7 with pyenv https://review.openstack.org/547126 | 13:11 |
| *** woodard has joined #openstack-ansible | 13:12 | |
| *** woodard has quit IRC | 13:13 | |
| *** woodard has joined #openstack-ansible | 13:14 | |
| *** armaan has quit IRC | 13:20 | |
| *** armaan has joined #openstack-ansible | 13:21 | |
| *** jwitko_ has quit IRC | 13:22 | |
| *** udesale has joined #openstack-ansible | 13:22 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build master: Drop unnecessary dependencies from role https://review.openstack.org/547140 | 13:25 |
| odyssey4me | gameon when you use limit to execute against containers, always include the lxc_hosts group in the limit | 13:26 |
| *** acormier has joined #openstack-ansible | 13:26 | |
| odyssey4me | sar please use a paste service for code/logs, instead of spamming the channel | 13:28 |
| sar | sorry | 13:28 |
| *** acormier has quit IRC | 13:29 | |
| *** acormier has joined #openstack-ansible | 13:29 | |
| odyssey4me | sar I think the question you're asking would probably be better served by finding out from the service teams in their channels. Once you know how to do the configuration bits you need, you can then apply it to any of the services using https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-advanced-config-override.html | 13:30 |
| openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: Add a warning in openstack_hostnames_ips.yml https://review.openstack.org/547421 | 13:30 |
| odyssey4me | sar then once you're comfortable with how that works, take a peek at https://evrard.me/group-and-host-variables-overriding-in-openstack-ansible.html for some fancy ways of implementing the overrides to host groups :) | 13:30 |
| sar | Thanks, i'll take it to the #openstack-cinder channel instead. And I think maybe this has to be solved by changing the nova client code that cinder uses, but i'm not sure. Just want some one else to verify :-) | 13:32 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_nova master: [DNM] - Testing openSUSE nova role https://review.openstack.org/495767 | 13:36 |
| *** sxc731 has quit IRC | 13:38 | |
| *** hamza21 has joined #openstack-ansible | 13:41 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Improve healthchecks https://review.openstack.org/547426 | 13:42 |
| evrardjp | does someone have a multi node to test this? ^ | 13:42 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Improve healthchecks https://review.openstack.org/547426 | 13:43 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Improve healthchecks https://review.openstack.org/547426 | 13:45 |
| evrardjp | mhayden: maybe ? | 13:45 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: [WIP] Test repo build w/trimmed heat requirements https://review.openstack.org/546331 | 13:45 |
| mhayden | evrardjp: i could probably test it later today | 13:45 |
| evrardjp | great. | 13:45 |
| evrardjp | thank you | 13:45 |
| evrardjp | if it's busted please paste a log. If it's fine, let's make it part of the deploy guide. | 13:46 |
| mhayden | oh, that reminds me -- i got the user agent stuff into haproxy's health checks, but we need to stop logging those with each service ;) | 13:47 |
| evrardjp | mhayden: what do you mena? | 13:49 |
| evrardjp | https://github.com/openstack/openstack-ansible/blob/master/inventory/group_vars/haproxy_all/haproxy.yml#L92 | 13:49 |
| evrardjp | where do you want to stop logging stuff? | 13:49 |
| mhayden | well, in the wsgi files in each service, i intend to put an ignore rule for health checks | 13:50 |
| mhayden | so those aren't logged | 13:50 |
| evrardjp | service? | 13:50 |
| evrardjp | good idea | 13:50 |
| mhayden | otherwise you log into check a keystone log and it's full of healthcheck logs | 13:50 |
| mhayden | makes it harder to find what you're looking for | 13:50 |
| mhayden | and takes up disk space | 13:50 |
| mhayden | and I/O | 13:50 |
| evrardjp | but does it matter for the next gen? | 13:50 |
| evrardjp | well ok anyway in the meantime | 13:50 |
| mhayden | haha | 13:51 |
| *** acormier has quit IRC | 13:52 | |
| *** armaan has quit IRC | 13:53 | |
| *** armaan has joined #openstack-ansible | 13:53 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Remove empty file https://review.openstack.org/547338 | 13:59 |
| *** aruns__ has quit IRC | 14:02 | |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible-tests master: Lint rule to ensure package operations are retried https://review.openstack.org/547432 | 14:04 |
| * mhayden waves at hughsaunders | 14:05 | |
| hughsaunders | sup mhayden | 14:05 |
| evrardjp | mhayden: how do you planned to do that -- I quickly check uwsgi config docs | 14:05 |
| mhayden | just adding a route | 14:06 |
| evrardjp | there is no filtering easily done, so you'd have to do something different in the python code. | 14:06 |
| mhayden | matching the useragent | 14:06 |
| evrardjp | Yeah. | 14:06 |
| evrardjp | that's ugly | 14:06 |
| evrardjp | :) | 14:06 |
| hughsaunders | but reducing healthcheck spam is a win | 14:06 |
| evrardjp | that's fair. | 14:06 |
| hughsaunders | hidden uglyness > spammy logs | 14:06 |
| hughsaunders | especially if one is a log grepper by profession | 14:07 |
| evrardjp | well not if we follow the spec of having nginx on host | 14:07 |
| evrardjp | I guess it whole depends on what we want to log with uwsgi | 14:07 |
| evrardjp | yeah it's probably better to have log that just api queries | 14:08 |
| evrardjp | ok I'm in! | 14:08 |
| mhayden | AND THERE WAS MUCH REJOICING | 14:08 |
| hughsaunders | thatwaseasy.gif | 14:08 |
| mattt | PTO: yeah, i was having weird container networking issues after an AIO reboot, but after enough reboots trying to diagnose it just started working :( the issue was that dnsmasq wasn't running tho, if i forced that to run everything worked | 14:08 |
| *** hamza21 has quit IRC | 14:10 | |
| PTO | mattt: i noticed that some ~1/3 containers did not obtain a 10.0.3.x ip, thus the container installation failed. I restarted the lxc-dnsmasq.serive and the lxc.service. Next, restarted all containers, and now the playbook is running | 14:11 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/newton: Fix cache update after initial apt_repository fail https://review.openstack.org/547433 | 14:12 |
| *** idlemind has joined #openstack-ansible | 14:16 | |
| *** hamza21 has joined #openstack-ansible | 14:16 | |
| *** sxc731 has joined #openstack-ansible | 14:17 | |
| *** esberglu has joined #openstack-ansible | 14:21 | |
| *** Sha000000 has joined #openstack-ansible | 14:23 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: Restore PowerVM repo setup https://review.openstack.org/547435 | 14:27 |
| *** udesale_ has joined #openstack-ansible | 14:28 | |
| *** udesale has quit IRC | 14:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: Restore PowerVM repo setup https://review.openstack.org/547438 | 14:30 |
| *** jwitko_ has joined #openstack-ansible | 14:32 | |
| *** fusmu has quit IRC | 14:33 | |
| *** fusmu_ has joined #openstack-ansible | 14:33 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_nova master: Skip logging for haproxy health checks https://review.openstack.org/547440 | 14:35 |
| *** acormier has joined #openstack-ansible | 14:35 | |
| mhayden | evrardjp / hughsaunders ^^ | 14:36 |
| *** acormier has quit IRC | 14:37 | |
| *** acormier has joined #openstack-ansible | 14:37 | |
| *** lbragstad has joined #openstack-ansible | 14:37 | |
| hughsaunders | mhayden: Thats much better than I was expecting, I thought you said you'd have to mess with teh python | 14:38 |
| mhayden | ain't nobody got time for that | 14:38 |
| hughsaunders | mhayden: now you just need to add in the MaaS poller and we're done. | 14:38 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_keystone master: Skip logging for haproxy health checks https://review.openstack.org/547442 | 14:39 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_glance master: Skip logging for haproxy health checks https://review.openstack.org/547443 | 14:39 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_cinder master: Skip logging for haproxy health checks https://review.openstack.org/547444 | 14:40 |
| mhayden | oh, we don't use uwsgi for neutron | 14:41 |
| mhayden | learned something new | 14:41 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_heat master: Skip logging for haproxy health checks https://review.openstack.org/547446 | 14:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install stable/newton: Fix cache update after initial apt_repository fail https://review.openstack.org/547448 | 14:43 |
| *** armaan has quit IRC | 14:44 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_octavia master: Skip logging for haproxy health checks https://review.openstack.org/547449 | 14:44 |
| *** armaan has joined #openstack-ansible | 14:45 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_ironic master: Skip logging for haproxy health checks https://review.openstack.org/547465 | 14:45 |
| *** kstev has joined #openstack-ansible | 14:46 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_sahara master: Skip logging for haproxy health checks https://review.openstack.org/547467 | 14:46 |
| mhayden | okay, i think that's all of them that use uwsgi | 14:46 |
| *** kstev1 has joined #openstack-ansible | 14:48 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [Docs] Migrate security into user guide https://review.openstack.org/547151 | 14:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/newton: Fix cache update after initial apt_repository fail https://review.openstack.org/547469 | 14:49 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Install vanilla Python 2.7 on CentOS 7 with pyenv https://review.openstack.org/547126 | 14:49 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Install vanilla Python 2.7 on CentOS 7 with pyenv https://review.openstack.org/547126 | 14:50 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [Docs] Guide users more https://review.openstack.org/547219 | 14:50 |
| *** kstev has quit IRC | 14:51 | |
| mhayden | odyssey4me: i think i worked the kinks out of https://review.openstack.org/545461 | 14:51 |
| vedin | Hi evrardjp, can you please have look to this issue >> http://paste.openstack.org/show/682960/ | 14:51 |
| *** acormier has quit IRC | 14:52 | |
| vedin | all openstack services deployed successfully, facing error in last | 14:53 |
| evrardjp | sure | 14:53 |
| evrardjp | vedin: could you do ansible -m setup utility_all ? | 14:54 |
| evrardjp | and re-run ? | 14:54 |
| evrardjp | you can only re-run the last part: openstack-ansible os-tempest-install.yml | 14:54 |
| evrardjp | and you're not even forced to do it | 14:55 |
| evrardjp | depends if you want to test with tempest or not :) | 14:55 |
| vedin | do I need to run this command "ansible -m setup utility_all" | 14:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server master: Explicitly ensure when apt cache updates happen https://review.openstack.org/547471 | 14:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server master: Explicitly ensure when apt cache updates happen https://review.openstack.org/547471 | 14:56 |
| *** hw_wutianwei_ has quit IRC | 14:58 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server master: Explicitly ensure when apt cache updates happen https://review.openstack.org/547471 | 15:01 |
| evrardjp | vedin: yes | 15:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547395 | 15:03 |
| *** masber has quit IRC | 15:05 | |
| vedin | evrardjp: Thanks a lot, finally I have completed the installation :) | 15:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547396 | 15:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547397 | 15:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Restore support for percona packages when using ppc64le https://review.openstack.org/547101 | 15:14 |
| *** armaan has quit IRC | 15:21 | |
| *** armaan has joined #openstack-ansible | 15:22 | |
| *** prometheanfire has joined #openstack-ansible | 15:23 | |
| *** sxc731 has quit IRC | 15:23 | |
| odyssey4me | um, it looks like OVS is now always installed - WTH? | 15:27 |
| odyssey4me | in the integrated build: http://logs.openstack.org/44/545844/1/check/openstack-ansible-deploy-aio_lxc-ubuntu-xenial/bff5fa1/logs/ara/result/a90901cd-55ac-4a70-aa0e-96e6cb286cea/ | 15:27 |
| cloudnull | mornings | 15:27 |
| *** dave-mccowan has joined #openstack-ansible | 15:29 | |
| prometheanfire | moin | 15:31 |
| *** weezS has joined #openstack-ansible | 15:32 | |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: fix the url https://review.openstack.org/547363 | 15:33 |
| cloudnull | odyssey4me: looks like https://github.com/openstack/openstack-ansible-os_neutron/blob/master/vars/main.yml#L35-L41 is always getting run | 15:33 |
| cloudnull | we might need to tune up https://github.com/openstack/openstack-ansible-os_neutron/blob/master/vars/main.yml#L19-L22 | 15:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: [TEST] Debug variables to figure out why OVS is installed https://review.openstack.org/547481 | 15:35 |
| cloudnull | we can probably just shrink that condition to using just the enabled plugins | 15:36 |
| cloudnull | something like "neutron_services['XXXXXXX'].service_en | bool or ... " | 15:36 |
| cloudnull | but that's quite curious | 15:36 |
| odyssey4me | there's clearly something activating of of the conditions in neutron_needs_openvswitch | 15:36 |
| cloudnull | or maybe something else we're installing is has a new ovs dependency | 15:36 |
| odyssey4me | ja, I'll do a patch to the integrated repo to debug using that one up there | 15:37 |
| cloudnull | in other news, Uros sent this to me last night https://github.com/urosorozel/os_ironic_inspector | 15:38 |
| cloudnull | and https://github.com/urosorozel/openstack-ansible/commit/6710eccd8d3c3537885eebbb5cd8bbdfa5f62e69 | 15:38 |
| cloudnull | looks like ironic inspector is now working :) | 15:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: [TEST] Debug variables to figure out why OVS is installed https://review.openstack.org/547482 | 15:39 |
| odyssey4me | what does ironic inspector do? | 15:40 |
| *** vedin has quit IRC | 15:41 | |
| cloudnull | it discovers hardware capabilities | 15:41 |
| *** kenperkins has joined #openstack-ansible | 15:41 | |
| cloudnull | in ironic you have to statically set everything when you enroll the node and create the flavors | 15:41 |
| cloudnull | the inspector makes that whole interaction suck less. | 15:41 |
| *** fusmu_ has quit IRC | 15:42 | |
| odyssey4me | nice work that - it'd be good to import it | 15:42 |
| cloudnull | agreed | 15:43 |
| *** MikeW has joined #openstack-ansible | 15:43 | |
| MikeW | Hey what piece of OSA controls which venvs get deployed to the repo servers? Is it decided by what exists in openstack_user_config and the env.d? | 15:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-nspawn_hosts master: add minimal functional tests https://review.openstack.org/547157 | 15:44 |
| cloudnull | MikeW: its a combination of what you have in the user config and which groups are populated for a given service. | 15:45 |
| cloudnull | specifically when https://github.com/openstack/openstack-ansible-repo_build/blob/master/defaults/main.yml#L67-L68 is enabled | 15:46 |
| cloudnull | which is the default | 15:46 |
| *** epalper has quit IRC | 15:46 | |
| cloudnull | if you want to `emerge world` you can set that for "no" and it'll build all of the known openstack venvs regardless of what's being deployed | 15:47 |
| odyssey4me | cloudnull if you have a bit of time, would you mind working through these? | 15:48 |
| odyssey4me | https://review.openstack.org/#/q/topic:bug/1750656+(status:open+OR+status:merged) | 15:48 |
| cloudnull | sure | 15:48 |
| MikeW | cloudnull Excellent thanks that's exactly what I needed... for some reason my magnum venv isn't getting built... though I see the other components of it are running correctly | 15:49 |
| odyssey4me | MikeW if it's not being built, then it's most likely that you don't have the correct host group populated | 15:50 |
| odyssey4me | the default is to look at your inventory, then build only what you need | 15:50 |
| odyssey4me | I confirmed just the other day that if the right host group is populated, it auto-builds as it should - I tested ocata and pike | 15:51 |
| odyssey4me | this was the host group: https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/conf.d/magnum.yml.aio#L1 | 15:51 |
| mattt | i have an AIO that i took from pike->queens, but my repo server keeps building ansible 2.3 and not 2.4, i'm a bit rusty on this stuff ... where do i begin looking? :) | 15:55 |
| *** Sha000000 has quit IRC | 15:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Restore support for percona packages when using ppc64le https://review.openstack.org/547101 | 15:58 |
| odyssey4me | mattt your repo server is building ansible because rally uses it - the ansible in the repo server is not what's used for your deployment | 16:00 |
| odyssey4me | mattt https://github.com/openstack/openstack-ansible-os_rally/blob/master/defaults/main.yml#L57 | 16:01 |
| mattt | odyssey4me: the problem is when i re-run bootstrap ansible ... despite us explicitly pulling in 2.4 it manages to be happy with 2.3 from the repo servers | 16:01 |
| odyssey4me | oh really? | 16:01 |
| odyssey4me | that would be a bug | 16:02 |
| odyssey4me | bootstrap ansible has a specific requirement: https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L25 | 16:02 |
| odyssey4me | the venvs is built with constraints: https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L163-L164 | 16:03 |
| odyssey4me | the constraints come from openstack's u-c: https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L115 | 16:03 |
| odyssey4me | and I don't see ansible in u-c | 16:04 |
| odyssey4me | mattt if you have the stdout for when the ansible bootstrap executes, could you paste.openstack.org it? | 16:04 |
| mattt | odyssey4me: http://paste.openstack.org/raw/682996/ -- we request 2.4.3.0 but it installs 2.3.2.0 ... i believe if the installation fails it will try again with --isolated | 16:04 |
| odyssey4me | I'm keen on seeing this part specifically: https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L170-L172 | 16:05 |
| odyssey4me | heh, oh this is due to some recent changes | 16:05 |
| odyssey4me | it makes sense now | 16:05 |
| odyssey4me | on the host /root/.pip/pip.conf has constraints set, doesn't it? | 16:06 |
| mattt | yep! | 16:06 |
| odyssey4me | ja, and the pip.conf settings are used regardless of CLI options | 16:06 |
| mattt | and that has ansible==2.3.2.0 | 16:06 |
| odyssey4me | I haven't totally confirmed it, but I think the configs are merged. | 16:06 |
| mattt | aaah | 16:06 |
| odyssey4me | yep, this is the give-away: Collecting ansible==2.3.2.0 (from -c http://172.29.236.100:8181/os-releases/17.0.0/ubuntu-16.04-x86_64/requirements_absolute_requirements.txt (line 91)) | 16:07 |
| admin0 | finally .. setup-openstack finished .without any errors | 16:07 |
| odyssey4me | mattt so this was from pike->queens? | 16:07 |
| mattt | odyssey4me: well i've just been on master since before queens was cut | 16:07 |
| *** SmearedBeard has quit IRC | 16:08 | |
| odyssey4me | ok, I didn't realise that change had made its way into queens | 16:08 |
| mattt | odyssey4me: wait wait, i'm getting confused | 16:08 |
| mattt | odyssey4me: not pike, queens->whatever :) | 16:08 |
| odyssey4me | ah, so queens -> master | 16:08 |
| odyssey4me | or an old master to a newer master | 16:09 |
| odyssey4me | this is what changed the behaviour: https://github.com/openstack/openstack-ansible-pip_install/commit/0f18e40fb61086b89cf12932cf9583adaf5b5609 | 16:09 |
| mattt | yeah old master to new master, just been riding master for a little while | 16:10 |
| mattt | i'll create a bug, not sure i'm in a situation to fix this atm ... since i don't understand that change | 16:11 |
| odyssey4me | thanks, let me know when you've filed the bug - I'm looking at options to resolve that, as it's pretty bad | 16:11 |
| *** armaan has quit IRC | 16:12 | |
| *** armaan has joined #openstack-ansible | 16:13 | |
| odyssey4me | I'm not sure if this is a pip bug actually, because the cli option for the constraint should be overriding the config file option | 16:13 |
| odyssey4me | although you can specify multiple constraints, so I think a merge is happening here | 16:14 |
| cloudnull | maybe it's treating all constraints equally and squashing? | 16:15 |
| cloudnull | admin0: great news! | 16:16 |
| odyssey4me | cloudnull I think it's chaining them, so it's treating it like I added more CLI --constraints options | 16:16 |
| *** lvdombrkr89 has joined #openstack-ansible | 16:18 | |
| evrardjp | odyssey4me: you told me when chaining, first match wins ? or last match whatever so it might just be an ordering issue? | 16:18 |
| mattt | odyssey4me: https://bugs.launchpad.net/openstack-ansible/+bug/1751316 | 16:19 |
| openstack | Launchpad bug 1751316 in openstack-ansible "bootstrap-ansible.sh not installing ansible==2.4.3.0" [Undecided,New] | 16:19 |
| odyssey4me | evrardjp no, in this case we give pip a requirement for ansible 2.4, but there's a constraint for 2.3 | 16:19 |
| openstackgerrit | git-harry proposed openstack/openstack-ansible-tests master: Add ansible-lint rule for apt_repository https://review.openstack.org/547497 | 16:19 |
| cloudnull | yes that makes sense. the pip config is an extension of the cli so it'd should be chaining them | 16:19 |
| mattt | why can you not format output in launchpad bugs, or can you? | 16:19 |
| odyssey4me | so for us to fix it we have two options - one is to move the config file out the way when we're bootstrapping, and the other is to build another constraint file and use it | 16:19 |
| cloudnull | mattt: mind pasting the pip.conf ? | 16:19 |
| odyssey4me | and using it in the CLI option would make it win, because it'd be first | 16:20 |
| * cloudnull maybe missed it | 16:20 | |
| cloudnull | we could pass in --isolated when bootstrapping | 16:20 |
| evrardjp | odyssey4me: yeah that's what I meant -- adding a new constraint for what we install to be in the winning order. | 16:20 |
| odyssey4me | cloudnull that doesn't help | 16:20 |
| mattt | cloudnull: added to the bug | 16:20 |
| *** lvdombrkr has quit IRC | 16:20 | |
| odyssey4me | the issue is not that it's using the internal repo - the issue is that it's given a constraint for the package it's installing | 16:20 |
| cloudnull | --isolated should make it ignore all config | 16:20 |
| cloudnull | oh | 16:21 |
| odyssey4me | ignore all config? | 16:21 |
| evrardjp | cloudnull: does isolated work with constraints? | 16:21 |
| odyssey4me | hmm, can try that | 16:21 |
| cloudnull | yea --isolated would ignore any and all config | 16:21 |
| odyssey4me | lemme fire up a quick test | 16:21 |
| cloudnull | you could pass in additional options when isolated is used | 16:22 |
| odyssey4me | mattt I'm picking up that bug, should have a patch shortly | 16:22 |
| evrardjp | I am using isolated into venvs on regular basis and it works, so I'd say yes, but not so sure if it takes the -c too . It works with -r at least :) | 16:22 |
| evrardjp | I think isolated is the best for bootstrap-ansible. | 16:22 |
| odyssey4me | yep, that'd be the simplest - so I'll give it a go | 16:22 |
| evrardjp | well maybe we should isolate all the things. | 16:22 |
| evrardjp | if we are always passing the right constraints | 16:23 |
| evrardjp | oh no that's a bad idea. | 16:23 |
| odyssey4me | isolate would kill the offline story though, because it uses pypi unless you give it a different index | 16:23 |
| evrardjp | yeah | 16:23 |
| mattt | thanks odyssey4me, it's not urgent for me, i just thought the issue was that we were mistakingly building ansible 2.3 on the repo servers | 16:23 |
| evrardjp | odyssey4me: agreed | 16:23 |
| evrardjp | mattt: this is a big thing, thanks! | 16:24 |
| odyssey4me | evrardjp I have an idea - lemme verify that it'll do the right things | 16:24 |
| *** sxc731 has joined #openstack-ansible | 16:28 | |
| mattt | evrardjp: np! let me know if i can provide any more info :) | 16:29 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-nspawn_hosts master: add minimal functional tests https://review.openstack.org/547157 | 16:30 |
| *** openstackgerrit has quit IRC | 16:33 | |
| *** openstackgerrit has joined #openstack-ansible | 16:36 | |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/newton: Fix cache update after initial apt_repository fail https://review.openstack.org/547448 | 16:36 |
| odyssey4me | cloudnull - I see https://review.openstack.org/#/c/547218/ which removes max_fail_percentage (cc evrardjp)... any thoughts on actually add ing any_error_fatal to replace that for all control plane plays? | 16:39 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547398 | 16:39 |
| cloudnull | odyssey4me: ++ | 16:39 |
| odyssey4me | as far as I'm concerned, if any part of your control plane is not contactable, or breaks, you want to know about it asap | 16:39 |
| cloudnull | makes sense to me. | 16:39 |
| odyssey4me | then for the non-control plane stuff, perhaps max_fail_percentage should stay, but be a configurable var | 16:40 |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: fix the url https://review.openstack.org/547363 | 16:40 |
| odyssey4me | so if 1 of your 100s of computes are down, you can still do stuff | 16:40 |
| cloudnull | The max fail percentages were intended to help large scale deployments, however it would seem they're causing more harm / confusion | 16:40 |
| odyssey4me | I suppose if one of your computes is down, you can just use a --limit to skip it | 16:41 |
| cloudnull | I've kinda of the opinion now that if 1 of 100 compute nodes are down folks should pass in a --limit !busted_node | 16:41 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/newton: Fix cache update after initial apt_repository fail https://review.openstack.org/547469 | 16:41 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547400 | 16:41 |
| odyssey4me | lol, great minds | 16:41 |
| cloudnull | haha, indeed | 16:41 |
| odyssey4me | ja, I would support doing that, and backporting it to newton... | 16:42 |
| odyssey4me | this is one of those hidden gems | 16:42 |
| odyssey4me | there should be a reno though to go with the patch | 16:42 |
| cloudnull | I can make a reno | 16:43 |
| cloudnull | do we still have time to get it to newton ? | 16:43 |
| cloudnull | i thought that's "dead jim" | 16:43 |
| odyssey4me | nope, we still have it - and some patches are still trickling into it | 16:44 |
| cloudnull | at the moment that pr to remove max fail is not passing due to our new found ovs issues | 16:44 |
| odyssey4me | but we're on the last legs | 16:44 |
| odyssey4me | upstream is done EOLing | 16:44 |
| *** sxc731 has left #openstack-ansible | 16:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Restore support for percona packages when using ppc64le https://review.openstack.org/547101 | 16:48 |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: fix the url https://review.openstack.org/547363 | 16:51 |
| *** udesale__ has joined #openstack-ansible | 16:52 | |
| *** udesale_ has quit IRC | 16:52 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Remove the "max_fail_percentage" option https://review.openstack.org/547218 | 16:59 |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: fix the url https://review.openstack.org/547363 | 16:59 |
| cloudnull | odyssey4me: ^ reno added | 16:59 |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547395 | 17:01 |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547396 | 17:01 |
| *** chyka has joined #openstack-ansible | 17:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547397 | 17:06 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Remove the "max_fail_percentage" option https://review.openstack.org/547218 | 17:06 |
| cloudnull | added a limit example | 17:06 |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: fix the doc link https://review.openstack.org/547517 | 17:06 |
| odyssey4me | cloudnull lgtm | 17:11 |
| cloudnull | it wont pass due to ovs shinnanigans | 17:11 |
| cloudnull | so we'll need to figure that out to make the integrated gate happy | 17:12 |
| *** dave-mccowan has quit IRC | 17:15 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: [TEST] Debug variables to figure out why OVS is installed https://review.openstack.org/547481 | 17:15 |
| *** dave-mccowan has joined #openstack-ansible | 17:16 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Fix cache update after initial apt_repository fail https://review.openstack.org/547433 | 17:19 |
| openstackgerrit | wangqi proposed openstack/openstack-ansible master: Replace http with https in app-resources.rst and delete useless url https://review.openstack.org/547517 | 17:20 |
| *** shardy is now known as shardy_afk | 17:20 | |
| *** jafeha has quit IRC | 17:24 | |
| *** lvdombrkr89 has quit IRC | 17:26 | |
| *** MikeW has quit IRC | 17:41 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/pike: Fix cache update after initial apt_repository fail https://review.openstack.org/547394 | 17:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Isolate the Ansible bootstrap https://review.openstack.org/547526 | 17:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-nspawn_hosts master: add minimal functional tests https://review.openstack.org/547157 | 17:46 |
| *** udesale_ has joined #openstack-ansible | 17:47 | |
| odyssey4me | mattt evrardjp cloudnull https://review.openstack.org/547526 seems to work well, and is the simplest way I could think of solving it | 17:48 |
| *** udesale__ has quit IRC | 17:50 | |
| *** mbuil has quit IRC | 17:55 | |
| *** omega has joined #openstack-ansible | 17:56 | |
| *** omega is now known as omegapoint | 17:56 | |
| cloudnull | odyssey4me: looks good to me :) | 17:56 |
| *** omegapoint has left #openstack-ansible | 17:56 | |
| *** omegapoint has joined #openstack-ansible | 17:56 | |
| *** hamza21 has quit IRC | 17:57 | |
| omegapoint | hey, can anyone help me out? i finished deploying OSA using the prod system template, and the networking works fine, however i can't create a provider network which can actually reach the internet through the gateway. the ARP requests arrive on the br-vlan (eth12) of the network containers but it won't get redirected towards the external network. | 18:03 |
| omegapoint | in other words, i can't ping my gateway from the qrouter namespace. I've read through all the troubleshooting guides and documentation there is, but i can't figure it out | 18:04 |
| mattt | odyssey4me: testing now | 18:04 |
| *** SimAloo has joined #openstack-ansible | 18:05 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Switch to using the imported nspawn roles https://review.openstack.org/547554 | 18:06 |
| *** jillr has quit IRC | 18:07 | |
| *** bradm has quit IRC | 18:07 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Isolate the Ansible bootstrap https://review.openstack.org/547526 | 18:09 |
| *** hamza21 has joined #openstack-ansible | 18:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Isolate the Ansible bootstrap https://review.openstack.org/547526 | 18:14 |
| shananigans | omegapoint: You should be able to use something like tcpdump on the infra host's(br-vlan and eth12 depending on your setup) to see if the arp requests are making it out of the container and host nic. | 18:28 |
| omegapoint | @shananigans i've tried and the arp requests go through eth12, br-vlan and the host NIC for br-vlan | 18:29 |
| shananigans | omegapoint: If it is, you will need to make sure the gateway_ip you set up in your subnet is set up on your network hardware with all the proper access. | 18:29 |
| *** lbragstad has quit IRC | 18:31 | |
| *** MikeW has joined #openstack-ansible | 18:31 | |
| omegapoint | @shananigans so the gateway needs to be accessible from the br-vlan bridge, right? | 18:31 |
| *** esberglu has quit IRC | 18:31 | |
| shananigans | omegapoint: correct. | 18:32 |
| MikeW | odyssey4me Does having a magnum container defined in the inventory suffice for repo venv builds or is there a specific field within the object required? | 18:32 |
| odyssey4me | MikeW just a member of the magnum_all group | 18:33 |
| *** pcaruana has quit IRC | 18:33 | |
| odyssey4me | so it can be a host or container | 18:33 |
| omegapoint | @shananigans alright, thanks | 18:33 |
| MikeW | odyssey4me I pulled the out with a quick copy not properly formatted... I think it's wrong? lol https://pastebin.com/kvr8Df91 | 18:35 |
| *** udesale_ has quit IRC | 18:36 | |
| odyssey4me | MikeW yep, empty group | 18:36 |
| odyssey4me | do you have an entry in conf.d/openstack_user_config for magnum-infra_hosts ? | 18:36 |
| *** berendt_ has joined #openstack-ansible | 18:37 | |
| *** neith_ has joined #openstack-ansible | 18:37 | |
| *** hogepodge_ has joined #openstack-ansible | 18:37 | |
| MikeW | No it's not empty which is what's confusing me https://pastebin.com/AZCGdPgh | 18:37 |
| odyssey4me | MikeW hmm, that is odd | 18:38 |
| *** gus_ has joined #openstack-ansible | 18:38 | |
| *** Tyrantelf_ has joined #openstack-ansible | 18:38 | |
| MikeW | odyssey4me Is there a tool for repairing the inventory? | 18:39 |
| *** kysse_ has joined #openstack-ansible | 18:39 | |
| MikeW | odyssey4me Well I know we ahve the python inventory management thing I can play with it perhaps to insert the proper info? | 18:40 |
| *** shardy_a1k has joined #openstack-ansible | 18:40 | |
| odyssey4me | MikeW is your /etc/openstack_deploy read-only, or out of space or something perhaps? | 18:40 |
| odyssey4me | check that the inventory.json file is not protected in some way | 18:41 |
| *** stuartgr has left #openstack-ansible | 18:42 | |
| odyssey4me | MikeW if you run the following, do you see a populated group in the resulting json output? /opt/ansible-runtime/bin/python inventory/dynamic_inventory.py --debug | 18:42 |
| odyssey4me | adjust the path as necessary for the branch you're using | 18:43 |
| *** logan_ has joined #openstack-ansible | 18:43 | |
| odyssey4me | MikeW there are several magnum groups, as they're nested layers - check the other groups? | 18:44 |
| *** tobberydberg has quit IRC | 18:44 | |
| *** berendt has quit IRC | 18:44 | |
| *** kysse has quit IRC | 18:44 | |
| *** dariko has quit IRC | 18:44 | |
| *** shardy_afk has quit IRC | 18:44 | |
| *** logan- has quit IRC | 18:44 | |
| *** neith has quit IRC | 18:44 | |
| *** hogepodge has quit IRC | 18:44 | |
| *** ArchiFleKs has quit IRC | 18:44 | |
| *** gus has quit IRC | 18:44 | |
| *** astellwag has quit IRC | 18:44 | |
| *** Tyrantelf has quit IRC | 18:44 | |
| *** berendt_ is now known as berendt | 18:44 | |
| *** neith_ is now known as neith | 18:44 | |
| *** hogepodge_ is now known as hogepodge | 18:44 | |
| MikeW | odyssey4me Other groups are populated the only one not is magnum_all | 18:45 |
| MikeW | So weird | 18:45 |
| odyssey4me | can you pastebin the set? | 18:45 |
| MikeW | Yeah sure | 18:46 |
| MikeW | https://pastebin.com/NCi1pC7N | 18:47 |
| *** logan_ is now known as logan- | 18:47 | |
| *** tobberydberg has joined #openstack-ansible | 18:48 | |
| odyssey4me | magnum_all has the child magnum, which has the containers | 18:48 |
| MikeW | odyssey4me Yeah I noticed that relationship, but didn't know if repo would follow that trail down... so it should be building the venvs then right? | 18:49 |
| odyssey4me | it should be, yes | 18:49 |
| odyssey4me | could you pastebin the requirements.txt from /var/www/repo/os-releases/... from the repo container? | 18:50 |
| *** dariko has joined #openstack-ansible | 18:51 | |
| MikeW | Yeah here you go: https://pastebin.com/EmVQG9f7 | 18:51 |
| *** TxGirlGeek has joined #openstack-ansible | 18:51 | |
| odyssey4me | magnum's there, so the requirements are right and there should be a wheel | 18:51 |
| *** astellwag has joined #openstack-ansible | 18:52 | |
| odyssey4me | ok, there should also be a venv_requirements.txt file there for magnum | 18:52 |
| *** ArchiFleKs has joined #openstack-ansible | 18:52 | |
| odyssey4me | something like this: http://rpc-repo.rackspace.com/os-releases/r16.1.0/ubuntu-16.04-x86_64/venv-build-options-os_magnum.txt | 18:52 |
| MikeW | Yeah I see that file as well | 18:52 |
| odyssey4me | ok, so the venv should be there? | 18:53 |
| odyssey4me | it should be in venvs/<tag>/<distro>/ | 18:54 |
| MikeW | Hmm... https://pastebin.com/VkHVm90P | 18:54 |
| odyssey4me | ok, interesting - right, have you tried forcing a venv rebuild then? | 18:54 |
| *** TxGirlGeek has quit IRC | 18:55 | |
| odyssey4me | openstack-ansible repo-build.yml -e repo_build_venv_rebuild=yes | 18:55 |
| MikeW | I was going to get it to pull down all venvs I'll just rebuild instead | 18:55 |
| MikeW | Is this a mistake by me? Do I need to set that var every time I add a new service? | 18:56 |
| odyssey4me | nope, you shouldn't have to - it should detect the new requirement, build the options file, then build the venv | 18:57 |
| *** lbragstad has joined #openstack-ansible | 18:57 | |
| *** ansmith has quit IRC | 18:57 | |
| odyssey4me | somehow that didn't happen here - not sure how | 18:57 |
| *** TxGirlGeek has joined #openstack-ansible | 18:59 | |
| *** TxGirlGeek has quit IRC | 19:03 | |
| cloudnull | odyssey4me: mind giving me a push here https://review.openstack.org/#/c/547157/ | 19:03 |
| odyssey4me | duh, I just realised why that openvswitch issue is happening | 19:03 |
| cloudnull | oh? | 19:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: neutron_needs_openvswitch should be jinja, not a string https://review.openstack.org/547650 | 19:05 |
| odyssey4me | cloudnull ^ that's why | 19:05 |
| cloudnull | oh... | 19:06 |
| cloudnull | that's going to need a backport to all the things | 19:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: neutron_needs_openvswitch should be jinja, not a string https://review.openstack.org/547650 | 19:06 |
| odyssey4me | doesn't look like it - looks like that patch was master only | 19:07 |
| *** adcpdk has quit IRC | 19:07 | |
| cloudnull | in queens https://github.com/openstack/openstack-ansible-os_neutron/blob/stable/queens/vars/main.yml#L19-L22 | 19:08 |
| cloudnull | looks like pike is ok | 19:08 |
| *** jillr has joined #openstack-ansible | 19:08 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: neutron_needs_openvswitch should be jinja, not a string https://review.openstack.org/547650 | 19:09 |
| odyssey4me | I think that should do it. | 19:09 |
| *** TxGirlGeek has joined #openstack-ansible | 19:10 | |
| odyssey4me | cloudnull are you gonna just get those in, then unlock the other tests over time? | 19:12 |
| cloudnull | yes. | 19:13 |
| cloudnull | i need to get the host test in | 19:13 |
| cloudnull | then the container ones. | 19:13 |
| cloudnull | then re-enable the host + contianer ones | 19:14 |
| cloudnull | i left the tests commented out so I can do it in phases | 19:14 |
| cloudnull | *the container tests** | 19:14 |
| odyssey4me | ok, pushing it through | 19:15 |
| odyssey4me | better to have something now than nothing | 19:15 |
| cloudnull | tyvm | 19:15 |
| * cloudnull doing the container_create now | 19:16 | |
| odyssey4me | cloudnull got a couple more of these to work through when you've got a gap: https://review.openstack.org/#/q/topic:bug/1750656+status:open | 19:18 |
| cloudnull | all lgtm | 19:22 |
| dmsimard | Hey, I don't have a blog post about making ARA work with OSA yet but I got an idea and posted about it: https://dmsimard.com/2018/02/23/rebranding-ansible-run-analysis-to-ara-records-ansible/ | 19:23 |
| dmsimard | Feel free to share ^ I'm being noisy on purpose :p | 19:23 |
| odyssey4me | hahaha | 19:24 |
| MikeW | odyssey4me Oh my god | 19:25 |
| MikeW | I built all the venvs... guess which one is missing | 19:25 |
| odyssey4me | MikeW ? | 19:25 |
| odyssey4me | oh, so you set it to do them all, and magnum is still missing? | 19:26 |
| MikeW | odyssey4me Sorry... I set the repo build to not be selective in which venvs it installs, and the only one that isn't in now is magnum | 19:26 |
| MikeW | Yeah | 19:26 |
| odyssey4me | lol, ok that's weird | 19:26 |
| odyssey4me | have you modified https://github.com/openstack/openstack-ansible/blob/master/playbooks/defaults/repo_packages/openstack_services.yml at all, or put overrides for any of those vars into you user_variables.yml ? | 19:27 |
| omegapoint | @shananigans figured it out, the switch configuration was wrong and not connected to the upstream network where the gateway was. (i'm not using interface bonding and vlan tagging as implied by the example deployment) | 19:29 |
| shananigans | nifty. glad to hear its working | 19:30 |
| *** TxGirlGeek has quit IRC | 19:30 | |
| odyssey4me | MikeW you can try building it yourself directly using: /opt/venv-build-script.sh /var/www/repo/os-releases/<tag>/<distro>/venv-build-options-os_magnum.txt | 19:31 |
| MikeW | odyssey4me Interesting the manual build fails, but I'm getting now failures and no errors from the playbook | 19:33 |
| MikeW | I'll keep digging and post back if it's something you'd be interested in... probably something wrong on my servers | 19:34 |
| odyssey4me | MikeW happy to hear feedback - I'm a bit concerned that the error is not bubbling up, so I'd like to know the root cause to see if there's a way we can catch it and make sure it bubbles up next time | 19:36 |
| cloudnull | dmsimard: retweeted the post :) | 19:41 |
| dmsimard | cloudnull: awww yeah, thanks <3 | 19:45 |
| dmsimard | I'll buy you maple syrup if you come to Vancouver :p | 19:46 |
| dmsimard | (maple syrup is more valuable than beer here) | 19:46 |
| *** TxGirlGeek has joined #openstack-ansible | 19:49 | |
| *** TxGirlGeek has quit IRC | 19:49 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:52 | |
| *** lvdombrkr89 has joined #openstack-ansible | 20:00 | |
| *** omegapoint has quit IRC | 20:01 | |
| *** omegapoint has joined #openstack-ansible | 20:01 | |
| *** TxGirlGeek has quit IRC | 20:01 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:02 | |
| *** TxGirlGeek has quit IRC | 20:05 | |
| *** pbandark has quit IRC | 20:05 | |
| *** omegapoint has quit IRC | 20:05 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-nspawn_container_create master: Add minimal functional gate https://review.openstack.org/547158 | 20:22 |
| cloudnull | odyssey4me: ^ if you're still around :) | 20:23 |
| cloudnull | dmsimard: lol! | 20:23 |
| cloudnull | will you be at the PTG? | 20:23 |
| dmsimard | cloudnull: nop, hence my suggestion for Vancouver | 20:25 |
| odyssey4me | cloudnull lgtm, although zuul has croaked - infra's busy working on restarting it | 20:25 |
| odyssey4me | cloudnull mind giving this a nudge? https://review.openstack.org/546259 | 20:26 |
| odyssey4me | and https://review.openstack.org/546147 | 20:27 |
| odyssey4me | lol, and https://review.openstack.org/545894 / https://review.openstack.org/545895 | 20:27 |
| odyssey4me | and finally https://review.openstack.org/547101 / https://review.openstack.org/546008 | 20:27 |
| *** weezS has quit IRC | 20:28 | |
| *** weezS has joined #openstack-ansible | 20:28 | |
| odyssey4me | https://media.giphy.com/media/7cTTE2Z1OmrFm/giphy.gif | 20:28 |
| *** weezS has quit IRC | 20:28 | |
| *** weezS has joined #openstack-ansible | 20:29 | |
| *** weezS has joined #openstack-ansible | 20:30 | |
| *** weezS has quit IRC | 20:30 | |
| *** weezS has joined #openstack-ansible | 20:30 | |
| *** weezS has quit IRC | 20:31 | |
| *** weezS has joined #openstack-ansible | 20:31 | |
| *** weezS has quit IRC | 20:32 | |
| *** weezS has joined #openstack-ansible | 20:32 | |
| *** omegapoint has joined #openstack-ansible | 20:32 | |
| *** pramodrj07 has joined #openstack-ansible | 20:34 | |
| cloudnull | done. | 20:38 |
| odyssey4me | cloudnull thx - I'm busy self-testing https://review.openstack.org/547650 given the zuul situation | 20:39 |
| *** pbandark has joined #openstack-ansible | 20:42 | |
| *** pbandark has quit IRC | 20:46 | |
| MikeW | odyssey4me I can't find a way around it... magnum is pulling down kubernetes which is pulling down websocket-client. When it tries to get websocket-client it can't find the right version | 20:46 |
| cloudnull | MikeW: is that a dep in their requirements file that's unversioned/broken ? | 20:47 |
| odyssey4me | MikeW ah, I know that issue | 20:47 |
| odyssey4me | which branch is this? | 20:47 |
| odyssey4me | or which OSA tag/branch are you using? | 20:47 |
| MikeW | I'm running ocata 15.1.6 or something | 20:47 |
| odyssey4me | aha, yeah - gimme a sec | 20:48 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_magnum/commit/fd806ccddec31e36234ce979906448501db47083 | 20:48 |
| *** pbandark has joined #openstack-ansible | 20:49 | |
| cloudnull | odyssey4me: for prez! | 20:49 |
| odyssey4me | the current head of the integrated build's stable/ocata branch includes that fix: https://github.com/openstack/openstack-ansible/commit/c5bbe694cb67fef3713ca3483cf2a49bcbd59f8f | 20:49 |
| MikeW | Oh thank god | 20:49 |
| MikeW | odyssey4me Has my vote! Thanks everyone | 20:49 |
| *** omegapoint has quit IRC | 20:49 | |
| odyssey4me | actually, it was pulled in just before the head: https://github.com/openstack/openstack-ansible/commit/4aba989d084d373a58a06af93d980279f3aba5f5 | 20:50 |
| odyssey4me | :) happy to help - a quick minor version upgrade will get you on your way | 20:50 |
| odyssey4me | it was fghaas who raised it just the other day, so thanks to both of you for raising it as feedback, so that we could fix it | 20:51 |
| *** omegapoint has joined #openstack-ansible | 20:55 | |
| *** omegapoint has quit IRC | 20:59 | |
| *** pramodrj07 has quit IRC | 21:02 | |
| *** hamza21 has quit IRC | 21:04 | |
| *** TxGirlGe_ has joined #openstack-ansible | 21:08 | |
| *** omegapoint has joined #openstack-ansible | 21:13 | |
| *** pbandark has quit IRC | 21:15 | |
| *** omegapoint has quit IRC | 21:18 | |
| *** omegapoint has joined #openstack-ansible | 21:20 | |
| odyssey4me | mhayden it would appear that we broken something somewhere: http://logs.openstack.org/97/547497/1/check/openstack-ansible-functional-centos-7/c5a4e2d/logs/ara/result/89507e4a-6c94-4d0c-9ac3-66bea266ada9/ | 21:23 |
| mhayden | okay well that's downright weird | 21:24 |
| mhayden | the key should still be present | 21:24 |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_hosts master: add minimal functional tests https://review.openstack.org/547157 | 21:25 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547402 | 21:25 |
| *** omegapoint has quit IRC | 21:25 | |
| mhayden | yeah, the percona-release installed fine | 21:25 |
| odyssey4me | heh, yeah - just found that: http://logs.openstack.org/97/547497/1/check/openstack-ansible-functional-centos-7/c5a4e2d/logs/ara/result/c1a60b22-b965-4ab0-9266-0b865737310a/ | 21:26 |
| mhayden | odyssey4me: i wonder if it has to do with this patch https://github.com/openstack/openstack-ansible-galera_server/commit/dbbcb56a62e4a97af3f63f970a34a3c815352ced | 21:26 |
| odyssey4me | yup, very likely - I already have one follow on from that one: https://review.openstack.org/547101 | 21:27 |
| odyssey4me | not a bad patch, but it did remove some things so there will be fall out | 21:27 |
| odyssey4me | mhayden cloudnull ok, I've confirmed that https://review.openstack.org/547650 unblocks our integrated gate - any chance for a review to hopefully get that out the way? | 21:29 |
| *** armaan has quit IRC | 21:30 | |
| *** armaan has joined #openstack-ansible | 21:30 | |
| *** NostawRm has quit IRC | 21:32 | |
| *** strobelight has joined #openstack-ansible | 21:32 | |
| *** NostawRm has joined #openstack-ansible | 21:32 | |
| mhayden | lookin' | 21:33 |
| *** MikeW has quit IRC | 21:34 | |
| mhayden | odyssey4me: done deal sir | 21:34 |
| *** dave-mcc_ has joined #openstack-ansible | 21:35 | |
| odyssey4me | tyvm | 21:35 |
| *** omegapoint has joined #openstack-ansible | 21:36 | |
| *** dave-mccowan has quit IRC | 21:36 | |
| *** omegapoint has quit IRC | 21:40 | |
| *** jwitko__ has joined #openstack-ansible | 21:41 | |
| *** armaan has quit IRC | 21:42 | |
| *** armaan has joined #openstack-ansible | 21:42 | |
| *** omegapoint has joined #openstack-ansible | 21:43 | |
| *** jwitko_ has quit IRC | 21:44 | |
| mhayden | odyssey4me: tiny typo in https://review.openstack.org/547554 | 21:46 |
| *** jwitko__ has quit IRC | 21:46 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Switch to using the imported nspawn roles https://review.openstack.org/547554 | 21:47 |
| odyssey4me | mhayden no there isn't ;) | 21:47 |
| mhayden | ___ ____ _ __ _____ | 21:48 |
| mhayden | / _ \| _ \| | \ \ / /__ \ | 21:48 |
| mhayden | | | | | |_) | | \ V / / / | 21:48 |
| mhayden | | |_| | _ <| |___| | |_| | 21:48 |
| mhayden | \___/|_| \_\_____|_| (_) | 21:48 |
| mhayden | gotta get down on friday | 21:48 |
| *** omegapoint has quit IRC | 21:49 | |
| odyssey4me | cloudnull not consistent, but concerning: http://logs.openstack.org/59/546259/1/gate/openstack-ansible-functional-ubuntu-xenial/88c8fbb/job-output.txt.gz#_2018-02-23_21_21_04_418711 | 21:51 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/queens: Fix cache update after initial apt_repository fail https://review.openstack.org/547399 | 21:51 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/ocata: Fix cache update after initial apt_repository fail https://review.openstack.org/547401 | 21:51 |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server master: Explicitly ensure when apt cache updates happen https://review.openstack.org/547471 | 21:51 |
| *** TxGirlGe_ has quit IRC | 21:53 | |
| *** TxGirlGeek has joined #openstack-ansible | 21:53 | |
| dmsimard | mhayden: Do you remember the orly owl ? Pepper ridge farm remembers. | 21:57 |
| mhayden | i do | 21:57 |
| *** omegapoint has joined #openstack-ansible | 22:01 | |
| *** strobelight has quit IRC | 22:06 | |
| *** omegapoint has quit IRC | 22:06 | |
| odyssey4me | cloudnull https://review.openstack.org/523218 really needs a revisit | 22:08 |
| cloudnull | it does | 22:09 |
| cloudnull | maybe i;; do that now | 22:09 |
| *** jwitko_ has joined #openstack-ansible | 22:11 | |
| *** bradm has joined #openstack-ansible | 22:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build master: [WIP] Make the git clone and wheel build optional https://review.openstack.org/537015 | 22:13 |
| *** omegapoint has joined #openstack-ansible | 22:16 | |
| *** omegapoint has quit IRC | 22:21 | |
| *** dariko has quit IRC | 22:22 | |
| *** omegapoint has joined #openstack-ansible | 22:25 | |
| *** masber has joined #openstack-ansible | 22:27 | |
| *** omegapoint has quit IRC | 22:29 | |
| *** masber has quit IRC | 22:30 | |
| *** masber has joined #openstack-ansible | 22:30 | |
| *** kstev1 has quit IRC | 22:33 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests master: Use OpenStack-Infra mirrors for tests https://review.openstack.org/547686 | 22:34 |
| *** dave-mcc_ has quit IRC | 22:34 | |
| *** omegapoint has joined #openstack-ansible | 22:41 | |
| *** omegapoint has quit IRC | 22:45 | |
| *** dave-mccowan has joined #openstack-ansible | 22:50 | |
| *** omegapoint has joined #openstack-ansible | 22:51 | |
| *** omegapoint has quit IRC | 22:56 | |
| *** weezS has quit IRC | 23:04 | |
| *** weezS has joined #openstack-ansible | 23:04 | |
| *** weezS has quit IRC | 23:04 | |
| *** weezS has joined #openstack-ansible | 23:05 | |
| *** weezS has joined #openstack-ansible | 23:06 | |
| *** omegapoint has joined #openstack-ansible | 23:06 | |
| *** weezS has joined #openstack-ansible | 23:06 | |
| *** masuberu has joined #openstack-ansible | 23:07 | |
| *** weezS has joined #openstack-ansible | 23:07 | |
| *** weezS has joined #openstack-ansible | 23:08 | |
| *** weezS has quit IRC | 23:08 | |
| *** weezS has joined #openstack-ansible | 23:09 | |
| *** masber has quit IRC | 23:09 | |
| *** weezS has joined #openstack-ansible | 23:09 | |
| *** weezS has quit IRC | 23:10 | |
| *** weezS has joined #openstack-ansible | 23:10 | |
| *** dave-mccowan has quit IRC | 23:11 | |
| *** omegapoint has quit IRC | 23:11 | |
| *** weezS has quit IRC | 23:11 | |
| *** weezS has joined #openstack-ansible | 23:11 | |
| *** weezS has quit IRC | 23:11 | |
| *** weezS has joined #openstack-ansible | 23:12 | |
| *** weezS has joined #openstack-ansible | 23:13 | |
| *** weezS has joined #openstack-ansible | 23:14 | |
| *** weezS has joined #openstack-ansible | 23:14 | |
| *** weezS has quit IRC | 23:15 | |
| *** weezS has joined #openstack-ansible | 23:15 | |
| *** weezS has joined #openstack-ansible | 23:16 | |
| *** weezS has quit IRC | 23:16 | |
| *** weezS has joined #openstack-ansible | 23:17 | |
| *** weezS has quit IRC | 23:17 | |
| *** omegapoint has joined #openstack-ansible | 23:18 | |
| *** omegapoint has quit IRC | 23:22 | |
| *** omegapoint has joined #openstack-ansible | 23:31 | |
| *** messy has joined #openstack-ansible | 23:35 | |
| *** omegapoint has quit IRC | 23:35 | |
| *** messy has quit IRC | 23:38 | |
| *** omegapoint has joined #openstack-ansible | 23:51 | |
| *** agrebennikov has joined #openstack-ansible | 23:57 | |
| *** omegapoint has quit IRC | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
