| opendevreview | James Denton proposed openstack/openstack-ansible-os_neutron master: Implements ironic_neutron_agent and baremetal ML2 driver https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/813011 | 03:38 |
|---|---|---|
| opendevreview | James Denton proposed openstack/openstack-ansible-os_neutron master: Implements ironic_neutron_agent and baremetal ML2 driver https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/813011 | 03:53 |
| opendevreview | Merged openstack/openstack-ansible-os_ironic master: Refactor galera_use_ssl behaviour https://review.opendev.org/c/openstack/openstack-ansible-os_ironic/+/810210 | 05:50 |
| *** odyssey4me is now known as Guest2184 | 07:34 | |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible-haproxy_server master: Fix typo for user supplied certificate variable https://review.opendev.org/c/openstack/openstack-ansible-haproxy_server/+/811979 | 09:56 |
| opendevreview | Merged openstack/openstack-ansible stable/stein: fix double-double-quotes https://review.opendev.org/c/openstack/openstack-ansible/+/813044 | 10:15 |
| opendevreview | Merged openstack/openstack-ansible-lxc_hosts stable/train: ensure 20listchanges is not in debian container https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/731625 | 10:27 |
| MrClayPole | I've recently completed an OSA upgrade from Rocky to Train. After the upgrade I was unable to deploy glance images as it was failing with the error "Incorrect configuration file: /etc/glance/rootwrap.conf". We currently used iSCSI LUNs to store glance images. This file was missing from /etc/glance. If I take this file from https://raw.githubusercontent.com/openstack/glance_store/stable/train/etc/glance/rootwrap.conf | 11:08 |
| MrClayPole | and add "/openstack/venvs/glance-20.2.6/etc/glance" to the "exec_dirs" it resolves the issue but I notice that rootwrap.conf isn't in the template dir from Stein onwards so I'm wondering if what I've done is the correct fix? | 11:08 |
| cardiff_dan | jamesdenton: hey James, we talked back on Tuesday about instances in the 10.0.3.0/24 IP range. I was just wondering whether a bug had been logged about it so I can keep track? | 12:11 |
| *** sshnaidm is now known as sshnaidm|afk | 12:17 | |
| jamesdenton | hi cardiff_dan - no, no bug, yet. Sorry. | 12:42 |
| cardiff_dan | no problem, we're putting this on our docs as a known issue for now so users don't get flummoxed :) | 12:44 |
| jamesdenton | understood. i don't think it's an OSA bug, necessarily, but a by-product of linux routing on-host | 12:46 |
| spatel | jamesdenton here you go, HA with connection mirroring - https://satishdotpatel.github.io/ha-with-keepalived-and-conntrackd/ | 12:47 |
| jamesdenton | next stop, OSA! | 12:51 |
| mgariepy | spatel, systemd ? not systemctl ? also missing a `e` there `systemd enable keepalivd` | 12:54 |
| opendevreview | Merged openstack/openstack-ansible master: Remove unnecessary pki step in haproxy install https://review.opendev.org/c/openstack/openstack-ansible/+/812361 | 12:56 |
| mgariepy | nice post :D | 12:57 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible stable/wallaby: Remove unnecessary pki step in haproxy install https://review.opendev.org/c/openstack/openstack-ansible/+/813099 | 13:00 |
| spatel | Thank you! damn it good catch let me fix it :) | 13:00 |
| mgariepy | also the cp command ends with a . which works.. but isn't needed | 13:01 |
| spatel | its my habit to do (.) but you are correct | 13:02 |
| spatel | fixed... | 13:03 |
| jamesdenton | tough crowd! :D | 13:04 |
| spatel | very tough..hehe | 13:05 |
| mgariepy | sorry | 13:05 |
| mgariepy | lol | 13:05 |
| jamesdenton | i have a backlog of posts. lots of notes. | 13:05 |
| spatel | glad that means you guys paying attention not just eyes rolling | 13:05 |
| jamesdenton | s/This post if/This post is/ | 13:06 |
| jamesdenton | if i must contribute something | 13:06 |
| jamesdenton | but yes, nice post! | 13:06 |
| spatel | i should share my creds with you guys to we all can fix stuff :) | 13:06 |
| jamesdenton | we just need to submit PRs :D | 13:07 |
| spatel | s/to/so/ | 13:07 |
| spatel | i think we should turn on conntrackd for OSA so during failover we don't lost any connection :) | 13:08 |
| mgariepy | it works much better than it was now :) | 13:08 |
| mgariepy | services expect to loose connection and reconnect now haha | 13:08 |
| spatel | How does ESTABLISHED connection of mysql get handled ? | 13:09 |
| spatel | if someone running long terraform and failover trigger what will happened ? | 13:10 |
| mgariepy | for mysql anyway. if the failover happens when the host crash. the mysql on it will die also.. and connection will need to be re-done on the new master. | 13:10 |
| spatel | if we don't need then sure we don't need to make it over complicated :) | 13:10 |
| mgariepy | how the mysql connection through haproxy would be migrated to the new server ? when it's controlled it's not much of an issue. | 13:12 |
| mgariepy | but if the host1 becomes unavailable the backup keepalived/haproxy node won't be able to connect to it. | 13:13 |
| spatel | hmm | 13:14 |
| opendevreview | James Denton proposed openstack/openstack-ansible-os_neutron master: Implement ironic_neutron_agent and baremetal driver https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/813011 | 13:24 |
| spatel | jamesdenton look like Ironic goodness going on :) | 13:30 |
| jamesdenton | there's a little bit of movement there. probably need to do some cleanup in that role | 13:31 |
| jamesdenton | and a blog post could help,m too | 13:31 |
| spatel | I would wait for that | 13:32 |
| spatel | I haven't seen anyone using Ironic with OSA (i may be wrong) | 13:32 |
| jamesdenton | i've been using it locally for at least the last... 9 months or so. Pretty sure we have some customers using it, too | 13:33 |
| spatel | why i haven't seen any doc on openstack-ansible page? | 13:33 |
| spatel | i think it need some special stuff to make it work like PXE lan etc | 13:34 |
| jamesdenton | well, someone needs to write the doc, i guess :D | 13:37 |
| jamesdenton | it's somewhere in my stack of notes | 13:37 |
| spatel | i will see if i get some free time for ironic, i want to learn that because we have so many pizza box in datacenter and one dedicated guy managing it so i would like to put them in openstack to manage hardware | 13:43 |
| jamesdenton | mgariepy i did confirm ipmitool was fully functional on this deployed baremetal instance | 14:44 |
| mgariepy | hopefully not on a shared ethenet network ;) | 14:45 |
| jamesdenton | well | 14:48 |
| jamesdenton | :D | 14:48 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible-os_tempest stable/wallaby: Pin neutron-tempest-plugin to v1.6.0 https://review.opendev.org/c/openstack/openstack-ansible-os_tempest/+/813198 | 14:48 |
| mgariepy | any idea how it's kept secured ? | 14:48 |
| jamesdenton | well, this was in my lab, not public cloud | 14:48 |
| jamesdenton | so no, i'm not sure | 14:48 |
| jamesdenton | local ipmi allowed, but access to another host requires auth. | 14:49 |
| mgariepy | yes indeed. | 14:49 |
| jamesdenton | so maybe there's username and password rotation in place, not sure | 14:49 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible stable/wallaby: Remove unnecessary pki step in haproxy install https://review.opendev.org/c/openstack/openstack-ansible/+/813099 | 14:49 |
| mgariepy | ¯\_(ツ)_/¯ well | 14:51 |
| mgariepy | ipmitool over lan needs auth unless cypher 0 can be used. | 14:55 |
| mgariepy | https://book.hacktricks.xyz/pentesting/623-udp-ipmi | 14:55 |
| opendevreview | James Denton proposed openstack/openstack-ansible-os_neutron master: Add support for openvswitch interface driver with OVN https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/813200 | 15:03 |
| opendevreview | Merged openstack/openstack-ansible-os_cinder master: setup.cfg: Replace dashes with underscores https://review.opendev.org/c/openstack/openstack-ansible-os_cinder/+/789712 | 17:20 |
| spatel | very ugly question, we have many kind of servers in openstack like gen8, gen9, Dell etc... is there a way from VM i can pull that information out related which hardware its running? | 17:56 |
| jamesdenton | i think that's all abstracted away | 18:36 |
| jamesdenton | anyone here using Trove? | 18:40 |
| spatel | I have never heard anyway talking about Trove here :) | 19:03 |
| jamesdenton | :) | 19:10 |
| opendevreview | Merged openstack/openstack-ansible master: Implements framework for ironic_neutron_agent and Neutron 'baremetal' plugin https://review.opendev.org/c/openstack/openstack-ansible/+/813006 | 19:16 |
| opendevreview | Merged openstack/openstack-ansible master: Add serial execution to all playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/805188 | 19:20 |
| jamesdenton | anyone seen an issue where a constraints file isn't built on the repo server? I can see global-constraints.txt, requirements.txt, source-constraints.txt, but not regular 'ol constraints.txt - | 20:29 |
| spatel | jamesdenton what is error related on compute nodes - Timed out waiting for nova-conductor. Is it running? | 21:18 |
| spatel | my nova-conductor service is running | 21:18 |
| spatel | does compute use rabbitMQ to talk to nova-conductor? | 21:19 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!