*** vnogin has joined #openstack-ansible | 00:08 | |
*** markvoelker has joined #openstack-ansible | 00:11 | |
openstackgerrit | Michael Johnson proposed openstack/openstack-ansible stable/queens: Update Octavia SHA to the 2.0.4 release https://review.openstack.org/635376 | 00:21 |
---|---|---|
*** dmsimard has quit IRC | 00:25 | |
*** sdake has quit IRC | 00:37 | |
*** hwoarang has quit IRC | 00:42 | |
*** tosky has quit IRC | 00:43 | |
*** markvoelker has quit IRC | 00:44 | |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: venv: use inventory_hostname instead of ansible_hostname https://review.openstack.org/635187 | 00:44 |
*** hwoarang has joined #openstack-ansible | 00:44 | |
*** dmsimard has joined #openstack-ansible | 00:57 | |
*** hwoarang has quit IRC | 00:58 | |
*** hwoarang has joined #openstack-ansible | 01:01 | |
*** vnogin has quit IRC | 01:07 | |
*** sdake has joined #openstack-ansible | 01:12 | |
*** TxGirlGeek has quit IRC | 01:13 | |
*** gyee has quit IRC | 01:22 | |
*** hwoarang has quit IRC | 01:29 | |
*** hwoarang has joined #openstack-ansible | 01:31 | |
*** sdake has quit IRC | 01:48 | |
*** sdake has joined #openstack-ansible | 01:50 | |
*** hwoarang has quit IRC | 02:05 | |
*** hwoarang has joined #openstack-ansible | 02:06 | |
*** sdake has quit IRC | 02:19 | |
*** cmart has quit IRC | 02:22 | |
*** fnpanic has quit IRC | 03:24 | |
*** hwoarang has quit IRC | 03:31 | |
*** hwoarang has joined #openstack-ansible | 03:32 | |
*** partlycloudy has joined #openstack-ansible | 03:34 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/ansible-role-systemd_service master: Build out the PrivateNetwork function for services https://review.openstack.org/635396 | 03:34 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/ansible-role-systemd_service master: Build out the PrivateNetwork function for services https://review.openstack.org/635396 | 03:37 |
openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible-os_octavia master: Update octavia configuration documentation https://review.openstack.org/635402 | 03:41 |
openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible-os_octavia master: Update octavia configuration documentation https://review.openstack.org/635402 | 03:42 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/ansible-role-systemd_service master: Build out the PrivateNetwork function for services https://review.openstack.org/635396 | 03:42 |
*** sdake has joined #openstack-ansible | 03:42 | |
*** sdake has quit IRC | 03:44 | |
*** gkadam has joined #openstack-ansible | 04:06 | |
*** sdake has joined #openstack-ansible | 04:10 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add logstash ingestion for collectd https://review.openstack.org/635418 | 04:22 |
*** udesale has joined #openstack-ansible | 04:25 | |
*** hwoarang has quit IRC | 05:15 | |
*** hwoarang has joined #openstack-ansible | 05:17 | |
*** sdake has quit IRC | 05:18 | |
*** chandankumar has joined #openstack-ansible | 05:45 | |
*** chandankumar is now known as chkumar|ruck | 05:46 | |
*** hwoarang has quit IRC | 06:04 | |
*** hwoarang has joined #openstack-ansible | 06:06 | |
*** markvoelker has joined #openstack-ansible | 06:07 | |
*** vnogin has joined #openstack-ansible | 06:08 | |
*** markvoelker has quit IRC | 06:11 | |
*** vnogin has quit IRC | 06:13 | |
*** sohny has joined #openstack-ansible | 06:30 | |
*** sohny has quit IRC | 06:41 | |
*** chkumar|ruck has quit IRC | 06:48 | |
*** chandankumar has joined #openstack-ansible | 06:48 | |
*** chandankumar is now known as chkumar|ruck | 06:49 | |
*** kopecmartin|off is now known as kopecmartin | 07:01 | |
*** pcaruana has joined #openstack-ansible | 07:02 | |
openstackgerrit | Martin Kopec proposed openstack/openstack-ansible-os_tempest master: Improve overview subpage https://review.openstack.org/633934 | 07:04 |
*** markvoelker has joined #openstack-ansible | 07:08 | |
*** hwoarang has quit IRC | 07:15 | |
*** slaweq has joined #openstack-ansible | 07:16 | |
*** hwoarang has joined #openstack-ansible | 07:18 | |
*** slaweq has quit IRC | 07:21 | |
*** hwoarang has quit IRC | 07:24 | |
*** hwoarang has joined #openstack-ansible | 07:25 | |
mkuf | Hello, I'm having an issues deploying policy.json for heat under 17.1.6. I'm adding heat_policy_overrides in user_variables but getting an error while running os-heat-install --tags heat-config see: http://paste.openstack.org/show/CZ5lmJXU5UgJJ7RExYXM/ | 07:27 |
mkuf | Deploying policy.json for nova in the same style works just fine. | 07:28 |
*** mnasiadka has joined #openstack-ansible | 07:39 | |
*** mnasiadka has left #openstack-ansible | 07:40 | |
*** rgogunskiy has joined #openstack-ansible | 07:41 | |
*** markvoelker has quit IRC | 07:41 | |
*** DanyC has joined #openstack-ansible | 07:41 | |
*** DanyC has quit IRC | 07:43 | |
*** DanyC has joined #openstack-ansible | 07:50 | |
*** DanyC has quit IRC | 07:53 | |
*** DanyC has joined #openstack-ansible | 07:53 | |
jrosser | mkuf: if you are able to dig a bit it would be very helpful, particularly to see if this has changed the behaviour for you https://github.com/openstack/openstack-ansible-os_heat/commit/61f2c903cfc18e7d37d202d11eec9013ffd3c963 | 07:56 |
*** hwoarang has quit IRC | 07:58 | |
*** vnogin has joined #openstack-ansible | 07:59 | |
*** hwoarang has joined #openstack-ansible | 07:59 | |
*** Emine has joined #openstack-ansible | 08:02 | |
*** DanyC has quit IRC | 08:12 | |
*** miloa has joined #openstack-ansible | 08:18 | |
*** slaweq has joined #openstack-ansible | 08:18 | |
*** shyamb has joined #openstack-ansible | 08:30 | |
*** shardy has joined #openstack-ansible | 08:32 | |
*** aedc has joined #openstack-ansible | 08:35 | |
*** slaweq has quit IRC | 08:36 | |
mkuf | jrosser: that fixed it, my overrides now get deployed for heat. it removed the default entries but i guess that's ok? | 08:37 |
*** markvoelker has joined #openstack-ansible | 08:38 | |
*** rjgibson has quit IRC | 08:44 | |
*** rjgibson has joined #openstack-ansible | 08:45 | |
*** tosky has joined #openstack-ansible | 08:46 | |
*** rgogunskiy has quit IRC | 08:47 | |
*** rgogunskiy has joined #openstack-ansible | 08:49 | |
*** chkumar|ruck has quit IRC | 08:53 | |
*** shyamb has quit IRC | 08:55 | |
*** markvoelker has quit IRC | 09:12 | |
CeeMac | morning all | 09:16 |
*** fnpanic has joined #openstack-ansible | 09:18 | |
fnpanic | hi | 09:19 |
fnpanic | good morning | 09:19 |
*** Dantalion has joined #openstack-ansible | 09:25 | |
*** DanyC has joined #openstack-ansible | 09:27 | |
*** DanyC has quit IRC | 09:31 | |
*** DanyC has joined #openstack-ansible | 09:33 | |
*** slaweq has joined #openstack-ansible | 09:33 | |
*** chandankumar has joined #openstack-ansible | 09:34 | |
*** chandankumar is now known as chkumar|ruck | 09:35 | |
*** shyamb has joined #openstack-ansible | 09:36 | |
*** jbadiapa has quit IRC | 09:47 | |
*** jbadiapa has joined #openstack-ansible | 09:48 | |
jrosser | mkuf: as i understand it the default policies are now carried in the code rather than a config file. you should only need the config file if you have to specify non-default policy | 09:48 |
jrosser | mkuf: there are a bunch of recent patches to remove old policy.json files as they are potentially mismatching the defaults in the code and creating breakages | 09:49 |
*** priteau has joined #openstack-ansible | 09:49 | |
*** chandankumar has joined #openstack-ansible | 09:50 | |
jrosser | chandankumar: \o/ https://review.openstack.org/#/c/633883/ | 09:50 |
chandankumar | jrosser: yes, thanks :-) | 09:51 |
*** chkumar|ruck has quit IRC | 09:51 | |
*** oanson has joined #openstack-ansible | 09:51 | |
*** chandankumar is now known as chkumar|ruck | 09:51 | |
mkuf | jrosser: great, thanks :) | 09:59 |
*** aedc has quit IRC | 09:59 | |
*** slaweq has quit IRC | 10:00 | |
*** shyamb has quit IRC | 10:05 | |
*** markvoelker has joined #openstack-ansible | 10:09 | |
*** shyamb has joined #openstack-ansible | 10:09 | |
CeeMac | quick question, when changes are required to a configuration file. Would you say it was better practice to update the user_variables file and run the appropriate playbook rather than manually editing the config file? | 10:10 |
CeeMac | also, i noticed on some of the documentation there are separate install and config tags, presumably the *-config tag cuts out a lot of the tasks in the playbook to just adjust the config? | 10:12 |
CeeMac | maybe not so quick a question >_> | 10:12 |
jrosser | never edit the config file | 10:12 |
jrosser | you should be able to add everything you need, using config overrides | 10:14 |
CeeMac | ok, thats a fairly straightforward mantra to adapt :) | 10:14 |
jrosser | and if you edit by hand it will all get undone when you do an upgrade | 10:14 |
CeeMac | ah, yeah, also a good point | 10:15 |
jrosser | CeeMac: this is handy documentation https://docs.openstack.org/openstack-ansible/rocky/reference/configuration/using-overrides.html | 10:18 |
*** aedc has joined #openstack-ansible | 10:18 | |
jrosser | openstack-ansible makes extensive use of config_template which means that there does not need to be a variable to override for every possible var in every config file, that would be totally unmanageable | 10:19 |
*** sm806 has quit IRC | 10:19 | |
*** sm806 has joined #openstack-ansible | 10:20 | |
CeeMac | thanks jrosser, i had that doc in my chrome history somewhere but couldn't find it. I'll bookmark it | 10:21 |
CeeMac | jrosser, wrt to tags e.g."The nova-config tag can be used to manage configuration" | 10:23 |
CeeMac | is that something i add at the end of the playbook line, similar to limiting hosts/groups? | 10:23 |
jrosser | --tags nova-config | 10:24 |
jrosser | and you can use that in conjunction with --limit as well | 10:24 |
CeeMac | cool | 10:24 |
CeeMac | presumably that makes config changes more efficient than running the whole playbook? | 10:25 |
jrosser | that is the idea, yes | 10:25 |
CeeMac | cool, thanks. Sorry for the n00b questions :) | 10:26 |
CeeMac | have been so focussed on actually getting a deployment to install over the last few months I haven't had much time to look at and absorb the processes for operational adjustment / management tasks | 10:27 |
openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer stable/rocky: Fixes gnocchi resource-type list being empty after setup https://review.openstack.org/635189 | 10:30 |
*** markvoelker has quit IRC | 10:41 | |
*** shyamb has quit IRC | 10:57 | |
*** fdegir has quit IRC | 10:57 | |
*** fdegir has joined #openstack-ansible | 10:58 | |
*** slaweq has joined #openstack-ansible | 11:13 | |
*** chkumar|ruck has quit IRC | 11:29 | |
*** chandankumar has joined #openstack-ansible | 11:30 | |
*** chandankumar is now known as chkumar|ruck | 11:30 | |
*** udesale has quit IRC | 11:33 | |
*** udesale has joined #openstack-ansible | 11:34 | |
*** markvoelker has joined #openstack-ansible | 11:38 | |
*** slaweq has quit IRC | 11:54 | |
*** electrofelix has joined #openstack-ansible | 11:58 | |
*** electrofelix has quit IRC | 12:04 | |
*** vnogin has quit IRC | 12:05 | |
*** electrofelix has joined #openstack-ansible | 12:05 | |
*** markvoelker has quit IRC | 12:11 | |
*** priteau has quit IRC | 12:12 | |
CeeMac | hmm | 12:12 |
CeeMac | i've got cinder_cinder_conf_overrides to take | 12:13 |
CeeMac | nova_nova_conf_overrides not so much | 12:13 |
CeeMac | playbook runs, no changes | 12:13 |
CeeMac | with the --tags nova-config | 12:13 |
CeeMac | do "normal" configuration overrides and config_templates conflict? | 12:15 |
jamesdenton | which playbook are you running? | 12:15 |
CeeMac | os-nova-install.yml | 12:15 |
CeeMac | and hi :) | 12:15 |
jamesdenton | good morning | 12:16 |
jamesdenton | you'd likely need to run the cinder playbook to lay down cinder_cinder_conf_overrides | 12:16 |
CeeMac | yep, got that bit working | 12:16 |
jamesdenton | oh i see, you're trying to lay down nova_nova_conf_overrides? | 12:16 |
CeeMac | indeed :) | 12:17 |
jamesdenton | gotcha | 12:17 |
CeeMac | i already have an override for nova_console_type in user_variables, not sure if the two methods are mutually exclusive? | 12:17 |
jrosser | use a config override when there isnt a role default for the thing you want to change | 12:19 |
jrosser | so for nova_console_type there is a default here https://github.com/openstack/openstack-ansible-os_nova/blob/master/defaults/main.yml#L321 | 12:19 |
jrosser | just put a var overriding that in your user_variables | 12:19 |
jrosser | if you find you need to change something that isn't expressed through role defaults, thats when to use a config override | 12:20 |
CeeMac | right | 12:21 |
CeeMac | and you can do both right? | 12:21 |
CeeMac | one doesnt preclude the other | 12:21 |
jrosser | and if you need to do it for specific hosts or groups, there is group/host vars in your /etc/openstack_deploy folder | 12:22 |
CeeMac | this is what I have in my user_variables.yml related to nova right now: http://paste.openstack.org/show/744665/ | 12:22 |
CeeMac | i used the same format for the cinder conf overrides and that ran through fine | 12:23 |
jrosser | if it is not applying that make sure that the task templating out the config file is not skipped | 12:23 |
jrosser | there may be issues in the code meaning it's not quite doing the right thing with --tags | 12:25 |
CeeMac | http://paste.openstack.org/show/744666/ | 12:25 |
CeeMac | ^ thats the output from the playbook run | 12:25 |
CeeMac | not sure which task relates to templating the config | 12:26 |
jamesdenton | yeah it's in nova_post_install, which doesn't look like it's being invoked | 12:26 |
CeeMac | presumably because the install is being skipped? | 12:27 |
CeeMac | would there be any issues running without the tag? | 12:27 |
CeeMac | or is this something that should work? | 12:28 |
fnpanic | hi eberyone | 12:28 |
CeeMac | hi | 12:28 |
fnpanic | i am now at the stage deploying stable/rocky with proxy... | 12:28 |
fnpanic | looks like it fails validating the ceph-client keys in the glance container | 12:29 |
fnpanic | do i need to disable this? | 12:29 |
fnpanic | or is it supposed to work? | 12:29 |
fnpanic | i am getting the error that the keyserver is not reachable | 12:30 |
jamesdenton | CeeMac You might temporarily be better off running it w/o the tag | 12:31 |
fnpanic | http://paste.openstack.org/show/744667/ | 12:31 |
CeeMac | jamesdenton, thanks, I'll do that | 12:31 |
fnpanic | so everything worked flawless till the setup-openstack playbook | 12:32 |
fnpanic | ceph_package_download_validate_certs: false | 12:32 |
fnpanic | can this solve the problem? | 12:32 |
fnpanic | or how is it called? | 12:33 |
jrosser | fnpanic: apt-key ansible module sadly does not respect proxy settings at all :( | 12:33 |
jrosser | remember what we had to do for galera and a few other things? https://review.openstack.org/#/c/625291/ | 12:36 |
CeeMac | jamesdenton, jrosser the config overrides are being picked up when I run the playbook without the tag, so thats good anyway | 12:38 |
jamesdenton | Good. There's a good chunk of tasks/includes not being executed with that tag. Will have to look at that | 12:38 |
jamesdenton | if you can submit a bug that could be helpful | 12:39 |
jamesdenton | (if there isn't one already there) | 12:39 |
fnpanic | jrosser: yeah, was crawling through the squid logs and see no connect | 12:40 |
fnpanic | can i ignore it with a variable? | 12:40 |
fnpanic | i remmeber the patches :-( | 12:40 |
*** priteau has joined #openstack-ansible | 12:43 | |
jrosser | fnpanic: it looks like you are deploying ceph? if so you'll need the binaries and therefore the repo keys | 12:43 |
CeeMac | i'll do a quick search through the bugs and report one if there isn't | 12:44 |
jrosser | fnpanic: you can make a patch if you like :) | 12:44 |
jrosser | it will be following exactly the same pattern as the galera_client one | 12:44 |
fnpanic | i am looking at the patches and see if i can make it work in some way | 12:46 |
*** sdake has joined #openstack-ansible | 12:47 | |
*** priteau has quit IRC | 12:47 | |
fnpanic | so setting ceph_package_download_validate_certs: false will not work? | 12:47 |
*** sdake has quit IRC | 12:51 | |
*** sdake has joined #openstack-ansible | 12:53 | |
openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: venv: use inventory_hostname instead of ansible_hostname https://review.openstack.org/635172 | 12:56 |
jamesdenton | Does anyone know if the files for the diagrams here still exists? https://docs.openstack.org/openstack-ansible/latest/reference/architecture/container-networking.html | 13:01 |
*** gkadam has quit IRC | 13:02 | |
*** priteau has joined #openstack-ansible | 13:06 | |
*** markvoelker has joined #openstack-ansible | 13:08 | |
*** stuartgr has quit IRC | 13:09 | |
*** stuartgr has joined #openstack-ansible | 13:12 | |
*** slaweq has joined #openstack-ansible | 13:17 | |
vollman | Anyone know why the os_nova basicserverops tempest test is not working on centos-7? | 13:19 |
chkumar|ruck | mnaser: cloudnull jrosser: odyssey4me https://review.openstack.org/#/c/634377/ https://review.openstack.org/633883 https://review.openstack.org/630695 https://review.openstack.org/#/c/633934/ are good to go, feel free to take a look thanks :-) | 13:26 |
mnaser | vollman: i think that was fixed with some work that jrosser did a while back? | 13:27 |
mnaser | chkumar|ruck: does tripleo ci actually use the cloned version from git/zuul? | 13:28 |
odyssey4me | hello everyone - back from almost a week away | 13:28 |
mnaser | odyssey4me: welcome back :) | 13:28 |
vollman | mnaser: Its still broken on master. I've been testing it since yesterday. Seems something changed that is causing an ssh timeout during the tempest test | 13:29 |
mnaser | vollman: sigh, okay, i will try to get to this today. i'm looking over chkumar|ruck's reviews | 13:29 |
odyssey4me | cloudnull jrosser I have discussed the plan for switching to the standalone repo for config_template from the plugins repo. I will push up a bunch of patches today, but -w them until they're all passing tests - they will all have to be in a working state otherwise we will break the world. | 13:29 |
chkumar|ruck | mnaser: yes, https://github.com/openstack/tripleo-quickstart/blob/master/quickstart-extras-requirements.txt | 13:30 |
odyssey4me | Oh, I meant to say that I've discussed it with evrardjp :) | 13:30 |
evrardjp | odyssey4me: you could have discussed it alone | 13:30 |
evrardjp | but that would be awkward | 13:30 |
vollman | mnaser: I'm going to continue to look into it also. just curious if someone knew what changed. I can manually create VMs and ssh to them so it is specific to centos and tempest | 13:30 |
chkumar|ruck | mnaser: once clonned we are setting the path here https://github.com/openstack/tripleo-quickstart/blob/master/ansible.cfg | 13:31 |
mnaser | chkumar|ruck: seems good then :D | 13:31 |
odyssey4me | mnaser nice to see you have some work in the wings to make use of consul and such - we should discuss that at the next PTG. It probably makes sense for us to switch from only using ansible to use other tools where they're best suited - for example, use ansible to prep hosts and push out the software, but use consul for config file changes. | 13:31 |
odyssey4me | However, we *may* wish to try and make that opt-in... although the tech debt would be considerable I think. | 13:31 |
mnaser | odyssey4me: yeah, right now it's nothing but an aio that i'm hacking on and seeing where we might fall short or not | 13:31 |
odyssey4me | evrardjp :p | 13:32 |
mnaser | more poc-ing rather than writing code around it | 13:32 |
mnaser | but yes i agree, i want to bring it up at the tpg | 13:32 |
fnpanic | jrosser: i did a patch for my local install following the pattern of this galera patch | 13:32 |
mnaser | ptg.. which is already in 3 months? time goes by.. | 13:32 |
odyssey4me | mnaser yep, I think that many would be interested in that PoC... and I think that cloudnull may be able to help get it done if we decide to go that route | 13:32 |
evrardjp | oh btw did you see that ansible finally embraces dw 's mitogen work | 13:33 |
mnaser | odyssey4me: a lot of interesting things i learned from it like how you can bootstrap a rabbitmq cluster using consul.. a lot of our code would be less complicated that way :P | 13:33 |
evrardjp | Forgot to say that here yesterday | 13:33 |
odyssey4me | spanice work on the collectd things :) looking forward to seeing a blog post or something :) | 13:33 |
mnaser | evrardjp: link? | 13:33 |
mnaser | vollman: when you say "I can manually create VMs and ssh" i assume not on centos? | 13:33 |
evrardjp | core meetings logs? Let me find htat | 13:33 |
odyssey4me | mnaser yes, absolutely - we discussed etcd as a means to achieve this in the past, but the new dependency was not something anyone was thrilled about | 13:34 |
odyssey4me | *spatel was the collectd comment targeted at... my hands aren't used to typing no more :p | 13:35 |
odyssey4me | evrardjp yeah, saw that - nice to see that progressing into real change in ansible | 13:35 |
mnaser | i have 2 things which are a super PITA but i'd appreciate reviews. i finished up os_placement support (https://review.openstack.org/#/c/618820/55) -- also with a functional test to make sure it works (placement doesnt use tempest). and os_mistral role (https://review.openstack.org/#/c/634991/) passing tempest tests on all source and distros | 13:35 |
mnaser | odyssey4me: yeah i think if we sell the value of it, it will make a lot more sense | 13:36 |
evrardjp | mnaser: https://meetbot.fedoraproject.org/ansible-meeting/2019-02-05/ansible_core_irc_meeting.2019-02-05-19.00.html | 13:36 |
vollman | mnaser: no on centos. to test I popped a centos-7 vm and ran run_tests.sh which fails with the ssh timeout. but then I can manually server create and ssh into that server, so all the plumbing looks correct | 13:36 |
mnaser | evrardjp: that's sweet | 13:37 |
odyssey4me | mnaser https://twitter.com/svg/status/1093064366021177344 | 13:37 |
mnaser | that's awesome | 13:37 |
CeeMac | jamesdenton, jrosser as discussed https://bugs.launchpad.net/openstack-ansible/+bug/1815043 | 13:38 |
openstack | Launchpad bug 1815043 in openstack-ansible "running os-nova-install.yml --tags nova-config does note generate config" [Undecided,New] | 13:38 |
jamesdenton | thx | 13:38 |
CeeMac | hi odyssey4me wb :) | 13:38 |
mnaser | vollman: https://review.openstack.org/#/c/633474/ ok i see the failure here | 13:38 |
CeeMac | out of interest, are any of you guys going to be at the openstack summit in denver? | 13:39 |
* mnaser will be | 13:40 | |
*** markvoelker has quit IRC | 13:41 | |
mnaser | Lease of 192.168.74.5 obtained, lease time 86400 | 13:42 |
mnaser | ok so for teh os_nova jobs, the vxlan net works fine | 13:42 |
mnaser | which means its a neutron/network thing on the flat/floating ip side | 13:42 |
mnaser | http://logs.openstack.org/74/633474/4/check/openstack-ansible-functional-centos-7/6bdccc6/logs/host/neutron/ | 13:44 |
mnaser | looks like only neutron-lxb service went up? | 13:44 |
vollman | rest of the services are running in the openstack1 container | 13:46 |
mnaser | that's not right is it | 13:46 |
*** ansmith has joined #openstack-ansible | 13:46 | |
odyssey4me | mnaser isn't that the issue jrosser and cloudnull were working on? | 13:46 |
mnaser | those agents should be running on the host | 13:46 |
odyssey4me | it only affects centos because *old* | 13:46 |
mnaser | odyssey4me: i think so? but i think they came to a conclusion and merged something to os_tempest to fix it | 13:46 |
mnaser | :(((( | 13:47 |
*** sdake has quit IRC | 13:47 | |
mnaser | im spinning up tests on a centos-7 vm to have a replicated environment | 13:47 |
*** KeithMnemonic has joined #openstack-ansible | 13:48 | |
odyssey4me | mnaser yep, but I think it would also need some inventory changes in os_nova and anything else that uses the same tempest test | 13:48 |
mnaser | odyssey4me: in integrated we run all agents on host afaik right? | 13:49 |
odyssey4me | mnaser yep | 13:49 |
mnaser | today on yet another reason why we should be gating with integrated D: | 13:49 |
odyssey4me | I think that https://review.openstack.org/634057 was meant to solve it, but didn't. | 13:49 |
mnaser | ill see what my vm looks like when im done | 13:49 |
odyssey4me | mnaser yep, tomorrow morning I'll put together an etherpad to plan that one out - we're pretty close, but need a few more tweaks to make it universal | 13:50 |
mnaser | odyssey4me: okay, i got help to fix all the ansible_hostname => inventory_hostname references | 13:51 |
mnaser | so that's all cleaned up | 13:51 |
odyssey4me | mnaser otherwise, this has the inventory changes needed: https://review.openstack.org/633732 | 13:51 |
odyssey4me | mnaser great - is that ll done and merged already? | 13:51 |
mnaser | odyssey4me: https://review.openstack.org/#/q/owner:%22Heba+Naser+%253Cheba%2540vexxhost.com%253E%22+venv mostly | 13:52 |
odyssey4me | mnaser haha, thanks again to heber for helping out | 13:52 |
mnaser | :) | 13:52 |
odyssey4me | *heba | 13:53 |
mnaser | typing is hard after vacationing | 13:53 |
mnaser | arxcruz, chkumar|ruck: i think we might have broken aodh's distro tests with our recent changes as this is broken -- https://review.openstack.org/#/c/635165/ | 13:53 |
odyssey4me | ok, let me get cracking with that config_template thing - it should be relatively straightforward... I'll start with just the AIO services and get those passing tests... then will do the rest | 13:54 |
odyssey4me | mnaser aodh will be fixed with https://review.openstack.org/632125 | 13:54 |
*** vnogin has joined #openstack-ansible | 13:55 | |
*** vnogin has quit IRC | 13:59 | |
chkumar|ruck | mnaser: I will take a look | 14:00 |
odyssey4me | mnaser chkumar|ruck to prove it works, I tested it with https://review.openstack.org/621183 | 14:01 |
odyssey4me | which incidentally needs another review ;) | 14:01 |
odyssey4me | oh mnaser I see that jrosser commented some diagnostic info about the centos issue for os_nova in https://review.openstack.org/588951 | 14:03 |
jrosser | odyssey4me: yes - though i've not managed to reproduce that again, would be great if someone else can try | 14:05 |
jrosser | it was very very obviously not natting the traffic | 14:05 |
*** chhagarw has joined #openstack-ansible | 14:06 | |
mnaser | i'm working on it | 14:07 |
* mnaser is in tc meeting | 14:08 | |
chkumar|ruck | odyssey4me: http://logs.openstack.org/94/630694/9/check/openstack-ansible-deploy-aio_metal_heat-centos-7/a655e4c/job-output.txt.gz#_2019-02-05_11_28_53_271932 | 14:08 |
chkumar|ruck | it is failing on aio jobs | 14:08 |
chkumar|ruck | The repository located at mirror.ord.rax.openstack.org is not a trusted or secure host and is being ignored. If this repository is available via HTTPS we recommend you use HTTPS instead, otherwise you may silence this warning and allow it anyway with '--trusted-host mirror.ord.rax.openstack.org'.\n The repository located at mirror.ord.rax.openstack.org is not a trusted or secure host and is being | 14:08 |
chkumar|ruck | ignored. I | 14:08 |
odyssey4me | chkumar|ruck easier to read: http://logs.openstack.org/94/630694/9/check/openstack-ansible-deploy-aio_metal_heat-centos-7/a655e4c/logs/ara-report/result/42de01d2-962a-401b-9e6b-31ecc5c60697/ | 14:09 |
chkumar|ruck | odyssey4me: yes, :-) | 14:09 |
odyssey4me | 'Could not find a version that satisfies the requirement heat (from versions: )' | 14:09 |
odyssey4me | chkumar|ruck hmm, that's odd - lemme look at that more closely | 14:11 |
chkumar|ruck | odyssey4me: sure | 14:11 |
* chkumar|ruck headed home | 14:11 | |
odyssey4me | weird - for some reason the constraints has 'openstack_heat==11.1.0.dev138' instead of 'heat...' | 14:12 |
odyssey4me | the egg name in http://logs.openstack.org/94/630694/9/check/openstack-ansible-deploy-aio_metal_heat-centos-7/a655e4c/logs/repo/requirements_constraints.txt.txt.gz is 'heat' | 14:12 |
odyssey4me | I wonder if something changed upstream | 14:12 |
chkumar|ruck | https://github.com/openstack/heat/commit/b9af0cf53f42f04f78dc4be935a83efd7a28998c -> but in entry points | 14:14 |
odyssey4me | chkumar|ruck I don't see how that relates? | 14:16 |
CeeMac | jamesdenton, see when an instance has a floating IP on an external/public network. does a/the router proxy-arp for the floating IPs, or does the network node handle arp responses? | 14:16 |
chkumar|ruck | odyssey4me: I have just passed this change, it is not related | 14:16 |
odyssey4me | https://github.com/openstack/heat/commit/8ae5f8657b2bdf841bdf114b49839700c2ee924f | 14:16 |
jamesdenton | CeeMac In a standard linuxbridge-based setup, there's really no proxy-arp involved. The floating IP is configured on the qg interface inside the router namespace. Normal ARP behavior applies | 14:17 |
odyssey4me | chkumar|ruck ok, so we need to change the egg name in os_heat, and in th eintegrated repo to match the new name | 14:18 |
chkumar|ruck | od doing that | 14:19 |
jamesdenton | CeeMac the qg->(veth)->tap->brq-bridge->bond1.X->bond1->eth0/2->switch | 14:19 |
odyssey4me | chkumar|ruck fantastic, thanks | 14:19 |
CeeMac | jamesdenton, ok thanks | 14:19 |
jamesdenton | or some variation | 14:19 |
CeeMac | just trying to work through external access workflow so i get my nat/routing/policy right | 14:19 |
CeeMac | i have a /22 for the external network, but want to allocate a /28 with a static prefix NAT on our edge firewall | 14:20 |
CeeMac | trying to work out if i need a static route, or if normal arping will resolve the packet path | 14:20 |
CeeMac | *sub /28 of the /22 | 14:20 |
jamesdenton | Hmm, well the expectation is that if you configure a /22 external provider network, then the edge device would have an SVI to match. Whether you implement NAT on the edge device is outside the scope of Neutron - it will be NAT'ing floating IPs to fixed IPs. | 14:21 |
jamesdenton | You can implement a /28 floating IP network with a /28 on the edge device. And if you outgrow that, stack an additional subnet on the existing network | 14:21 |
jamesdenton | not all device support that, though (looking at you, Cisco ASA) | 14:22 |
*** rgogunskiy has quit IRC | 14:22 | |
openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_heat master: Fixed the egg name of heat to openstack_heat https://review.openstack.org/635518 | 14:22 |
CeeMac | yeah. i've kind of purloined a network i'm using for something else for this RC/POC and wanted to restrict its impact to the top end of the range to avoid conflict | 14:23 |
CeeMac | "You can implement a /28 floating IP network with a /28 on the edge device" as in set up a new provider network for floating IPs seperate to the existing one? | 14:24 |
evrardjp | mnaser: osa has a plan for py3, you can ask odyssey4me :p | 14:24 |
mnaser | evrardjp: i missed the fosdem ptg | 14:26 |
mnaser | :( | 14:26 |
jamesdenton | Well, sort of. You have a provider interface (e.g. physnet1) that can potentially service multiple external provider networks using different VLAN IDs (if it's a trunk, that is). So for each network, you would have a corresponding SVI or L3 interface on the edge device. BUT - let's say you have a single network using a /28. You can stack an additional subnet on that network using its own CIDR, but it would be on the | 14:27 |
jamesdenton | same VLAN. On a router, this might appear as a 'secondary' IP on the existing L3 interface. | 14:27 |
jamesdenton | So if you outgrow the existing subnet, but want to continue using the same provider network/vlan, you simply add a new subnet | 14:27 |
CeeMac | ah, ok, yeah i'm doing that already in this network | 14:28 |
CeeMac | for the "other" use case | 14:28 |
evrardjp | mnaser: lol | 14:29 |
CeeMac | so i would just need to update the "public" network in neutron to the /28 and changed its GW to the additional IP on the edge SVI | 14:29 |
*** irclogbot_0 has joined #openstack-ansible | 14:29 | |
CeeMac | or create a new network, new router, and plumb my tenant network in to that | 14:30 |
CeeMac | was thinking i could use a floating IP pool, but apparently thats only available for compute v2 network | 14:30 |
odyssey4me | mnaser once the python build simplficaction is done, we can switch each service that supports py3 to py3 individually - right now it's all or nothing | 14:30 |
CeeMac | which is presumably earlier or later than the compute i'm running ...... | 14:30 |
odyssey4me | once that's done, we can start pulling out all the py2-isms, which sill reduce our deps | 14:31 |
odyssey4me | *will | 14:31 |
mnaser | odyssey4me: oooo i see, that makes sense | 14:31 |
mnaser | i dunno how we can deal with centos for that tho :( | 14:31 |
mnaser | chkumar|ruck: how is tripleo doing py3 testing? | 14:31 |
mnaser | considering rhel8 isnt out yet, or we don't know when its gonna be out | 14:32 |
odyssey4me | mnaser yeah, I figure that with the changes in place we can leave centos on py2 and switch suse/ubuntu to py3 | 14:32 |
odyssey4me | it'll be easy enough with a map, I think | 14:32 |
jamesdenton | CeeMac You you create an 'external' provider network, the IP range specified in the 'allocation pool' of the respective subnet(s) == the available floating IPs that can be associated | 14:34 |
jamesdenton | *When you create... | 14:34 |
CeeMac | ohhhhhhhhh | 14:35 |
CeeMac | see, that makes sense | 14:35 |
jamesdenton | For an external provider network, be sure to disable DHCP on the subnet if you don't plan attaching VMs directly. Then you should get anywhere from 11-13 usable IPs | 14:35 |
CeeMac | i'd been thinking that just going to be used for router addresses | 14:35 |
*** sdake has joined #openstack-ansible | 14:35 | |
jamesdenton | you have to subtract the network, broadcast, gateway and any VRRP addrs (if they apply) | 14:36 |
CeeMac | sure | 14:36 |
jamesdenton | Well, it is also true that each router you create will also grab one of those IPs | 14:36 |
CeeMac | i only needed a handful of IPs for basic testing | 14:36 |
jamesdenton | the router needs one, and then the rest are for floats. But each router you add subtracts one | 14:36 |
jrosser | you need to put the external VIP somewhere too | 14:37 |
jrosser | don't forget that | 14:37 |
CeeMac | totally obvious now in hindsite | 14:37 |
CeeMac | jrosser, for the LB? | 14:37 |
jamesdenton | yeah, the external LB VIPs needs to be routable, but hopefully it comes from some other network | 14:37 |
CeeMac | or you mean VRRP | 14:37 |
CeeMac | yeah, its on a seperate box/network | 14:38 |
jamesdenton | i assume thats what jrosser is referring to | 14:38 |
*** markvoelker has joined #openstack-ansible | 14:38 | |
CeeMac | after the shennanigans i had with the last iteration of the environment i split the roles out | 14:38 |
CeeMac | and following much sage advise from yourselves of course :) | 14:39 |
jrosser | in the simplest possible case the external VIP could live in the provider network outside the allocation pool | 14:40 |
CeeMac | jrosser, yeah, that makes sense | 14:41 |
jrosser | but you might want to future proof yourself (or allow router ACL for example) by creating a specific subnet for api endpoints | 14:41 |
CeeMac | i've put the external VIP and a completely seperate DMZ for simplicity | 14:41 |
jrosser | really up to you how you want to cut that | 14:41 |
CeeMac | *on a | 14:42 |
CeeMac | so the provider network is purely for neutron | 14:42 |
CeeMac | can you have multiple allocation pools in a subnet? | 14:42 |
CeeMac | yes, apparently | 14:43 |
CeeMac | sorry, that was a lazy question, my dashboard was literally on the other screen | 14:43 |
jamesdenton | yep | 14:45 |
CeeMac | but there doesn't appear to be a way to choose which allocation to assign an IP from | 14:46 |
CeeMac | hmm | 14:46 |
CeeMac | I'll manually create few floating IPs and assign them up to the project for testing | 14:47 |
CeeMac | in production i'll be looking at exposing public addresses directly to neutron anyhow | 14:47 |
jamesdenton | yeah it's automatic. BUT, you can choose a particular IP using some flags on the create command | 14:48 |
CeeMac | i'm doing everything through the UI right now, as I imagine thats how most of our current customer base would be driving it (when its released) | 14:50 |
jamesdenton | Sure. The user shouldn't really care which IP they get - only that they get one and it works | 14:51 |
*** sdake has quit IRC | 14:54 | |
CeeMac | yep | 14:54 |
*** aedc has quit IRC | 15:02 | |
odyssey4me | mnaser cloudnull chkumar|ruck FYI, for the config_template transition: https://bugs.launchpad.net/openstack-ansible/+bug/1791258 | 15:06 |
openstack | Launchpad bug 1791258 in openstack-ansible "Make use of ansible-config_template repo" [Medium,Confirmed] - Assigned to Jesse Pretorius (jesse-pretorius) | 15:06 |
*** sdake has joined #openstack-ansible | 15:08 | |
*** markvoelker has quit IRC | 15:11 | |
openstackgerrit | James Denton proposed openstack/openstack-ansible master: [docs] Apply provider network config on per-group basis https://review.openstack.org/635013 | 15:11 |
*** brad[] has quit IRC | 15:13 | |
*** brad[] has joined #openstack-ansible | 15:17 | |
odyssey4me | mnaser FYI evrardjp did a great ob of making the whole openstack_release version thing simpler - see https://review.openstack.org/#/q/topic:easier_osa_releases - it does depend on the bug fix in https://review.openstack.org/633759 | 15:25 |
*** slaweq has quit IRC | 15:27 | |
mnaser | odyssey4me: does the second bug mean that we don't always use OSA'd ansible unless we're inside the directory? | 15:27 |
mnaser | because afaik maybe we want a release note cause a lot of folks might rely on just running it anywhere? | 15:27 |
odyssey4me | mnaser it was always designed to only use the wrapper if you execute 'openstack-ansible' or if you execute 'ansible-playbook' from inside the OSA git clone | 15:28 |
odyssey4me | the fact that it now uses the wrapper everywhere is a bug | 15:28 |
mnaser | but because of this bug it worked anywhere | 15:28 |
mnaser | fair | 15:28 |
mnaser | i wonder how long that bug has been a bug | 15:28 |
odyssey4me | well yeah, perhaps we should add a 'fix' release note | 15:28 |
odyssey4me | it's only rocky IIRC | 15:29 |
odyssey4me | yep: https://review.openstack.org/#/c/582633/ | 15:29 |
odyssey4me | ok, evrardjp mind if I adjust the patch to add a reno? | 15:29 |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Add telemetry distro plugin install for aodh https://review.openstack.org/632125 | 15:30 |
*** slaweq has joined #openstack-ansible | 15:39 | |
admin0 | what process/agent is respoinsible to re-create the iptables rules ? is it the neutron-l3-agent ? | 15:39 |
mnaser | odyssey4me: wanna make a follow up fix release note and i can +w this to get moving? | 15:39 |
odyssey4me | mnaser yeah, I can do that instead - that'll prevent wasting resources | 15:42 |
openstackgerrit | Merged openstack/openstack-ansible-os_glance master: venv: use inventory_hostname instead of ansible_hostname https://review.openstack.org/635177 | 15:44 |
mnaser | +w'd odyssey4me | 15:48 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add release note for wrapper fix https://review.openstack.org/635547 | 15:52 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add release note for wrapper fix https://review.openstack.org/635547 | 15:52 |
odyssey4me | admin0 it would depend which rules you mean - neutron managed those relating to neutron-managed networks, whereas lxc-syste,-manage manages those for the containers | 15:54 |
mnaser | odyssey4me: i've been randomly noticing those failures here and there | 15:56 |
mnaser | http://paste.openstack.org/show/744688/ | 15:57 |
mnaser | happening mostly on my local run_tests | 15:57 |
odyssey4me | mnaser yeah, chkumar|ruck reported that too - but I've not been able to reproduce it | 15:59 |
odyssey4me | I suspect it's due to the upgrade/downgrade process that the role tests do with pip which is all a nasty ball of mess... if you can figure out how to reproduce it reliably, then we can sticky tape it. I'd rather just be rid of all that and use the integrate repo though. | 16:00 |
odyssey4me | I've only heard reports from centos. :/ | 16:00 |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Add user and password for secure image download (optional) https://review.openstack.org/625266 | 16:00 |
odyssey4me | we did a good job of isolating the integrated repo build pip installs into venvs exclusively, but didn't do that for the role tests | 16:01 |
mnaser | yeah i think that was it | 16:02 |
mnaser | maybe we'll wait it out | 16:02 |
*** aedc has joined #openstack-ansible | 16:03 | |
*** TxGirlGeek has joined #openstack-ansible | 16:07 | |
*** markvoelker has joined #openstack-ansible | 16:08 | |
*** miloa has quit IRC | 16:10 | |
*** spatel has joined #openstack-ansible | 16:12 | |
spatel | cloudnull: yt? | 16:13 |
*** gyee has joined #openstack-ansible | 16:13 | |
spatel | does OSAD enable rabbitmq_management plugin by default ? | 16:13 |
CeeMac | well. That went surprisingly well. | 16:14 |
cloudnull | spatel yes | 16:15 |
* cloudnull believes we still do | 16:15 | |
spatel | cloudnull: thanks i can see its listening on 15672 | 16:17 |
openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Enable overriding the service setup host python interpreter https://review.openstack.org/621183 | 16:20 |
*** pcaruana has quit IRC | 16:21 | |
admin0 | is there a way to validate that the platform has updated to 18.1.2 .. how to validate this | 16:26 |
admin0 | this is a question for the change-management form :) | 16:26 |
admin0 | on the* | 16:26 |
spatel | cloudnull: what is the admin username password for management i am looking at secret file and found rabbitmq_monitoring_password: 98d6997c62aae72411bf07d1fd | 16:28 |
spatel | what would be the username? | 16:28 |
*** slaweq has quit IRC | 16:29 | |
odyssey4me | admin0 every service venv is tagged according to the OSA tag - so /openstack/venvs/<service>-<tag>/ | 16:30 |
odyssey4me | so the easiest way to verify that all the software deployed is to check those exist, and that the services are actually using that venv to run the service | 16:31 |
odyssey4me | of course it'd be nice to put together something that does that automagically | 16:31 |
spatel | curl -i -u guest:guest http://ostack-infra-01_rabbit_mq_container-190a0568:15672/api/whoami got error HTTP/1.1 401 Unauthorized | 16:32 |
spatel | let me try different account | 16:32 |
odyssey4me | spatel by default we delete the guest account | 16:32 |
spatel | what would be the user account to access management plugin? | 16:33 |
*** udesale has quit IRC | 16:33 | |
odyssey4me | spatel the plugins installed is governed by this var: https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/defaults/main.yml#L93-L95 | 16:33 |
odyssey4me | spatel also, the user for monitoring is created here: https://github.com/openstack/openstack-ansible/blob/master/playbooks/rabbitmq-install.yml#L69-L84 | 16:35 |
odyssey4me | so the default username for monitoring is 'monitoring' | 16:35 |
spatel | monitoring - odyssey4me thanks | 16:36 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/rocky: Enable overriding the service setup host python interpreter https://review.openstack.org/635564 | 16:37 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/rocky: Enable overriding the service setup host python interpreter https://review.openstack.org/635564 | 16:38 |
*** chkumar|ruck has quit IRC | 16:38 | |
*** chandankumar has joined #openstack-ansible | 16:39 | |
*** markvoelker has quit IRC | 16:41 | |
CeeMac | speaking of monitoring. | 16:42 |
CeeMac | what are people using for monitoring? | 16:42 |
CeeMac | I | 16:42 |
CeeMac | I've noticed there is an osa-ops project | 16:43 |
CeeMac | and monasca | 16:43 |
CeeMac | so many options | 16:43 |
CeeMac | not sure how mature either/both of them are? | 16:43 |
odyssey4me | CeeMac osa-ops is a bit of an incubator for us, where anyone can add anything - that said, the various metric gathering things in there are being used in production by several users | 16:45 |
odyssey4me | monasca's use in the OSA community is considerably less, I think - there is a team using it and maintaining it though, and the pics they've posted are pretty nice | 16:46 |
odyssey4me | as always with any open source software - you need to pick what will be best for you to understand and maintain, and you should test it in a test environment before using it in production | 16:46 |
openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible-os_octavia master: Update octavia configuration documentation https://review.openstack.org/635402 | 16:48 |
CeeMac | thanks odyssey4me | 16:48 |
cloudnull | ceemac: we (Rackspace) use quite a bit from the ops repo. | 16:49 |
cloudnull | elk_6x, osquery, skydive, etc | 16:49 |
CeeMac | i think i saw an osa role for monasca, could work out if it installed the pre-reqs too though (influxdb, kafka etc) | 16:49 |
odyssey4me | mnaser did you take a peek at https://review.openstack.org/#/q/topic:easier_osa_releases - that eliminates the need for us to maintain a version-in-code | 16:49 |
cloudnull | spatel just anounced he's been doing somethings with collectd, which come to find out cern is also using. | 16:49 |
CeeMac | cloudnull, yeah i'd eyeballed elk already | 16:50 |
spatel | collectd+influxdb+grafana i found better tools for monitoring... | 16:50 |
CeeMac | been looking at implementing that for a while in general, so would be good to use this as an excuse to finally get it up and running | 16:50 |
CeeMac | spatel, what did you find was better? | 16:51 |
spatel | last few days i am playing with it and soon will post documents about it how i did it... | 16:51 |
spatel | grafana has very cool graphing and dynamic dashboard | 16:52 |
CeeMac | in the short term, i'll probably look at throwin up a grafana for plugging gnocchi into | 16:52 |
spatel | CeeMac: for example.. https://ibb.co/DMRCwFh | 16:53 |
CeeMac | but would probably want something else in the longer term | 16:53 |
spatel | it can go far deeper... this is just begining.. | 16:53 |
spatel | beginning* | 16:53 |
CeeMac | oooo purdy :) | 16:53 |
spatel | https://ibb.co/42xXZzT | 16:53 |
chandankumar | odyssey4me: http://logs.openstack.org/18/635518/1/check/openstack-ansible-functional-centos-7/a01b697/logs/ara-report/result/75c5fd08-6e24-4069-b95a-28b40484dfa6/ | 16:53 |
chandankumar | odyssey4me: I think we need to install libopenmpi-dev getting from this one https://stackoverflow.com/questions/28440834/error-when-installing-mpi4py | 16:54 |
CeeMac | look forward to seeing your writeup :) | 16:54 |
odyssey4me | chandankumar yeah, looks like missing deps | 16:55 |
spatel | https://ibb.co/Pr2dP9K | 16:55 |
spatel | this is how i monitoring OSAD | 16:55 |
CeeMac | cloudnull, skydive looks neat | 16:55 |
noonedeadpunk | odyssey4me: Hi, as promised - re-worked ceilometer to be similar with smart-resources https://review.openstack.org/#/c/634431/ | 16:56 |
odyssey4me | noonedeadpunk excellent, thanks - I'll review in the morning with a fresh head | 16:56 |
cloudnull | noonedeadpunk nice! | 17:00 |
noonedeadpunk | yep, sure. Sorry for the delay - made once had some time... | 17:00 |
odyssey4me | jrosser you mentioned that ansible-config_template is broken when using it for json files - could you elaborate? perhaps cloudnull could help figure out a fix | 17:01 |
cloudnull | ^ ? | 17:02 |
jrosser | one moment | 17:02 |
odyssey4me | cloudnull I'm going to get https://etherpad.openstack.org/p/osa-config-template-migration done ASAP, so that we're not maintaining two of the same thing | 17:02 |
odyssey4me | If someone has the time to work out what's changed in the plugins repo and hasn't been done in the ansible-config_template repo, I'd appreciate it | 17:03 |
odyssey4me | also, there are tests in both, but I don't think they line up | 17:03 |
jrosser | odyssey4me: cloudnull has a patch with that diff in it right now, which is a bit eye watering | 17:04 |
jrosser | re. config_template and json, this was from earlier http://paste.openstack.org/show/CZ5lmJXU5UgJJ7RExYXM/ | 17:05 |
jrosser | and that went away with this applied https://github.com/openstack/openstack-ansible-os_heat/commit/61f2c903cfc18e7d37d202d11eec9013ffd3c963 | 17:05 |
jrosser | but i think we maybe are talking about two different things | 17:07 |
*** sdake has quit IRC | 17:07 | |
jrosser | iirc this came up in the context of the ELK deploy most likley | 17:07 |
*** sdake has joined #openstack-ansible | 17:10 | |
cloudnull | i just recently resolved an issue where config template was not registering a "change" which as causing some issues in the plugin, however, im not sure that's related to anything breaking with json | 17:11 |
odyssey4me | ok, maybe there is no bug then - but it does look like there's a test differential... it seems like there's a bit more being tested in the plugins repo vs the config_template repo | 17:12 |
cloudnull | that very well could be | 17:13 |
odyssey4me | the implementation is quite confusing though to my dumb self, so I'd appreciate someone more in tune with config_template's implementation to implement changes to the independent repo to get those posted up | 17:13 |
odyssey4me | ideally before tomorrow, and passing CI, so that I can get on with getting the migration going | 17:13 |
odyssey4me | alternatively I can just go with what we already have and we can improve it later | 17:14 |
jrosser | well this is a starting point https://review.openstack.org/#/c/634857/ ? | 17:14 |
spatel | in logs i am getting error " Error Opening Queue [metering.sample] details: 401 - Unauthorized (ACCESS_REFUSED - access to queue 'metering.... "auto"}')" | 17:14 |
odyssey4me | jrosser that's the reverse? | 17:15 |
jrosser | oh duh of course | 17:15 |
spatel | i didn't find metering.sample queue in rabbitmqctl list_queues | 17:15 |
openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_nova master: Add nova_user_pip_packages variable https://review.openstack.org/635579 | 17:19 |
*** aedc has quit IRC | 17:21 | |
*** aedc has joined #openstack-ansible | 17:22 | |
openstackgerrit | Merged openstack/openstack-ansible stable/pike: Bump version to 16.0.26 https://review.openstack.org/633751 | 17:30 |
openstackgerrit | Merged openstack/openstack-ansible stable/queens: Bump version to 17.1.8 https://review.openstack.org/633725 | 17:30 |
openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Bump version to 18.1.4 https://review.openstack.org/633720 | 17:30 |
*** markvoelker has joined #openstack-ansible | 17:38 | |
*** aedc has quit IRC | 17:39 | |
*** priteau has quit IRC | 17:45 | |
*** chhagarw has quit IRC | 17:49 | |
*** slaweq has joined #openstack-ansible | 17:50 | |
*** kopecmartin is now known as kopecmartin|off | 18:00 | |
partlycloudy | hello folks, i always get this error (https://pasted.tech/l/8dae8b16aecf0566) when I run setup-openstack.yml in a linux screen session. | 18:07 |
partlycloudy | To fix it, i need to delete every thing under /etc/openstack_deploy/ansible_facts and re-run the playbook. | 18:08 |
odyssey4me | partlycloudy there is a known issue when using screen - I personally use tmux... I don't think anyone's worked out how to fix the screen issue | 18:09 |
odyssey4me | somehow ansible picks up some of the screen related env vars and stores them in facts, and that breaks things | 18:10 |
jrosser | You end up with TERMCAP in a variable/fact which something then tries to template | 18:10 |
*** markvoelker has quit IRC | 18:11 | |
jrosser | Perhaps unsetting the env var whilst in screen will help | 18:11 |
partlycloudy | thank you guys! now i am just moving on after removing the ansible_facts. i will try unsetting env var in screen next time, or switch to tmux. | 18:13 |
jrosser | Long time ansible issue it would seem https://github.com/ansible/ansible/issues/12191 | 18:15 |
*** aedc has joined #openstack-ansible | 18:17 | |
openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer stable/rocky: Fixes gnocchi resource-type list being empty after setup https://review.openstack.org/635189 | 18:21 |
*** DanyC has quit IRC | 18:24 | |
*** DanyC has joined #openstack-ansible | 18:25 | |
*** DanyC has quit IRC | 18:29 | |
openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_nova master: Add nova_user_pip_packages variable https://review.openstack.org/635579 | 18:29 |
*** aedc has quit IRC | 18:32 | |
openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Cleanup files and templates using smart sources https://review.openstack.org/634431 | 18:33 |
*** aedc has joined #openstack-ansible | 18:37 | |
*** aedc has quit IRC | 18:40 | |
*** aedc has joined #openstack-ansible | 18:41 | |
openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_nova master: Add nova_user_pip_packages variable https://review.openstack.org/635579 | 18:41 |
openstackgerrit | Merged openstack/openstack-ansible-tests master: Add placement_install_method for distro installations https://review.openstack.org/635281 | 18:42 |
*** aedc has quit IRC | 18:43 | |
*** slaweq has quit IRC | 18:49 | |
*** sdake has quit IRC | 18:58 | |
*** vnogin has joined #openstack-ansible | 19:00 | |
*** vnogin has quit IRC | 19:04 | |
openstackgerrit | James Denton proposed openstack/openstack-ansible master: [docs] Apply provider network config on per-group basis https://review.openstack.org/635013 | 19:05 |
*** hamzaachi has joined #openstack-ansible | 19:07 | |
*** markvoelker has joined #openstack-ansible | 19:08 | |
openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer stable/rocky: Fixes gnocchi resource-type list being empty after setup https://review.openstack.org/635189 | 19:13 |
*** TxGirlGeek has quit IRC | 19:17 | |
*** TxGirlGeek has joined #openstack-ansible | 19:19 | |
openstackgerrit | Jacob Wagner proposed openstack/openstack-ansible-ops master: Add ability to deploy designate with BIND9 servers https://review.openstack.org/635611 | 19:39 |
*** markvoelker has quit IRC | 19:42 | |
*** sdake has joined #openstack-ansible | 19:46 | |
*** sdake has quit IRC | 19:49 | |
*** priteau has joined #openstack-ansible | 19:49 | |
*** sdake has joined #openstack-ansible | 19:51 | |
*** sdake has quit IRC | 19:53 | |
*** sdake has joined #openstack-ansible | 19:53 | |
*** priteau has quit IRC | 19:58 | |
*** weezS has joined #openstack-ansible | 20:08 | |
*** sdake has quit IRC | 20:31 | |
*** sdake has joined #openstack-ansible | 20:32 | |
*** markvoelker has joined #openstack-ansible | 20:39 | |
*** priteau has joined #openstack-ansible | 20:42 | |
*** weezS has quit IRC | 20:52 | |
*** priteau has quit IRC | 20:55 | |
*** priteau has joined #openstack-ansible | 21:08 | |
*** markvoelker has quit IRC | 21:11 | |
*** hamzaachi has quit IRC | 21:11 | |
*** DanyC has joined #openstack-ansible | 21:13 | |
*** DanyC has quit IRC | 21:14 | |
*** DanyC has joined #openstack-ansible | 21:14 | |
*** DanyC has quit IRC | 21:18 | |
*** sdake has quit IRC | 21:29 | |
*** ansmith has quit IRC | 21:31 | |
*** vnogin has joined #openstack-ansible | 21:33 | |
*** sdake has joined #openstack-ansible | 21:37 | |
openstackgerrit | Michael Vollman proposed openstack/openstack-ansible-os_nova master: Add an ip address to eth12 in OSA test containers https://review.openstack.org/635654 | 21:38 |
*** priteau has quit IRC | 21:40 | |
*** vnogin has quit IRC | 21:42 | |
*** priteau has joined #openstack-ansible | 21:48 | |
*** priteau has quit IRC | 21:49 | |
*** sdake has quit IRC | 21:51 | |
*** sdake has joined #openstack-ansible | 21:54 | |
*** spatel has quit IRC | 21:54 | |
*** sdake has quit IRC | 21:56 | |
*** sdake has joined #openstack-ansible | 21:58 | |
*** markvoelker has joined #openstack-ansible | 22:08 | |
*** DanyC has joined #openstack-ansible | 22:15 | |
*** DanyC has quit IRC | 22:19 | |
*** DanyC has joined #openstack-ansible | 22:20 | |
openstackgerrit | Merged openstack/openstack-ansible master: Define OSA clone dir in the openstack-ansible.sh script https://review.openstack.org/633759 | 22:29 |
openstackgerrit | Merged openstack/openstack-ansible master: Add release note for wrapper fix https://review.openstack.org/635547 | 22:29 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_glance master: Correct notification driver https://review.openstack.org/635672 | 22:29 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_blazar master: Correct notification driver https://review.openstack.org/635673 | 22:32 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_cinder master: Correct notification driver https://review.openstack.org/635674 | 22:32 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_designate master: Correct notification driver https://review.openstack.org/635675 | 22:32 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_glance master: Correct notification driver https://review.openstack.org/635676 | 22:32 |
*** sdake has quit IRC | 22:34 | |
*** sdake has joined #openstack-ansible | 22:36 | |
*** sdake has quit IRC | 22:36 | |
*** DanyC_ has joined #openstack-ansible | 22:37 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_heat master: Correct notification driver https://review.openstack.org/635678 | 22:37 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_keystone master: Correct notification driver https://review.openstack.org/635679 | 22:38 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_magnum master: Correct notification driver https://review.openstack.org/635680 | 22:38 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_sahara master: Correct notification driver https://review.openstack.org/635681 | 22:38 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_tacker master: Correct notification driver https://review.openstack.org/635682 | 22:38 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_trove master: Correct notification driver https://review.openstack.org/635683 | 22:39 |
*** DanyC has quit IRC | 22:40 | |
*** markvoelker has quit IRC | 22:42 | |
*** tosky has quit IRC | 22:52 | |
*** hwoarang has quit IRC | 22:54 | |
*** hwoarang has joined #openstack-ansible | 22:59 | |
*** Emine has quit IRC | 22:59 | |
*** ansmith has joined #openstack-ansible | 23:16 | |
*** TxGirlGeek has quit IRC | 23:21 | |
*** sdake has joined #openstack-ansible | 23:24 | |
*** markvoelker has joined #openstack-ansible | 23:39 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!