| *** kevko has quit IRC | 00:06 | |
| *** diurnalist has quit IRC | 00:10 | |
| *** diurnalist has joined #openstack-kolla | 00:35 | |
| *** mnasiadka has quit IRC | 01:19 | |
| *** alanmeadows has quit IRC | 01:20 | |
| *** mnasiadka has joined #openstack-kolla | 01:22 | |
| *** alanmeadows has joined #openstack-kolla | 01:23 | |
| *** zijlboot has joined #openstack-kolla | 01:39 | |
| *** zijlboot has quit IRC | 01:44 | |
| *** Limech has quit IRC | 02:05 | |
| *** Limech has joined #openstack-kolla | 02:06 | |
| *** ricolin has joined #openstack-kolla | 02:41 | |
| *** bsanjeewa has joined #openstack-kolla | 02:47 | |
| *** jcmdln has joined #openstack-kolla | 02:57 | |
| *** k_mouza has joined #openstack-kolla | 03:34 | |
| *** k_mouza has quit IRC | 03:39 | |
| *** abdysn has joined #openstack-kolla | 03:56 | |
| *** abdysn has quit IRC | 03:56 | |
| *** also_stingrayza has joined #openstack-kolla | 04:19 | |
| *** stingrayza has quit IRC | 04:23 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-kolla | 04:33 | |
| *** wuchunyang has joined #openstack-kolla | 04:36 | |
| *** bsanjeewa has quit IRC | 04:55 | |
| *** bsanjeewa has joined #openstack-kolla | 04:59 | |
| *** wuchunyang has quit IRC | 05:15 | |
| *** vishalmanchanda has joined #openstack-kolla | 05:37 | |
| *** cah_link has joined #openstack-kolla | 05:38 | |
| *** seco_ has joined #openstack-kolla | 05:47 | |
| *** ysirndjuro has quit IRC | 05:54 | |
| *** ysirndjuro has joined #openstack-kolla | 06:02 | |
| *** diurnalist has quit IRC | 06:11 | |
| *** e0ne has quit IRC | 06:11 | |
| *** ysirndjuro has quit IRC | 06:16 | |
| *** gfidente|afk has quit IRC | 06:30 | |
| *** born2bake has joined #openstack-kolla | 07:01 | |
| *** bsanjeewa has quit IRC | 07:11 | |
| *** bengates has joined #openstack-kolla | 07:14 | |
| *** bengates has quit IRC | 07:14 | |
| *** bengates has joined #openstack-kolla | 07:15 | |
| *** threestrands has quit IRC | 07:41 | |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe master: Update feature flags https://review.opendev.org/734064 | 07:45 |
|---|---|---|
| yoctozepto | morning | 07:46 |
| mgoddard | morning | 07:46 |
| sorin-mihai | morning | 07:54 |
| *** also_stingrayza is now known as stingrayza | 07:54 | |
| *** e0ne has joined #openstack-kolla | 07:56 | |
| *** Torel has joined #openstack-kolla | 07:57 | |
| *** e0ne has quit IRC | 07:58 | |
| *** e0ne has joined #openstack-kolla | 07:58 | |
| openstackgerrit | Merged openstack/kayobe master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/730089 | 08:00 |
| *** Torel has quit IRC | 08:16 | |
| *** Torel has joined #openstack-kolla | 08:21 | |
| eliaswimmer | Good morning | 08:26 |
| eliaswimmer | When doing a kayobe overcloud service deploy, the task "heat : Running Heat bootstrap container" fails with "Conflict occurred attempting to store project - it is not permitted to have two projects acting as domains with the same name: heat_user_domain." | 08:28 |
| eliaswimmer | Any hint how I can fix this? | 08:28 |
| eliaswimmer | And an other problem is that I have local chronyd and docker chronyd daemons. How can I remove the containerized versions? | 08:32 |
| *** kevko has joined #openstack-kolla | 08:36 | |
| andrein | good morning! | 08:48 |
| sorin-mihai | eliaswimmer, i'm not entirely sure what the effect would be if yo do it like this, in my case it seems to be working fine, but i'm not using kayobe. maybe someone else can confirm/infirm or suggest a different method. you can disable chrony in config using enable_chrony: "no", then 'docker stop chrony' if it's already running, eventually 'docker rm chrony' to remove the container. before doing this, i'd first make sure that | 08:49 |
| sorin-mihai | chrony is running properly in (all) the node(s) | 08:49 |
| eliaswimmer | thx you, I will do so. I just hoped that kayobe/kolla is stopping the containers | 08:50 |
| andrein | my company is evaluating the idea of outsourcing the deployment and maintenance of our openstack cloud. About a month ago, I saw a presentation by Gaël Therond at the kolla klub, but I can't find his contact info. Is he active on IRC? Are there other companies that you could recommend that offer this sort of service? | 08:51 |
| eliaswimmer | I currently have support by https://www.stackhpc.com/ | 08:56 |
| andrein | mgoddard: you work at stackhpc, right? | 09:01 |
| mgoddard | hi andrein, indeed I do. I'm a little busy at the moment, but my colleague jovial[m] may be able to help | 09:02 |
| mgoddard | andrein: or if you pass on an email address I could make an introduction | 09:03 |
| jovial[m] | eliaswimmer: I've found this bug report: https://bugs.launchpad.net/kolla/+bug/1839824, does that look similar? | 09:11 |
| openstack | Launchpad bug 1839824 in kolla "kolla-ansible fails to deploy heat if the heat_user_domain already exists in keystone" [Medium,In progress] - Assigned to Fabian Bermel (fbermel) | 09:11 |
| *** priteau has joined #openstack-kolla | 09:13 | |
| eliaswimmer | jovial[m]: yes, that looks quite similar. I disable heat now. And will enable it again after first successful run of kayobe | 09:15 |
| *** e0ne_ has joined #openstack-kolla | 09:16 | |
| *** e0ne has quit IRC | 09:16 | |
| priteau | eliaswimmer: Are you deploying a new OpenStack or taking over an existing one? | 09:17 |
| eliaswimmer | priteau: a new one | 09:18 |
| eliaswimmer | train with centos8 | 09:18 |
| priteau | It's odd that you are running into this with a fresh install. Did you customise heat config at all? | 09:20 |
| eliaswimmer | but it may be that the error changed over time, as I run it several times, there was a connection issue with one of the machines before | 09:20 |
| priteau | I assume your keystone deployment is functional? Can you list domains? | 09:21 |
| *** Torel has quit IRC | 09:22 | |
| *** Torel has joined #openstack-kolla | 09:23 | |
| jovial[m] | eliaswimmer: I'd be interested to see the output of: `openstack domain list | grep heat | awk '{print $4}'` | 09:32 |
| eliaswimmer | ok, one moment I have to install openstack cli | 09:32 |
| eliaswimmer | output is heat_user_domain | 09:34 |
| jovial[m] | hmm, possibly not that bug then? Could you put the complete error message in a paste bin? | 09:37 |
| eliaswimmer | I will | 09:38 |
| *** priteau has quit IRC | 09:39 | |
| *** priteau has joined #openstack-kolla | 09:41 | |
| *** e0ne has joined #openstack-kolla | 09:49 | |
| *** e0ne_ has quit IRC | 09:49 | |
| mgoddard | andrein: and to be fair to Gaël, his IRC nick is Fl1nt | 09:57 |
| andrein | Thanks! | 09:58 |
| *** wuchunyang has joined #openstack-kolla | 10:14 | |
| wuchunyang | Radosław Piliszek hi | 10:14 |
| mgoddard | wuchunyang: == yoctozepto | 10:15 |
| wuchunyang | yoctozepto | 10:15 |
| wuchunyang | thanks mgoddard | 10:15 |
| wuchunyang | mgoddard , hi, i have a problem , when host os is centos 8 and kolla-distro is ubuntu, k-a will check failed because no /etc/timezone | 10:17 |
| wuchunyang | should we create this file on centos ? | 10:18 |
| *** vishalmanchanda has quit IRC | 10:19 | |
| eliaswimmer | had the same issue | 10:30 |
| eliaswimmer | I fixed it with timezonectl --set-timezone ... | 10:31 |
| wuchunyang | i create the /etc/timezone file | 10:31 |
| eliaswimmer | jovial[m]: we fixed it, it was keystone making troubles | 10:33 |
| *** e0ne has quit IRC | 10:33 | |
| eliaswimmer | a certain percentage of the requests where broken, restart helped | 10:33 |
| *** zijlboot has joined #openstack-kolla | 10:46 | |
| *** k_mouza has joined #openstack-kolla | 10:46 | |
| *** zijlboot has quit IRC | 10:50 | |
| *** wuchunyang has quit IRC | 10:53 | |
| *** wuchunyang has joined #openstack-kolla | 10:54 | |
| *** wuchunyang has quit IRC | 10:58 | |
| *** Limech has quit IRC | 11:02 | |
| *** Limech has joined #openstack-kolla | 11:02 | |
| yoctozepto | aaand he's gone | 11:05 |
| yoctozepto | wuchunyang: we don't really support mixing host and container distros | 11:05 |
| yoctozepto | so I can see how it went unnoticed so far | 11:06 |
| yoctozepto | since ubuntu/debian expects this file, we might want to simulate it, but I don't really see the point, better use matching distros to avoid opening a pandora box | 11:06 |
| *** zijlboot has joined #openstack-kolla | 11:08 | |
| *** zijlboot_ has joined #openstack-kolla | 11:10 | |
| *** zijlboot has quit IRC | 11:13 | |
| jovial[m] | eliaswimmer: glad to hear it :) | 11:14 |
| *** cah_link1 has joined #openstack-kolla | 11:19 | |
| *** cah_link has quit IRC | 11:21 | |
| *** cah_link1 is now known as cah_link | 11:21 | |
| *** wuchunyang has joined #openstack-kolla | 11:26 | |
| *** shyamb has joined #openstack-kolla | 11:55 | |
| shyamb | Hi | 11:58 |
| shyamb | Any developer guide available for beginner in kolla ansible? | 11:59 |
| shyamb | thanks | 12:00 |
| *** wuchunyang has quit IRC | 12:01 | |
| mgoddard | shyamb: https://docs.openstack.org/kolla-ansible/latest/user/quickstart.html | 12:07 |
| shyamb | mgoddard: thanks. I got this one: https://github.com/openstack/kolla-ansible/blob/master/doc/source/contributor/adding-a-new-service.rst | 12:07 |
| shyamb | Any thoughts on 'Ussuri' release timelines for kolla? | 12:08 |
| mgoddard | shyamb: there is also https://docs.google.com/presentation/d/12n5TmheuvafcHIH6UnIokerP4nqtMm0m_w9RwEgCUSc/edit#slide=id.p1 | 12:08 |
| mgoddard | shyamb: hopefully in the next few weeks | 12:09 |
| shyamb | mgoddard: Okay, thank you. | 12:09 |
| shyamb | Presentation looks private. I just requested the access | 12:09 |
| shyamb | mgoddard: Do I need specific login account to access it? | 12:11 |
| mgoddard | shyamb: try again | 12:11 |
| shyamb | mgoddard: worked! Thanks a lot. | 12:11 |
| *** e0ne has joined #openstack-kolla | 12:13 | |
| shyamb | mgoddard: This is nicely written doc. Thanks. | 12:15 |
| mgoddard | :) | 12:16 |
| *** e0ne has quit IRC | 12:17 | |
| *** ricolin has quit IRC | 12:19 | |
| openstackgerrit | Will Szumski proposed openstack/kayobe master: Update image regexes https://review.opendev.org/735181 | 12:21 |
| openstackgerrit | Will Szumski proposed openstack/kayobe master: WIP: Update feature flag defaults https://review.opendev.org/735329 | 12:21 |
| *** e0ne has joined #openstack-kolla | 12:27 | |
| *** kevko has quit IRC | 12:32 | |
| *** shyamb has quit IRC | 12:33 | |
| *** wuchunyang has joined #openstack-kolla | 12:33 | |
| openstackgerrit | Will Szumski proposed openstack/kayobe master: Revert "Use OpenStack Ussuri release" https://review.opendev.org/735331 | 12:41 |
| openstackgerrit | Will Szumski proposed openstack/kayobe master: CI: set previous_release to ussuri https://review.opendev.org/735333 | 12:51 |
| openstackgerrit | Will Szumski proposed openstack/kayobe master: Remove python 2 cleanup https://review.opendev.org/735334 | 12:54 |
| *** jonaspaulo has joined #openstack-kolla | 12:55 | |
| *** jonaspaulo has quit IRC | 12:55 | |
| *** Torel has quit IRC | 12:56 | |
| *** Torel has joined #openstack-kolla | 12:57 | |
| *** wuchunyang has quit IRC | 13:02 | |
| *** lvdombrkr has joined #openstack-kolla | 13:11 | |
| *** Torel has quit IRC | 13:17 | |
| *** born2bake has quit IRC | 13:18 | |
| *** Torel has joined #openstack-kolla | 13:20 | |
| *** noxoid has quit IRC | 13:32 | |
| *** noxoid has joined #openstack-kolla | 13:33 | |
| *** vishalmanchanda has joined #openstack-kolla | 13:41 | |
| dking | Regarding kayobe overclouds, it seems that it might require IPs for the admin_oc network to be set explicitly. Is there a way to have Kayobe generate the IP from the network's _allocation_pool if none is given? | 13:44 |
| mgoddard | dking: I'm not sure I understand. IPs are allocated from that pool, and added to network-allocation.yml | 13:46 |
| mgoddard | kplant: you should be able to see those slides now | 13:48 |
| dking | Okay. I was hoping that would be the case. However, on my recent run of "kayobe overcloud provision", on task "Wait for SSH access to the nodes", it times out with "Timeout when waiting for :22", which seems like ansible_host is empty. Furthermore, I don't see network-allocation.yml updated. | 13:48 |
| kplant | mgoddard: thxc | 13:48 |
| maharg101 | on stable/train, can see measurements in gnocchi, but they're not being archived to the ceph backend. See http://paste.openstack.org/show/794692/ - using all default metrics, measurements, archive policies etc. Example: disk.root.size, using ceilometer-low archive policy. I should expect to see objects in the ceph rbd pool, but there are none. No obvious errors in the logs. Grateful for any | 13:48 |
| maharg101 | clues. | 13:48 |
| dking | How might I go about determining why an IP wasn't allocated, or why there's no update to network-allocation.yml? Is that done by .../roles/ip-allocation/library/ip_allocation.py? | 13:56 |
| priteau | dking: that could be because the host wasn't configured to be on the network you expected, or you limited the ansible run too much | 13:59 |
| priteau | When Kayobe runs and reaches the point where it is allocating IPs for your host, do you see the network you're expecting? | 14:00 |
| dking | I'm thinking the later is the issue. I don't see the play being run to allocate the IPs. I had previously run the deploy commands up to, but not including "kayobe overcloud provision". I added the controller hosts and wanted to immediately run "kayobe overcloud provision". Which command runs the ip_allocation role? | 14:02 |
| openstackgerrit | Merged openstack/kolla-ansible stable/stein: Add missing become to some VMWare tasks https://review.opendev.org/735075 | 14:05 |
| mgoddard | dking: if you run 'kayobe overcloud inventory discover' it should generate the IP allocations | 14:05 |
| dking | Ultimately, I would like to find the least amount of time, and hopefully the fewest commands, to add a set of controller nodes and have them in the state after "kayobe overcloud host configure" | 14:06 |
| dking | Okay. That might be what I'm missing in this step. That should have been obvious. Thank you. | 14:07 |
| *** kevko has joined #openstack-kolla | 14:15 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/train: Add missing become to some VMWare tasks https://review.opendev.org/735074 | 14:19 |
| openstackgerrit | Merged openstack/kolla-ansible stable/ussuri: Add missing become to some VMWare tasks https://review.opendev.org/735073 | 14:20 |
| *** k_mouza has quit IRC | 14:20 | |
| *** k_mouza has joined #openstack-kolla | 14:20 | |
| *** k_mouza has quit IRC | 14:24 | |
| *** kevko has quit IRC | 14:27 | |
| openstackgerrit | Gaël THEROND proposed openstack/kolla master: Change lang package block order. Base image lang pack installation block need to be reordered for offline installation. Closes-bug: 1883233 https://review.opendev.org/735346 | 14:29 |
| openstack | bug 1883233 in kolla "lang package installation happens before mirrors get overridden" [Medium,Triaged] https://launchpad.net/bugs/1883233 | 14:29 |
| *** Fl1nt has joined #openstack-kolla | 14:33 | |
| Fl1nt | Good afternoon folks! | 14:33 |
| priteau | Congratulations on your first patch Fl1nt :-) | 14:36 |
| Fl1nt | Actually not the first one, but the first one within my own company, that's a small one but at least it's a begining, I'd to refresh my knowledges about how things behave with git/gerrit/review etc ^^ | 14:39 |
| Fl1nt | thanks a lot ^^ | 14:39 |
| Fl1nt | let's hope it help :p | 14:39 |
| Fl1nt | More patchs are planned as I chased them down on our current deployment ^^ | 14:39 |
| *** diurnalist has joined #openstack-kolla | 14:40 | |
| Fl1nt | arff, seems like shift+enter isn't recognized by review as a linebreak ^^ | 14:40 |
| *** k_mouza has joined #openstack-kolla | 14:45 | |
| *** wuchunyang has joined #openstack-kolla | 14:45 | |
| *** bsanjeewa has joined #openstack-kolla | 14:49 | |
| lvdombrkr | Hello guys. very quick question. i have deployed network agents to controller but now decide to keep them on compute node instead of controller | 14:49 |
| lvdombrkr | how in proper way remove containers from controleer node - neutron_metadata_agent, neutron_l3_agent | 14:51 |
| lvdombrkr | , neutron_dhcp_agent, neutron_openvswitch_agent | 14:51 |
| lvdombrkr | , openvswitch_vswitchd, openvswitch_db | 14:51 |
| Fl1nt | @lvdombrkr, wanna do DVR mode? | 14:51 |
| lvdombrkr | Fl1nt: provider networks | 14:52 |
| Fl1nt | oh ok, do you have many networks? | 14:53 |
| lvdombrkr | Fl1nt: each compute node will have saparate network | 14:53 |
| lvdombrkr | independent | 14:53 |
| lvdombrkr | nodes are in different DC | 14:54 |
| Fl1nt | so you're filtering your VM launch on each request? | 14:54 |
| Fl1nt | well, easier way is to deploy new config, then remove the remaining containers. | 14:55 |
| lvdombrkr | Fl1nt: yes | 14:55 |
| lvdombrkr | Fl1nt, you mean destroy and re-deploy? | 14:55 |
| Fl1nt | no redeploy and then destroy ^^ | 14:56 |
| Fl1nt | if you destroy your network things before having them installed on your compute nodes, it will not be cool for you ^^ | 14:56 |
| *** cah_link has quit IRC | 15:00 | |
| lvdombrkr | Fl1nt: im not sure that i catch your idea. i already deployed agents to compute node but they also still persist on controller http://paste.openstack.org/raw/794695/ | 15:02 |
| lvdombrkr | i want to remove from controller them with best practises | 15:03 |
| Fl1nt | aaaah | 15:03 |
| Fl1nt | you'll need to remove them the usual openstack way and then destroy the containers | 15:03 |
| lvdombrkr | Fl1nt: ok i see.. i hopened thet there is some better ways | 15:04 |
| lvdombrkr | :D | 15:04 |
| dking | For Kayobe, what network is used for the IP that is set by the deploy image? I thought that it would be admin_oc_net_name, since that is what it seems that Ansible is using to connect to the host. However, I just got an overcloud node on BMC, and it looks like the IP is from a different network. | 15:08 |
| openstackgerrit | Merged openstack/kayobe master: Update feature flags https://review.opendev.org/734064 | 15:09 |
| mgoddard | dking: provisioning happens using the provision_oc network. Has provisioning finished or is it still in the IPA ramdisk? | 15:13 |
| dking | It fails "Wait for SSH access to the nodes" as the IP it's attempting to connect to is not setup on that node. So, admin_oc_net_name must be the same as provision_oc_net_name ? | 15:14 |
| dking | I don't think it's booted by PXE. So, I think it's whatever it installed locally. The state in bifrost is "active", and it has rebooted. It does seem like the IP there is from the provision_oc_net_name, though it isn't one listed in network-allocations.yml. | 15:17 |
| mgoddard | dking: could be from DHCP from inspector? | 15:19 |
| mgoddard | if your configdrive didn't parse it will fall back to dhcp | 15:19 |
| mgoddard | cloud-init logs may help | 15:19 |
| dking | Hmm.. That would make sense. Let me check those logs. | 15:22 |
| *** Torel has quit IRC | 15:24 | |
| dking | Odd. It actually gets the correct IP at one point in the logs. The line with "stages.py[INFO]: Applying network configuration from ds bringup=False: {'version': 1, 'config': [{'type': 'vlan',..." has the correct IP. It's not what's setup, but I suppose that it at least got into cloud-init somehow. | 15:35 |
| openstackgerrit | Gaël THEROND proposed openstack/kolla master: Change lang package block order https://review.opendev.org/735346 | 15:43 |
| dking | It also wrote the network files correctly. I might check with the cloud-init folks. | 15:44 |
| *** bsanjeewa_ has joined #openstack-kolla | 15:44 | |
| *** k_mouza has quit IRC | 15:45 | |
| *** k_mouza has joined #openstack-kolla | 15:45 | |
| *** bsanjeewa has quit IRC | 15:48 | |
| *** bengates has quit IRC | 15:48 | |
| *** k_mouza has quit IRC | 15:49 | |
| *** k_mouza has joined #openstack-kolla | 15:49 | |
| Fl1nt | @priteau, fixed the commit message according to your suggestions and checked on my side why didn't it used breakline. | 15:50 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe stable/ussuri: Update feature flags https://review.opendev.org/735365 | 15:50 |
| *** seco_ has quit IRC | 15:50 | |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: CI: Update IPA images during upgrade https://review.opendev.org/735182 | 16:01 |
| *** k_mouza has quit IRC | 16:11 | |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: IPA: Switch to IPA builder and CentOS 8 https://review.opendev.org/735115 | 16:24 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: CI: Update IPA images during upgrade https://review.opendev.org/735182 | 16:25 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: Fix seed VM interface ordering https://review.opendev.org/734040 | 16:32 |
| *** wuchunyang has quit IRC | 16:45 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/train: CI: add missing base jobs https://review.opendev.org/734795 | 16:47 |
| dking | I'm not sure what part Kayobe might play into this or not, but looking further, I see that with cloud-init, the network-scripts are setup as I would expect. However, they just aren't set. When I try to bring up the VLAN manually, I get an error that "(device eth0 not available because profile is not compatible with device (mismatching interface name))". Would that ring a bell for anybody? | 16:59 |
| priteau | Sorry dking, never seen this before | 17:13 |
| r3ap3r | mgoddard: yoctozepto: I added comments to https://bugs.launchpad.net/kolla-ansible/train/+bug/1878412 and https://bugs.launchpad.net/kolla-ansible/+bug/1799689 if it helps at all? If not, let me know. ;D | 17:14 |
| openstack | Launchpad bug 1878412 in kolla-ansible train "CentOS 8 Failure of Restart elasticsearch container" [High,In progress] - Assigned to Mark Goddard (mgoddard) | 17:14 |
| openstack | Launchpad bug 1799689 in kolla-ansible ussuri "Rejecting mapping update to [.kibana] as the final mapping would have more than 1 type: [doc, config]" [Medium,Fix committed] - Assigned to Radosław Piliszek (yoctozepto) | 17:14 |
| yoctozepto | r3ap3r: thx, feedback is welcome | 17:15 |
| yoctozepto | r3ap3r: I've just approved a set of elastikibana patches for train | 17:16 |
| yoctozepto | fyi | 17:16 |
| dking | priteau: Thanks for the response. If I find anything relevant, I'll post back here. | 17:17 |
| *** bsanjeewa_ has quit IRC | 17:19 | |
| r3ap3r | yoctozepto: cool. Glad things appear to be moving smoothly. Hoping to try an upgrade to Ussuri this weekend but we will see how that goes. | 17:21 |
| priteau | dking: this appears to be an error message from NetworkManager. We disable NM in kayobe, so there must be something wrong with your setup | 17:25 |
| priteau | What version of kayobe are you using & are you using the latest HEAD of the stable branch? | 17:26 |
| sorin-mihai | i'm getting a weird error when i try to create a new instance "Error: No valid host was found.". it's weird because all the other instances work just fine and nova doesn't seem to be down. any idea where i should look for the cause? whatever i could find on google regarding this error doesn't really help | 17:27 |
| priteau | sorin-mihai: check nova-scheduler and nova-conductor logs | 17:28 |
| sorin-mihai | right. "Got no allocation candidates from the Placement API. This could be due to insufficient resources or a temporary occurrence as compute nodes start up." might be caused by over provisioning? | 17:29 |
| sorin-mihai | that was from nova-scheduler and the next one is from nova-conductor http://paste.openstack.org/raw/e6wFWVkwjuhuETSLS1dF/ | 17:32 |
| sorin-mihai | all the instances have a total of 98 vCPUs, there are 96 physical CPU threads, i had the impression that by default it would be a 16:1 ratio for CPU threads | 17:35 |
| dking | priteau: I'm very glad to hear you say that! I was actually about to try disabling NetworkManager. I'm using Kayboe version 7.0.1.dev77, which I pulled from stable/train. | 17:36 |
| dking | priteau: I should also overcloud_host_image_workaround_cloud_init_enabled enabled, if that makes a difference. And I just did a git pull, and it is the current stable/train. | 17:42 |
| priteau | overcloud_host_image_workaround_cloud_init_enabled is for an old centos 7.5 bug, it shouldn't be relevant anymore (and was removed from master a few days ago) | 17:43 |
| priteau | However if you are using the latest stable/train, it should disable NM in all situations. So if your NM is still enabled there is a problem. | 17:44 |
| priteau | dking: When you run `kayobe overcloud host configure -t network`, what happens during task "Ensure NetworkManager is disabled"? | 17:45 |
| dking | I'm trying that now. | 17:49 |
| dking | priteau: It never gets there. It fails on "Gathering Facts" in play "Ensure networking is configured" because it can't connect to the nodes. | 17:51 |
| dking | It can't connect directly to any of them yet. It's only ever been able to boot the deploy image through bifrost. The server becomes active and reboots, but as it's using NM, and since that can't boot the VLAN properly, it never gets an IP it to which Kayobe can connect. | 17:53 |
| priteau | Ah, I misunderstood your problem. | 17:53 |
| priteau | I wonder if your interfaces are getting renamed across boots | 17:54 |
| dking | I have seen an issue like this once before when I had to swap out NetworkManager for a similar reason. | 17:54 |
| dking | I'm not sure, but the interface names as they are now are what I would expect them to be. | 17:55 |
| priteau | Is this centos8? | 17:55 |
| *** eliaswimmer has quit IRC | 17:57 | |
| dking | Yes, everything at the moment should be centos8, including the ansible host, the seed host, and the deploy image. The node is currently running centos8. | 17:59 |
| dking | Okay, I looked it up, and I had this same problem before. It's looking like the problem is that I'm having trouble provisioning because the deploy image is having trouble bringing up the VLAN with NetworkManager, and should probably be using network-scripts. Is there something that can be done to make it do so? | 18:03 |
| *** seco_ has joined #openstack-kolla | 18:05 | |
| *** priteau has quit IRC | 18:06 | |
| *** bsanjeewa_ has joined #openstack-kolla | 18:07 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/train: CentOS 8: Support Elasticsearch & Kibana 6.x https://review.opendev.org/730390 | 18:13 |
| openstackgerrit | Merged openstack/kolla-ansible stable/train: [elasticsearch] Update config for 6.x https://review.opendev.org/730391 | 18:13 |
| *** stingrayza has quit IRC | 18:14 | |
| *** kevko has joined #openstack-kolla | 18:19 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/train: [elasticsearch] Add migration for Kibana 6.x index https://review.opendev.org/730392 | 18:19 |
| *** irclogbot_3 has quit IRC | 18:29 | |
| *** bsanjeewa has joined #openstack-kolla | 18:32 | |
| *** irclogbot_2 has joined #openstack-kolla | 18:32 | |
| *** bsanjeewa_ has quit IRC | 18:34 | |
| Fl1nt | @mgoddard, thanks a lot for ELK 6 release on Train and CentOS 8, perfect timing! | 18:36 |
| *** eliaswimmer has joined #openstack-kolla | 18:47 | |
| *** bsanjeewa_ has joined #openstack-kolla | 18:48 | |
| *** bsanjeewa has quit IRC | 18:52 | |
| *** e0ne_ has joined #openstack-kolla | 19:08 | |
| *** e0ne has quit IRC | 19:08 | |
| *** kevko has quit IRC | 19:08 | |
| *** lvdombrkr has quit IRC | 19:08 | |
| *** stingrayza has joined #openstack-kolla | 19:10 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Move NFV reqs installation to where it belongs https://review.opendev.org/735386 | 19:35 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible stable/ussuri: CI: Move NFV reqs installation to where it belongs https://review.opendev.org/735387 | 19:35 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible stable/train: CI: Move NFV reqs installation to where it belongs https://review.opendev.org/735388 | 19:37 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible stable/stein: CI: Move NFV reqs installation to where it belongs https://review.opendev.org/735389 | 19:38 |
| *** vishalmanchanda has quit IRC | 19:51 | |
| *** born2bake has joined #openstack-kolla | 19:53 | |
| *** bsanjeewa has joined #openstack-kolla | 20:00 | |
| *** bsanjeewa_ has quit IRC | 20:02 | |
| *** seco_ has quit IRC | 20:07 | |
| admin0 | https://docs.openstack.org/kolla-ansible/latest/user/quickstart.html -- python-selinux is not there | 20:26 |
| admin0 | E: Unable to locate package python-selinux | 20:26 |
| admin0 | plz ignore :) | 20:28 |
| *** bsanjeewa_ has joined #openstack-kolla | 20:38 | |
| *** bsanjeewa has quit IRC | 20:41 | |
| *** e0ne_ has quit IRC | 20:49 | |
| *** jovial[m] has quit IRC | 20:49 | |
| *** bsanjeewa_ has quit IRC | 20:53 | |
| admin0 | i seem to be getting stuck at name: Generate /etc/hosts for all of the nodes | 20:54 |
| admin0 | ASK [baremetal : Generate /etc/hosts for all of the nodes] | 20:54 |
| admin0 | i checked an old bug report https://review.opendev.org/#/c/434940/ | 20:55 |
| patchbot | patch 434940 - kolla-ansible - Fix bootstrap-servers with invalid api_interface (MERGED) - 3 patch sets | 20:55 |
| admin0 | but my network interface matches api_interface | 20:55 |
| admin0 | any ideas what i could be doing wrong ? | 20:56 |
| -openstackstatus- NOTICE: The Gerrit service on review.opendev.org is going offline momentarily at 21:00 UTC for project rename maintenance, but should return within a few minutes: http://lists.opendev.org/pipermail/service-announce/2020-June/000004.html | 20:57 | |
| admin0 | here is the ansible log: https://gist.github.com/a1git/9e58e56866aa49f3f9bea455b093b6ae | 21:05 |
| -openstackstatus- NOTICE: gerrit is being taken offline for emergency cleanup, will return to service again shortly | 21:59 | |
| *** Fl1nt has quit IRC | 22:04 | |
| kplant | can you paste the failure? | 22:11 |
| *** Limech has quit IRC | 22:31 | |
| -openstackstatus- NOTICE: The Gerrit service on review.opendev.org is available again | 22:49 | |
| *** eliaswimmer has quit IRC | 22:54 | |
| *** born2bake has quit IRC | 23:02 | |
| *** dasp has quit IRC | 23:57 | |
| *** bsanjeewa_ has joined #openstack-kolla | 23:59 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!