*** watersj has joined #openstack-ansible | 00:33 | |
*** cshen has joined #openstack-ansible | 00:43 | |
*** cshen has quit IRC | 00:48 | |
*** weifan has joined #openstack-ansible | 00:54 | |
*** weifan has quit IRC | 00:54 | |
watersj | why does this insist to connect as root openstack-ansible -u ubuntu -bK setup-hosts.yml | 01:02 |
---|---|---|
*** gary_perkins has quit IRC | 01:40 | |
*** cshen has joined #openstack-ansible | 01:58 | |
*** cshen has quit IRC | 02:02 | |
*** cshen has joined #openstack-ansible | 02:08 | |
*** cshen has quit IRC | 02:13 | |
*** cshen has joined #openstack-ansible | 02:18 | |
*** cshen has quit IRC | 02:23 | |
*** Dwane has quit IRC | 02:55 | |
*** Dwane has joined #openstack-ansible | 02:55 | |
openstackgerrit | Merged openstack/openstack-ansible master: Change glance haproxy balancing method https://review.opendev.org/718533 | 03:07 |
*** Dwane has quit IRC | 03:14 | |
*** gary_perkins has joined #openstack-ansible | 03:45 | |
*** Dwane has joined #openstack-ansible | 03:46 | |
*** rh-jelabarre has quit IRC | 04:02 | |
*** openstack has joined #openstack-ansible | 04:23 | |
*** ChanServ sets mode: +o openstack | 04:23 | |
*** cshen has quit IRC | 04:24 | |
*** evrardjp has quit IRC | 04:35 | |
*** evrardjp has joined #openstack-ansible | 04:36 | |
*** rohit02 has joined #openstack-ansible | 04:36 | |
rohit02 | hi team,can we integrate external ceph in openstack-ansible for cinder and nova services? | 04:37 |
*** gyee has quit IRC | 04:40 | |
*** udesale has joined #openstack-ansible | 04:40 | |
*** dpaclt has joined #openstack-ansible | 04:52 | |
*** cp- has quit IRC | 04:57 | |
*** cshen has joined #openstack-ansible | 05:00 | |
*** cshen has quit IRC | 05:05 | |
noonedeadpunk | watersj: we never tested/intended to use that way, as anyway you will need root for most of the files. Also there was smth weird when using delegate_to and become (but not 100% sure) | 05:15 |
noonedeadpunk | rohit02: sure thing you can! | 05:15 |
noonedeadpunk | fridtjof[m]: hm, and does gateway ip is provided in the subnet? I think that actually config drive should be able to set gateway, as it sets network configuration (like ip)? | 05:18 |
rohit02 | @noonedeadpunk thanx....can u help me with docs for this in official document unable to find these external ceph configuration | 05:21 |
noonedeadpunk | rohit02: so actually you need to set ceph_mons to the list of ips with your monitors and set openstack_config: false | 05:24 |
noonedeadpunk | and do not set ceph-osd, ceph-mons, etc groups in openstack_user_config | 05:25 |
noonedeadpunk | another option, is just to provide ceph config and auth keys as variables | 05:25 |
noonedeadpunk | then you don't need to set ceph_mons variable in user_variables | 05:26 |
noonedeadpunk | but I think this feature misses this patch to be functional https://review.opendev.org/#/c/724118/ | 05:27 |
noonedeadpunk | I'm not sure if we really have this documented well but eventually we should cover this with docs )and help with that is appreciated:)) | 05:29 |
noonedeadpunk | rohit02: so we install ceph via ceph-ansible with that playbook https://opendev.org/openstack/openstack-ansible/src/branch/master/playbooks/ceph-install.yml | 05:35 |
noonedeadpunk | so as you may see it runs against ceph-mon, ceph-osd and ceph-mds hosts. If these groups are not defined - ceph won't be isntalled. | 05:36 |
noonedeadpunk | and ceph_mons variable is used by https://opendev.org/openstack/openstack-ansible-ceph_client/src/branch/master/tasks/ceph_get_mon_host.yml to get config and keyrings from it | 05:37 |
rohit02 | i will tried with openstack-ansible deployment with default ceph everything works fine but cinder volume service for ceph always down with error:cluster not found | 05:41 |
*** cshen has joined #openstack-ansible | 05:56 | |
noonedeadpunk | I think you should define cinder_backends correctly... | 06:20 |
noonedeadpunk | actully do you have connectivity from cinder_volume container to ceph? | 06:21 |
noonedeadpunk | you may try to debug this with ceph-client by running ceph -s --user cinder from cinder-volume container | 06:22 |
*** rohit02 has quit IRC | 06:27 | |
*** starborn has joined #openstack-ansible | 06:41 | |
*** Dwane has quit IRC | 06:42 | |
*** Dwane has joined #openstack-ansible | 06:42 | |
*** Dwane has quit IRC | 06:47 | |
*** miloa has joined #openstack-ansible | 06:49 | |
*** jbadiapa has joined #openstack-ansible | 06:59 | |
*** yolanda has joined #openstack-ansible | 07:14 | |
*** dpaclt has quit IRC | 07:21 | |
*** tosky has joined #openstack-ansible | 07:30 | |
*** rpittau|afk is now known as rpittau | 07:31 | |
*** gokhani has joined #openstack-ansible | 07:35 | |
*** openstackstatus has quit IRC | 07:46 | |
*** openstack has joined #openstack-ansible | 07:50 | |
*** ChanServ sets mode: +o openstack | 07:50 | |
*** DanyC has joined #openstack-ansible | 07:51 | |
*** DanyC has quit IRC | 07:56 | |
*** gokhani43 has joined #openstack-ansible | 08:04 | |
*** gokhani has quit IRC | 08:08 | |
fridtjof[m] | noonedeadpunk: so choosing "enable gateway" in horizon, and specifying an IP should work? I think I just got confused by this earlier: https://wiki.openstack.org/wiki/Gateway | 08:16 |
noonedeadpunk | Yeah, if you haven't this, you won't have default route even with regular cloud-init | 08:17 |
*** gokhani43 has quit IRC | 08:33 | |
*** andrelf has joined #openstack-ansible | 08:33 | |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Drop monitoring of nova-compute process https://review.opendev.org/725798 | 08:45 |
*** admin0 has quit IRC | 08:46 | |
*** DanyC has joined #openstack-ansible | 08:49 | |
*** DanyC has quit IRC | 08:50 | |
*** DanyC has joined #openstack-ansible | 08:50 | |
openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening master: Remove CI for fedora-latest https://review.opendev.org/725800 | 09:01 |
*** gokhani has joined #openstack-ansible | 09:01 | |
fridtjof[m] | nice! it works. I was a bit scared of trying it, in the unlikely case it meant something different and tried setting up some router at that IP (leading to conflicts and potential connection loss) | 09:04 |
fridtjof[m] | thanks a lot, noonedeadpunk :) | 09:04 |
*** DanyC has quit IRC | 09:06 | |
*** DanyC has joined #openstack-ansible | 09:07 | |
*** dpaclt has joined #openstack-ansible | 09:08 | |
*** DanyC has joined #openstack-ansible | 09:08 | |
*** DanyC has joined #openstack-ansible | 09:09 | |
*** DanyC has joined #openstack-ansible | 09:09 | |
*** DanyC has joined #openstack-ansible | 09:10 | |
*** DanyC has joined #openstack-ansible | 09:11 | |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Move openSUSE to experimental status https://review.opendev.org/725598 | 09:12 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Get ceilometer_lib_dir in more reliable way https://review.opendev.org/725158 | 09:12 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari master: Use openstack_venv_python_executable https://review.opendev.org/725803 | 09:22 |
*** gokhani has quit IRC | 09:32 | |
*** gokhani has joined #openstack-ansible | 09:33 | |
*** andrelf has quit IRC | 09:35 | |
*** rpittau is now known as rpittau|bbl | 09:44 | |
*** owalsh has quit IRC | 09:46 | |
*** owalsh has joined #openstack-ansible | 10:09 | |
*** andrelf has joined #openstack-ansible | 10:10 | |
andrelf | in openstack-ansible in which path is the keystone comand in keystone lxc container? im getting a strange error when doing "openstack user show --domain teste user" and would like to use the keystone user-list but can't find't in keystone or utility lxc containers. | 10:13 |
*** cshen has quit IRC | 10:15 | |
*** cshen has joined #openstack-ansible | 10:15 | |
andrelf | i was able to bind some comands inside /openstack/venvs/keystone-20.1.0.dev201/bin/ but now keystone client. | 10:17 |
jrosser | is the standalone keystone command deprecated? | 10:18 |
gokhani | hi folks, is there anyone who install ironic with OSA ? | 10:19 |
*** maharg101 has quit IRC | 10:19 | |
*** maharg101 has joined #openstack-ansible | 10:19 | |
*** cshen has quit IRC | 10:20 | |
andrelf | jrosser i wound't think its deprecated but i wil try to find out | 10:25 |
jrosser | this suggests that at some point it was marked for deprecation https://docs.openstack.org/python-keystoneclient/2.3.2/man/keystone.html# | 10:28 |
andrelf | jrosser yep i found this https://specs.openstack.org/openstack/keystone-specs/specs/keystoneclient/mitaka/deprecate-cli.html so i suposse that is the problem... ok now i do need to find when the openstack cli is giving "get() takes 1 positional argument but 2 were given" with openstack user show --domain teste user1 | 10:28 |
jrosser | you can add --debg | 10:29 |
jrosser | --debug | 10:29 |
andrelf | jrosser thanks i will try to find the source in that huge debug output heheh :D | 10:30 |
*** pcaruana has quit IRC | 10:31 | |
andrelf | i'm trying to add ldap domain for identity but testing first making the changes in each container before doing the change in osa-deployment and recreate the keystone and horizon container | 10:32 |
*** mathlin has joined #openstack-ansible | 10:34 | |
jrosser | andrelf: you shouldnt need to recreate the containers | 10:38 |
jrosser | if you update the necessary variables and config overrides then the new config will be deployed | 10:38 |
andrelf | jrosser oh ok thanks. it seams that that error "TypeError: get() takes 1 positional argument but 2 were given" is reported when try to GET /v3/users | 10:42 |
andrelf | i will probably try to change in osa-deployment and see if it changes more containers than keystone. | 10:44 |
*** pcaruana has joined #openstack-ansible | 10:50 | |
*** cshen has joined #openstack-ansible | 10:55 | |
*** pcaruana has quit IRC | 10:57 | |
*** cshen has quit IRC | 11:00 | |
*** cshen has joined #openstack-ansible | 11:00 | |
*** pcaruana has joined #openstack-ansible | 11:02 | |
*** watersj has quit IRC | 11:05 | |
*** watersj has joined #openstack-ansible | 11:06 | |
watersj | noonedeadpunk, thx for response. I switch root w/sshkeys and able run deploys now | 11:07 |
*** dpaclt has quit IRC | 11:11 | |
*** ygk_12345 has joined #openstack-ansible | 11:29 | |
ygk_12345 | hi all | 11:29 |
ygk_12345 | can someone tell me how to disable the variable value "need_online_data_migrations" for cinder playbook ? | 11:30 |
andrelf | in keystone-wsgi-public.service journalctl logs i can only find ERROR ldappool ldap.SERVER_DOWN: {'desc': "Can't contact LDAP server", 'errno': 22, 'info': 'Invalid argument'}. how can i get more debug from ldappool to see what is the invalid argument? | 11:43 |
*** rh-jelabarre has joined #openstack-ansible | 12:02 | |
*** rh-jelabarre has quit IRC | 12:02 | |
*** rh-jelabarre has joined #openstack-ansible | 12:03 | |
*** rpittau|bbl is now known as rpittau | 12:18 | |
mgariepy | andrelf, you can set the debug_level in your ldap section | 12:22 |
mgariepy | andrelf, either 255 or 4095.(it's ldap debug numbers) | 12:23 |
*** udesale_ has joined #openstack-ansible | 12:23 | |
andrelf | mgariepy ok thanks i will try that | 12:23 |
mgariepy | but server down it might be that it's only unable to contact the ldap server. | 12:24 |
mgariepy | you can install ldap-utils and try to query the ldap server from your keystone container | 12:24 |
*** udesale has quit IRC | 12:26 | |
andrelf | mgariepy i have already tried to use ldapsearch and it works. i supoose that the problem is in the configurations that im putting in /etc/keystone/domains/keystone.test.conf but i will try with debug and see if i can find the invalid argument | 12:27 |
mgariepy | you can use the same arguments you are passing in the config file. | 12:28 |
mgariepy | which os / version are you on ? | 12:28 |
andrelf | mgariepy the debug_level helped because it was a problem in URL name.... :( | 12:33 |
*** Dwane has joined #openstack-ansible | 12:44 | |
*** cp- has joined #openstack-ansible | 13:01 | |
ygk_12345 | hi all | 13:03 |
ygk_12345 | when I am upgrading from neutron from rocky to stein I am receving this error in the neutron playbook at the db expand step | 13:04 |
ygk_12345 | ImportError: cannot import name _model_query | 13:04 |
ygk_12345 | File "/openstack/venvs/neutron-19.0.10/lib/python2.7/site-packages/neutron_fwaas/db/firewall/v2/firewall_db_v2.py", line 20, in <module> | 13:04 |
ygk_12345 | from neutron.db import _model_query as model_query | 13:04 |
ygk_12345 | can someone help me what I am missing here ? | 13:04 |
ygk_12345 | root@c1b-neutron-server-container-8575bd90:~# ./neutron-db-manage upgrade --expand command is failing | 13:05 |
ygk_12345 | is it the issue of neutron firewall version 2 ? I have set it to firewall_v2 in the neutron pluygins section in user_variables file | 13:06 |
ygk_12345 | Uploaded file: https://uploads.kiwiirc.com/files/5a300a97f91e1d6e78b07a7c832390a2/pasted.txt | 13:12 |
ygk_12345 | Uploaded file: https://uploads.kiwiirc.com/files/299546fdcfff29d0f6e93307b9614de1/pasted.txt | 13:17 |
ygk_12345 | is anyone here who can help me please | 13:24 |
*** itandops has joined #openstack-ansible | 13:26 | |
*** gokhani has quit IRC | 13:26 | |
mgariepy | andrelf, glad it works now :) | 13:29 |
ygk_12345 | is this a bug ? | 13:31 |
* noonedeadpunk wondering if fwaas is supported | 13:38 | |
ygk_12345 | noonedeadpunk it worked in our another setup | 13:39 |
noonedeadpunk | it works on stein? | 13:39 |
ygk_12345 | yes an in train as well | 13:39 |
ygk_12345 | i did it as part of upgrade from rocky--stein---train | 13:40 |
noonedeadpunk | tbh I never had hands-on with fwaas except heard it has lack of support... | 13:40 |
ygk_12345 | hmm | 13:40 |
noonedeadpunk | Tbh you'd better ask about it in #openstack-neutron | 13:41 |
ygk_12345 | ok | 13:41 |
noonedeadpunk | they might gave some good advice | 13:41 |
ygk_12345 | hmm ok | 13:41 |
CeeMac | I seem to recall there being some changes introduced between rocky and stein | 13:41 |
ygk_12345 | one is the firewall version to v2 | 13:42 |
noonedeadpunk | I recall some about lbaas only:( | 13:42 |
ygk_12345 | in the plugins section of user_variables | 13:42 |
CeeMac | Fwaas skills are short in #openstack-neutron now I think but worth asking for sure | 13:42 |
CeeMac | Jamesdenton wrote a patch to fix something as well but I can't remember what now it's a while since I looked at it | 13:43 |
jamesdenton | hmm | 13:43 |
CeeMac | You didn't remember last time i mentioned it to you either jamesdenton if that helps? :) | 13:44 |
jamesdenton | :D | 13:44 |
jamesdenton | i recall updating docs and maybe tests, but that's all i remember | 13:44 |
CeeMac | I'm not at my desk right now otherwise I'd check my notes | 13:44 |
ygk_12345 | ok | 13:44 |
*** maharg101 has quit IRC | 13:45 | |
jamesdenton | So, _model_query.py exists in Rocky, but not Stein | 13:48 |
jamesdenton | and it looks like neutron_fwaas (rocky) references it, stein does not | 13:50 |
jamesdenton | so maybe we're pulling in an older neutron_fwaas | 13:50 |
itsjg | Hello! Would you happen to know if openstack-ansible officially supports aarch64 (ARM)? I've seen hints in the code that support is there, but upon running the setup-openstack.yml playbook on a ARM compute host, I'm getting errors building the python packages (the "Install python packages into the venv" step). | 13:51 |
jrosser | itsjg: we have had a large contribution in the past to get OSA working on arm, and there was a company who got a proper HA multinode deployment working | 13:52 |
itsjg | Ahh, neat! | 13:53 |
jrosser | however that has not translated into a CI job though, so i cannot comment on the current state of support or if anything is broken | 13:53 |
itsjg | Does that deployment require different playbooks? | 13:53 |
itsjg | Ohh, I see | 13:53 |
jrosser | no on aarch64/ubuntu it should "just work" | 13:53 |
itsjg | Gotcha, thanks | 13:53 |
jamesdenton | @ygk_12345 can you check your /opt/openstack-ansible/playbooks/defaults/repo_packages/openstack_services.yml and let me know the value for neutron_fwaas_git_install_branch? | 13:53 |
jrosser | it will be slow to install initially becasue there are no pre-built wheels on pypy for aarch64 | 13:54 |
itsjg | Thanks jrosser | 13:55 |
jrosser | itsjg: i did a bunch of the fixes needed for arm so feel free to ask if you get stuck, but bear in mind i have no access to hardware to test anything | 13:55 |
itsjg | I believe my issue at the moment is attempting to install on a CentOS host. I'll try the Ubuntu route. | 13:55 |
jrosser | oh ok, what i did was entirely ubuntu bionic based | 13:55 |
jrosser | i think it should be not so difficult to spin up an aarch64 all-in-one, thats where i'd start | 13:56 |
itsjg | Thanks! | 14:00 |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: [DNM] Test arm64 https://review.opendev.org/618305 | 14:06 |
itsjg | If I can be of any assistance with testing, please do let me know I'd be happy to help | 14:09 |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: [DNM] Test arm64 https://review.opendev.org/618305 | 14:12 |
ygk_12345 | jamesdenton let me check that | 14:14 |
ygk_12345 | jamesdenton it is this neutron_fwaas_git_track_branch: stable/stein | 14:15 |
jamesdenton | there's also neutron_fwaas_git_install_branch, right? | 14:15 |
ygk_12345 | jamesdenton neutron_fwaas_git_install_branch: 648d725da0de99b5104196a2888fe2490a57cf47 # HEAD as of 29.01.2020 | 14:15 |
jamesdenton | right, ok | 14:15 |
jamesdenton | based on your logs, it looks like your neutron_fwaas is still rocky | 14:16 |
ygk_12345 | ok what should I do now ? | 14:16 |
ygk_12345 | i have deleted the old neutron container and recreated it | 14:18 |
jamesdenton | just now? | 14:18 |
ygk_12345 | yes | 14:20 |
ygk_12345 | and I ran the os-neutron playbook but still the error exists | 14:21 |
jamesdenton | and c1b_neutron_server_container-8575bd90 is the one you recreated? | 14:22 |
*** jhesketh has quit IRC | 14:22 | |
ygk_12345 | yes | 14:22 |
ygk_12345 | yes | 14:22 |
ygk_12345 | jamesdenton yes | 14:26 |
jamesdenton | can you do me a favor and post the first 50 lines of /openstack/venvs/neutron-19.0.10/lib/python2.7/site-packages/neut | 14:30 |
jamesdenton | ron_fwaas/db/firewall/v2/firewall_db_v2.py to pastebin? | 14:30 |
ygk_12345 | jamesdenton sure. please hold on | 14:30 |
ygk_12345 | jamesdenton http://paste.openstack.org/show/793198/ | 14:32 |
jamesdenton | thx | 14:32 |
*** maharg101 has joined #openstack-ansible | 14:35 | |
*** udesale_ has quit IRC | 14:38 | |
openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_nova master: Add ignore_msrs=1 https://review.opendev.org/725880 | 14:49 |
*** jhesketh has joined #openstack-ansible | 14:51 | |
*** maharg101 has quit IRC | 14:57 | |
openstackgerrit | Merged openstack/ansible-hardening master: Remove CI for fedora-latest https://review.opendev.org/725800 | 15:02 |
*** maharg101 has joined #openstack-ansible | 15:08 | |
*** starborn has quit IRC | 15:12 | |
openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening stable/train: Remove CI for fedora-latest https://review.opendev.org/725897 | 15:26 |
openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening stable/stein: Remove CI for fedora-latest https://review.opendev.org/725898 | 15:26 |
openstackgerrit | Andrew Bonney proposed openstack/openstack-ansible-ops master: logstash: add cisco ios/nexus syslog parsing to pipeline https://review.opendev.org/725899 | 15:35 |
*** fghaas has joined #openstack-ansible | 15:38 | |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-tests master: [WIP] Use galera server role to install galera client https://review.opendev.org/725902 | 15:41 |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-galera_server master: [WIP] combine galera client role into galera_server https://review.opendev.org/725903 | 15:42 |
jrosser | noonedeadpunk: ^ i did a giant hack on the galera_server role to make it also install the client | 15:42 |
jrosser | it's very ugly and needs refactoring, if we decide it is a good idea | 15:43 |
noonedeadpunk | jrosser: what is that?:) https://review.opendev.org/#/c/725903/1/file-galera | 15:44 |
jrosser | ah crap | 15:45 |
jrosser | curse of git add -A | 15:45 |
noonedeadpunk | I see:) | 15:46 |
jrosser | i did some local tests but didnt have my key to push the patch so had to git diff > scp > git apply it elsewhere | 15:46 |
noonedeadpunk | Actually I think we should install server also conditionally... | 15:46 |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-galera_server master: [WIP] combine galera client role into galera_server https://review.opendev.org/725903 | 15:46 |
noonedeadpunk | Or you mean we should leave galera_client role as well to run it against utility | 15:46 |
jrosser | i basically copied the tasks over from client into server | 15:47 |
noonedeadpunk | and galera_server to be indepenedent? | 15:47 |
jrosser | yeah well this is what we need to decide | 15:47 |
jrosser | if the installation of client or server is controlled by some booleans | 15:47 |
jrosser | or a tasks_from: install_client | 15:47 |
jrosser | because the client install is hardly and tasks at all and quite lightweight | 15:48 |
noonedeadpunk | and then we should update client tests pretty heavily | 15:48 |
noonedeadpunk | so they do not interfer with server | 15:48 |
jrosser | yeah we can merge the client and server tests all together | 15:49 |
noonedeadpunk | let me think about it for a while? | 15:49 |
jrosser | sure yes | 15:49 |
jrosser | i thought i'd just do a POC as something to think about - it's throwaway if need be | 15:50 |
noonedeadpunk | It's probably right direction | 15:50 |
noonedeadpunk | but not sure about futher steps yet:) | 15:50 |
noonedeadpunk | as running galera-server role against utility just to install client may be overwhelming in terms of role execution speed | 15:51 |
noonedeadpunk | or we eventually just need to create galera_client_install.yml and galera_server_install.yml in role and include correspondive tasks there | 15:52 |
jrosser | yes, thats why i left galera_client_install.yml alone with adding the repos, which is duplicate | 15:53 |
*** gyee has joined #openstack-ansible | 15:53 | |
jrosser | i think that tasks_from: galera_client_install.yml should work right now | 15:53 |
jrosser | and that would make it skip the server install becasue main.yml would not run | 15:54 |
jrosser | but i'm sure there is stuff i have missed there | 15:54 |
*** gokhani has joined #openstack-ansible | 15:56 | |
*** cshen has quit IRC | 16:03 | |
*** fghaas has quit IRC | 16:06 | |
noonedeadpunk | jrosser: btw, have you found galera_wsrep_provider path for centos? | 16:08 |
mgariepy | https://review.opendev.org/#/c/724721/3/vars/redhat-8.yml | 16:08 |
jrosser | now that is an interesting question | 16:08 |
mgariepy | noonedeadpunk, there. | 16:08 |
mgariepy | galera-4 pkg stuff. | 16:09 |
noonedeadpunk | oh, lets' try? | 16:09 |
jrosser | we want the path for galera-4 on centos7 don't we? | 16:09 |
noonedeadpunk | I think yes | 16:09 |
jrosser | i think i may have deleted my vm for that sadly | 16:09 |
mgariepy | probably indeed, should follow the pkg we install i guess.. | 16:09 |
jrosser | can we just query the rpm? | 16:10 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Bump MariaDB version to 10.4 https://review.opendev.org/717830 | 16:10 |
jrosser | <tool> <options> <rpm> -> answer :) | 16:10 |
noonedeadpunk | I'd rather just wait half an hour to see if path that mgariepy gave would suite :p | 16:11 |
mgariepy | lol | 16:11 |
openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening stable/train: Remove CI for fedora-latest https://review.opendev.org/725897 | 16:17 |
*** gokhani has quit IRC | 16:23 | |
*** rpittau is now known as rpittau|afk | 16:24 | |
*** cshen has joined #openstack-ansible | 16:30 | |
*** cshen has quit IRC | 16:35 | |
noonedeadpunk | fwiw - keystone has major security flaw covered with patches https://review.opendev.org/#/c/725891/ and https://review.opendev.org/#/c/725890/ | 16:35 |
*** evrardjp has quit IRC | 16:36 | |
*** evrardjp has joined #openstack-ansible | 16:36 | |
noonedeadpunk | so lets merge https://review.opendev.org/#/c/725275/ to unlock stein bump | 16:37 |
*** itandops has quit IRC | 16:39 | |
*** itandops has joined #openstack-ansible | 16:39 | |
mnaser | noonedeadpunk: +w, please feel free to ping for reviews | 16:41 |
*** miloa has quit IRC | 16:43 | |
noonedeadpunk | mnaser: than would be great to put +w on some more https://review.opendev.org/#/q/owner:+noonedeadpunk%2540ya.ru+label:Verified+status:open | 16:44 |
noonedeadpunk | ah, I see you already are:) | 16:44 |
mnaser | :) | 16:47 |
*** ygk_12345 has quit IRC | 16:55 | |
openstackgerrit | Merged openstack/openstack-ansible-haproxy_server master: Clarify documentation for deploying mutliple haproxy without keepalived https://review.opendev.org/723834 | 16:58 |
watersj | keepalived is getting virtual address of: 169.254.2.1/24 not ip address I assigned in openstack_user_config internal_lb_vip_address: 10.1.1.100 | 16:58 |
watersj | its default value, so I need to add these haproxy_keepalived_external_vip_cidr haproxy_keepalived_internal_vip_cidr | 17:00 |
noonedeadpunk | I think you need to define both of these things. | 17:02 |
noonedeadpunk | based on cidr interface is going to be defined iirc | 17:02 |
watersj | noonedeadpunk, yes i'm seeing that now. tracing it through user_variables.yml now | 17:02 |
*** andrea15 has joined #openstack-ansible | 17:08 | |
*** irclogbot_2 has quit IRC | 17:20 | |
andrelf | i have enable keystone_ldap for one domain (the default one continues to be sql) and enable horizon_keystone_multidomain_suportt in user_variables.yaml. after aplying all ansible steps, i can now use "openstack user show" and get the information of the ldap user but when i try to use the horizon dashboard it return error 500 (it was working before | 17:22 |
andrelf | with admin user). from what im seing in apache2 logs, from horizon container, it seems as problem in django "Internal Server Error: /auth/login/" and one of the last entries in the stack error is "ValueError: not enough values to unpack (expected 2, got 1)". any ideia how can i troubleshoot this? | 17:22 |
*** irclogbot_1 has joined #openstack-ansible | 17:22 | |
openstackgerrit | Merged openstack/openstack-ansible-tests master: Set apparmor statement based on LXC version used https://review.opendev.org/718998 | 17:25 |
*** sshnaidm is now known as sshnaidm|afk | 17:26 | |
openstackgerrit | Merged openstack/ansible-hardening stable/train: Remove CI for fedora-latest https://review.opendev.org/725897 | 17:28 |
noonedeadpunk | andrelf: sounds pretty common. I think we had patch for that recently | 17:30 |
noonedeadpunk | https://review.opendev.org/#/c/719934/ | 17:30 |
andrelf | noonedeadpunk how should i apply that patch and see if it works? | 17:34 |
noonedeadpunk | yeah, you can cherry-pick it or just manually apply | 17:34 |
*** isakgicu has joined #openstack-ansible | 17:40 | |
andrelf | noonedeadpunk so going to /etc/ansible/roles/os_keystone and cherry pick to that location right? | 17:45 |
isakgicu | hey guys, how hard would it be to add a second neutron node to an existing OpenStack cluster, and set haproxy to HA mode? | 17:45 |
noonedeadpunk | andrelf: /etc/ansible/roles/os_horizon and copy cherry-pick command from donwload menu on gerrit | 17:46 |
andrelf | yep i have done that but it gives some conflits. | 17:47 |
noonedeadpunk | andrelf: what release do you have? that patch is for train... | 17:47 |
noonedeadpunk | isakgicu: I think it's pretty trivial - like add host in openstack_user_config.yml, run haproxy-install.yml and os-neutron-isntall.yml | 17:48 |
andrelf | i think i have the lastest in master branch in openstack-ansible | 17:48 |
noonedeadpunk | ofc setup-hosts.yml | 17:48 |
noonedeadpunk | hm I guess master should have it included | 17:49 |
noonedeadpunk | https://review.opendev.org/#/c/607474/ has been merged 3 weeks ago... | 17:49 |
isakgicu | how about uptime? would it affect the activity of current neutron node? | 17:49 |
andrelf | and i confirmed that os_keystorne is also using master branch | 17:50 |
noonedeadpunk | isakgicu: I don't think it should.. Like node removal will cause l3 routers which are pointed to it to be down. But can't come to situation when downtime is possible while adding | 17:51 |
noonedeadpunk | except you have some non-standart driver | 17:51 |
noonedeadpunk | andrelf: but you need to check os_horizon, not os_keystone | 17:52 |
*** hamzy_ has quit IRC | 17:52 | |
andrelf | yep sorry ... long day lol | 17:52 |
noonedeadpunk | so if it's master then just try to git pull | 17:53 |
*** isakgicu has quit IRC | 17:54 | |
*** hamzy has joined #openstack-ansible | 17:54 | |
*** aedc_ has joined #openstack-ansible | 17:55 | |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest stable/rocky: Make functional tests voting https://review.opendev.org/723268 | 17:57 |
*** aedc has quit IRC | 17:58 | |
jrosser | i have a couple more haproxy patches which are at the bottom of the stack to review https://review.opendev.org/#/c/722419 https://review.opendev.org/#/c/722420 | 17:58 |
*** aedc_ has quit IRC | 18:01 | |
noonedeadpunk | they just were -W so didn't place vote on them | 18:01 |
andrelf | noonedeadpunki have done the git pull in os_horizon and can confirm the chjanges are there | 18:02 |
andrelf | applying with openstack-ansible steps them | 18:03 |
noonedeadpunk | I think you'll need to define horizon_keystone_multidomain_choices so that another domain appeared in the list | 18:03 |
andrelf | ok i will check the doc what it says about that | 18:04 |
openstackgerrit | Merged openstack/openstack-ansible stable/train: Bump SHAs for stable/train https://review.opendev.org/725115 | 18:04 |
noonedeadpunk | train recheck time! | 18:04 |
openstackgerrit | Merged openstack/openstack-ansible-repo_server stable/stein: Remove git deamon functionality https://review.opendev.org/725275 | 18:07 |
*** fghaas has joined #openstack-ansible | 18:10 | |
openstackgerrit | Merged openstack/openstack-ansible-repo_server stable/stein: Remove apt-cacher-ng cleanup tasks https://review.opendev.org/725276 | 18:15 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Get ceilometer_lib_dir in more reliable way https://review.opendev.org/725158 | 18:17 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Get ceilometer_lib_dir in more reliable way https://review.opendev.org/725158 | 18:21 |
openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Use more reliable way to get mysql version https://review.opendev.org/725269 | 18:22 |
*** cshen has joined #openstack-ansible | 18:30 | |
*** cshen has quit IRC | 18:35 | |
openstackgerrit | Merged openstack/openstack-ansible-os_masakari master: Use openstack_venv_python_executable https://review.opendev.org/725803 | 18:37 |
openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Fix rados linking for py3 https://review.opendev.org/725154 | 18:38 |
*** hamzy_ has joined #openstack-ansible | 18:39 | |
*** andrelf has quit IRC | 18:40 | |
*** hamzy has quit IRC | 18:42 | |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_gnocchi stable/train: Fix rados linking for py3 https://review.opendev.org/725965 | 18:46 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari stable/train: Use openstack_venv_python_executable https://review.opendev.org/725966 | 18:46 |
*** Dwane has quit IRC | 18:54 | |
*** fghaas has quit IRC | 18:55 | |
openstackgerrit | Merged openstack/openstack-ansible master: [doc] Adjust deploy-guide to install py3 https://review.opendev.org/724836 | 18:56 |
*** gokhani has joined #openstack-ansible | 19:09 | |
openstackgerrit | Merged openstack/openstack-ansible-ceph_client master: Do not try to connect to mons when everything is provided https://review.opendev.org/724118 | 19:19 |
openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Use horizon native wsgi app https://review.opendev.org/725775 | 19:21 |
*** gokhani has quit IRC | 19:33 | |
*** yolanda has quit IRC | 19:33 | |
*** gokhani has joined #openstack-ansible | 19:36 | |
openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Updated from OpenStack Ansible Tests https://review.opendev.org/709482 | 19:39 |
*** gokhani has quit IRC | 19:47 | |
*** jamesdenton has quit IRC | 19:48 | |
*** cshen has joined #openstack-ansible | 20:01 | |
*** jamesdenton has joined #openstack-ansible | 20:04 | |
*** cshen has quit IRC | 20:06 | |
*** jamesden_ has joined #openstack-ansible | 20:08 | |
*** spatel has joined #openstack-ansible | 20:10 | |
*** jamesdenton has quit IRC | 20:11 | |
*** jamesdenton has joined #openstack-ansible | 20:21 | |
*** jamesden_ has quit IRC | 20:22 | |
*** watersj has quit IRC | 21:06 | |
*** watersj has joined #openstack-ansible | 21:12 | |
spatel | noonedeadpunk: do you know OSA is ready for CentOS 8? (Train release) | 21:15 |
spatel | i would like to test if you guys need help to test CentOS 8? | 21:15 |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_gnocchi stable/train: Fix rados linking for py3 https://review.opendev.org/725965 | 21:23 |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-ceph_client stable/train: Do not try to connect to mons when everything is provided https://review.opendev.org/725998 | 21:23 |
*** watersj has quit IRC | 21:24 | |
openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible stable/train: [doc] Adjust deploy-guide to install py3 https://review.opendev.org/725999 | 21:24 |
noonedeadpunk | spatel: not yet | 21:42 |
noonedeadpunk | once we have it ready try to not forget to ping you for testing) | 21:43 |
*** spatel has quit IRC | 21:45 | |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: [doc] Adjust deploy-guide to install py3 https://review.opendev.org/725999 | 21:45 |
*** spatel has joined #openstack-ansible | 21:46 | |
noonedeadpunk | jrosser: btw, with centos 7.8 I guess libselinux for py3 should appear? | 21:46 |
noonedeadpunk | So we won't need to use custom repo fro that | 21:46 |
spatel | noonedeadpunk: in new datacenter i was thinking to use CentOS 8 but look like we are not ready | 21:48 |
*** _mmethot_ has joined #openstack-ansible | 21:49 | |
spatel | noonedeadpunk: if you want i can run test on CentOS 8 (with Train or ussuri) | 21:50 |
spatel | I think ussuri would be good idea to go with CentOS 8.x | 21:50 |
openstackgerrit | Merged openstack/ansible-hardening stable/train: Remove duplicated Zuul config file https://review.opendev.org/724064 | 21:51 |
*** mmethot_ has quit IRC | 21:51 | |
noonedeadpunk | spatel: it's good but centos8 is pretty tricky, so not really easy to implement | 21:52 |
spatel | yes specially dnf (which i am deal with now days) | 21:53 |
spatel | noonedeadpunk: anyway good to know, i will keep going with 7.x | 21:55 |
*** _mmethot_ has quit IRC | 21:56 | |
*** mmethot has joined #openstack-ansible | 21:56 | |
noonedeadpunk | yeah, I guess we kinda stopped at dnf priorities... | 21:59 |
noonedeadpunk | or smth like that | 21:59 |
mgariepy | spatel, do you have some cycle to help with the centos8 work ? | 22:00 |
spatel | mgariepy: sure, i can put sometime.. | 22:00 |
spatel | working from home so getting some free time also :) | 22:01 |
mgariepy | https://review.opendev.org/#/q/topic:osa/el8+(status:open+OR+status:merged) | 22:01 |
mgariepy | there are a few patch there | 22:01 |
*** cshen has joined #openstack-ansible | 22:02 | |
*** maaritamm has quit IRC | 22:02 | |
spatel | okay.. let me check | 22:03 |
mgariepy | no lxc support yet. the cors repo are there for lxc3. | 22:04 |
spatel | lets say i want to start testing then should i be using latest repo? | 22:06 |
mgariepy | yep start from master and apply the patch on top. | 22:06 |
*** cshen has quit IRC | 22:07 | |
spatel | mgariepy: sounds good, i will start deploying latest master on CentOS 8 and will keep you posted with bugs or if any patch | 22:07 |
mgariepy | there are bugs | 22:08 |
spatel | I am sure because CentOS 8 is many way new product.. | 22:09 |
spatel | also I found very few folks using CentOS with OSA :( even 2 year ago when i start deploying OSA on centOS during that time CentOS was experimental OS | 22:11 |
spatel | Anyway let me see how far i can make it with CentOS 8.x | 22:12 |
*** threestrands has joined #openstack-ansible | 22:20 | |
openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server stable/train: Add NODENAME in rabbitmq env https://review.opendev.org/706949 | 22:26 |
*** spatel has quit IRC | 22:28 | |
openstackgerrit | Merged openstack/openstack-ansible-os_rally stable/train: Updated from OpenStack Ansible Tests https://review.opendev.org/690371 | 22:44 |
openstackgerrit | Merged openstack/openstack-ansible-os_panko stable/train: Updated from OpenStack Ansible Tests https://review.opendev.org/690369 | 22:47 |
*** tosky has quit IRC | 22:56 | |
*** rh-jelabarre has quit IRC | 23:07 | |
openstackgerrit | Merged openstack/openstack-ansible-os_glance stable/train: Updated from OpenStack Ansible Tests https://review.opendev.org/690356 | 23:07 |
*** itandops has quit IRC | 23:25 | |
openstackgerrit | Merged openstack/openstack-ansible-os_ironic stable/train: Updated from OpenStack Ansible Tests https://review.opendev.org/690361 | 23:36 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!