| *** ysandeep|out is now known as ysandeep | 01:47 | |
| *** ysandeep is now known as ysandeep|afk | 03:15 | |
| *** akahat|ruck is now known as akahat | 03:46 | |
| *** ysandeep|afk is now known as ysandeep | 03:55 | |
| *** ysandeep is now known as ysandeep|afk | 04:32 | |
| *** chandankumar is now known as chkumar|rover | 04:32 | |
| *** ysandeep|afk is now known as ysandeep | 05:07 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump OpenStack-Ansible master https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 06:46 |
|---|---|---|
| noonedeadpunk | mornings | 06:49 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump OpenStack-Ansible master https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 06:53 |
| *** frenzyfriday is now known as frenzyfriday|ruck | 07:01 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump OpenStack-Ansible master https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 07:09 |
| noonedeadpunk | it not reall good morning though :( | 07:11 |
| derekokeeffe85 | Hi again, so running the neutron role keeps failing with this https://paste.openstack.org/show/bt3BTxVjOMgXhAILmtKG/ that was happening before because of a misconfig on that interface but I have fixed that since. It's the internal_lb_vip address, the external_lb_vip address is just a dns to a second IP on the same link as it's all internal so could this be an issue? | 08:20 |
| *** ysandeep is now known as ysandeep|lunch | 08:44 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_keystone master: Improve way of cache backend selection https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/859054 | 08:49 |
| noonedeadpunk | derekokeeffe85: it should be fine to have external/internal address on same network | 08:52 |
| noonedeadpunk | But can you reach internal VIP from compute1 with ping or smth like that? | 08:52 |
| derekokeeffe85 | Yep I can ping it and ssh to it using both IP's. | 09:20 |
| *** ysandeep|lunch is now known as ysandeep | 09:50 | |
| ThiagoCMC | Hey folks! I'm trying OSA `master` (latest fixes for Ironic) on Ubuntu 20.04, 'os-horizon-install.yml' TASK 'os_horizon : Registering panels' prints lots of warnings, directories (`/conf`, `local_settings.d`) reporting "is a directory". | 10:58 |
| ThiagoCMC | Then, when I try to login, Horizon's page print "Something went wrong!" error. | 10:59 |
| ThiagoCMC | Any magic? lol | 10:59 |
| ThiagoCMC | Output: https://pastebin.com/AyB1FWvM | 11:04 |
| ThiagoCMC | I manually created the apparently missing directories in the horizon container, os-horizon-install.yml seem ok, but still can't login. | 11:28 |
| ThiagoCMC | Here's the `journalctl -f` logs from horizon container: https://pastebin.com/7RjayYn9 | 11:39 |
| noonedeadpunk | derekokeeffe85: ok, interesting. And haproxy does listen on 8181? | 11:51 |
| noonedeadpunk | ThiagoCMC: tbh no idea. smth might have changed on master and we might need to adjsut deployment to fit it... | 11:52 |
| noonedeadpunk | and you do have ironic, heat, magnum, masakari and zun? | 11:54 |
| ThiagoCMC | Oh, ok... I might have to then stick with 'stable/yoga' and try to cherry-pick fixes for Ironic in a local branch. | 12:00 |
| noonedeadpunk | tbh I'm not sure what;s going on there - looks like quite valid bug to me. But I didn't know about it | 12:02 |
| ThiagoCMC | Yeah! Mostly! Magnum is not being used now, the others, yes (Ironic trying for first time). Zun could be an alternative for baremetal deployment (if Ironic idea goes south) | 12:02 |
| ThiagoCMC | Ok, no problem... | 12:03 |
| ThiagoCMC | Thanks! :-) | 12:03 |
| noonedeadpunk | I'd say zun is the one that will likely go south given how project being maintained | 12:05 |
| noonedeadpunk | ThiagoCMC: or you can figure out what's wrong with horizon and propose fix ;) | 12:06 |
| ThiagoCMC | Never tried zun... Wanna see what it does. And yeah, getting ready for a deep dive on OSA over the weekend! | 12:07 |
| ThiagoCMC | Oh, another thing, just curious... No plans to add 'senlin_ui' to OSA? :-P | 12:12 |
| noonedeadpunk | super easy thing to add. nobody asked :) | 12:13 |
| noonedeadpunk | I can give even some example of how other plugins were added so you can try to do it yourself ;) | 12:14 |
| ThiagoCMC | Precisely what I have in mind! :-D | 12:14 |
| *** frenzyfriday|ruck is now known as frenzyfriday|ruck|food | 12:26 | |
| ThiagoCMC | So, what's out there for containers? I really like LXD, sad that 'nova-lxd' died. It could potentially be used for both Containers amd QEMU/KVM, since LXD supports both seamlessly. I think it would be awesome to create LXD Containers straight on baremetal with OpenStack, as if they were regular Nova Instances... | 12:27 |
| noonedeadpunk | yeah, I guess then you need zun indeed | 12:34 |
| noonedeadpunk | if it's jsut about containers | 12:34 |
| noonedeadpunk | or smth like murano, but I guess it has died completely | 12:34 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump OpenStack-Ansible master https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 12:40 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump OpenStack-Ansible master https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 12:41 |
| ThiagoCMC | Well, but if Zun isn't properly maintained, then, I would prefer Ironic and deploy LXD on them, and try to figure out the network on the fly lol | 12:42 |
| noonedeadpunk | ThiagoCMC: example on adding adjutant dashboard https://opendev.org/openstack/openstack-ansible-os_horizon/commit/0634bf19b4b8091a1c0ee91e5dc8e7de390677e8 | 12:42 |
| ThiagoCMC | Awesome! | 12:43 |
| noonedeadpunk | And https://opendev.org/openstack/openstack-ansible/commit/416f96434410f47fabed30013694e44cf09b75b1 also needed | 12:44 |
| ThiagoCMC | It seems very simple | 12:46 |
| noonedeadpunk | it really is | 12:47 |
| noonedeadpunk | no rocket since involved :) | 12:48 |
| noonedeadpunk | *science | 12:48 |
| ThiagoCMC | ^_^ | 12:48 |
| ThiagoCMC | I'm no expert but I'd like to contribute with Prs, if I can / manage to do something cool! | 12:49 |
| ThiagoCMC | On the other hand, resurrecting 'nova-lxd' won't be easy. lol | 12:52 |
| *** frenzyfriday|ruck|food is now known as frenzyfriday|ruck | 13:11 | |
| derekokeeffe85 | noonedeadpunk I ran that again and it ran through!! Only change I made was to the netplan I added our name servers to the bridge | 13:32 |
| spatel | jamesdenton, i have this public ip subnet range, now i want to add few more ip in this range how should i add interleave ip ? for example i want to add .138 .139 and .140 new 3 ip address in existing subnet https://paste.opendev.org/show/bqSsIyBBgWjQXhZ7Uo5n/ | 13:53 |
| jamesdenton | am i seeing that in the /24 you only have 3 IPs available for allocation? .9, .10, .11? | 13:54 |
| jamesdenton | You can edit the subnet to add a second allocation range or .138-.140 | 13:54 |
| jamesdenton | *of | 13:55 |
| spatel | Yes.. this is very special case :) | 14:02 |
| *** ysandeep is now known as ysandeep|out | 14:02 | |
| spatel | how do i add extra range.. like 49.xx.xx.109-111,49,xx.xx.138-140 like this? | 14:03 |
| spatel | let me try and see | 14:04 |
| anskiy | explicitly setting IP-address, when creating a port overrides allocation range, at least via SDK | 14:05 |
| anskiy | this should work (from `openstack port create --help`): `--fixed-ip subnet=<subnet>,ip-address=<ip-address>` | 14:06 |
| jamesdenton | spatel https://docs.openstack.org/python-openstackclient/yoga/cli/command-objects/subnet.html | 14:34 |
| jamesdenton | Allocation pool IP addresses for this subnet e.g.: start=192.168.199.2,end=192.168.199.254 (repeat option to add multiple IP addresses) | 14:35 |
| spatel | +1 got it | 14:35 |
| spatel | can i add single ip in allocation pool? | 14:37 |
| spatel | start=1.1.1.1,end=1.1.1.1 ? | 14:37 |
| jamesdenton | good question, i think so? | 14:39 |
| jamesdenton | keep in mind there may be some reserved for DHCPO | 14:39 |
| jamesdenton | DHCP servers, i mean | 14:39 |
| spatel | oh wait.. so you are saying if i add 3 new public ip in subnet and one will get use by DHCP? | 14:42 |
| spatel | that is bad.. | 14:42 |
| spatel | i thought adding extra subnet allocation will just update existing DHCP range and not spawn DHCP agent to reserve ip | 14:42 |
| spatel | I just added 3 extra ip and they all available for floating IP.. phew! so that is good | 14:44 |
| spatel | That means start=1.1.1.1,end=1.1.1.1 should work | 14:44 |
| anskiy | there is reservation for one IP-address from each subnet. And yes, you can have subnets with only one IP-address in allocation. | 15:21 |
| cloudnull | what's good ? | 15:47 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Fix ansible-openstack-collection version https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 15:47 |
| noonedeadpunk | gates are broken. Is it good enough ?:) | 16:13 |
| damiandabrowski | noonedeadpunk: ouh, they seem to be broken since yesterday. Did you have a chance to check it out? | 16:35 |
| damiandabrowski | https://paste.openstack.org/show/bGhRxb9hwsJEZ2140dQT/ | 16:35 |
| damiandabrowski | there's something wrong with wheel | 16:35 |
| damiandabrowski | s | 16:35 |
| noonedeadpunk | yeah I did | 16:35 |
| noonedeadpunk | 858981 maybe will fix it... | 16:35 |
| noonedeadpunk | But full bump on master also fails because of memcahced that requires extra package now | 16:36 |
| damiandabrowski | okok great, fingers crossed then | 16:38 |
| opendevreview | Merged openstack/openstack-ansible master: Fix ansible-openstack-collection version https://review.opendev.org/c/openstack/openstack-ansible/+/858981 | 21:29 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!