ignaziocassano | hello, enabling security groups log, I need to specify a file where security logs are logged, but the openvswitch agent give me errors: 021-07-06 08:31:14.750 7 ERROR neutron FileNotFoundError: [Errno 2] No such file or directory: '/var/log/neutron/security_groups.log | 06:48 |
---|---|---|
*** rpittau|afk is now known as rpittau | 07:30 | |
ignaziocassano | hello, I cannot configure neutron security groups logs with kolla. If I spcify e log file //var/log/neutro/security_groups.log) openvswitch agent restarts because it cannot find the log file | 07:38 |
mgoddard | ignaziocassano: kolla-ansible writes all logs to /var/log/kolla/. So use /var/log/kolla/neutron/security_groups.log | 07:42 |
ignaziocassano | ok | 07:44 |
ignaziocassano | I will try it | 07:44 |
ignaziocassano | thanks | 07:44 |
mgoddard | mchlumsky: we're not too strict about blueprint approvals. I would say the main barriers are 1. general agreement to accept adjutant as a new project, 2. finding two cores willing & able to review the code | 07:46 |
mgoddard | mchlumsky: we're in quite a quiet period right now, with some of the main cores away | 07:47 |
kevko | morning | 07:52 |
kevko | mgoddard: found a bug in masakari debian images -> masakari_monitors is trying to read /dev/shm/something ..but this is owned by another uuid/gid (corosync_pacemaker) | 07:55 |
mgoddard | kevko: is there an easy fix? | 07:56 |
kevko | mgoddard: issue is that RPMs on centos are creating hacluster haclient uuid and gid 189, debian is doing dynamic allocation of uuid ..so, there is no guaranteed that uuid are same .. | 07:56 |
kevko | kevko: so in both debian images (hacluster_pacemaker, masakari_hostmonitor) there is need to install pacemaker-cli-utils ..which depends on pacemaker-common ..which is creating users and gids .. so two different containers ..two different not-controlable uuids/gids .. common /dev/shm/* -> not working :( | 07:58 |
kevko | mgoddard: well, I've fixed it by installing pacemaker-common (which is meta package) to base image ..and it's working | 07:58 |
kevko | mgoddard: so ancestor of both images is same ..so users are same .. | 07:59 |
kevko | mgoddard: another solution could be to use multiple FROM in masakari image ... FROM hacluster_base (where it could be installed pacemaker-common just for user creation) and another FROM openstack_base which masakari_hostmonitors using | 08:00 |
kevko | mgoddard: but i wanted to ask which is preferable ..or if there is another solution | 08:00 |
kevko | mgoddard: i think there could be also possible to add user to kolla/common/config to users block ..but it has to be system user ..so < 1000 (if i am correct) ..and there is no support in kolla | 08:01 |
kevko | mgoddard: so, issue is quite simple ..but every solution has some cons and pros | 08:02 |
kevko | and i don't know how to handle ..which solution is best | 08:02 |
mgoddard | kevko: certainly I'd prefer to avoid installing pacemaker in every image :) | 08:02 |
mgoddard | it sounds similar to the issue we had with /dev/kvm | 08:03 |
kevko | well, i've installed only pacemaker-common ... (but only because i used adduser --system .... from postinst in debian ..) | 08:03 |
kevko | mgoddard: i think it should be too enough just create user ..without package of course | 08:03 |
mgoddard | +1 | 08:04 |
mgoddard | the user can be created in both images with the same ID before the package is installed | 08:04 |
kevko | yes, this is working ..i've tested already ..but it has to be system uuid | 08:04 |
kevko | and it has to be in base .. | 08:05 |
kevko | because hacluster_base is FROM base and masakari_hostmonitors is FROM openstack_base which is FROM base << | 08:05 |
kevko | or just that two images ? | 08:06 |
kevko | i think creation in base image is OK as there are already all users | 08:07 |
kevko | mgoddard: regarding issue with kvm ...do you think you can find review ID ? | 08:08 |
mgoddard | kevko: I61235b51e2e1325b8a9b4f85bf634f663c7ec3cc | 08:09 |
kevko | mgoddard: it's different issue ..but it's similar in terms of permissions | 08:13 |
kevko | mgoddard: will fill a bug, propose change and reno for Michal :) | 08:14 |
mgoddard | thanks | 08:21 |
opendevreview | Mark Goddard proposed openstack/kolla master: ironic-neutron-agent: enable for Debuntu binary https://review.opendev.org/c/openstack/kolla/+/799611 | 08:57 |
Fl1nt | Howdy everyone! | 09:13 |
mgoddard | Hi Fl1nt | 09:15 |
Fl1nt | how are you dear people of the IRC ? | 09:15 |
mgoddard | all good thanks | 09:16 |
mgoddard | it's pretty quiet at the moment | 09:16 |
mgoddard | and you? | 09:16 |
Fl1nt | I'm pretty good thanks ^^ Enjoying summer time :D | 09:16 |
mgoddard | You might be interested in this: https://review.opendev.org/c/openstack/kolla-ansible/+/799229 | 09:17 |
Fl1nt | looking at it ^^ | 09:17 |
Fl1nt | woooh what a long hell of a patchset ^^ | 09:18 |
Fl1nt | interesting, I'll review it. | 09:19 |
mgoddard | great | 09:19 |
Fl1nt | on a side note: is there any documentation about how to request for/perform a cherry pick with kolla/k-a? prometheus exporter k-a template was fixed a year ago but didn't made it to train branch, I'll like to cherry pick it from master, but I didn't find any gerrit/kolla documentation about this process | 09:21 |
Fl1nt | mgoddard, wooh, this podman patchset is a high quality one from what I can see, neat and clean, missing few things but yet that's an impressive work. | 09:24 |
mgoddard | Fl1nt: cherry-pick is in gerrit-ui. Three dots button in top-right, then Cherry pick from drop-down | 09:34 |
mgoddard | glad to hear you like the podman code. I haven't looked at it yet | 09:34 |
opendevreview | Mark Goddard proposed openstack/kolla master: ironic-neutron-agent: enable for Debuntu binary https://review.opendev.org/c/openstack/kolla/+/799611 | 09:38 |
Fl1nt | it still need to finish the python libraries/plugins but yet I see where the contributor want to go. | 09:50 |
Fl1nt | oh ok, thanks! | 09:50 |
opendevreview | Gaël THEROND proposed openstack/kolla-ansible stable/train: Fix prometheus-elasticsearch-exporter flags https://review.opendev.org/c/openstack/kolla-ansible/+/799581 | 09:53 |
opendevreview | Gaël THEROND proposed openstack/kolla master: Fix prometheus-elasticsearch-exporter container. https://review.opendev.org/c/openstack/kolla/+/799620 | 10:04 |
opendevreview | Skylar Tristan Kelty proposed openstack/kayobe master: Add support for apt proxy setting https://review.opendev.org/c/openstack/kayobe/+/799629 | 10:39 |
hrw | morning | 10:59 |
opendevreview | Marcin Juszkiewicz proposed openstack/kolla master: drop leftovers of RHEL support https://review.opendev.org/c/openstack/kolla/+/785569 | 11:04 |
Fl1nt | hrw, o/ | 11:08 |
opendevreview | Marcin Juszkiewicz proposed openstack/kolla master: Debian/Ubuntu: handle APT keys in proper way https://review.opendev.org/c/openstack/kolla/+/784923 | 11:54 |
*** kmasterson is now known as Guest5 | 13:34 | |
*** osmanlicilegi is now known as Guest4 | 13:34 | |
mchlumsky | mgoddard Thanks for the Adjutant answer. I'll do some lobbying during the meeting tomorrow. :) | 13:54 |
parallax | https://review.opendev.org/c/openstack/kolla-ansible/+/782735 - good to go? | 14:22 |
*** Guest5 is now known as kmasterson` | 14:32 | |
*** kmasterson` is now known as kmasterson | 14:32 | |
opendevreview | Merged openstack/kolla-ansible stable/victoria: docs: Add note about internal VIP when HAProxy is disabled https://review.opendev.org/c/openstack/kolla-ansible/+/799154 | 15:09 |
opendevreview | Merged openstack/kolla-ansible stable/wallaby: docs: Add note about internal VIP when HAProxy is disabled https://review.opendev.org/c/openstack/kolla-ansible/+/799153 | 15:12 |
opendevreview | Merged openstack/kolla-ansible stable/ussuri: docs: Add note about internal VIP when HAProxy is disabled https://review.opendev.org/c/openstack/kolla-ansible/+/799155 | 15:12 |
*** hrww is now known as hrw | 16:37 | |
*** rpittau is now known as rpittau|afk | 16:39 | |
*** hrww_ is now known as hrw | 16:59 | |
*** mgoddard- is now known as mgoddard | 17:26 | |
*** hrww is now known as hrw | 18:32 | |
Redcavalier | Hey, I got a quick question : If I were to shutdown my host machine, is kolla putting in place a mechanism to gracefully shutdown the VMs in the container, like libvirt-guests? | 20:17 |
Redcavalier | I'm asking here because I've found some discussions on google that happened around 2018, but nothing in actual documentation indicates it was implemented. | 20:20 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!