| *** Tengu has quit IRC | 00:04 | |
| *** Tengu has joined #openstack-kolla | 00:06 | |
| *** k_mouza has quit IRC | 00:41 | |
| *** k_mouza has joined #openstack-kolla | 00:43 | |
| *** jbadiapa has quit IRC | 01:37 | |
| *** dking_desktop has joined #openstack-kolla | 01:57 | |
| *** k_mouza has quit IRC | 02:11 | |
| *** jcmdln has joined #openstack-kolla | 02:26 | |
| *** xinliang has quit IRC | 02:55 | |
| *** xinliang has joined #openstack-kolla | 02:58 | |
| *** cah_link has joined #openstack-kolla | 03:23 | |
| *** xinliang has quit IRC | 03:33 | |
| *** cah_link1 has joined #openstack-kolla | 03:39 | |
| *** cah_link has quit IRC | 03:40 | |
| *** xinliang has joined #openstack-kolla | 03:41 | |
| *** cah_link1 has quit IRC | 03:43 | |
| *** cah_link has joined #openstack-kolla | 03:43 | |
| *** cah_link has quit IRC | 03:47 | |
| *** cah_link has joined #openstack-kolla | 03:48 | |
| *** cah_link1 has joined #openstack-kolla | 04:23 | |
| *** cah_link has quit IRC | 04:24 | |
| *** cah_link1 is now known as cah_link | 04:24 | |
| *** cah_link1 has joined #openstack-kolla | 04:27 | |
| *** cah_link has quit IRC | 04:29 | |
| *** cah_link1 is now known as cah_link | 04:29 | |
| *** skramaja has joined #openstack-kolla | 04:38 | |
| *** Tony31 has joined #openstack-kolla | 04:42 | |
| *** cah_link has quit IRC | 04:47 | |
| Tony31 | Hi is anyone here yet? | 05:08 |
|---|---|---|
| *** evrardjp has quit IRC | 05:34 | |
| *** evrardjp has joined #openstack-kolla | 05:34 | |
| *** shyamb has joined #openstack-kolla | 05:54 | |
| *** numans has quit IRC | 06:02 | |
| *** skramaja has quit IRC | 06:04 | |
| yoctozepto | Tony31: that was very early, we only really start opening our eyes at that time :-) | 06:31 |
| Tony31 | no excuse ;) | 06:31 |
| Tony31 | is it possible to use `host_vars` to set a variable for a host? My problem is that, the controller VM is not PXEboot, so it does not have a PXE interface and the command `kayobe overcloud host configure` fails because it validates this interface. So I wanted to set this for the controller host only: `provision_oc_net_name: inside` but I tried tons | 06:33 |
| Tony31 | of things and none of them resolved. | 06:33 |
| Tony31 | About 30 mins ago I changed this setting in network-allocation.yml :/ | 06:33 |
| *** shyamb has quit IRC | 07:02 | |
| *** shyamb has joined #openstack-kolla | 07:03 | |
| *** jbadiapa has joined #openstack-kolla | 07:05 | |
| *** ricolin has quit IRC | 07:11 | |
| *** shyamb has quit IRC | 07:22 | |
| *** cah_link has joined #openstack-kolla | 07:22 | |
| *** shyamb has joined #openstack-kolla | 07:35 | |
| hrw | morning | 07:41 |
| hrw | yoctozepto: infra passed basic CI :D | 07:41 |
| *** bengates has joined #openstack-kolla | 07:49 | |
| *** rpittau|afk is now known as rpittau | 08:10 | |
| yoctozepto | hrw: yay | 08:14 |
| yoctozepto | what was wrong then? | 08:15 |
| hrw | dockerfile rewriting | 08:15 |
| yoctozepto | hmm, oddly it somehow worked locally | 08:15 |
| hrw | yep | 08:15 |
| hrw | k-a change will be tricky | 08:15 |
| hrw | and then merging all after c8 work | 08:15 |
| yoctozepto | Tony31: need to wait for mnasiadka and/or mgoddard, question too specific to kayobe | 08:16 |
| mnasiadka | morning | 08:16 |
| *** carli has joined #openstack-kolla | 08:17 | |
| Tony31 | good morning :) | 08:17 |
| Tony31 | TBH I don't like just waking up and having things demanded of me, so apologies for that | 08:17 |
| *** gfidente has joined #openstack-kolla | 08:18 | |
| mnasiadka | Tony31: no problem, you can override the variables on host_vars/group_vars level - so it should work | 08:19 |
| Tony31 | wouldnt work :( | 08:20 |
| mnasiadka | Tony31: under inventory/host_vars/host_name/network.yml just put the overrides | 08:20 |
| Tony31 | I was using host_vars/hostname.yml as well as host_vars/compute0 and host_vars/compute0.yml and I also copied that to the kolla directory.... but having read your update now I think I Was doing it wrong | 08:20 |
| Tony31 | I'll try it - although I am halfway through deploying at the moment and it's failing waiting for mariadb | 08:21 |
| Tony31 | it shows an error that its waiting to connect to hostname-int.domain:3306 but hostname-int.domain is not dns resolvabe and there are no /etc/host entries for it. I just put in a host entry - how is this part supposed to work? | 08:22 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: Create 'infra' type of images https://review.opendev.org/698753 | 08:22 |
| hrw | moved aarch64 job to ubuntu | 08:22 |
| *** tonythomas has joined #openstack-kolla | 08:23 | |
| *** negronjl has quit IRC | 08:23 | |
| Tony31 | the /etc/hosts didnt resolve that issue | 08:24 |
| Tony31 | this host will not be reachable from the ansible host... `TASK [mariadb : Waiting for MariaDB service to be ready through VIP] **` | 08:25 |
| *** negronjl has joined #openstack-kolla | 08:27 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla master: Add collectd-pmu package for Intel PMU (C7 only) https://review.opendev.org/704560 | 08:27 |
| *** shyamb has quit IRC | 08:27 | |
| Tony31 | another problem I am seeing is the controller is not getting configured on the storage network. for tripleo, when an instance is built, the controller logs into the iscsi target and installs the image into it. Then the iscsi is moved to the host so the instance can boot from it. Which kayobe `network` is this for? I guess I have this listed | 08:28 |
| Tony31 | incorrectly in the networks section, so it's using another network instead of my storage net. | 08:28 |
| Tony31 | mnasiadka `inventory/host_vars/hostname/network.yml` is also not working | 08:30 |
| Tony31 | my hostname is juc-kcont1-prd for the controller | 08:30 |
| mnasiadka | so it's inventory/host_vars/juc-kcont1-prd/network.yml? | 08:31 |
| Tony31 | wait, now there's only one error instead of two | 08:31 |
| mnasiadka | and are you sure your controller in inventory/overcloud is juc-kcont1-prd? | 08:31 |
| Tony31 | so it looks like it's half working - i'll check spelling | 08:31 |
| mnasiadka | not by ip or something? | 08:31 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla master: Add collectd-pmu package for Intel PMU (C7 only) https://review.opendev.org/704560 | 08:31 |
| Tony31 | The error is `failed: [juc-kcont1-prd] (item={u'var_name': u'kolla_inspector_dnsmasq_interface', u'description': u'Bare metal introspection network interface name', u'required': True}) => {"ansible_loop_var": "item", "changed": false, "item": {"description": "Bare metal introspection network interface name", "required": true, "var_name": | 08:32 |
| Tony31 | "kolla_inspector_dnsmasq_interface"}, "msg": "The Kolla Ansible variable kolla_inspector_dnsmasq_interface (Bare metal introspection network interface name) is invalid. Value: \"<undefined>\".\n"}` | 08:32 |
| Tony31 | do you know which network does this relate to in `networks.yml` ? I thought this was `inspection_net_name:` | 08:32 |
| Tony31 | mnasiadka I am not using the seed baremetal inventory, so that file was empty... is it wrong? | 08:33 |
| mnasiadka | Tony31: well, you have no overcloud nodes to deploy openstack to? :) | 08:36 |
| Tony31 | :) | 08:36 |
| Tony31 | understood - I put in the hostname | 08:36 |
| *** ivve has joined #openstack-kolla | 08:37 | |
| mnasiadka | Tony31: one moment, so you had this node in inventory/hosts, right? under controllers group? | 08:39 |
| mnasiadka | Tony31: or the seed/seed-hypervisor group? | 08:39 |
| Tony31 | inventory/hosts as a controller | 08:39 |
| yoctozepto | mnasiadka, hrw: https://review.opendev.org/704560 | 08:40 |
| patchbot | patch 704560 - kolla - Add collectd-pmu package for Intel PMU (C7 only) - 5 patch sets | 08:40 |
| Tony31 | in the controllers group | 08:40 |
| mnasiadka | ok, the seed/seed-hypervisor groups are empty? | 08:40 |
| Tony31 | no, [seed] has the seed in there. but i manage the VMs (seed / compute / controller / ansible) manually in terms of installing the OS and setting up SSH etc | 08:41 |
| Tony31 | seed hypervisor is empty | 08:41 |
| mnasiadka | yoctozepto: once zuul passes I'll push it :) | 08:41 |
| mnasiadka | Tony31: well, if you don't need to install OS on anything, I suppose you don't need the seed | 08:41 |
| mnasiadka | but I might be wrong, I'm not so knowledgable in kayobe as Mark is | 08:42 |
| Tony31 | i wanted to use the seed to build out the real physical compute systems later | 08:42 |
| Tony31 | seed seems to be working ok | 08:42 |
| mnasiadka | ah ok | 08:42 |
| mnasiadka | Tony31: so, it would be good when sharing errors to include the task/play name it fails on - it's easier for us to help you :) | 08:43 |
| Tony31 | i can't get passed this error though `failed: [juc-kcont1-prd] (item={u'var_name': u'kolla_inspector_dnsmasq_interface', u'description': u'Bare metal introspection network interface name', u'required': True}) => {"ansible_loop_var": "item", "changed": false, "item": {"description": "Bare metal introspection network interface name", "required": | 08:43 |
| Tony31 | true, "var_name": "kolla_inspector_dnsmasq_interface"}, "msg": "The Kolla Ansible variable kolla_inspector_dnsmasq_interface (Bare metal introspection network interface name) is invalid. Value: \"<undefined>\".\n"}` | 08:43 |
| Tony31 | `TASK [Validate overcloud host Kolla Ansible network configuration] **` | 08:43 |
| Tony31 | I created that host_vars file and I have : | 08:43 |
| Tony31 | `provision_oc_net_name: inside | 08:43 |
| Tony31 | initially I had an error about the provision network - I dont get that error anymore | 08:44 |
| Tony31 | I'm not sure if the `inspection_net_name:` is the right one to be using... maybe it is not since I am getting the error but I can't link that error to a network | 08:44 |
| mnasiadka | Tony31: do you have kolla_enable_ironic set to True? | 08:44 |
| Tony31 | where would it be set? | 08:45 |
| Tony31 | i can check | 08:45 |
| mnasiadka | in kolla.yml | 08:45 |
| Tony31 | no - the only thing enabled in the kolla.yml file is the ansible bug thing | 08:45 |
| Tony31 | `workaround_ansible_issue_8743: yes` | 08:46 |
| mnasiadka | anyway - just set kolla_inspector_dnsmasq_interface variable to some interface | 08:51 |
| mnasiadka | yoctozepto: we still haven't fixed fluentplugin-elasticsearch on Ubuntu/rocky? | 08:54 |
| yoctozepto | mnasiadka: yeah, probably | 08:54 |
| *** k_mouza has joined #openstack-kolla | 09:00 | |
| *** k_mouza has quit IRC | 09:05 | |
| *** dougsz has joined #openstack-kolla | 09:06 | |
| Tony31 | where's `kolla_inspector_dnsmasq_interface` ? is there a kayobe variable for this ? | 09:17 |
| Tony31 | I searched the kayobe directory and these are the results: | 09:17 |
| mgoddard | morning | 09:17 |
| mgoddard | evening Tony31 :) | 09:17 |
| hrw | https://review.opendev.org/#/c/706428/1 - brexit everywhere... | 09:17 |
| patchbot | patch 706428 - opendev/system-config - Replace nb03.openstack.org - 1 patch set | 09:17 |
| Tony31 | [cv-user@juc-kach1-prd kayobe]$ grep -rn -e "kolla_inspector_dnsmasq_interface" | 09:18 |
| Tony31 | "kolla_inspector_dnsmasq_interface" | 09:18 |
| Tony31 | venvs/kayobe/share/kayobe/ansible/roles/kolla-ansible/tests/test-extras.yml:27: kolla_inspector_dnsmasq_interface: "eth9" | 09:18 |
| Tony31 | all results are in the `venv` | 09:18 |
| Tony31 | was expecting a `kayobe-config` | 09:19 |
| hrw | yoctozepto: +2 | 09:19 |
| mgoddard | Tony31: bonus 'feature' of kayobe is that it enables ironic by default. we should probably change that | 09:21 |
| mgoddard | Tony31: try setting `kolla_enable_ironic: no` in your etc/kayobe/kolla.yml | 09:22 |
| *** shyamb has joined #openstack-kolla | 09:22 | |
| *** shyamb has quit IRC | 09:24 | |
| *** shyamb has joined #openstack-kolla | 09:25 | |
| openstackgerrit | Noboru Iwamatsu proposed openstack/kolla-ansible master: Adapt to Octavia Certificate Configuration Guide. https://review.opendev.org/706203 | 09:28 |
| Tony31 | cheers - so now ironic is turned off, do I need to manually install all my hosts? Or can I pxe some? | 09:31 |
| *** k_mouza has joined #openstack-kolla | 09:32 | |
| Tony31 | hrw - under sea cables are always breaking... I'm always getting issues going from WA to Sri Lanka. Sometimes, the only path is through USA and across Europe. | 09:35 |
| Tony31 | Capacity is a common issue also these days. Not enough of it | 09:35 |
| hrw | Tony31: cables... | 09:36 |
| Tony31 | even if there are mutliple under sea cables, we cant expect to use them. It depends on who owns the cables and whom they are leased to / or which ISPs are paying "rent" to use them. | 09:37 |
| hrw | Tony31: I remember times when it was easier to cross a street with data than send it over network | 09:37 |
| hrw | Tony31: as data had to pass around country | 09:37 |
| Tony31 | Today thats mostly true for cloud migrations :) When there are TB of data for example. I'm lucky as I've only been doing this about 15 years. | 09:38 |
| hrw | :) | 09:38 |
| hrw | Tony31: that's why Amazon still has truck-as-a-service for migrating data to the cloud | 09:39 |
| Tony31 | :) | 09:40 |
| Tony31 | I need to speak with you guys about a bunch of things: | 09:40 |
| Tony31 | 1. the ` kayobe overcloud host configure` is now passing with ironic = off. | 09:40 |
| Tony31 | 2. can I no longer use ironic now? | 09:41 |
| Tony31 | (no real issue if not) I just need to plan it out | 09:41 |
| Tony31 | when trying to deploy the overcloud, the compute node is not being touched.. not sure why. May be it's set up after the controller is deployed | 09:41 |
| Tony31 | 4. The controller is not getting the storage network configured into it and I am not sure why... I need the storage network to be able to build instances. | 09:42 |
| hrw | you lost point 3. | 09:42 |
| Tony31 | 3. when trying to deploy the overcloud, the compute node is not being touched.. not sure why. May be it's set up after the controller is deployed :) | 09:42 |
| Tony31 | for point 3, in inventory/overcloud when I was trying to use baremetal it had put both controller and compute node under `[controllers]` I removed both of these when I deleted the 2 x baremetal nodes from the seed | 09:44 |
| Tony31 | for point 4, which network is used for building instances from glance? | 09:44 |
| Tony31 | 5. `kayobe overcloud service deploy` fails because it's waiting for mariadb and fails with error about connecting to the host:3306. How is DNS managed here? Do I need to create the dns entry or host entry manually? | 09:46 |
| mgoddard | 1. good | 09:47 |
| mgoddard | 2. no ironic for baremetal instances, but yes for deploying controllers/hypervisors via the seed | 09:48 |
| mgoddard | 3. sounds like the inventory is not setup correctly. Compute nodes needs to be in the compute group | 09:48 |
| Tony31 | I thought ironic was used for deploying hypervisors? | 09:48 |
| mgoddard | it is | 09:48 |
| mgoddard | but that is running in the seed | 09:49 |
| Tony31 | sorry for my confusion - I've just turned it off in kolla.yml | 09:49 |
| mgoddard | kolla_enable_ironic allows you to deploy another ironic on the controllers which can be used to deploy baremetal compute nodes (not hypervisors, users get bare metal) | 09:49 |
| Tony31 | ok. thanks for that - I still am not quite sure but it will probably make sense when I see it later | 09:50 |
| Tony31 | :) | 09:50 |
| Tony31 | about the inventory, i have this: `[baremetal-compute]# Add baremetal compute nodes here if required.juc-kcomp1-prd` | 09:50 |
| Tony31 | the baremetal-compute is the only "compute" section | 09:50 |
| Tony31 | is it correct? | 09:50 |
| Tony31 | deployment now fails with `TASK [nova-cell : Checking available compute nodes in inventory] ************************************************************************************************************************* | 09:51 |
| Tony31 | disabled.\n"} | 09:51 |
| Tony31 | `¯\_(ツ)_/¯` | 09:51 |
| mgoddard | ah, no | 09:52 |
| mgoddard | just [compute] | 09:52 |
| Tony31 | ah - it's not there. i'll add it | 09:52 |
| hrw | hm. we have not merged yet erlang/rabbitmq changes, right? | 09:52 |
| mgoddard | [baremetal-compute] is for baremetal compute nodes, not hypervisors | 09:53 |
| mgoddard | Tony31: have you seen https://docs.google.com/presentation/d/1L6KHoqcQi835Idix2xO_XzlgSn7nUk5DTr71A7dZ5vk ? | 09:53 |
| mgoddard | in particular, https://docs.google.com/presentation/d/1L6KHoqcQi835Idix2xO_XzlgSn7nUk5DTr71A7dZ5vk/edit#slide=id.g347582824b_0_453 | 09:53 |
| hrw | ok, have to create train patch from https://review.opendev.org/695356 one... | 09:54 |
| patchbot | patch 695356 - kolla - rabbitmq: update to latest 3.7.x (+ erlang 22.x) (ABANDONED) - 16 patch sets | 09:54 |
| Tony31 | mgoddard I hadnt seen that... | 09:54 |
| Tony31 | My understanding was "compute" was a hypervisor for the instances | 09:54 |
| mgoddard | yes | 09:55 |
| mgoddard | which is what you want | 09:55 |
| Tony31 | yes | 09:55 |
| Tony31 | but the ACH doesnt deploy it via seed? | 09:56 |
| Tony31 | the controller deploys it? | 09:56 |
| Tony31 | wait | 09:56 |
| Tony31 | are you saying `baremetal-compute` is for deploying instances from openstack, not as VMs but onto the entire bare-metal ? | 09:56 |
| mgoddard | yes | 09:57 |
| Tony31 | ah I understand now | 09:57 |
| Tony31 | thank you | 09:57 |
| mgoddard | np | 09:57 |
| yoctozepto | hrw: train patch? | 09:59 |
| hrw | yoctozepto: it will land on master first but I need it in train | 10:01 |
| yoctozepto | hrw: mgoddard will block you | 10:01 |
| hrw | yoctozepto: Linaro Developer Cloud migrates to Train so we need patched libvirt so we need to add repo | 10:01 |
| yoctozepto | libvirt? you mentioned erlang+rmq | 10:01 |
| hrw | yoctozepto: once patch is in gerrit I have infra to using them on CI :D | 10:02 |
| hrw | yoctozepto: repo info taken from that patch ;d | 10:02 |
| yoctozepto | hrw: ah | 10:03 |
| *** Tony312 has joined #openstack-kolla | 10:03 | |
| Tony312 | i got kicked out and i cant log in because my session hasnt timed out yet | 10:03 |
| *** Tony31 has quit IRC | 10:03 | |
| hrw | yoctozepto: https://paste.centos.org/view/f907f0ed | 10:04 |
| Tony312 | cant hang around right now - will try and come back online later but if i cant then I'll send an email. Thanks again guys for all your help. | 10:05 |
| *** Tony312 has quit IRC | 10:10 | |
| *** devfaz has quit IRC | 10:10 | |
| mnasiadka | yoctozepto: https://review.opendev.org/#/c/706055 +w or not +w? :) | 10:10 |
| patchbot | patch 706055 - kolla-ansible - CI: Replace cinder-lvm scenario with zun scenario - 1 patch set | 10:10 |
| mnasiadka | it's a bit blocking a nice queue of centos8 changes | 10:11 |
| yoctozepto | mnasiadka: +w but merge me the kolla one too | 10:11 |
| yoctozepto | mnasiadka: https://review.opendev.org/706059 | 10:11 |
| patchbot | patch 706059 - kolla - CI: Replace cinder-lvm scenario with zun scenario - 7 patch sets | 10:11 |
| *** theZorro266 has quit IRC | 10:15 | |
| mnasiadka | yoctozepto: the overlapping subnet thing - I suppose you'll follow up? | 10:17 |
| yoctozepto | mnasiadka: I have no idea why it popped there | 10:17 |
| yoctozepto | mnasiadka: it's on ansible side anyway | 10:18 |
| yoctozepto | mnasiadka: looks random (it worked in previous run above) | 10:18 |
| yoctozepto | mnasiadka: so just merge | 10:19 |
| mnasiadka | mgoddard: ^^ - keen on +2+w? | 10:21 |
| mgoddard | kolla needs to go first | 10:22 |
| yoctozepto | mgoddard: yeah, you are asked about kolla | 10:23 |
| mgoddard | done | 10:24 |
| yoctozepto | mgoddard: tyvm | 10:24 |
| mgoddard | yvw | 10:24 |
| *** jovial[m] has quit IRC | 10:27 | |
| *** aleccoder has quit IRC | 10:27 | |
| yoctozepto | mgoddard, mnasiadka, hrw: though I notice kolla is not gating on k-a jobs in this very change | 10:32 |
| yoctozepto | arar | 10:32 |
| mgoddard | it should gate on aio source jobs | 10:32 |
| yoctozepto | mgoddard: it does not, probably another zuul bug eh | 10:33 |
| mgoddard | yoctozepto: '\..' listed in irrelevant files | 10:34 |
| mgoddard | .zuul/ | 10:34 |
| yoctozepto | mgoddard: ah, true | 10:34 |
| mgoddard | maybe we should rename it to zuul.d | 10:37 |
| mgoddard | would be more consistent | 10:37 |
| mgoddard | and less obtuse | 10:37 |
| *** shyamb has quit IRC | 10:38 | |
| mnasiadka | so just go for it, simple change :) | 10:40 |
| hrw | meh. infra images will be untestable in k-a on aarch64 ;( | 10:40 |
| hrw | 2h needed to just build and we do not publish them | 10:40 |
| hrw | mnasiadka: and pile of merge conflicts on gerrit ;D | 10:41 |
| mnasiadka | hrw: good pile is not bad :) | 10:41 |
| mnasiadka | hrw: so why don't you publish them from Linaro CI? :D | 10:41 |
| hrw | mnasiadka: I do not think that you want to check which tag is latest? | 10:42 |
| hrw | mnasiadka: linaro/debian-source-base:ussuri-235, linaro/debian-source-base:train-ldc-2, linaro/debian-source-base:rocky-lcd-12 | 10:42 |
| hrw | mnasiadka: linaro/debian-source-base:ussuri-235, linaro/debian-source-base:train-ldc-2, linaro/debian-source-base:rocky-ldc-12 | 10:42 |
| mnasiadka | hrw: I was rather thinking of publishing them on Docker Hub with our normal approach? | 10:43 |
| hrw | mnasiadka: will you share docker credentials? | 10:43 |
| hrw | I would not | 10:43 |
| hrw | mnasiadka: also we do weekly builds not daily | 10:44 |
| hrw | and if they fail to build I complain upstream and send patches. if they fail to publish I just do not care | 10:44 |
| mnasiadka | this is not the best process I guess :) | 10:45 |
| *** devfaz has joined #openstack-kolla | 10:45 | |
| *** priteau has joined #openstack-kolla | 10:46 | |
| yoctozepto | re BadRequestException: 400: Client Error for url: http://192.0.2.10:9696/v2.0/subnets, Invalid input for operation: Requested subnet with cidr: 10.0.0.0/24 for network: e7c1bd4a-a9bc-40bf-beaa-8a27f6ead1b0 overlaps with another subnet. | 10:47 |
| yoctozepto | it's an effect of "Lost connection to MySQL server during query" in neutron-server | 10:47 |
| yoctozepto | it seemingly "forgot" it did the insert | 10:47 |
| hrw | mnasiadka: fine for us | 10:48 |
| *** admin0 has quit IRC | 10:49 | |
| priteau | Hi. I am looking into a problem with CloudKitty (Stein) returning only 10,000 dataframes entries when using InfluxDB as the storage backend. | 10:51 |
| priteau | This appears to be directly linked to the InfluxDB configuration set by Kolla Ansible: https://opendev.org/openstack/kolla-ansible/src/branch/master/ansible/roles/influxdb/templates/influxdb.conf.j2#L45 | 10:51 |
| priteau | Upstream InfluxDB has changed the default to 0 (unlimited) a while ago: https://docs.influxdata.com/influxdb/v1.7/administration/config/#max-row-limit-0 | 10:51 |
| priteau | Anyone know the potential impact of changing this setting in Kolla Ansible? dougsz? | 10:52 |
| mnasiadka | hrw: around dockerhub creds - zuul has those | 10:54 |
| hrw | mnasiadka: I am aware | 10:54 |
| priteau | Our default setting of 10000 rows hasn't changed since the initial import of the influxdb role in 2016 | 10:56 |
| dougsz | priteau, I have no objection to matching the upstream configuration. | 10:56 |
| priteau | Any potential impact on Monasca? | 10:57 |
| dougsz | We have generally been tuning infludb config via an override | 10:57 |
| dougsz | I think query-timeout is probably most important - which we leave untouched in KA by default | 10:58 |
| hrw | https://paste.centos.org/view/a4dfef72 is quite ugly for master ;( | 10:58 |
| * hrw off | 11:00 | |
| *** shyamb has joined #openstack-kolla | 11:00 | |
| dougsz | priteau: I guess this is Cloud Kitty doing some mega query rather than in chunks? | 11:01 |
| priteau | There is code to paginate but it doesn't appear to be used in this version | 11:04 |
| *** shyam89 has joined #openstack-kolla | 11:05 | |
| *** mleone87 has joined #openstack-kolla | 11:06 | |
| *** shyamb has quit IRC | 11:06 | |
| mnasiadka | priteau: looking at the docs I think we can safely remove that setting - and then it will take the default (0 in newer influxdb and 10000 in previous versions) | 11:06 |
| yoctozepto | +1 for aligning with upstream | 11:08 |
| mnasiadka | and it's in the http backend options, so I guess it doesn't change anything in the database | 11:11 |
| *** aleccoder has joined #openstack-kolla | 11:12 | |
| priteau | Indeed I think it's only for queries, I was just wondering if it could break some InfluxDB clients by making queries return massive amounts of data. | 11:12 |
| priteau | It fixes the issue for CloudKitty at least. | 11:13 |
| *** rpittau is now known as rpittau|bbl | 11:18 | |
| *** shyamb has joined #openstack-kolla | 11:25 | |
| *** shyam89 has quit IRC | 11:28 | |
| openstackgerrit | Merged openstack/kolla master: Add collectd-pmu package for Intel PMU (C7 only) https://review.opendev.org/704560 | 11:41 |
| *** shyam89 has joined #openstack-kolla | 11:50 | |
| *** shyamb has quit IRC | 11:53 | |
| *** xinliang has quit IRC | 11:54 | |
| *** kplant has joined #openstack-kolla | 12:02 | |
| *** shyamb has joined #openstack-kolla | 12:09 | |
| *** shyam89 has quit IRC | 12:11 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Debug init-runonce https://review.opendev.org/706482 | 12:18 |
| yoctozepto | mgoddard, mnasiadka: making init-runonce less of a black box ^ | 12:19 |
| mnasiadka | yoctozepto: maybe I'm a bit idealistic - but wouldn't it be better to rewrite it to Ansible? | 12:20 |
| yoctozepto | mnasiadka: would be, and to ditch the openstack python client in favor of ansible modules yeah | 12:21 |
| mgoddard | there was a patch to do that | 12:21 |
| yoctozepto | let's find it | 12:21 |
| yoctozepto | https://review.opendev.org/523072 | 12:22 |
| patchbot | patch 523072 - kolla-ansible - ansiblize the init-runonce script - 10 patch sets | 12:22 |
| yoctozepto | "Otherwise.. if it isn't broken..." | 12:23 |
| mnasiadka | well, shade is not needed anymore | 12:23 |
| yoctozepto | mnasiadka: well, it's openstacksdk now | 12:24 |
| mnasiadka | well, we all know it's broken-ish | 12:25 |
| mgoddard | tbh, ansible openstack modules are more of a black box than the openstack CLI | 12:34 |
| mgoddard | errors are rarely helpful | 12:34 |
| mgoddard | and I'm not sure we should give the impression this is for production use | 12:34 |
| mgoddard | yoctozepto, mnasiadka: ^ | 12:34 |
| yoctozepto | mgoddard: for now let's just add debugging caps to the script and use them | 12:35 |
| mnasiadka | well, I wouldn't even give the impression for users using init-runonce script - especially the beginning ones :) | 12:35 |
| mnasiadka | I would prefer them to use some doc to setup basic image/network structures | 12:35 |
| mnasiadka | and move init-runonce under tests/ | 12:36 |
| mgoddard | we could do that | 12:36 |
| mgoddard | although temp symlink would be nice since kayobe uses it | 12:36 |
| mnasiadka | yeah, like a normal deprecation process | 12:36 |
| mnasiadka | generally we have a lot of things under tools/ that give the impression that a user could use it | 12:38 |
| *** k_mouza has quit IRC | 12:39 | |
| *** k_mouza_ has joined #openstack-kolla | 12:39 | |
| yoctozepto | mnasiadka: yeah, like kolla-ansible | 12:41 |
| yoctozepto | who would use that | 12:41 |
| mnasiadka | I wasn't really mentioning it | 12:41 |
| mnasiadka | but setup-compute-libvirt.yaml? | 12:41 |
| yoctozepto | mnasiadka: setup_gate :-) | 12:41 |
| mnasiadka | yeah, I'm a newbie and will run ./setup_gate.sh | 12:42 |
| mnasiadka | I hope it will deploy a gateway | 12:42 |
| mnasiadka | :D | 12:42 |
| yoctozepto | virt-guest-shutdown.target | 12:42 |
| yoctozepto | troll directory | 12:42 |
| mnasiadka | I'll analyse what is there, move most of the stuff to contrib/ (because I feel it belongs there) | 12:42 |
| yoctozepto | mnasiadka: or even in the "recycle bin" | 12:44 |
| mnasiadka | contrib/trash | 12:44 |
| mnasiadka | :) | 12:44 |
| *** ktibi has joined #openstack-kolla | 12:52 | |
| mnasiadka | yoctozepto, mgoddard: do we use init-vpn anywhere? :) | 13:07 |
| *** shyamb has quit IRC | 13:09 | |
| *** rpittau|bbl is now known as rpittau | 13:14 | |
| *** hongbin has joined #openstack-kolla | 13:15 | |
| *** Luzi has joined #openstack-kolla | 13:16 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: Tidy up tools/ subdirectory https://review.opendev.org/706496 | 13:17 |
| openstackgerrit | Merged openstack/kolla master: CI: Replace cinder-lvm scenario with zun scenario https://review.opendev.org/706059 | 13:21 |
| *** mleone87 has quit IRC | 13:31 | |
| mgoddard | mnasiadka: not that I'm aware of | 13:34 |
| *** mleone87 has joined #openstack-kolla | 13:36 | |
| mgoddard | mnasiadka: can you remove W-1 from https://review.opendev.org/#/c/706055/1 | 13:36 |
| patchbot | patch 706055 - kolla-ansible - CI: Replace cinder-lvm scenario with zun scenario - 1 patch set | 13:36 |
| mnasiadka | mgoddard: done | 13:37 |
| mgoddard | thanks | 13:37 |
| yoctozepto | mnasiadka: vpn? whaaa? what does it do? | 13:40 |
| mnasiadka | yoctozepto: init vpnaas it seems | 13:42 |
| *** ktibi has quit IRC | 13:43 | |
| *** ktibi has joined #openstack-kolla | 13:45 | |
| *** hongbin has quit IRC | 13:54 | |
| yoctozepto | mnasiadka: verrry interesting | 13:59 |
| *** yan0s has joined #openstack-kolla | 14:00 | |
| openstackgerrit | Merged openstack/kolla master: Add ssh clients to Ironic Conductor container https://review.opendev.org/706149 | 14:02 |
| mnasiadka | yoctozepto: not really, given vpnaas is dead-ish | 14:05 |
| *** jovial[m] has joined #openstack-kolla | 14:11 | |
| yoctozepto | mnasiadka: oh, well | 14:15 |
| *** pbing19 has joined #openstack-kolla | 14:18 | |
| *** rlljorge has joined #openstack-kolla | 14:28 | |
| rlljorge | Hello, I am trying enable fwaas but I docker neutron_server not start with error Error loading class by alias: NoMatches: No 'neutron.service_plugins' driver found, looking for 'firewall' .... the full log in http://paste.openstack.org/show/789292/ | 14:29 |
| *** ktibi has quit IRC | 14:39 | |
| yoctozepto | rlljorge: which release of kolla-ansible against which release of kolla? | 14:40 |
| yoctozepto | looks like wrong version used | 14:40 |
| rlljorge | yoctozepto kolla-ansible 8.0.0 | 14:42 |
| yoctozepto | rlljorge: why 8.0.0? and not 8.1.0? https://pypi.org/project/kolla-ansible/#history | 14:42 |
| yoctozepto | I assume you run against stein? | 14:43 |
| rlljorge | yoctozepto is Stein release | 14:43 |
| rlljorge | O will upgrade to 8.1.0 now | 14:43 |
| *** Luzi has quit IRC | 14:45 | |
| yoctozepto | mgoddard, mnasiadka: keepalived is the current ci job rapist; it looks like we have a high chance of failing if keepalived ever loses its vip because services do not seem to like mariadb connections breaking | 14:52 |
| *** ktibi has joined #openstack-kolla | 14:57 | |
| mgoddard | language yoctozepto | 15:03 |
| *** rpittau is now known as rpittau|brb | 15:04 | |
| yoctozepto | mgoddard: where? ;o | 15:04 |
| mgoddard | not conducive to an inclusive community | 15:05 |
| mgoddard | any idea why we are seeing this now and not previously? | 15:06 |
| *** pbing19 has quit IRC | 15:07 | |
| yoctozepto | mgoddard: had to lookup the smart "c" word | 15:07 |
| mgoddard | sorry | 15:07 |
| yoctozepto | mgoddard: no idea | 15:07 |
| yoctozepto | it happened now and then though | 15:07 |
| yoctozepto | so it's not something new | 15:07 |
| *** TrevorV has joined #openstack-kolla | 15:08 | |
| yoctozepto | but having fixed mariadb and keystone we are left with keepalived :-) | 15:08 |
| mgoddard | why would keepalived lose it's vip? upgrade? | 15:08 |
| yoctozepto | mgoddard: nah, haproxy check returning 1 or even better - signal 15 | 15:09 |
| mnasiadka | yoctozepto: can we for now introduce an option to disable this bloody hcheck script and investigate what is happenning with ghis fine piece of artwork code in the background? | 15:09 |
| yoctozepto | on upgrade we restart services anyway | 15:09 |
| mnasiadka | Is bloody an inclusive wording? | 15:09 |
| yoctozepto | mnasiadka: I would rate it the same as the "r" word | 15:10 |
| yoctozepto | mnasiadka: usually safe but not quite :-) | 15:10 |
| *** ktibi has quit IRC | 15:10 | |
| mnasiadka | We need an inclusive rating bot | 15:10 |
| yoctozepto | yeah, that would clear up things; surely our language standards in Poland are much more relaxed... | 15:11 |
| yoctozepto | mnasiadka: you propose to disable just in CI? how's it going to help? | 15:11 |
| mnasiadka | yoctozepto: I think mgoddard would be very freightened if we adapted to the style 80% of Polish people speak :) | 15:12 |
| mnasiadka | yoctozepto: what is the alternative? Add logging to the script and check what is happening? | 15:12 |
| yoctozepto | mnasiadka: for sure, thankfully we are in those 20%, aren't we? ;D | 15:12 |
| mgoddard | bloody is fine IMO | 15:13 |
| yoctozepto | mnasiadka: it connects to haproxy via unix socket | 15:13 |
| mgoddard | but I am no expert | 15:13 |
| yoctozepto | not much going on there | 15:13 |
| mnasiadka | Not mentioning that this script is a bit lame | 15:13 |
| yoctozepto | well, it would move vip if you killed haproxy | 15:13 |
| mnasiadka | Maybe this unix socket is the problem? | 15:13 |
| yoctozepto | yeah, could be haproxy lame with handling it | 15:13 |
| mnasiadka | Now it’s moving the vip without haproxy being dead | 15:14 |
| yoctozepto | yeah, indeed | 15:14 |
| mnasiadka | Just unresponsive | 15:14 |
| mnasiadka | In a way | 15:14 |
| yoctozepto | yeah, possibly | 15:14 |
| yoctozepto | at least via socket | 15:14 |
| yoctozepto | wonder why sometimes signal 15 and sometimes return 1 | 15:14 |
| yoctozepto | signal 15 is for timeout | 15:15 |
| yoctozepto | but you should be getting 1 for timeout anyways | 15:15 |
| yoctozepto | eh | 15:15 |
| mnasiadka | Weird | 15:15 |
| *** ivve has quit IRC | 15:15 | |
| mnasiadka | Second thing how can we improve oslo.db connection handling? Are there any tunables? | 15:15 |
| yoctozepto | yeah, it looks like that socket is not functioning the way one imagined | 15:16 |
| yoctozepto | no idea, I guess normally you would just repeat the command | 15:16 |
| yoctozepto | though in the case of neutron discussed just today | 15:16 |
| yoctozepto | it created the subnet | 15:16 |
| yoctozepto | failed to notify client | 15:17 |
| yoctozepto | client retried | 15:17 |
| yoctozepto | and faild due to existing already | 15:17 |
| yoctozepto | real drama | 15:17 |
| mnasiadka | Raise a bug in neutron :) | 15:19 |
| *** rpittau|brb is now known as rpittau | 15:20 | |
| yoctozepto | mnasiadka: or oslo or openstack client or all at once | 15:23 |
| yoctozepto | mnasiadka: the thing is I don't have logs of client | 15:24 |
| yoctozepto | only server's | 15:24 |
| yoctozepto | mgoddard, mnasiadka: https://review.opendev.org/706055 w+1 ? | 15:27 |
| patchbot | patch 706055 - kolla-ansible - CI: Replace cinder-lvm scenario with zun scenario - 1 patch set | 15:27 |
| mgoddard | done | 15:29 |
| yoctozepto | mgoddard: danke schön! | 15:31 |
| *** vmixor has joined #openstack-kolla | 15:32 | |
| mgoddard | yoctozepto: can we discuss https://review.opendev.org/#/c/699888/ ? | 15:33 |
| patchbot | patch 699888 - kolla-ansible - Copy CA into containers. - 23 patch sets | 15:33 |
| mgoddard | "need to ping mgoddard on that" | 15:33 |
| yoctozepto | mgoddard: yeah, go ahead :-) | 15:33 |
| mgoddard | you pinged? | 15:33 |
| yoctozepto | oh, sorry, I was writing a reply but I can't see even a draft | 15:33 |
| yoctozepto | mgoddard: let me write here then eh | 15:34 |
| mgoddard | given we have a few patches chained with +2a, wdyt about merge and refactor? | 15:34 |
| yoctozepto | the code in patch is generally ok but a bit incostinent | 15:34 |
| yoctozepto | inconsistent* | 15:34 |
| mgoddard | how so? | 15:35 |
| mgoddard | ordering of conditions? | 15:35 |
| mgoddard | hmm, also some missing group check | 15:36 |
| yoctozepto | yeah | 15:36 |
| yoctozepto | and comments for docs | 15:36 |
| yoctozepto | so it needs a followup | 15:36 |
| yoctozepto | i'm ok with +2 as it's no backport material | 15:36 |
| yoctozepto | but we need a followup to keep quality | 15:36 |
| mgoddard | I'm sure James will follow up if asked | 15:37 |
| yoctozepto | mgoddard: ok then, this applies also to releasenote :D | 15:37 |
| yoctozepto | no need to write about issue with py2.7.9 in ussuri... | 15:37 |
| *** igordc has joined #openstack-kolla | 15:37 | |
| mgoddard | true | 15:40 |
| yoctozepto | mgoddard: so, what do we do? | 15:40 |
| mgoddard | I don't mind. We can modify the note in a follow up | 15:41 |
| openstackgerrit | Pierre Riteau proposed openstack/kolla-ansible master: Use InfluxDB default [http]/max-row-limit setting https://review.opendev.org/706547 | 15:42 |
| yoctozepto | mgoddard: done then | 15:43 |
| yoctozepto | mgoddard: your turn | 15:43 |
| mgoddard | yoctozepto: oh yeah, there was this one: https://review.opendev.org/#/c/700788/ | 15:48 |
| patchbot | patch 700788 - kolla-ansible - Use kolla_toolbox to execute REST methods - 18 patch sets | 15:48 |
| mgoddard | do we really want to make it a fix? | 15:48 |
| yoctozepto | mgoddard: you might be right it's really not worth the effort | 15:49 |
| generalfuzz | hi - I will update note and refactor copying certs into role in https://review.opendev.org/#/c/664516/ | 15:52 |
| patchbot | patch 664516 - kolla-ansible - Add support for encrypting backend HAProxy traffic - 9 patch sets | 15:52 |
| *** kplant has quit IRC | 15:56 | |
| *** kplant has joined #openstack-kolla | 16:01 | |
| yoctozepto | generalfuzz: thanks :-) | 16:06 |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Support for custom API paste files https://review.opendev.org/695054 | 16:08 |
| *** ktibi has joined #openstack-kolla | 16:26 | |
| *** pbing19 has joined #openstack-kolla | 16:28 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Debug init-runonce https://review.opendev.org/706482 | 16:34 |
| *** bengates has quit IRC | 16:39 | |
| *** yan0s has quit IRC | 16:43 | |
| *** carli has quit IRC | 16:43 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla stable/train: CI: Replace cinder-lvm scenario with zun scenario https://review.opendev.org/706563 | 16:45 |
| *** ktibi has quit IRC | 16:46 | |
| *** ktibi has joined #openstack-kolla | 16:53 | |
| openstackgerrit | Merged openstack/kolla-ansible master: CI: Replace cinder-lvm scenario with zun scenario https://review.opendev.org/706055 | 16:54 |
| *** hongbin has joined #openstack-kolla | 16:55 | |
| *** hongbin has quit IRC | 16:59 | |
| *** rpittau is now known as rpittau|afk | 17:02 | |
| *** noxoid_ is now known as noxoid | 17:06 | |
| mgoddard | mnasiadka: I thought of another improvement to ext ceph user experience. Prechecks for the expected configs (conf.conf, keyrings) | 17:13 |
| mgoddard | mnasiadka: will be especially useful following change of defaults for keys | 17:13 |
| mnasiadka | mgoddard: good, that’s on my list :) | 17:13 |
| mgoddard | mnasiadka: also just spotted a few hard coded ceph.client.gnocchi.keyring | 17:15 |
| mnasiadka | mgoddard: ouch, will look for them on Monday - if you can raise a bug I would be grateful :) | 17:16 |
| *** ktibi has quit IRC | 17:17 | |
| yoctozepto | mgoddard, mnasiadka: are you aware whether sahara works in kolla? | 17:17 |
| yoctozepto | (or even in general?) | 17:17 |
| mgoddard | mnasiadka: https://bugs.launchpad.net/kolla-ansible/+bug/1862393 | 17:18 |
| openstack | Launchpad bug 1862393 in kolla-ansible ussuri "Some hard coded ceph.client.gnocchi.keyring left" [Undecided,New] | 17:18 |
| mgoddard | yoctozepto: I spent some time with it a few years ago, but not recently | 17:18 |
| yoctozepto | mgoddard: w/ kolla? | 17:18 |
| mgoddard | yes | 17:18 |
| yoctozepto | mgoddard: thanks, raises spirits :-) | 17:19 |
| mgoddard | haven't seen many patches since then though | 17:19 |
| mgoddard | maybe it still works | 17:19 |
| yoctozepto | will see | 17:19 |
| yoctozepto | mnasiadka, mgoddard: are we backport ceph-ansible to train? | 17:20 |
| mgoddard | you mean the CI? | 17:20 |
| yoctozepto | asking because backporting in kolla would be quite easy before I do mine with zun | 17:20 |
| yoctozepto | but k-a is a bit of pain | 17:20 |
| yoctozepto | mgoddard: CI | 17:20 |
| mgoddard | probably a bit overkill to do two four ceph jobs | 17:20 |
| yoctozepto | we ain't doing ceph-ansible port for train :-) | 17:21 |
| yoctozepto | mnasiadka: you agree? I do | 17:21 |
| yoctozepto | then I backport zun and make backporting c-a painful | 17:21 |
| yoctozepto | looks like mnasiadka is on the move already | 17:21 |
| yoctozepto | I see nobody promoted sahara away from 3rd tier | 17:22 |
| mnasiadka | Less work for me - I’m happier :) | 17:22 |
| yoctozepto | mnasiadka: all right, then agreed | 17:22 |
| openstackgerrit | Merged openstack/kolla-ansible master: CentOS 8: Add deploy jobs in CI https://review.opendev.org/693544 | 17:24 |
| openstackgerrit | Merged openstack/kolla-ansible master: Copy CA into containers. https://review.opendev.org/699888 | 17:25 |
| *** oncall-pokemon has joined #openstack-kolla | 17:25 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Enable etcd in ceph and cinder-lvm scenarios https://review.opendev.org/697485 | 17:25 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible stable/train: CI: Test Zun with Cinder LVM backend (iSCSI) https://review.opendev.org/706587 | 17:29 |
| yoctozepto | had to backport this first ^ | 17:29 |
| yoctozepto | train CI did not include cinder lvm for zun | 17:29 |
| yoctozepto | well, let's make it consistent | 17:29 |
| oncall-pokemon | with the kolla prometheus. what would cause the keys `openstack_nova_total_vms` to not be populated? | 17:33 |
| openstackgerrit | Merged openstack/kolla-ansible master: Use kolla_toolbox to execute REST methods https://review.opendev.org/700788 | 17:33 |
| openstackgerrit | Merged openstack/kolla-ansible master: Generate self signed TLS certificates https://review.opendev.org/701323 | 17:33 |
| *** evrardjp has quit IRC | 17:34 | |
| *** evrardjp has joined #openstack-kolla | 17:34 | |
| *** mleone87 has quit IRC | 17:35 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible stable/train: CI: Test Zun with Cinder LVM backend (iSCSI) https://review.opendev.org/706587 | 17:35 |
| *** ivve has joined #openstack-kolla | 17:38 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible stable/train: CI: Replace cinder-lvm scenario with zun scenario https://review.opendev.org/706591 | 17:39 |
| *** k_mouza_ has quit IRC | 17:44 | |
| mgoddard | yoctozepto: did you see this thread: http://lists.openstack.org/pipermail/openstack-discuss/2020-January/012071.html | 17:44 |
| mgoddard | references OSA galera config: https://opendev.org/openstack/openstack-ansible/src/branch/master/inventory/group_vars/haproxy/haproxy.yml#L48-L49 | 17:44 |
| mgoddard | thought you might be interested as current galera tormentor | 17:45 |
| yoctozepto | mgoddard: thanks, I think I saw that but then forgot, might come in handy indeed | 17:46 |
| yoctozepto | mgoddard: osa also uses other means of liveness discovery - our script is limited and might allow a catching up instance to serve requests (or rather rejections :-) ) | 17:50 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: Docs: show other versions https://review.opendev.org/706595 | 17:55 |
| *** priteau has quit IRC | 17:56 | |
| *** dougsz has quit IRC | 18:02 | |
| *** rlljorge has quit IRC | 18:09 | |
| *** k_mouza has joined #openstack-kolla | 18:14 | |
| *** igordc has quit IRC | 18:18 | |
| *** k_mouza has quit IRC | 18:19 | |
| yoctozepto | mgoddard, mnasiadka: for lolz: k-a master ci b0rken with c8 failing on bootstrap not able to find python2... TASK [Gather facts] ************************************************************ | 18:32 |
| yoctozepto | mgoddard, mnasiadka: oddly zuul ran jobs from definitions on master but used old files... | 18:35 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Debug init-runonce https://review.opendev.org/706482 | 18:35 |
| openstackgerrit | Merged openstack/kolla master: CentOS 8: Add deploy jobs in CI https://review.opendev.org/705952 | 18:36 |
| *** igordc has joined #openstack-kolla | 18:49 | |
| *** k_mouza has joined #openstack-kolla | 18:50 | |
| *** k_mouza has quit IRC | 18:51 | |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Support for custom API paste files https://review.opendev.org/695054 | 18:58 |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Support for custom API paste files https://review.opendev.org/695054 | 19:02 |
| *** gfidente is now known as gfidente|afk | 19:11 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: Tidy up tools/ subdirectory https://review.opendev.org/706496 | 19:16 |
| dking_desktop | This is probably better suited for the neutron channel, but would anybody know how I should configure neutron_external_interface if I'm using my bond (bond0) for all my VLANs, but I also need to be able to connect to the controller to receive untagged requests? | 19:19 |
| *** igordc has quit IRC | 19:19 | |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Adds support of CADF notifying for other APIs next to keystone. https://review.opendev.org/674579 | 19:20 |
| openstackgerrit | Merged openstack/kayobe stable/rocky: Fix multiple CI failures https://review.opendev.org/704774 | 19:26 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Use internal API endpoints in overcloud API interaction https://review.opendev.org/704620 | 19:26 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Stop running kolla-target-venv in seed hypervisor host upgrade https://review.opendev.org/704614 | 19:26 |
| noxoid | curious, has anyone deployed standard ironic (not kayobe) in a multi-region environment? it looks like the k-a playbooks for ironic are heavily dependent on enable_keystone being set to yes which is not the case in multi-region environments | 19:35 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Use mariabackup for database backups https://review.opendev.org/704624 | 19:37 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Don't list tasks for config dump https://review.opendev.org/704609 | 19:37 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Allow Kayobe to wipe virtualized disks https://review.opendev.org/704612 | 19:37 |
| noxoid | ah yea its been reported https://bugs.launchpad.net/kolla-ansible/+bug/1809216 | 19:37 |
| openstack | Launchpad bug 1809216 in kolla-ansible "Ironic in other regions do not register endpoints" [Medium,Triaged] | 19:37 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: Remove kolla-ceph https://review.opendev.org/704309 | 19:39 |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Adds support of CADF notifying for other APIs next to keystone. https://review.opendev.org/674579 | 19:41 |
| *** pbing19 has quit IRC | 19:57 | |
| *** cah_link has quit IRC | 20:08 | |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Adds support of CADF notifying for other APIs next to keystone. https://review.opendev.org/674579 | 20:09 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Fix kayobe overcloud provision without seed in limit https://review.opendev.org/704616 | 20:10 |
| openstackgerrit | Merged openstack/kayobe stable/stein: Fix use of IPA images in Bifrost without a checksum https://review.opendev.org/704618 | 20:10 |
| openstackgerrit | Zhuo Zhen proposed openstack/kolla-ansible master: Configurable logrotate rotation interval and count https://review.opendev.org/706616 | 20:12 |
| openstackgerrit | Viktor Michalek proposed openstack/kolla-ansible master: Added support for custom logging.conf https://review.opendev.org/679812 | 20:17 |
| *** kplant has quit IRC | 20:42 | |
| *** tonythomas has quit IRC | 20:44 | |
| *** primeministerp has joined #openstack-kolla | 20:46 | |
| *** primeministerp has joined #openstack-kolla | 20:47 | |
| *** primeministerp has joined #openstack-kolla | 20:48 | |
| *** primeministerp has quit IRC | 20:48 | |
| *** primeministerp has joined #openstack-kolla | 20:48 | |
| *** primeministerp has quit IRC | 20:49 | |
| *** primeministerp has joined #openstack-kolla | 20:50 | |
| *** k_mouza has joined #openstack-kolla | 20:52 | |
| *** k_mouza has quit IRC | 20:56 | |
| openstackgerrit | James Kirsch proposed openstack/kolla-ansible master: Service configuration urls should be constructed using kolla_internal_fqdn https://review.opendev.org/706627 | 21:43 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/rocky: Stop running kolla-target-venv in seed hypervisor host upgrade https://review.opendev.org/704615 | 21:51 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/rocky: Allow Kayobe to wipe virtualized disks https://review.opendev.org/704613 | 21:51 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/rocky: Fix kayobe overcloud provision without seed in limit https://review.opendev.org/704617 | 21:52 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/rocky: Fix use of IPA images in Bifrost without a checksum https://review.opendev.org/704619 | 21:52 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/rocky: Use internal API endpoints in overcloud API interaction https://review.opendev.org/704621 | 21:52 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/rocky: Don't list tasks for config dump https://review.opendev.org/704611 | 21:52 |
| *** jbadiapa has quit IRC | 21:55 | |
| *** primeministerp has quit IRC | 22:06 | |
| *** vmixor has quit IRC | 22:12 | |
| *** TrevorV has quit IRC | 22:57 | |
| *** oncall-pokemon has quit IRC | 23:15 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!