| *** k_mouza has joined #openstack-kolla | 00:01 | |
| *** k_mouza has quit IRC | 00:06 | |
| *** brinzhang_ has joined #openstack-kolla | 00:15 | |
| *** brinzhang0 has quit IRC | 00:18 | |
| *** k_mouza has joined #openstack-kolla | 00:40 | |
| *** k_mouza has quit IRC | 00:45 | |
| *** e0ne has joined #openstack-kolla | 00:55 | |
| *** e0ne has quit IRC | 01:00 | |
| *** e0ne has joined #openstack-kolla | 01:06 | |
| *** e0ne has quit IRC | 01:10 | |
| *** e0ne has joined #openstack-kolla | 01:16 | |
| *** e0ne has quit IRC | 01:20 | |
| *** e0ne has joined #openstack-kolla | 01:25 | |
| *** e0ne has quit IRC | 01:30 | |
| *** hongbin has joined #openstack-kolla | 01:31 | |
| *** e0ne has joined #openstack-kolla | 01:35 | |
| *** e0ne has quit IRC | 01:40 | |
| *** e0ne has joined #openstack-kolla | 01:45 | |
| *** e0ne has quit IRC | 01:50 | |
| *** brinzhang0 has joined #openstack-kolla | 02:30 | |
| *** brinzhang_ has quit IRC | 02:33 | |
| *** williampiv has joined #openstack-kolla | 03:30 | |
| *** e0ne has joined #openstack-kolla | 03:55 | |
| *** e0ne has quit IRC | 04:00 | |
| *** williampiv has quit IRC | 04:11 | |
| *** e0ne has joined #openstack-kolla | 04:16 | |
| *** e0ne has quit IRC | 04:21 | |
| *** e0ne has joined #openstack-kolla | 04:26 | |
| *** e0ne has quit IRC | 04:30 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-kolla | 04:33 | |
| *** hongbin has quit IRC | 04:36 | |
| *** e0ne has joined #openstack-kolla | 04:36 | |
| *** hongbin has joined #openstack-kolla | 04:40 | |
| *** e0ne has quit IRC | 04:40 | |
| *** hongbin has quit IRC | 04:44 | |
| *** e0ne has joined #openstack-kolla | 04:46 | |
| *** e0ne has quit IRC | 04:51 | |
| *** brinzhang_ has joined #openstack-kolla | 04:57 | |
| *** brinzhang0 has quit IRC | 05:00 | |
| *** bersace has joined #openstack-kolla | 05:01 | |
| *** e0ne has joined #openstack-kolla | 05:06 | |
| *** e0ne has quit IRC | 05:11 | |
| *** e0ne has joined #openstack-kolla | 05:16 | |
| *** williampiv has joined #openstack-kolla | 05:17 | |
| *** shyamb has joined #openstack-kolla | 05:19 | |
| *** e0ne has quit IRC | 05:21 | |
| *** e0ne has joined #openstack-kolla | 05:36 | |
| *** e0ne has quit IRC | 05:41 | |
| *** e0ne has joined #openstack-kolla | 05:46 | |
| *** bengates has joined #openstack-kolla | 05:48 | |
| *** bengates has quit IRC | 05:51 | |
| *** e0ne has quit IRC | 05:51 | |
| *** williampiv has quit IRC | 05:59 | |
| *** cah_link has joined #openstack-kolla | 06:01 | |
| *** abdysn has joined #openstack-kolla | 06:05 | |
| *** e0ne has joined #openstack-kolla | 06:06 | |
| *** Abdallahyas has joined #openstack-kolla | 06:07 | |
| *** e0ne has quit IRC | 06:10 | |
| *** abdysn has quit IRC | 06:11 | |
| *** ricolin has joined #openstack-kolla | 06:16 | |
| *** e0ne has joined #openstack-kolla | 06:26 | |
| *** e0ne has quit IRC | 06:31 | |
| *** shyamb has quit IRC | 06:32 | |
| yoctozepto | morning | 06:35 |
|---|---|---|
| *** e0ne has joined #openstack-kolla | 06:37 | |
| *** e0ne has quit IRC | 06:41 | |
| *** wuchunyang has joined #openstack-kolla | 06:43 | |
| recyclehero | yoctozepto: morning | 06:46 |
| recyclehero | here we have fridays as weekend I was dying this two days :D | 06:46 |
| *** e0ne has joined #openstack-kolla | 06:47 | |
| recyclehero | deployed had no nova compute. getting to the root of problem. precheck, after parsing ini inventory I get Parsed /home/alistar/Desktop/Projects/Openstack/multinode.ini inventory source with ini plugin | 06:50 |
| recyclehero | [WARNING]: Could not match supplied host pattern, ignoring: enable_nova_True | 06:50 |
| *** e0ne has quit IRC | 06:51 | |
| * recyclehero dying here | 06:54 | |
| *** e0ne has joined #openstack-kolla | 06:57 | |
| *** nikparasyr has joined #openstack-kolla | 07:01 | |
| *** e0ne has quit IRC | 07:01 | |
| yoctozepto | recyclehero: where are you from? | 07:05 |
| recyclehero | sanctioned Iran, building servers from old computers! | 07:06 |
| recyclehero | though this gov should be sanctioned | 07:07 |
| yoctozepto | recyclehero: oh :-( I heard tough times in there | 07:07 |
| yoctozepto | recyclehero: did you perhaps set enable_nova to no or enable_openstack_core to no? (it should default to yes via enable_openstack_core defaulting to yes) | 07:08 |
| recyclehero | yes, its getting very expensive. i dont know how people survive. its getting harder each day. us election affects us more than us people | 07:09 |
| recyclehero | yoctozepto: no they are both true | 07:09 |
| recyclehero | let me 2 check | 07:09 |
| recyclehero | core is true and enable nove get evaluated to true | 07:10 |
| recyclehero | yoctozepto: the playbook also behave more complex just upon nova but I dont know what caould be causing this. | 07:11 |
| recyclehero | do you think this precheck warning later shows itself as no compute? | 07:11 |
| *** e0ne has joined #openstack-kolla | 07:16 | |
| recyclehero | globals.yml diff ===> Your paste can be seen here: https://dpaste.com/8MDM8N7R5 | 07:18 |
| recyclehero | multinode.ini diff ==> Your paste can be seen here: https://dpaste.com/6RKDTKMQ8 | 07:20 |
| *** kevko has quit IRC | 07:20 | |
| *** e0ne has quit IRC | 07:21 | |
| *** bengates has joined #openstack-kolla | 07:23 | |
| *** mrunge has joined #openstack-kolla | 07:24 | |
| *** mrunge has quit IRC | 07:25 | |
| recyclehero | what I understood is that host are added a group enable_servicename_true from the site.yml task named 'group host based on cinfig' | 07:31 |
| recyclehero | then it goes roles to host by checking inventory patterns + &enable_servicename_true | 07:32 |
| recyclehero | for nova cause its complicated it import another playbook nova.yml | 07:32 |
| recyclehero | and to the role appling there | 07:33 |
| *** mrunge has joined #openstack-kolla | 07:35 | |
| recyclehero | now i will try to add them manually to a group named enable_nova_True | 07:35 |
| *** e0ne has joined #openstack-kolla | 07:36 | |
| *** maharg101 has joined #openstack-kolla | 07:47 | |
| *** born2bake has joined #openstack-kolla | 07:51 | |
| *** Abdallahyas has quit IRC | 08:02 | |
| *** abdysn has joined #openstack-kolla | 08:02 | |
| recyclehero | God please give me compute power! | 08:06 |
| wuchunyang | it seems that enable_nova is disabled | 08:10 |
| recyclehero | wuchunyang: 10:45 < recyclehero> globals.yml diff ===> Your paste can be seen here: https://dpaste.com/8MDM8N7R5 | 08:11 |
| recyclehero | 10:47 < recyclehero> multinode.ini diff ==> Your paste can be seen here: https://dpaste.com/6RKDTKMQ8 | 08:11 |
| recyclehero | yoctozepto: I get into kolla_docker.py and saw ur name on two FIXME. | 08:12 |
| yoctozepto | recyclehero: yeah, that's me | 08:13 |
| yoctozepto | recyclehero: please share the output of kolla-ansible with at two levels of verbosity (-vv) set | 08:14 |
| yoctozepto | otherwise we can't really help | 08:14 |
| recyclehero | yoctozepto: ok. precheck, deploy what? | 08:15 |
| yoctozepto | recyclehero: what fails | 08:15 |
| recyclehero | it didnt failed on deploy. I deployed then I destroyed it. I even had horizon made users, project. couldnt lunch instances. | 08:15 |
| recyclehero | and didnt see any nova-compute docker container running | 08:16 |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible master: CI: enable designate in magnum CI job https://review.opendev.org/754456 | 08:16 |
| recyclehero | it should one hypervisor with QEMU which was the only host and no compute hosts available in horizon | 08:18 |
| wuchunyang | recyclehero try to deploy nova only. kolla-ansible deploy --tags nova | 08:18 |
| wuchunyang | will it skip all tasks ? | 08:18 |
| *** shyamb has joined #openstack-kolla | 08:19 | |
| recyclehero | ok, I will do that now paste stdout | 08:19 |
| *** abdysn has quit IRC | 08:23 | |
| *** abdysn has joined #openstack-kolla | 08:23 | |
| recyclehero | wuchunyang: kolla-ansible -i multinode.ini deploy --tags nova > ~/Desktop/deleteme 2>&1 | 08:23 |
| wuchunyang | ok | 08:23 |
| recyclehero | Your paste can be seen here: https://dpaste.com/3H9DFZQSY | 08:27 |
| *** k_mouza has joined #openstack-kolla | 08:27 | |
| wuchunyang | what's your ansible version ? | 08:28 |
| recyclehero | its iansible 2.9.13 | 08:29 |
| recyclehero | in a virtual env | 08:29 |
| wuchunyang | failed: [172.16.0.1] (item={'key': 'nova-api', 'value': {'container_name': 'nova_api', 'group': 'nova-api', 'image': '172.16.0.121:4000/kolla/debian-source-nova-api:ussuri', 'enabled': True, 'privileged': True, 'volumes': ['/etc/kolla/nova-api/:/var/lib/kolla/config_files/:ro', '/etc/localtime:/etc/localtime:ro', '/etc/timezone:/etc/timezone:ro', '/lib/modules:/lib/modules:ro', 'kolla_logs:/var/log/kolla/', ''], 'dimensions': | 08:30 |
| wuchunyang | {}, 'haproxy': {'nova_api': {'enabled': True, 'mode': 'http', 'external': False, 'port': '8774', 'listen_port': '8774'}, 'nova_api_external': {'enabled': True, 'mode': 'http', 'external': True, 'port': '8774', 'listen_port': '8774'}, 'nova_metadata': {'enabled': True, 'mode': 'http', 'external': False, 'port': '8775', 'listen_port': '8775'}, 'nova_metadata_external': {'enabled': True, 'mode': 'http', 'external': True, 'port': | 08:30 |
| wuchunyang | '8775', 'listen_port': '8775'}, 'nova_rdp': {'enabled': False, 'mode': 'http', 'external': False, 'port': '8001', 'host_group': 'hyperv'}}}}) => {"ansible_loop_var": "item", "changed": false, "checksum": "05e0a4f9b84f5977fbb676854a12ebc7e40a2314", "item": {"key": "nova-api", "value": {"container_name": "nova_api", "dimensions": {}, "enabled": true, "group": "nova-api", "haproxy": {"nova_api": {"enabled": true, "external": fals | 08:30 |
| wuchunyang | e, "listen_port": "8774", "mode": "http", "port": "8774"}, "nova_api_external": {"enabled": true, "external": true, "listen_port": "8774", "mode": "http", "port": "8774"}, "nova_metadata": {"enabled": true, "external": false, "listen_port": "8775", "mode": "http", "port": "8775"}, "nova_metadata_external": {"enabled": true, "external": true, "listen_port": "8775", "mode": "http", "port": "8775"}, "nova_rdp": {"enabled": false, | 08:30 |
| wuchunyang | "external": false, "host_group": "hyperv", "mode": "http", "port": "8001"}}, "image": "172.16.0.121:4000/kolla/debian-source-nova-api:ussuri", "privileged": true, "volumes": ["/etc/kolla/nova-api/:/var/lib/kolla/config_files/:ro", "/etc/localtime:/etc/localtime:ro", "/etc/timezone:/etc/timezone:ro", "/lib/modules:/lib/modules:ro", "kolla_logs:/var/log/kolla/", ""]}}, "msg": "Destination directory /etc/kolla/haproxy/services.d | 08:30 |
| wuchunyang | does not exist"} | 08:30 |
| wuchunyang | ok: [172.16.0.1] => (item=enable_nova_True) at line 136 | 08:31 |
| wuchunyang | enable_nova is true | 08:31 |
| wuchunyang | but your tasks failed | 08:31 |
| wuchunyang | Destination directory /etc/kolla/haproxy/services.d does not exis | 08:32 |
| *** k_mouza has quit IRC | 08:32 | |
| recyclehero | wuchunyang: I will teke care of that. but on the last deploy that I disabled haproxy since I have one node now. thanks | 08:33 |
| wuchunyang | but your haproxy is enabled now . | 08:34 |
| wuchunyang | it there haproxy container on hosts ? | 08:36 |
| mnasiadka | morning | 08:42 |
| *** hjensas_ is now known as hjensas | 08:52 | |
| *** k_mouza has joined #openstack-kolla | 09:01 | |
| *** shyam89 has joined #openstack-kolla | 09:11 | |
| *** shyamb has quit IRC | 09:13 | |
| LinPeiWen | morning | 09:29 |
| *** zzzeek has quit IRC | 09:30 | |
| *** zzzeek has joined #openstack-kolla | 09:32 | |
| LinPeiWen | mnasiadka： I would like to ask, is there any plan to merge this patch in the Victoria version? https://review.opendev.org/#/c/746384/ | 09:34 |
| patchbot | patch 746384 - kolla-ansible - Add support for GlusterFS NFS Manila backend - 17 patch sets | 09:34 |
| *** gfidente has joined #openstack-kolla | 09:35 | |
| mgoddard | morning | 09:38 |
| mnasiadka | LinPeiWen: I'll add this to my review list, now busy with something else | 09:40 |
| LinPeiWen | will do . | 09:42 |
| *** bersace has quit IRC | 09:46 | |
| LinPeiWen | mnasiadka: nice :] | 09:46 |
| *** abdysn has quit IRC | 09:52 | |
| *** abdysn has joined #openstack-kolla | 09:56 | |
| *** wuchunyang has quit IRC | 10:06 | |
| *** Tengu has joined #openstack-kolla | 10:10 | |
| *** Abdallahyas has joined #openstack-kolla | 10:17 | |
| *** abdysn has quit IRC | 10:20 | |
| *** shyam89 has quit IRC | 10:21 | |
| *** stingrayza has joined #openstack-kolla | 10:23 | |
| osmanlicilegi | morning | 10:23 |
| *** also_stingrayza has quit IRC | 10:26 | |
| recyclehero | guys deployed one more time. it succeded but cant launch instance | 11:09 |
| recyclehero | no hypervisor | 11:09 |
| recyclehero | https://file.io/UB2QRNW3vaw2 | 11:15 |
| recyclehero | above is output of kolla-ansible -i multinode.ini -vv deploy > /home/alistar/Desktop/deleteme 2>&1 | 11:16 |
| *** abdysn has joined #openstack-kolla | 11:21 | |
| *** Abdallahyas has quit IRC | 11:24 | |
| *** shyamb has joined #openstack-kolla | 11:26 | |
| openstackgerrit | Merged openstack/kolla master: Add missing heat deps for deb-binary https://review.opendev.org/753944 | 11:30 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: [baremetal]: Use $releasever in docker-ce repo https://review.opendev.org/754691 | 11:32 |
| *** k_mouza has quit IRC | 11:53 | |
| *** k_mouza has joined #openstack-kolla | 11:55 | |
| * recyclehero crying | 11:59 | |
| recyclehero | its not possible to get shell inside nova_libvirt container? docker exec -it nova_libvirt | 12:05 |
| recyclehero | i think it only changes the command prompt | 12:05 |
| recyclehero | maybe it has a no login shell | 12:05 |
| recyclehero | lots of problem and no way to debug | 12:05 |
| recyclehero | my fault can connect | 12:08 |
| recyclehero | finally see something | 12:08 |
| recyclehero | kvm is not supported by /usr/bin/qemu-sestem-riscv64 on this host | 12:09 |
| *** JamesBenson has joined #openstack-kolla | 12:12 | |
| *** shyamb has quit IRC | 12:14 | |
| recyclehero | actuaaly its bunch of simillar errors | 12:14 |
| recyclehero | kvm is not ..... /bin/qemu.....mips | 12:14 |
| recyclehero | aarch64 | 12:15 |
| recyclehero | its in the file /var/log/libvirt/libvirtd.log | 12:17 |
| recyclehero | its start with a failure on dnsmasq | 12:17 |
| recyclehero | then all of these messages | 12:17 |
| recyclehero | virQEMUCapsCacheLookupDefault:5050 : invalid argument : kvm is not supported by 'usr/bin/qemu-system-[architechure]' on this host | 12:18 |
| *** mrunge has quit IRC | 12:24 | |
| *** mrunge has joined #openstack-kolla | 12:27 | |
| recyclehero | heading to openstack cause I think its out of scope of kolla | 12:37 |
| recyclehero | thanks chanell | 12:37 |
| *** k_mouza has quit IRC | 12:41 | |
| hrw | recyclehero: "docker exec -it nova_libvirt bash" | 12:43 |
| hrw | recyclehero: trying to run openstack on riscv? | 12:43 |
| recyclehero | hrw no riscv was one of them it printed the same error for all archs | 12:44 |
| recyclehero | hrw: I am in thre and getting the logs out | 12:45 |
| recyclehero | the compute host get unrecognized as stated in logs | 12:45 |
| recyclehero | i dont know why | 12:45 |
| hrw | recyclehero: so you do boring x86-64 one? | 12:47 |
| hrw | meh. | 12:47 |
| * hrw -> food preparing | 12:47 | |
| recyclehero | yeah boring x86-64 doesnt even works | 12:48 |
| *** k_mouza has joined #openstack-kolla | 12:48 | |
| recyclehero | here are the logs if anyone care to look for nova-compute | 12:48 |
| recyclehero | https://file.io/MvVxAiOx6Gzi | 12:49 |
| *** k_mouza has quit IRC | 12:49 | |
| *** k_mouza has joined #openstack-kolla | 12:49 | |
| recyclehero | libvirt: https://file.io/ZilF0FBMPVN2 | 12:51 |
| *** Luzi has joined #openstack-kolla | 12:55 | |
| *** LinPeiWen has quit IRC | 13:10 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/ussuri: Fix keystone-startup.sh https://review.opendev.org/754392 | 13:14 |
| *** wuchunyang has joined #openstack-kolla | 13:19 | |
| *** TrevorV has joined #openstack-kolla | 13:52 | |
| recyclehero | do I need the network node If I only want a flat network? | 13:53 |
| recyclehero | I am going to simplify a bit then destory and redeploy and if wont work move to 1-centos 2-ubuntu | 13:54 |
| *** Luzi has quit IRC | 13:54 | |
| *** hongbin has joined #openstack-kolla | 13:54 | |
| yoctozepto | recyclehero: network node is used for haproxy as well | 13:55 |
| *** gmann_pto is now known as gmann | 13:55 | |
| recyclehero | and if I remove haproxy there are lots of other things as I saw in the playbooks so thats not possbile | 13:55 |
| yoctozepto | recyclehero: well, haproxy is optional | 13:56 |
| yoctozepto | but nevertheless strongly recommended | 13:56 |
| *** shyamb has joined #openstack-kolla | 13:56 | |
| recyclehero | :)) I cant lunch instances so dont recommend it yet | 13:56 |
| recyclehero | I remember some red warning while deploying relating to firewalls. can it be the cause. since debian dont have that ubuntu firewall | 13:57 |
| recyclehero | ? | 13:57 |
| recyclehero | yoctozepto: back to network node, so if I enable_haproxy to false I can leave [network] empty? I want to use linux bridge instead of ovs then | 14:00 |
| yoctozepto | recyclehero: you would also need to disable l3 and dhcp agents | 14:00 |
| yoctozepto | but they might self-disable as well | 14:00 |
| recyclehero | I think dhcp should be there though | 14:01 |
| recyclehero | so l3 should be disable in globals | 14:01 |
| recyclehero | and one last thing to redeploy i should run destory and then bootstrap-prechecks-deploy | 14:02 |
| recyclehero | do i need reconfigure to? | 14:02 |
| *** jbadiapa has joined #openstack-kolla | 14:26 | |
| *** shyamb has quit IRC | 14:36 | |
| *** kevko has joined #openstack-kolla | 14:41 | |
| *** shyamb has joined #openstack-kolla | 14:43 | |
| *** shyamb has quit IRC | 14:45 | |
| *** kevko has quit IRC | 14:45 | |
| *** shyamb has joined #openstack-kolla | 14:59 | |
| *** nikparasyr has left #openstack-kolla | 15:03 | |
| *** wuchunyang has quit IRC | 15:18 | |
| *** shyamb has quit IRC | 15:21 | |
| *** shyamb has joined #openstack-kolla | 15:24 | |
| *** shyamb has quit IRC | 15:30 | |
| *** bengates has quit IRC | 16:08 | |
| *** bengates has joined #openstack-kolla | 16:09 | |
| yoctozepto | recyclehero: no, reconfigure is pretty much the same as deploy | 16:11 |
| yoctozepto | you can later use it interchangeably as of now | 16:11 |
| yoctozepto | we might change this in future | 16:11 |
| *** bengates has quit IRC | 16:14 | |
| *** abdysn has quit IRC | 16:21 | |
| *** k_mouza has quit IRC | 16:28 | |
| *** hrw has quit IRC | 16:35 | |
| *** hrw has joined #openstack-kolla | 16:37 | |
| *** k_mouza has joined #openstack-kolla | 16:37 | |
| *** cah_link has quit IRC | 16:41 | |
| *** k_mouza has quit IRC | 16:42 | |
| *** gfidente is now known as gfidente|afk | 16:59 | |
| *** jonaspaulo has joined #openstack-kolla | 17:10 | |
| openstackgerrit | Merged openstack/kolla-ansible master: Reduce the use of SQLAlchemy connection pooling https://review.opendev.org/749632 | 17:14 |
| *** shyamb has joined #openstack-kolla | 17:17 | |
| *** jonaspaulo has quit IRC | 17:24 | |
| *** k_mouza has joined #openstack-kolla | 17:25 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: [baremetal]: Use $releasever in docker-ce repo https://review.opendev.org/754691 | 17:27 |
| *** k_mouza has quit IRC | 17:29 | |
| *** brinzhang0 has joined #openstack-kolla | 17:30 | |
| *** shyamb has quit IRC | 17:32 | |
| *** brinzhang_ has quit IRC | 17:33 | |
| *** k_mouza has joined #openstack-kolla | 17:37 | |
| *** k_mouza has quit IRC | 17:41 | |
| *** e0ne has quit IRC | 17:47 | |
| *** k_mouza has joined #openstack-kolla | 18:06 | |
| *** k_mouza has quit IRC | 18:10 | |
| *** k_mouza has joined #openstack-kolla | 18:11 | |
| *** k_mouza has quit IRC | 18:15 | |
| oklhost | Hm... somebody around who has a clue about neutron? I've a provider network, lets say vlan 1, 10.0.0.0/24, gw 10.0.0.1 as well a floating ip network, vlan 2, 10.0.1.0/24 gw 10.0.1.1. Instances get a IP in vlan 1, but I wan't users to get a floating ip in vlan 2 optional. I created a router in admin project for vlan 2 with port in vlan 1. The port gets 10.0.0.1 automatically, I can assign a floating ip to my instance, but then the network | 18:16 |
| oklhost | runs pretty strange, sometime the floating ip is reachable sometime the provider network ip, ssh connections are dying. I think there is an ip conflict? because the port get's the vlan 1 gw ip, right? Is this setup totally wrong?! | 18:16 |
| *** k_mouza has joined #openstack-kolla | 18:20 | |
| *** k_mouza has quit IRC | 18:25 | |
| *** k_mouza has joined #openstack-kolla | 18:31 | |
| *** k_mouza has quit IRC | 18:36 | |
| yoctozepto | mgoddard, mnasiadka: based on folks reports' it's going to be a keystone ride to stabilise the release :-) | 18:38 |
| mnasiadka | yoctozepto: not really | 18:39 |
| yoctozepto | mnasiadka: their times look much above the new bound | 18:39 |
| mnasiadka | yoctozepto: this guy Joel in the bug - he has crontab entry on one out of three controllers | 18:39 |
| yoctozepto | mnasiadka: must be some flaw in generation | 18:39 |
| mnasiadka | So that’s expected to have it stale :) | 18:40 |
| mnasiadka | yoctozepto: might be, but I can’t reproduce that flaw | 18:40 |
| yoctozepto | well, losing one controller for its rotation day shouldn't get us in such a mess | 18:41 |
| yoctozepto | mnasiadka: remind me; is the new keystone rotation code train+ now? | 18:43 |
| yoctozepto | or only ussuri+ | 18:43 |
| yoctozepto | or victoria | 18:43 |
| mnasiadka | That’s another thing, if you have three controllers and default settings - cron job ends up only on two | 18:43 |
| mnasiadka | yoctozepto: train+, it’s rather the check for stale that is giving us headaches :) | 18:44 |
| *** k_mouza has joined #openstack-kolla | 18:45 | |
| yoctozepto | mnasiadka: sorting issue? wrong timestamp used? | 18:48 |
| yoctozepto | mnasiadka: why you think so? | 18:48 |
| mnasiadka | yoctozepto: wrong variable was used, so it fixed majority of issues | 18:49 |
| mnasiadka | yoctozepto: i mean the latest patch | 18:49 |
| mnasiadka | yoctozepto: now question how to make crontab more reliable | 18:49 |
| *** k_mouza has quit IRC | 18:50 | |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible master: Implement automatic deploy of octavia https://review.opendev.org/740180 | 18:51 |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible master: octavia: generate certificates automatically https://review.opendev.org/754280 | 18:51 |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible master: CI: octavia: create and test a load balancer https://review.opendev.org/754285 | 18:51 |
| mgoddard | mnasiadka, yoctozepto: I have long thought we need a small etcd based app to do distributed key rotation | 18:52 |
| yoctozepto | mnasiadka: yeah, that was real pain yeah | 18:52 |
| mnasiadka | mgoddard: that wouldn’t be bad | 18:53 |
| yoctozepto | mgoddard: any spec of that? I believe right crontab is enough for this rare event ;-) | 18:53 |
| yoctozepto | just need to sit down to it | 18:53 |
| yoctozepto | oh how I wish the days were longer | 18:54 |
| mgoddard | yoctozepto: well you need to avoid under-rotation and over-rotation | 18:54 |
| yoctozepto | :-) | 18:54 |
| mgoddard | difficult with a static cron where nodes are unreliable | 18:54 |
| mgoddard | keystone should build this thing | 18:54 |
| mgoddard | every deploy tool needs it | 18:55 |
| mgoddard | anyway, off now | 18:55 |
| yoctozepto | mgoddard: I thought we need to minimise over-rotation rather than avoid it? | 18:55 |
| mgoddard | no spec I'm aware of | 18:55 |
| yoctozepto | just this qq and you are free mgoddard | 18:55 |
| yoctozepto | have a nice evening | 18:55 |
| *** maharg101 has quit IRC | 18:55 | |
| mgoddard | avoid if possible. otherwise keys go invalid | 18:55 |
| yoctozepto | ah, you mean this over-rotation | 18:55 |
| yoctozepto | all clear | 18:56 |
| yoctozepto | mgoddard, mnasiadka: we can use our newly made hacluster for that ;d | 18:56 |
| mgoddard | lol | 18:56 |
| mgoddard | over-kill | 18:57 |
| yoctozepto | mgoddard: ideal! ;p | 18:57 |
| mgoddard | \o | 18:57 |
| mnasiadka | yoctozepto: unreliable :) | 18:57 |
| *** e0ne has joined #openstack-kolla | 18:57 | |
| yoctozepto | well, it can ensure a single copy of the rotator and distributor | 18:57 |
| yoctozepto | mnasiadka: etcd can fail too ;p | 18:57 |
| yoctozepto | for me it is pretty reliable | 18:58 |
| mnasiadka | yoctozepto: etcd doesn’t spawn an army of processes to try to heal itself and fail miserably :) | 18:58 |
| yoctozepto | anyway, till tomorrow! | 18:58 |
| mnasiadka | yoctozepto: just laughing | 18:58 |
| mnasiadka | Yeah, tomorrow | 18:58 |
| yoctozepto | well, the issue is usually in the controllers used by pacemaker rather than pacemaker/corosync itself | 18:59 |
| *** hongbin has quit IRC | 19:29 | |
| *** hongbin has joined #openstack-kolla | 19:35 | |
| *** dking has joined #openstack-kolla | 19:38 | |
| *** e0ne has quit IRC | 19:41 | |
| *** k_mouza has joined #openstack-kolla | 19:43 | |
| *** k_mouza has quit IRC | 19:48 | |
| *** maharg101 has joined #openstack-kolla | 19:52 | |
| *** maharg101 has quit IRC | 20:01 | |
| *** hongbin has quit IRC | 20:07 | |
| *** k_mouza has joined #openstack-kolla | 20:19 | |
| *** hongbin has joined #openstack-kolla | 20:21 | |
| *** k_mouza has quit IRC | 20:23 | |
| *** jbadiapa has quit IRC | 21:41 | |
| *** JamesBenson has quit IRC | 22:09 | |
| *** recyclehero has quit IRC | 22:10 | |
| *** recyclehero has joined #openstack-kolla | 22:17 | |
| *** born2bake has quit IRC | 23:24 | |
| *** zzzeek has quit IRC | 23:44 | |
| *** zzzeek has joined #openstack-kolla | 23:46 | |
| *** TrevorV has quit IRC | 23:52 | |
| *** maharg101 has joined #openstack-kolla | 23:58 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!