cloudnull | is it presently possible to run keystone without apache? | 00:38 |
---|---|---|
cloudnull | as it turns out, yes I can :D https://paste.openstack.org/show/bE1BddHQyL2FOILiL8d5/ | 00:48 |
cloudnull | just rm /etc/apache2/sites-enabled/keystone-httpd.conf update the wsgi file, win! | 00:49 |
cloudnull | looks like its possible to do the same with horizon too 😃 | 01:12 |
* cloudnull livin' that apache free life https://paste.openstack.org/show/bH5mX19o0Ww9QxrNCmgB/ | 01:16 | |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add python filter to eliminate deprecation warnings https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856362 | 01:57 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add the ability to define CSRF_TRUSTED_ORIGINS https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856363 | 02:09 |
*** ysandeep|out is now known as ysandeep | 02:27 | |
opendevreview | Neil Hanlon proposed openstack/openstack-ansible-plugins master: Introduce variables for rocky linux 9 support in gluster https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/852183 | 02:53 |
jamesdenton | cloudnull I see you've been Skyline'ing | 03:05 |
cloudnull | i have :D | 03:39 |
cloudnull | its neat | 03:39 |
cloudnull | have you used i t? | 03:40 |
jamesdenton | ehhh, i feel like i deployed it soon after it was unveiled, and i remember it being a bit of a bear at the time | 03:40 |
*** ysandeep is now known as ysandeep|afk | 03:41 | |
jamesdenton | i think i liked it? :D i've slept since then, i don't remember | 03:41 |
cloudnull | i ran it tonight without all of their docker nonsense. and it worked really well. | 03:41 |
jamesdenton | ahh cool | 03:41 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add uwsgi option to horizon https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856393 | 04:58 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add the ability to define CSRF_TRUSTED_ORIGINS https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856363 | 05:02 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-rabbitmq_server master: Update rabbitmq to 3.10.7-1 https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/855985 | 05:04 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add uwsgi option to horizon https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856393 | 05:05 |
*** ysandeep|afk is now known as ysandeep | 06:18 | |
opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Require usage of service_token_roles https://review.opendev.org/c/openstack/openstack-ansible/+/856437 | 08:10 |
*** frenzy_friday is now known as frenzyfriday|afk | 09:09 | |
*** frenzyfriday|afk is now known as frenzyfriday | 09:54 | |
dokeeffe85 | Hi jamesdenton, me again :) so I did as you suggested and pulled the stable/yoga release after running lxc-container-destroy.yml to remove the containers, I then foollowed through the instructions again as normal and ran the first and second playbooks. The first was fine, the second got past the first error I was seeing but did hit the second error again ([Get list of repo packages]). https://paste.openstack.org/show/bvQMIcLKsuuswTBfYjwv/ | 11:34 |
noonedeadpunk | seems you can reproduce that quite reliably | 11:43 |
noonedeadpunk | though I really haven;t seen such issue anywhere in the sandbox | 11:43 |
noonedeadpunk | andrewbonney: do you use glusterfs for repo or you switched to cephfs or smth like that? | 11:44 |
noonedeadpunk | dokeeffe85: how much controllers you have? | 11:44 |
andrewbonney | We're using glusterfs for repo since Yoga. Didn't hit any issues across 3 deployments during upgrades | 11:44 |
noonedeadpunk | I'm not sure I do have not broken sandbox to check it out | 11:46 |
dokeeffe85 | noonedeadpunk Just one controller | 11:46 |
dokeeffe85 | also not sure if I'm "tagging" you when replying | 11:46 |
noonedeadpunk | you are | 11:47 |
noonedeadpunk | dokeeffe85: eventually, I'm not sure why even gluster would be needed for 1 controller... | 11:52 |
jamesdenton | dokeeffe85 i experienced a similar issue last night after an initial failed run of the playbook, too. https://paste.opendev.org/show/b5fDKzstK6OKydwo1eCY/. I was able to work past it by executing 'umount /var/www/repo' across all of the repo hosts/containers, though | 11:52 |
noonedeadpunk | maybe we have some issue with conditionals when gluster is not being installed but it tries to be mounted | 11:52 |
jamesdenton | if i have some time today i'll see if i can recreate the failure conditions, which may help identify what is actually happening here | 11:53 |
dokeeffe85 | Oh ok jamesdenton, so I'll umount on all the containers and run setup-infrastructure.yml again from the beginning? | 11:54 |
noonedeadpunk | also interesting, that despite error to reload, I do see gluster be still mounted in my aio | 11:54 |
jamesdenton | it should only exist in the repo containers, but yes, give that a go | 11:54 |
noonedeadpunk | and it's only reload that fails | 11:55 |
noonedeadpunk | dokeeffe85: what do you have as `mount` output inside repo-contasiner? | 11:55 |
noonedeadpunk | we should again cloudnull I believe as it's the block that fails :D https://opendev.org/openstack/ansible-role-systemd_mount/src/branch/master/tasks/systemd_mounts.yml#L75-L105 | 11:57 |
noonedeadpunk | *blame | 11:57 |
noonedeadpunk | dokeeffe85: what's your output of `systemctl status var-www-repo.mount` ? | 11:59 |
noonedeadpunk | as from what I see, this should not bring troubles. Thus, `Get list of repo packages` task might get failed due to completely unrelated reasons | 12:00 |
noonedeadpunk | dokeeffe85: eventually your paste trimmed exactly the way, that don't allow to understand what's going on | 12:01 |
dokeeffe85 | noonedeadpunk this is the output of 'mount' (sorry I pasted in the wrong content first) https://paste.openstack.org/show/bztsfi1vbcKWjVFXagco/ | 12:03 |
noonedeadpunk | ok, so mount is present at the end | 12:03 |
dokeeffe85 | and the other output https://paste.openstack.org/show/bGuyZsNk9EQYhdEWpUWL/ | 12:04 |
dokeeffe85 | Need to pop away for a bit but will be back, thanks so far to you both | 12:04 |
noonedeadpunk | dokeeffe85: are you sure that keepalived is configured properly and internal_vip is UP and can be reached? | 12:04 |
noonedeadpunk | As I believe, that 10.37.110.100 can't be accessed from utility container | 12:05 |
noonedeadpunk | and that's the IP that should be on br-mgmt on controller itself | 12:06 |
NeilHanlon | well, turns out rocky was a lot easier with those centos-release* packages :) Just need this one patch https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/852183 | 13:21 |
NeilHanlon | aio_lxc succeeds on rocky 9 with that | 13:22 |
spatel | NeilHanlon Sweet!!! | 13:51 |
dokeeffe85 | sorry noonedeadpunk & jamesdenton I got pulled into two meetings and have to join a call now. I will be back tomorrow and also get that info together for you noonedeadpunk. Thanks both | 14:27 |
*** ysandeep is now known as ysandeep|out | 14:54 | |
*** dviroel is now known as dviroel|lunch | 14:57 | |
*** dviroel|lunch is now known as dviroel | 15:46 | |
cloudnull | noonedeadpunk hahaha - https://opendev.org/openstack/ansible-role-systemd_mount/blame/branch/master/tasks/systemd_mounts.yml#L75 - 5 years , it was a good run | 16:28 |
cloudnull | :D | 16:28 |
noonedeadpunk | It actually still works, so good work ;) | 16:29 |
cloudnull | hahahaha - its more luck than anything else | 16:33 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add uwsgi option to horizon https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856393 | 16:42 |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_horizon master: Add python filter to eliminate deprecation warnings https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/856362 | 16:53 |
spatel | is cloudnull back ? :D | 19:09 |
cloudnull | no. | 19:18 |
cloudnull | but I'm using OSA atm :D | 19:18 |
supamatt | OSA at the new gig? or personally? ;) | 19:22 |
cloudnull | both ;) | 19:26 |
cloudnull | new gig is a ubuntu shop, so osa fits nicely :D | 19:27 |
supamatt | indeed it does.. kolla and ubuntu tho is pretty nice | 19:27 |
supamatt | that's what my shop uses | 19:27 |
cloudnull | nice | 19:28 |
cloudnull | we're running osa almost all baremetal on debian 11 -- so far so great. | 19:28 |
* cloudnull trying to eliminate apache :D | 19:29 | |
opendevreview | Kevin Carter proposed openstack/openstack-ansible-os_keystone master: Add the option to deploy keystone without apache https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/856563 | 19:35 |
cloudnull | with the horizon change and keystone changes https://review.opendev.org/q/topic:use_uwsgi+status:open -- no more apache :D | 19:42 |
*** dviroel is now known as dviroel|out | 21:57 | |
opendevreview | Damian DÄ…browski proposed openstack/openstack-ansible master: Improve Ceph production example https://review.opendev.org/c/openstack/openstack-ansible/+/856566 | 22:01 |
opendevreview | Damian DÄ…browski proposed openstack/openstack-ansible master: Improve Ceph production example https://review.opendev.org/c/openstack/openstack-ansible/+/856566 | 22:04 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!