Friday, 2018-09-07

*** spatel has joined #openstack-ansible00:18
*** gyee has quit IRC00:21
*** threestrands has joined #openstack-ansible00:32
*** threestrands has quit IRC00:32
*** threestrands has joined #openstack-ansible00:32
*** spsurya has quit IRC00:35
*** spartakos has quit IRC00:45
*** dave-mccowan has joined #openstack-ansible00:50
*** spatel has quit IRC00:55
*** hamzy has joined #openstack-ansible00:55
*** spatel has joined #openstack-ansible00:56
*** Emine has quit IRC01:11
*** Emine has joined #openstack-ansible01:11
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_zaqar master: add the project source code repository  https://review.openstack.org/60059501:13
*** markvoelker has joined #openstack-ansible01:20
openstackgerritlvxianguo proposed openstack/openstack-ansible-openstack_hosts master: add the project source code repository  https://review.openstack.org/60059601:22
*** markvoelker has quit IRC01:24
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_monasca-ui master: add the project source code repository  https://review.openstack.org/60059701:26
openstackgerritlvxianguo proposed openstack/openstack-ansible-rsyslog_server master: add the project source code repository  https://review.openstack.org/60059801:30
openstackgerritlvxianguo proposed openstack/openstack-ansible-lxc_container_create master: add the project source code repository  https://review.openstack.org/60060001:33
openstackgerritlvxianguo proposed openstack/openstack-ansible-rabbitmq_server master: add the project source code repository  https://review.openstack.org/60060101:36
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_nova master: add the project source code repository  https://review.openstack.org/60060201:40
*** markvoelker has joined #openstack-ansible01:41
*** EmilienM has left #openstack-ansible01:47
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_monasca-agent master: add the project source code repository  https://review.openstack.org/60060401:48
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_nova master: add the project source code repository  https://review.openstack.org/60060201:49
spatelcloudnull: ^^01:49
openstackgerritlvxianguo proposed openstack/openstack-ansible-repo_server master: add the project source code repository  https://review.openstack.org/60060501:56
openstackgerritMerged openstack/openstack-ansible master: Use a common python build/install role  https://review.openstack.org/57902101:59
openstackgerritlvxianguo proposed openstack/openstack-ansible-repo_build master: add the project source code repository  https://review.openstack.org/60060702:00
openstackgerritMerged openstack/openstack-ansible-nspawn_hosts master: Allow user-defined extra distro packages  https://review.openstack.org/59414202:04
openstackgerritlvxianguo proposed openstack/openstack-ansible-rsyslog_client master: add the project source code repository  https://review.openstack.org/60060902:08
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_masakari master: add the project source code repository  https://review.openstack.org/60061102:11
*** Bhujay has joined #openstack-ansible02:12
openstackgerritlvxianguo proposed openstack/openstack-ansible-apt_package_pinning master: add the project source code repository  https://review.openstack.org/60061202:14
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_sahara master: add the project source code repository  https://review.openstack.org/60061502:28
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_swift master: add the project source code repository  https://review.openstack.org/60061702:30
*** macza has joined #openstack-ansible02:32
openstackgerritlvxianguo proposed openstack/openstack-ansible-pip_install master: add the project source code repository  https://review.openstack.org/60061802:36
openstackgerritlvxianguo proposed openstack/openstack-ansible-os_molteniron master: add the project source code repository  https://review.openstack.org/60061902:38
openstackgerritlvxianguo proposed openstack/openstack-ansible-lxc_hosts master: add the project source code repository  https://review.openstack.org/60062002:42
*** Bhujay has quit IRC02:56
*** macza has quit IRC02:57
*** macza has joined #openstack-ansible02:58
*** spatel has quit IRC03:18
openstackgerritfupingxie proposed openstack/openstack-ansible-os_neutron master: switch documentation job to new PTI  https://review.openstack.org/60063003:19
*** gkadam has joined #openstack-ansible03:21
*** macza_ has joined #openstack-ansible03:22
*** macza has quit IRC03:22
openstackgerritfupingxie proposed openstack/openstack-ansible-os_octavia master: switch documentation job to new PTI  https://review.openstack.org/60063103:24
*** dave-mccowan has quit IRC03:25
openstackgerritMerged openstack/openstack-ansible-os_nova master: Revert region name option to os_region_name for Cinder  https://review.openstack.org/59657603:32
*** nicolasbock has quit IRC03:36
*** poopcat has quit IRC04:01
*** udesale has joined #openstack-ansible04:05
*** spsurya has joined #openstack-ansible04:10
*** Bhujay has joined #openstack-ansible04:17
*** Bhujay has quit IRC04:18
*** Bhujay has joined #openstack-ansible04:19
*** spartakos has joined #openstack-ansible04:29
openstackgerritMerged openstack/openstack-ansible master: Move bionic jobs to voting, add release note.  https://review.openstack.org/59492604:29
openstackgerritMerged openstack/openstack-ansible master: Update docs & upgrade scripts for Stein  https://review.openstack.org/59856204:38
*** spartakos has quit IRC04:57
*** fghaas has joined #openstack-ansible05:03
*** KeithMnemonic1 has joined #openstack-ansible05:04
*** macza_ has quit IRC05:06
*** KeithMnemonic has quit IRC05:07
*** spartakos has joined #openstack-ansible05:12
*** shyamb has joined #openstack-ansible05:13
*** shyamb has quit IRC05:17
*** fghaas has quit IRC05:21
*** shyamb has joined #openstack-ansible05:34
*** aicilamd has quit IRC05:41
*** gkadam_ has joined #openstack-ansible05:46
*** gkadam has quit IRC05:48
*** biffi has joined #openstack-ansible06:00
*** pcaruana has joined #openstack-ansible06:13
*** lvdombrkr has joined #openstack-ansible06:19
*** rgogunskiy has joined #openstack-ansible06:24
*** shyamb has quit IRC06:24
*** shyamb has joined #openstack-ansible06:24
*** lvdombrkr89 has joined #openstack-ansible06:24
*** lvdombrkr has quit IRC06:27
*** lvdombrkr89 has quit IRC06:28
*** lvdombrkr has joined #openstack-ansible06:28
*** fghaas has joined #openstack-ansible06:42
*** macza has joined #openstack-ansible06:42
evrardjpgood morning06:43
*** macza has quit IRC06:46
evrardjpwow rc1 of 2.7 is available and we are not yet in 2.606:47
evrardjpshould we jump?06:47
jrosserevrardjp: Given how we’ve had bug fixes to “older” Ansibles denied merging when those versions were still very much in use in OSA, imho it pays to keep up with their releases06:55
evrardjpjrosser: yeah, ofc, that's not what I mean -- My goal was to move to 2.606:56
evrardjpsoon06:56
evrardjpbut now I am thinking we should skip06:56
jrosserPerhaps what I mean there for today is that we should be on 2.6 for master at the very least06:56
evrardjpon that I agree06:56
evrardjpwell in fact I think we should be on 2.6 in Rocky but it didn't happen06:57
evrardjptoo many things to track06:57
jrosserThere are some ansible-next jobs somewhere aren’t there?06:57
*** shyamb has quit IRC07:00
jrosserevrardjp: my bionic reno patch won’t cherry pick to rocky - I suspect there may be a bunch of other changes which haven’t gone back too.07:00
evrardjpjrosser: yes, in experimental... probably need some adaptations, but yeah07:00
jrosserNot bionic specific, probably fiddling of the test config07:01
*** blinkiz has quit IRC07:01
evrardjpjrosser: git-deps?07:03
evrardjphttps://github.com/aspiers/git-deps07:03
hwoaranggood morning07:04
evrardjpmorning hwoarang07:04
*** shyamb has joined #openstack-ansible07:10
*** shyamb has quit IRC07:34
hwoarangmight actually make sense to dedicate some time during the stein release to see if we can upstream anything from our -plugins repo. that will make moving to newer ansible much easier07:38
jrosserhaving only one copy of config_template that we maintain, regardless of where it is would be nice too07:45
jrosserthere are currently two and they are diverged code07:45
*** radeks has joined #openstack-ansible07:50
*** hamzaachi has joined #openstack-ansible07:51
openstackgerritwangqi proposed openstack/openstack-ansible-os_cinder master: Drop un-used packages from role  https://review.openstack.org/60066907:53
openstackgerritwangqi proposed openstack/openstack-ansible-os_masakari master: Drop un-used packages from role  https://review.openstack.org/60067007:56
*** tosky has joined #openstack-ansible07:59
*** shardy has joined #openstack-ansible08:08
*** electrofelix has joined #openstack-ansible08:14
*** ethfci has joined #openstack-ansible08:16
*** luksky has joined #openstack-ansible08:16
*** spartakos has quit IRC08:18
*** olivierb has joined #openstack-ansible08:29
jrosserevrardjp: i'm a bit stuck - spent an hour trying to untangle all changes that have been made in zuul.d between master and stable/rocky, it's a real mess08:32
*** DanyC has joined #openstack-ansible08:32
*** threestrands has quit IRC08:33
*** hamzaachi has quit IRC08:34
*** hamzaachi has joined #openstack-ansible08:35
*** shyamb has joined #openstack-ansible08:45
odyssey4meevrardjp: what is holding us back from using the independent repo config_template? It's been in place for months now, but not used08:49
evrardjpnothing -- just time to change that: it needs change in meta, and a change in plugins path08:50
odyssey4mewell, can you please either put what needs doing in a bug, or get to doing it - right now you're the only one, as far as I know, who knows what needs to be done08:51
evrardjpodyssey4me: I hope I can hack it around during ptg08:52
evrardjpyeah will do08:52
odyssey4meI know cloudnull made an attempt in https://review.openstack.org/581166 - but that way of putting it in didn't seem to match what you had told me previously08:54
odyssey4mecores, I need reviews for https://review.openstack.org/#/c/599682/ - that will need backporting to rocky to make queens->rocky works08:55
evrardjpodyssey4me: https://bugs.launchpad.net/openstack-ansible/+bug/179125808:57
openstackLaunchpad bug 1791258 in openstack-ansible "Make use of ansible-config_template repo" [Undecided,New]08:57
odyssey4metyvm evrardjp hwoarang09:05
odyssey4meevrardjp: oh I see - we need to put it in the plugins path, then change all the roles to require it as a meta dep, then we can remove it from the plugins path09:06
evrardjpodyssey4me: well the tricky part is not there09:09
evrardjpthe tricky part is that at the same time we have to remove from plugins and add from <new location> to prevent conflicts09:10
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Update ansible to latest stable 2.6.1  https://review.openstack.org/58116609:11
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Update ansible to latest stable 2.6.x  https://review.openstack.org/58116609:11
odyssey4meevrardjp: surely the plugins are resolved in order of the path - first one wins?09:13
odyssey4meso if we add the new one first in the path, then we're good09:13
hwoarangi am curious about the healthcheck playbooks. are we supposed to run them after initial deployment? or on regular basis? how do you normally use them?09:14
odyssey4mehwoarang: currently I'm using them to validate a deployment - they clean up after themselves, assuming everything worked, so it's good for that09:18
odyssey4meit's not perfect, I think there are some gaps - but it's really nice to have something understandable to health check the environment09:19
hwoarangodyssey4me: ok thank you. the healthcheck-hosts misses tags. so it have any?09:19
odyssey4mewe should put some docs in place09:19
odyssey4meyeah, some of the tagging needs to be made more consistent09:19
hwoarangok09:19
odyssey4mefor example 'healthchecks-teardown-heat' vs 'healthchecks-heat-install'... if it was consistently 'healthchecks-<service>-<action>' it would be more intuitive09:20
odyssey4meservice and action could be the other way around, of course09:21
evrardjpodyssey4me: I think I need to work on the plugins story -- having a poc in my place. I will try to do that in the plane.09:22
evrardjpI think we should do one step at a time, first the config_template bringing, then upgrading ansible :)09:23
evrardjpwe never know what we might hit.09:23
evrardjpAnd in terms of priorities, I think it's not so simple.09:23
odyssey4meevrardjp: I think the plugins repo has 2.6 compatibility now, so I'll modify the patch to remove the config_template changes09:23
evrardjpit loads the things in a certain order that can help for having dependencies, but I am not sure how it deals with reloading on existing09:24
odyssey4mehttps://github.com/openstack/openstack-ansible-plugins/commit/5c70fb05d8e93a639c5e875f2b28270d541de7a809:24
evrardjpthat's why I will try locally with ANSIBLE_DEBUG09:24
evrardjpyeah focusing on 2.6 or 2.7 (because 2.7 rc1 was just tagged)09:24
evrardjpis a good idea09:24
evrardjpmore important than bringing config_template09:24
evrardjpbut yeah, reducing this tech debt, and having more ppl around config_template sounds very good in the long run09:25
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Update ansible to latest stable 2.6.x  https://review.openstack.org/58116609:26
hwoarangoh oh09:27
odyssey4meQuite honestly, if this works - I'd rather we release Rocky with 2.6 than 2.5.09:27
odyssey4me2.5 will go unmaintained soon, and I'd rather be fixing 2.6 bugs in early rocky releases than be stuck with an unmaintained ansible.09:28
noonedeadpunkmornings everyone09:37
noonedeadpunkjust fyi: pattern in ansible find module MUST be a list. Otherwise, it will be splitted by "," once found (even if it's used inside regexp)09:39
*** nicolasbock has joined #openstack-ansible09:39
jrossercan i get a second opinion on what should be done about resolving the differences between the contents of zuul.d between master and rocky? Backports don't seem straightforward09:40
*** nicolasbock has quit IRC09:40
hwoarangjrosser: why aren't backports any option? is that because commits in zuul.d also change files outside of that dir?09:43
openstackgerritDmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_horizon master: Pattern for find module was optimized  https://review.openstack.org/60068109:43
jrossersome of them are the "switch tempest to/from master" which also fiddled in zuul.d09:43
hwoarangif backport is not an option then maybe make a single commit an explain why it is not a backport in the commit message09:44
hwoarangmaybe we also need this tempest change in rocky too?09:44
jrosserthen the entire thing has been refactored wholly, and that commit wont backport either09:44
hwoarangok then prob a single commit then09:44
jrosserperhaps - thats why i want a seond opinion - i don't understand that stuff well enough09:44
openstackgerritMerged openstack/openstack-ansible-os_almanach stable/rocky: Update branch to point to Rocky  https://review.openstack.org/59977909:48
noonedeadpunkcores, if you have some time, can you say what do you think about https://review.openstack.org/#/c/600072/ ? The question is, wether ?ssl=0 is required at the end of the transport, as in queens it breaks messaging in ceilometer09:48
*** shyamb has quit IRC09:54
odyssey4mejrosser: yeah, I think ideally we'll need to port all the job changes back in a single commit - with reference to the sources09:55
openstackgerritMerged openstack/openstack-ansible-os_watcher stable/rocky: Update branch to point to Rocky  https://review.openstack.org/59981209:56
odyssey4meI've got to go afk for a bit, but will be happy to look at a review when I get back - or push up a review if you're stuck.09:58
*** shyamb has joined #openstack-ansible09:59
jrosserok, the thing that is concerning me is that there could well be other stuff on master thats not been backported that should be09:59
odyssey4mejrosser: yep, that's why I typically backport immediately after merge - it keeps these things easier to track10:00
*** macza has joined #openstack-ansible10:00
jrosserthe problem is here it's taken a couple of weeks to get my patch merged, during which time the branches have diverged10:01
*** macza has quit IRC10:04
*** shyamb has quit IRC10:04
openstackgerritMerged openstack/openstack-ansible-os_zaqar stable/rocky: Update branch to point to Rocky  https://review.openstack.org/59981310:21
*** shyamb has joined #openstack-ansible10:29
evrardjpodyssey4me: agreed on 2.6 story10:31
evrardjpin R.10:31
evrardjpI haven't proposed rc3 yet due to mnaser patches pending10:31
*** shyamb has quit IRC10:36
*** radeks_ has joined #openstack-ansible10:38
*** radeks has quit IRC10:41
*** dave-mccowan has joined #openstack-ansible10:52
*** shyamb has joined #openstack-ansible11:12
*** udesale has quit IRC11:16
*** jamesdenton has joined #openstack-ansible11:20
*** shyamb has quit IRC11:24
*** shyamb has joined #openstack-ansible11:24
*** nicolasbock has joined #openstack-ansible11:29
*** lvdombrkr has quit IRC11:30
*** hamzaachi has quit IRC11:31
spotzhwoarang: Your batch from this morning are reviewed, off to ride11:37
*** lvdombrkr has joined #openstack-ansible11:47
*** lvdombrkr89 has joined #openstack-ansible11:50
*** lvdombrkr has quit IRC11:52
odyssey4mehmm, any idea why all opensuse tests are hitting a retry limit? there are tons of them11:55
odyssey4meum, that 2.6 patch is failing with tmpfiles failures when trying to attach to containers: http://logs.openstack.org/66/581166/10/check/openstack-ansible-deploy-aio_lxc-ubuntu-bionic/44fbd2b/logs/ara-report/result/414867d8-f578-40b8-a288-4fa3b2083324/11:56
odyssey4meno containers just works11:57
jrosserodyssey4me: it's before it trys to attach - thats just trying to get the lxc version11:57
jrosserso its as if its ignored the delegate to on the task11:57
odyssey4mesame for bionic and xenial: http://logs.openstack.org/66/581166/10/check/openstack-ansible-deploy-aio_lxc-ubuntu-xenial/c4c7c9a/logs/ara-report/result/5d0aab33-8598-4b5b-84d2-54ca6a644561/11:57
jrosseri was just starting to dig, it's here https://github.com/openstack/openstack-ansible-lxc_container_create/blob/master/tasks/main.yml#L16-L2211:58
odyssey4mehmm, I wonder whether that's an issue with the strategy or connection plugin11:58
odyssey4mejrosser: did you give up on porting those zuul job changes back? if so, then I'll have a go while you try and figure out the issue with ansible 2.612:19
jrosserodyssey4me: oh sorry, yes, :( got n+1 simultaneous bonfires to fight today12:21
odyssey4mejrosser: okie dokey, getting on it now12:21
*** lvdombrkr89 has quit IRC12:22
*** lvdombrkr has joined #openstack-ansible12:22
jrosserodyssey4me: it was really having a good idea about which of these were/were not important for rocky https://github.com/openstack/openstack-ansible/commits/master/zuul.d12:22
openstackgerritMerged openstack/openstack-ansible-openstack_openrc stable/ocata: templates: openrc: Use 'true' and 'false' for insecure env variables  https://review.openstack.org/60005612:28
mgariepygood morning everyone12:30
noonedeadpunkmorning mgariepy12:30
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Refactor jobs to use project-templates  https://review.openstack.org/60072412:35
odyssey4melooks like https://github.com/openstack/openstack-ansible/commit/50712078a0c4a2763db6e7cbb844a17b8ff246a9#diff-db6e7f5498c58bb0e01d1c9d866c88ab is the missing link, and that's for stein only, so we need to leave it out12:35
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Move bionic jobs to voting, add release note.  https://review.openstack.org/60072512:36
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Refactor jobs to use project-templates  https://review.openstack.org/60072412:38
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Add aio host conf for barbican role  https://review.openstack.org/59963712:38
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Refactor jobs to use project-templates  https://review.openstack.org/60072412:39
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: zuul: Add ceph jobs for distribution installations  https://review.openstack.org/60072612:39
odyssey4methere we go - with those patches, master and rocky will have the same tests12:40
odyssey4mebarring that new documentation PTI template12:40
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Add aio host conf for barbican role  https://review.openstack.org/59963712:41
odyssey4mejrosser: mind reviewing https://review.openstack.org/599597 quickly?12:45
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/queens: Add aio host conf for barbican role  https://review.openstack.org/59964112:48
*** shyamb has quit IRC12:49
odyssey4meFrankZhang: I hope you don't mind, I've updated https://review.openstack.org/#/q/I10a07d5e4b2b3fbf122a79f15e27a9ca34799221 - for rocky, to rebase on top of the backport which brings the project-templates back, and for queens to re-fit the job config according to how it was done in queens.12:49
*** gkadam_ has quit IRC12:50
odyssey4mecombined backports should be avoided where possible12:51
odyssey4me(it makes it much harder to compare branch commit histories)12:51
odyssey4mejrosser: to your earlier question about what we want as jobs for master/rocky... ideally, especially before M1, we want them to be like for like... it makes porting back a lot easier12:52
odyssey4meI've always tried to keep master and the latest stable branch as close as possible until M1 across all repositories - that way any bugs found are easily solved in master and easily ported back without having to do a full re-implementation.12:53
odyssey4meOne would hope that by M1 we have a relatively stable code-base and can allow more divergance as features start getting put in.12:54
*** hamzaachi has joined #openstack-ansible12:56
*** ansmith has joined #openstack-ansible13:04
openstackgerritAndy Smith proposed openstack/openstack-ansible master: Add qdrouterd role for oslo.messaging backend deployment  https://review.openstack.org/57026113:07
odyssey4meAlthough I'm totally breaking that mould right now with the python build simplification work.13:07
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Ensure package cache is updated for nspawn containers  https://review.openstack.org/58843013:10
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Minimal(ist) network config for nspawn gating  https://review.openstack.org/59323013:10
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Add nspawn to experimental jobs  https://review.openstack.org/60073913:10
*** spatel has joined #openstack-ansible13:15
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Clean up role tests  https://review.openstack.org/60074113:16
*** gkadam has joined #openstack-ansible13:19
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Clean up role tests  https://review.openstack.org/60074113:22
spatelFolks, I have created group of compute nodes and i want to find out how much compute power i have in X and Y group is there a easy way in horizon?13:24
*** goldenfri has joined #openstack-ansible13:26
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: import zuul job settings from project-config  https://review.openstack.org/59423313:27
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: switch documentation job to new PTI  https://review.openstack.org/59423413:27
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Clean up role tests  https://review.openstack.org/60074113:29
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Clean up role tests  https://review.openstack.org/60074113:30
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: Clean up role tests  https://review.openstack.org/60075913:32
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican master: Clean up role tests  https://review.openstack.org/60076113:34
*** macza has joined #openstack-ansible13:36
*** goldenfr_ has joined #openstack-ansible13:40
*** goldenfr_ is now known as goldenfri_13:41
*** macza has quit IRC13:41
*** goldenfri has quit IRC13:44
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Clean up role tests  https://review.openstack.org/60076813:45
*** aicilamd has joined #openstack-ansible13:45
*** gkadam has quit IRC13:47
*** hamzaachi has quit IRC13:49
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate master: Clean up role tests  https://review.openstack.org/60077013:49
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi master: Clean up role tests  https://review.openstack.org/60077413:51
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat master: Clean up role tests  https://review.openstack.org/60077813:53
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon master: Clean up role tests  https://review.openstack.org/60077913:55
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic master: Clean up role tests  https://review.openstack.org/60078013:58
*** spatel has quit IRC14:05
*** spatel has joined #openstack-ansible14:06
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone master: Clean up role tests  https://review.openstack.org/60078114:07
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum master: Clean up role tests  https://review.openstack.org/60078214:11
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_masakari master: Clean up role tests  https://review.openstack.org/60078414:13
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_monasca-agent master: Clean up role tests  https://review.openstack.org/60078614:16
*** lbragstad is now known as elbragstad14:18
mnaserodyssey4me: i'm loving those cleanups14:20
odyssey4memnaser: yeah, loads of cruft left over - will do the rest after this conf call14:23
odyssey4meI intend to port those back to rocky too, so that we have an easier time porting job changes and such.14:23
*** biffi has quit IRC14:26
kaiokmofunctional-opensuse-423 is failing with "RETRY_LIMIT" among many patches. any specific reason for this to happen?14:28
hwoaranginfra mirrors do not seem up2date i am afraid14:30
*** pcaruana has quit IRC14:33
nicolasbockhwoarang, how are they synced?14:34
mnasernicolasbock: you can search openstack-infra/system-config14:35
mnaseror talk to infra-core in #openstack-infra14:35
openstackgerritMerged openstack/openstack-ansible stable/queens: Make sure the branch to track is in YAML  https://review.openstack.org/59959714:35
nicolasbockThanks mnaser14:36
hwoarangnicolasbock: rsync from a mirror on kernel.org14:40
hwoarangbut that mirror is also outdated. i have no clue what's going on and i see no info from the opensuse side apart from rsync.opensuse.org hw being broken14:40
hwoarangbut downloads.opensuse.org is updated so there must be a valid mirror somewhere14:41
nicolasbockhmm, you mean that rsync.opensuse.org doesn't even reply to ICMP?14:42
*** lvdombrkr89 has joined #openstack-ansible14:50
noonedeadpunkevrardjp: if you're still here, is ceilometer patch https://review.openstack.org/#/c/600072/ ok for you now?14:51
*** goldenfri_ has quit IRC14:52
*** spartakos has joined #openstack-ansible14:52
noonedeadpunkI've checked in zuul logs, that construction of url is correct14:52
*** lvdombrkr has quit IRC14:52
evrardjpnoonedeadpunk: that is very pleasant to read :)14:55
evrardjpnoonedeadpunk: so ?ssl=0 doesn't break notifications messaging_urls?14:56
FrankZhangodyssey4me: thanks for the update, appreciate it14:57
evrardjpansmith: do you mind having a look at https://review.openstack.org/#/c/600072/ ? it is to know if messaging_urls can receive the query ?ssl=014:59
noonedeadpunkevrardjp: actually I still don't have straight answer for this, as on queens it breaks transport_url as well and all ceilometer components stops sending notifications at all. I may set workflow -1 for now and doble check it. But I hoped that it has been already tested when ?ssl was implemented....14:59
*** goldenfri has joined #openstack-ansible14:59
evrardjpnoonedeadpunk: oh weird, transport url seems to be accepting query parameters15:00
ansmithevrardjp, sure thing, i will check it out15:00
evrardjphttps://docs.openstack.org/oslo.messaging/latest/configuration/opts.html#DEFAULT.transport_url15:00
evrardjpnoonedeadpunk: ^15:00
evrardjpansmith: thanks you're the man -- and quite the messenger!15:00
evrardjpansmith: Ok that was bad, but it's friday.15:01
evrardjpthanks !15:01
ansmithhappy friday :)15:01
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible stable/rocky: scripts: Fix getting external role SHA when working on master  https://review.openstack.org/60079415:01
*** gkadam has joined #openstack-ansible15:01
evrardjpnoonedeadpunk: if it doesn't work like this, we can replace the parameter "use_ssl" by "query_parameters", and pass things appropriately, depending on the backend driver -- should we consider multiple backends15:02
goldenfriHi, I've been beating my head against a wall trying to understand the eth12 networking on a simple infra, compute and storage test setup, would be be possible for someone to have a look at my network and userconfig to see if I am doing something dumb?15:03
*** gyee has joined #openstack-ansible15:06
goldenfricompute1 - http://paste.openstack.org/show/729690/15:07
goldenfriInfra - http://paste.openstack.org/show/729691/15:07
goldenfriuser_config - http://paste.openstack.org/show/729692/15:07
goldenfriIf anyone has time to take a peak, thank you!15:07
*** fghaas has quit IRC15:07
evrardjpmnaser: spotz should we settle for a date for the OSA team dinner? right now not many votes and wednesday is winning. We can then book something and say "Who wants to come, comes" :)15:08
evrardjpI propose wednesday therefore15:08
evrardjpWith extra people from helm/loci we could have a nice booking if we do it early I guess :)15:08
spotzYeah evrardjp mnaser - only3 OSA stuff filled out the form but we're up to like 8 in the groupme15:08
mnaseri personally wouldn't be able to do wednesday because of prior stuff, but all the other days are open for me :X15:09
mnaseralso i promised to get a schedule up but silly google apps restricted sharing calendars15:09
mnaserand changing that settting takes 48 hours to apply15:09
mnaserTHANKS GOOGLE15:09
mnaseradded my self on that list15:09
evrardjpwell you're the leader, but board game night is sacred for me :D15:10
evrardjptuesday would be great then15:10
noonedeadpunkevrardjp: yeah, proably this is worth doing. but the question is how to set it. Then we need more modifications of this role with adding some extra var. I'd suggest not to set ?ssl=015:10
evrardjpnoonedeadpunk: so you mean it's not working with ?ssl=0 right now?15:11
noonedeadpunkAnd I've checked one more time - with ?ssl=0 I didn't recieve any notification in ceilometer queue for 10 minutes15:11
noonedeadpunkevrardjp: yep, exactly.15:11
evrardjpok so we have to remove it -- right now the patch will break things15:11
evrardjpyeah so I'd say we change that query parameter15:11
noonedeadpunkevrardjp: actually they might be already broken for some time, as without patch we still have ?ssl=015:12
evrardjpnoonedeadpunk: not for messaging_urls15:12
evrardjpit's only for transport so it seems15:12
evrardjp(talking about current state in master)15:13
noonedeadpunkevrardjp: setting it for transport_url we don't have any notification from pollsters...15:13
noonedeadpunkyeah, I'm going just to set empty value for disabled ssl.15:14
evrardjpnoonedeadpunk: funny it seems that it's not set in Q: neither in messaging_urls nor transport_url15:14
noonedeadpunkevrardjp: you're right - it was added when moving to hybrid messaging15:15
spotzDepending on when board games start we could eat first then play, but Tuesday works15:15
noonedeadpunkok, I may really deploy AIO and test it for master. And till then workflow -115:16
noonedeadpunkprobably smth has changed since Q15:16
evrardjpnoonedeadpunk: yeah, we implemented the move to oslo messaging with ansmith -- that's why I was asking him. I think it's worth saying "query parameters" and where appropriate we pass said query parameters :)15:17
evrardjpso right now we could just say nothing, and when appropriate, passing the right values.15:17
evrardjpspotz: my vote goes on tuesday -- as I am not aware of any activity there.15:18
spotzevrardjp: Well the main happy hour event is Tuesday but I'm dieting:)15:18
evrardjpoh... I see15:18
evrardjpyeah I'd rather be there too -- do you know when it ends?15:19
spotzevrardjp: Let me see if it's in Kendall's email15:21
*** shardy has quit IRC15:21
noonedeadpunkevrardjp: yeah, its the way.15:22
spotzOk official even times -  networking reception Tuesday, 5:00-7:00pm , trivia night Wednesday, 8:30-10:30pm and game night Thursday, 8:30-11:30pm15:23
spotzI'm gonna go RSVP for trivia:)15:23
*** komish has joined #openstack-ansible15:24
evrardjpansmith: oh you won't be able to be at the PTG? (I saw remote on the agenda)15:25
*** kwu83tw has joined #openstack-ansible15:25
*** fghaas has joined #openstack-ansible15:25
ansmithevrardjp, yes, unfortunately i will not be there, i would like to attend remotely if possible15:25
*** gkadam has quit IRC15:26
*** shyamb has joined #openstack-ansible15:26
*** fghaas has quit IRC15:27
noonedeadpunkevrardjp: but it doesn't resolve question for query use_ssl=0 for transport_url15:28
evrardjpnoonedeadpunk: well we could simplify that discussion in two steps: 1) do exactly what we do now with the flexibility and simplicity you brought 2) change the query parameters when we feel necessary15:29
ansmithevrardjp, noonedeadpunk, looking at how ceilometer uses messaging_urls versus the base transport_url to understand differences if any15:29
mnaserso how do we feel about having it myabe thursday before game night i guess15:30
evrardjpmnaser: I am fine with that15:30
evrardjplet's just leave early and be back early, or eat close :D15:31
mnasersounds good to meeEeEe15:31
evrardjpjoke aside, it's not up to me,I am just giving opinion as there is only 3 votes :p15:31
evrardjpor 415:31
evrardjpI hope to make the flame burn brighter, but it doesn't seem to work :D15:32
noonedeadpunkansmith: the problem now is not in messaging_urls, but with transport_url. I really don't know as for now whether smth has changed in ceilometer in Rocky (comparing to Queens) regarding messaging, but in queens ceilometer stops sending any notifications, once has ?ssl=0 in transport_url15:33
evrardjpmnaser: I'd like to have a discussion on where we are for the rc3 release during ptg -- if possible hack things around to get there during ptg :)15:36
mnaserevrardjp: yeah that's cool, if you wanna add that to the proposals and ill try to put that early15:37
mnaserwe'll have an empty gate too :P15:37
spotzmnaser evrardjp - Not sure how close the places I put down are, but I think we should be fine. There wasn't really anything decent by the hotel15:37
*** luksky has quit IRC15:38
noonedeadpunkansmith so currently the question is how safe is it to remove ssl=0 from transport_url?15:41
*** kwu83tw has quit IRC15:42
ansmithnoonedeadpunk, i believe it should be safe as if query param is not there rabbit driver will use ssl option in [oslo_messaging_rabbit] section15:43
evrardjpansmith: noonedeadpunk oh yeah I seem to remember that :)15:44
*** aicilamd has quit IRC15:44
ansmithnoonedeadpunk, i will check the driver but i believe query param is present, driver will use that instead15:44
evrardjpit would simplify things at the end then15:44
*** ppetit has joined #openstack-ansible15:44
*** rgogunskiy has quit IRC15:45
evrardjphello ppetit :)15:45
*** kwu83tw has joined #openstack-ansible15:45
ppetithello evrardjp !15:45
*** macza has joined #openstack-ansible15:46
*** shyamb has quit IRC15:47
ppetitI am confused about release versioning. For example, the doc says to checkout tag 17.1.0 for queens instead of stable/queens why?15:48
ppetitI thought that tags have more recent commits but it's not the case apparently...15:49
ppetitIf someone can shed some light on this... It'd be appreciated.15:50
*** macza_ has joined #openstack-ansible15:50
*** macza has quit IRC15:51
*** lvdombrkr89 has quit IRC15:57
*** Bhujay has quit IRC15:58
openstackgerritDmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Changed messaging_urls for notification agent  https://review.openstack.org/60007216:03
*** dcdamien has quit IRC16:05
*** olivierb has quit IRC16:24
*** spartakos has quit IRC16:27
*** ppetit has quit IRC16:29
*** DanyC has quit IRC16:31
evrardjpppetit if you watch the logs later: we advise people to consume from tags generally, as branches are moving targets. However, stable branches have always been nice to me. Master is a different beast.16:40
noonedeadpunkStable branches might be tricky - thay can cause services upgrade when you don't expect this to happen16:53
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Cater for galera bootstrap without a master  https://review.openstack.org/60081816:54
noonedeadpunkhave a good weekends and ptg!)17:03
*** noonedeadpunk has quit IRC17:09
openstackgerritCam J. Loader proposed openstack/openstack-ansible-ops master: update ironic scripts  https://review.openstack.org/59584917:20
*** hamzaachi has joined #openstack-ansible17:21
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Clean up role tests  https://review.openstack.org/60082517:35
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron master: Clean up role tests  https://review.openstack.org/60082517:38
*** spartakos has joined #openstack-ansible17:39
*** goldenfri has quit IRC17:42
*** goldenfri_ has joined #openstack-ansible17:43
*** IAMJES has joined #openstack-ansible17:48
odyssey4meo/ IAMJES17:49
odyssey4mewhat's up, how can we help?17:49
* odyssey4me will try - it's friday evening, and I'm not necessarily useful depending on the topic17:49
IAMJESwell, from what i am reading is that this version of openstack can be installed on a 'clean' ubuntu 16.04 install with ssh17:50
openstackgerritGerman Eichberger proposed openstack/openstack-ansible-os_octavia master: Fixes too low security group rules quota  https://review.openstack.org/60082617:50
odyssey4meIAMJES: close to it, yes - ansible just needs the public ssh key you're using, and it can connect... then you need your networking/storage pre-setup, depending on what you're using17:51
IAMJESi have a small cluster of dell 1950 gen 3 and 2950 gen 3 servers i want to use to setup a environment...17:51
IAMJESif i can setup a 'prod' environment then i would get better hardware17:52
IAMJESthis url is for a production deployment? https://docs.openstack.org/project-deploy-guide/openstack-ansible/rocky/deploymenthost.html17:54
openstackgerritGerman Eichberger proposed openstack/openstack-ansible-os_octavia master: Fixes too low security group rules quota  https://review.openstack.org/60082617:54
odyssey4meIAMJES:  you may find the sample configs in https://docs.openstack.org/openstack-ansible/latest/user/index.html handy then :)17:55
*** electrofelix has quit IRC17:55
odyssey4mewell, perhaps https://docs.openstack.org/openstack-ansible/rocky/user/index.html is better :p17:55
odyssey4mebut yes, the deploy guide is a good read for understanding17:56
*** electrofelix has joined #openstack-ansible17:56
odyssey4mepossibly best to also peruse: https://docs.openstack.org/openstack-ansible/rocky/17:56
odyssey4meIAMJES: https://docs.openstack.org/openstack-ansible/rocky/reference/architecture/index.html for example is good background reading17:59
*** electrofelix has quit IRC17:59
IAMJESi have been looking at a high level, though right now my needs are small.. i will most likely deploy over 9 or 10 servers18:00
odyssey4meIAMJES: in that case, keep it simple - try to stick the the defaults where you can, as it'll give you a quicker chance of success in the short term18:01
*** goldenfri_ is now known as goldenfri18:01
odyssey4meideally, single out a beefy host to use for testing, as you'll need to do a lot of it - https://github.com/openstack/openstack-ansible-ops/tree/master/multi-node-aio is very useful for that if you have a beefy host18:01
IAMJESrome wasnt built in a day!18:01
IAMJESall the servers have 2tb and min 32gb ram and 2 x4 @318:03
odyssey4meIAMJES: I would recommend, especially while figuring things out, that you get a really good kick system going. You may find http://www.openstackfaq.com/ useful - admin0 uses that for his documentation. The stuff might sometimes be a bit dated, but it's good for reference.18:04
IAMJESon this AIO from github, once i get into it, is it terribly difficult to run a command to join the stack instead of pxe?18:06
odyssey4meIAMJES: I don't understand the question.18:07
IAMJESin the link for the all in one, how or where do i go to tell the compute nodes to join the stack without using PXE ?18:08
odyssey4meIAMJES: oh, the MNAIO is a testing tool for a single host - not a production deployment thing... it has a fixed set of VM's nodes which all join in18:10
odyssey4meit's there for testing and retesting quickly18:11
odyssey4meyou can change it up to pretend it's the deployment you want, once you know ansible and the code well enough18:11
*** IAMJES has quit IRC18:12
*** IAMJES has joined #openstack-ansible18:13
IAMJESsorry about that, what did i miss?18:13
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Clean up role tests  https://review.openstack.org/60083818:14
odyssey4me19:10 odyssey4me: IAMJES: oh, the MNAIO is a testing tool for a single host - not a production deployment thing... it has a fixed set of VM's nodes which all join in18:14
odyssey4me19:11 odyssey4me: it's there for testing and retesting quickly18:14
odyssey4me19:11 odyssey4me: you can change it up to pretend it's the deployment you want, once you know ansible and the code well enough18:14
IAMJESif i am reading this right, then all the commands are executed as root? https://docs.openstack.org/project-deploy-guide/openstack-ansible/rocky/deploymenthost.html18:24
odyssey4meIAMJES: yes, for the moment we do everything as root - until now non-root user execution has not been priority enough for anyone to solve18:27
*** hamzaachi has quit IRC18:30
*** fghaas has joined #openstack-ansible18:32
odyssey4meI'm out for the w/end. Have a great evening everyone!18:52
*** goldenfri has quit IRC18:55
*** macza_ has quit IRC18:56
*** hamzaachi has joined #openstack-ansible19:03
*** fghaas has quit IRC19:06
IAMJESi am looking at the openstack_user_config.yml - and what i would like to accomplish is for everything to use one network and virtual networks on top of the physical layer19:17
IAMJESthe whole thing is commented out and all the options are confusing19:18
*** macza has joined #openstack-ansible19:23
IAMJESi just want all the nodes to work over a 192.168.1.x19:23
IAMJESall of these options are vlan on top of eth0 ? cidr_networks:    container: 172.29.236.0/22    tunnel: 172.29.240.0/22    storage: 172.29.244.0/2219:26
*** jbadiapa has quit IRC19:36
IAMJESwhich file am i supposed to update to tell the playbooks the nodes to use?19:43
jrosserIAMJES: the OSA tooling expects that br-vlan, br-mgmt, br-storage etc exist on your hosts19:43
jrosserit is up to you to decide how that gets done to meet your particular use case, you need to prep the hosts to meet the pre-requsites in the documentation19:44
jrosserthose CIDR networks can just be vlans on eth0 as you say19:45
jrosserIAMJES: this is where you map the logical functions to as many/few physical hosts as you want to https://docs.openstack.org/project-deploy-guide/openstack-ansible/newton/app-config-prod.html19:46
jrosserargh....19:46
*** hamzaachi has quit IRC19:46
*** macza has quit IRC19:47
jrossersorry this is more up to date https://docs.openstack.org/openstack-ansible/rocky/user/prod/example.html see "environment layout"19:48
*** spartakos has quit IRC19:49
*** radeks__ has joined #openstack-ansible19:53
*** radeks_ has quit IRC19:54
IAMJESi am not able to get ansible to see the inventory.ini ---  Unable to parse /etc/openstack_deploy/inventory.ini as an inventory source20:01
*** DanyC has joined #openstack-ansible20:14
*** hamzaachi has joined #openstack-ansible20:16
*** fghaas has joined #openstack-ansible20:28
openstackgerritDavid Wilde proposed openstack/openstack-ansible-ops master: Deploy ELK in MNAIO  https://review.openstack.org/58661220:30
*** ansmith has quit IRC20:42
jrosserIAMJES: perhaps you could put some of your commands and the output on paste.openstack.org, might be some clues there20:42
*** fghaas has quit IRC20:42
*** luksky has joined #openstack-ansible20:54
*** spartakos has joined #openstack-ansible21:01
*** spartakos has quit IRC21:02
*** KeithMnemonic1 has quit IRC21:32
*** IAMJES has quit IRC21:37
*** tosky has quit IRC21:40
*** radeks__ has quit IRC21:41
*** spatel has quit IRC21:45
*** spatel has joined #openstack-ansible21:49
*** spartakos has joined #openstack-ansible21:57
*** luksky has quit IRC22:01
openstackgerritAntony Messerli proposed openstack/openstack-ansible-ops master: Move bootstrap-ansible and passwords to run_osa tag  https://review.openstack.org/60087522:01
*** kukacz_ has quit IRC22:02
*** kukacz_ has joined #openstack-ansible22:04
*** joseg has joined #openstack-ansible22:10
*** komish has quit IRC22:11
*** spartakos has quit IRC22:14
*** spartakos has joined #openstack-ansible22:15
*** spatel has quit IRC22:16
*** spartakos has quit IRC22:23
*** spartakos has joined #openstack-ansible22:28
*** dave-mccowan has quit IRC22:38
*** hamzaachi has quit IRC23:03
*** ansmith has joined #openstack-ansible23:16
*** spartakos has quit IRC23:17

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!