Tuesday, 2017-09-05

*** drifterza has joined #openstack-ansible00:22
*** drifterza has quit IRC00:23
*** thorst_afk has joined #openstack-ansible00:24
*** markvoelker has joined #openstack-ansible00:26
*** chas has joined #openstack-ansible00:49
*** chas has quit IRC00:53
*** thorst_afk has quit IRC00:55
*** markvoelker has quit IRC00:59
*** dxiri has quit IRC01:05
*** thorst_afk has joined #openstack-ansible01:39
*** thorst_afk has quit IRC01:39
*** dxiri has joined #openstack-ansible01:50
*** dxiri has quit IRC01:54
*** markvoelker has joined #openstack-ansible01:56
*** gouthamr has quit IRC02:08
*** gkadam has joined #openstack-ansible02:15
*** thorst_afk has joined #openstack-ansible02:17
*** gkadam has quit IRC02:17
*** thorst_afk has quit IRC02:17
*** gkadam has joined #openstack-ansible02:18
*** markvoelker has quit IRC02:30
*** dxiri has joined #openstack-ansible02:40
*** chas has joined #openstack-ansible02:49
*** chas has quit IRC02:54
*** thorst_afk has joined #openstack-ansible03:18
*** thorst_afk has quit IRC03:23
openstackgerritLogan V proposed openstack/openstack-ansible master: Set ceph_mons in group_vars  https://review.openstack.org/47032303:28
*** germs1 has quit IRC03:52
*** germs has joined #openstack-ansible03:53
*** germs1 has joined #openstack-ansible03:55
*** germs has quit IRC03:58
*** poopcat has quit IRC04:05
*** germs1 has quit IRC04:13
*** udesale has joined #openstack-ansible04:18
*** thorst_afk has joined #openstack-ansible04:19
*** thorst_afk has quit IRC04:23
*** drifterza has joined #openstack-ansible04:26
*** markvoelker has joined #openstack-ansible04:27
*** drifterza has quit IRC04:33
*** kristaps_ has quit IRC04:37
*** kristaps_ has joined #openstack-ansible04:37
*** chas has joined #openstack-ansible04:50
*** pcaruana has joined #openstack-ansible04:52
*** chas has quit IRC04:55
*** markvoelker has quit IRC05:00
*** thorst_afk has joined #openstack-ansible05:20
*** dxiri has quit IRC05:24
*** thorst_afk has quit IRC05:24
*** dxiri has joined #openstack-ansible05:25
*** dxiri has quit IRC05:29
*** a0x3e8 has joined #openstack-ansible05:40
*** lagman has joined #openstack-ansible05:41
*** yifei has joined #openstack-ansible05:55
*** markvoelker has joined #openstack-ansible05:58
*** jvidal has joined #openstack-ansible05:58
*** jvidal has quit IRC05:59
*** huxinhui_ has joined #openstack-ansible06:01
*** Oku_OS-away is now known as Oku_OS06:12
lagmanHi, Im trying to setup openstak's networking, and used the openstack example for test environment as a guid, but instead of vlans, I used physical interfaces, My network setup works, i have access to internal network, but I cant access the provider network. Do i have to setup a manual bridge between the network container and the second interface to which it is connected to my provider network(i.e internet).06:13
*** drifterza has joined #openstack-ansible06:17
*** thorst_afk has joined #openstack-ansible06:20
*** chas has joined #openstack-ansible06:22
*** thorst_afk has quit IRC06:25
*** markvoelker has quit IRC06:31
*** PTO has joined #openstack-ansible06:48
neithlagman: which test env?06:51
lagmanneith: the one explained on the openstack documents appendix A https://is.gd/ZnyRnt06:55
neithlagman: no word of the tenant network in this doc06:56
neithlagman: you could setup a flat network for vm external net06:57
neithlagman: something like https://docs.openstack.org/newton/install-guide-ubuntu/launch-instance-networks-provider.html06:58
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible stable/pike: roles: bootstrap-host: Fix typo in postup/postdown options for SUSE  https://review.openstack.org/50070007:14
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible stable/pike: roles: bootstrap-host: Remove useless 'break' command from case statement  https://review.openstack.org/50070107:14
lagmanneith: thanks for your answer07:17
*** thorst_afk has joined #openstack-ansible07:22
*** thorst_afk has quit IRC07:26
hwoaranggood morning07:27
hwoarangsome roles still have 'trusty' voting jobs. do we still need these?07:27
*** armaan has joined #openstack-ansible07:40
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible-os_nova stable/pike: Updated from OpenStack Ansible Tests  https://review.openstack.org/49689907:45
andymccrhwoarang: should only be on newton which will EOL in some months anyway07:57
andymccralso infra may take away trusty infra at some point - then we'll have to remove them.07:57
*** pbandark has joined #openstack-ansible07:59
andymccrhttps://review.openstack.org/#/c/500569/ & https://review.openstack.org/#/c/500574/ look ready to go!08:01
*** udesale has quit IRC08:03
*** shardy has joined #openstack-ansible08:03
*** ArchiFleKs has joined #openstack-ansible08:05
*** ArchiFleKs has quit IRC08:05
*** udesale has joined #openstack-ansible08:05
andymccrthat ceph centos7 job is really a pita.08:07
*** vnogin has joined #openstack-ansible08:15
*** hk135 has joined #openstack-ansible08:16
hk135Hi all, I have been looking around but I have been unable to find the answer, I was wondering if anyone could tell me how to configure neutrons bridge_mappings?08:17
hk135or point me to any documentation I might have missed08:17
*** ArchiFleKs has joined #openstack-ansible08:19
openstackgerritMerged openstack/openstack-ansible-os_swift master: Updated from OpenStack Ansible Tests  https://review.openstack.org/49728008:22
*** thorst_afk has joined #openstack-ansible08:22
*** ArchiFleKs has quit IRC08:26
*** thorst_afk has quit IRC08:27
*** udesale has quit IRC08:27
*** udesale has joined #openstack-ansible08:27
*** markvoelker has joined #openstack-ansible08:28
*** vnogin has quit IRC08:30
*** vnogin has joined #openstack-ansible08:31
PTOIm trying to deploy opstack using ansible and i have followed the very nice guide. However, I cant figure out how to get haproxy working. Am i supposed to manually configure the external lb ip on the nodes?08:35
neithPTO no, haproxy is configured by OSA if you chose so, otherwise yes08:36
PTOI have configured an external_lb_vip_address and haproxy_hosts. Everyting seems to deploy successfully, but the external_lb_vip_address is not responding08:39
*** electrofelix has joined #openstack-ansible08:40
PTOShould the external_lb_vip_address  be listed in the "ip addr show" command?08:43
neithanyone has a clue why setup-openstack.yml fails on:08:47
neithhttps://www.irccloud.com/pastebin/cqyZeFRC/08:47
neithPTO no08:49
neithPTO keepalived handles the  external_lb_vip_address08:49
PTOneith got it, and keepalived looks good. What to do when the e.g. horizon is not accessible on the external_lb_vip_address?08:53
*** markvoelker has quit IRC09:01
*** vnogin has quit IRC09:07
openstackgerritlidong proposed openstack/openstack-ansible-os_neutron master: Fix a typo  https://review.openstack.org/50074409:14
*** rodolof has joined #openstack-ansible09:16
*** armaan has quit IRC09:20
hwoarangandymccr: maybe not yet because we have some roles failing on pike :(09:22
*** vnogin has joined #openstack-ansible09:22
hwoaranghttps://review.openstack.org/#/c/496899/ https://review.openstack.org/#/c/496671/ and others09:22
hwoarangocata is having issues too :/09:23
*** thorst_afk has joined #openstack-ansible09:23
hwoarangi've also seen a report on 500http errors on ocata in the mailing list09:25
andymccryeah that fkn swift job - cant replicate the issue shown in those gates09:28
*** thorst_afk has quit IRC09:28
andymccrits not the lib-ec stuff, since we disabled that for swift tests.09:28
*** armaan has joined #openstack-ansible09:29
*** yifei has quit IRC09:39
*** shardy has quit IRC09:55
*** shardy has joined #openstack-ansible09:56
*** markvoelker has joined #openstack-ansible09:58
andymccrmhayden: could we have a dud repo for the ceph stuff? It fails on repo build: http://logs.openstack.org/69/500569/1/check/gate-openstack-ansible-openstack-ansible-ceph-centos-7-nv/eb33ee3/console.html - that doenst happen in my AIO so im guessing something is broken there.09:58
andymccr*doesnt - so im struggling to recreate10:00
andymccrthat upgrade nova job is weird - i'll look into that, its getting api db access errors.10:02
*** gparaskevas has joined #openstack-ansible10:14
*** stuartgr has joined #openstack-ansible10:16
*** thorst_afk has joined #openstack-ansible10:24
*** PTO has quit IRC10:25
*** thorst_afk has quit IRC10:28
*** markvoelker has quit IRC10:31
*** vnogin has quit IRC10:35
*** udesale__ has joined #openstack-ansible10:37
*** vnogin has joined #openstack-ansible10:38
*** vnogin has quit IRC10:39
*** udesale has quit IRC10:39
*** vnogin has joined #openstack-ansible10:39
*** udesale__ has quit IRC10:40
*** udesale has joined #openstack-ansible10:40
*** Oku_OS is now known as Oku_OS-away10:47
*** Oku_OS-away is now known as Oku_OS10:53
*** gkadam has quit IRC10:57
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible master: scripts: scripts-library.sh: Fix tracepath logging  https://review.openstack.org/50031211:08
*** smatzek has joined #openstack-ansible11:21
*** thorst_afk has joined #openstack-ansible11:25
*** vnogin has quit IRC11:25
*** jamesdenton has joined #openstack-ansible11:26
*** shardy is now known as shardy_lunch11:26
*** vnogin has joined #openstack-ansible11:27
*** markvoelker has joined #openstack-ansible11:29
*** thorst_afk has quit IRC11:33
*** jafeha has quit IRC11:33
*** epalper has joined #openstack-ansible11:44
*** DanyC has joined #openstack-ansible11:46
epalperhi everyone, i am trying to run xci in ha mode and seeing an error with os_neutron role "TASK [os_neutron : Copy common neutron config] *********************************11:46
epalperTuesday 05 September 2017  08:04:18 +0000 (0:00:00.760)       1:21:29.223 *****11:46
epalperchanged: [controller01_neutron_server_container-06db9dd1] => (item={'dest': '/etc/neutron/neutron.conf', 'src': 'neutron.conf.j2', 'config_type': 'ini', 'config_overrides': {}})11:46
epalperAn exception occurred during task execution. To see the full traceback, use -vvv. The error was: TypeError: a bytes-like object is required, not 'str'11:46
epalperfatal: [controller01_neutron_server_container-06db9dd1]: FAILED! => {"failed": true, "msg": "Unexpected failure during module execution.", "stdout": ""}"11:46
epalperHere is the full log: https://paste.ubuntu.com/25470969/11:47
epalperand verbose output: https://paste.ubuntu.com/25471609/11:47
epalpercan someone help me to find out the root cause of this issue ?11:48
hwoarangepalper: are you by any chance using python3.4?11:51
epalperhwoarang: I haven't changed any settings done by osa.11:52
hwoarangosa doesn't do much on that front11:53
hwoarangif you have python3 will use it (at least the version you have in opnfv)11:53
hwoarangso do you happen to have python3.4 around?11:53
hwoarangi presume this is the OSA that opnfv gives you?11:55
epalperhwoarang: i thought osa installs python on the target hosts and use it11:56
jamesdentonepalper Are you doing a config override in user_variables.yml?11:56
epalperjamesdenton: yes, it is overridden11:58
jamesdentonSo there was a bug related to this: https://bugs.launchpad.net/openstack-ansible/+bug/171134711:59
openstackLaunchpad bug 1711347 in openstack-ansible "Config file override mechanism broken" [Wishlist,New]11:59
epalperit's overwritten with: https://paste.ubuntu.com/25471824/11:59
jamesdentonYou might check to see if you're patched11:59
*** thorst_afk has joined #openstack-ansible12:00
jamesdentonPike should be patched. Master is not.12:01
*** markvoelker has quit IRC12:01
*** shardy_lunch is now known as shardy12:01
epalperjamesdenton: I'm using master12:04
openstackgerritTaseer Ahmed proposed openstack/openstack-ansible-specs master: Blueprint for Congress integration with OSA  https://review.openstack.org/49906612:04
epalpercan you cherrypick this change to master ?12:04
jamesdentonepalper You may be able to get away with the changes I made here for the time being, but I expect the dev team to get proper python3 support in this cycle. https://github.com/busterswt/openstack-ansible-plugins/commit/9253a1a47a8ee48182b7b44916c5efc5c707923712:05
admin0hi all .. i have a strange issue in one of my cluster.  as an admin, i can list images using horizon and via cli ( openstack image list ) .. as a tenant, i can list images via horizon, but not via the cli .. openstack image list gives a 404 ..  how do I troubleshoot ?12:05
*** dave-mccowan has joined #openstack-ansible12:05
epalperjamesdenton: thanks. let me try to use this one in my a-r-r file and see if it fixes the problem12:08
*** PTO has joined #openstack-ansible12:08
PTOWhat is the correct way to make a change to an openstack-ansible deployment? E.g. Reconfigure haproxy external interface?12:10
jamesdentonadmin0 Can you try to run with debug? openstack --debug image list12:11
*** markvoelker has joined #openstack-ansible12:16
*** huxinhui_ has quit IRC12:18
admin0jamesdenton, https://gist.github.com/a1git/783e0444ad3308c953c8a516a61d1e2112:19
admin0jamesdenton,  successful one as admin from utility container: https://gist.github.com/a1git/c9476c6d205b5b6b612acba7bdd7cfe012:20
admin0via tenant login to horizon, no issues12:21
admin0stable/ocata12:21
openstackgerritOpenStack Proposal Bot proposed openstack/openstack-ansible-os_swift master: Updated from global requirements  https://review.openstack.org/49523912:21
*** udesale has quit IRC12:25
*** armaan has quit IRC12:34
*** woodard has quit IRC12:35
*** armaan has joined #openstack-ansible12:35
*** woodard has joined #openstack-ansible12:35
*** epalper has quit IRC12:36
jamesdentonadmin0 are you using a local instance of openstack-client? What version?12:38
*** yifei has joined #openstack-ansible12:43
mgariepymorning everyone12:47
*** dmsimard is now known as dmsimard|afk12:47
*** mgariepy has quit IRC12:53
*** mgariepy has joined #openstack-ansible12:54
*** esberglu has joined #openstack-ansible12:57
*** persia has joined #openstack-ansible12:58
*** kylek3h has joined #openstack-ansible12:58
*** mardim has joined #openstack-ansible13:05
*** galstrom_zzz is now known as galstrom13:05
*** armaan has quit IRC13:07
*** gouthamr has joined #openstack-ansible13:14
*** shardy is now known as shardy_afk13:17
*** firebat has quit IRC13:23
admin0jamesdenton, i did a pip install python-openstackclient13:24
admin0 openstack --version13:24
admin0openstack 3.12.013:24
admin0oops13:24
admin0sorry .. multi line paste13:24
*** bhujay has quit IRC13:27
*** mrch has joined #openstack-ansible13:28
*** jwitko has joined #openstack-ansible13:29
admin0jamesdenton, you think version is an issue .. i see osa version is 3.8.113:29
jamesdentonTry matching the version on your local machine to whats in the utility container13:30
admin0yes .. trying to do that now13:30
*** PTO has quit IRC13:31
*** zerick has quit IRC13:31
*** zerick has joined #openstack-ansible13:33
admin0jamesdenton, didn't work ..13:35
admin0i guess i will stop 2 of the glance apis, set the remaining one to debug and dig further13:36
jamesdentonwell, it was a long shot. good luck13:36
mhaydenandymccr: taking a gander at ceph13:44
andymccrmhayden: thanks! I did a build and it worked fine but our upstream jobs are all failing on repobuild - so i think it must be a ceph repo that isnt grafting properly with the work we've done to setup ceph repos upstream?13:45
mhaydenweird13:46
mhaydenis this newton?13:46
evrardjp_oh andymccr and mhayden are you checking at the ceph pike ?13:46
andymccrmhayden: nope i did master - but its failing on pike and master.13:46
mhaydenweird13:46
mhaydenthe /etc directory isn't available in the job reports13:46
hwoaranghmm13:47
mhaydeni was wondering if the yum repo was misconfigured13:47
evrardjp_Error unpacking rpm package python2-requests-2.11.1-1.el7.noarch13:47
odyssey4memhayden you only did the etc stuff for the role tests, not the integrated build13:47
mhaydenalso hwoarang -- feel free to edit any and all patches i write at any time ;)13:47
hwoarangshouldn't the a-r-r files in role stable/branches have stable/$foo instead of 'master' as version?13:47
evrardjp_or Error unpacking rpm package python2-urllib3-1.16-1.el7.noarch13:47
mhaydenodyssey4me: oh... right13:47
mhayden:/13:47
*** shardy_afk is now known as shardy13:47
*** drifterza has quit IRC13:47
mhaydeni guess i should hop on that13:47
andymccryeah so this is master's bug: http://logs.openstack.org/74/500574/1/check/gate-openstack-ansible-openstack-ansible-ceph-centos-7-nv/ddb0c30/console.html#_2017-09-04_16_04_07_16991013:47
evrardjp_hwoarang: yes13:47
odyssey4mehwoarang yep, mostly13:47
andymccron pike its a bit diff: http://logs.openstack.org/69/500569/1/check/gate-openstack-ansible-openstack-ansible-ceph-centos-7-nv/eb33ee3/console.html#_2017-09-04_15_10_59_23231513:48
evrardjp_http://logs.openstack.org/periodic/periodic-openstack-ansible-deploy-ceph-pike-centos-7/03f89c7/console.html#_2017-09-05_07_20_58_47526613:48
odyssey4mefor OSA roles at least13:48
hwoarangbut they don't...13:48
hwoarang:)13:48
evrardjp_https://github.com/openstack/openstack-ansible/blob/stable/pike/ansible-role-requirements.yml ?13:48
odyssey4mehwoarang ORLY?13:48
hwoaranghmm13:48
mhaydenandymccr: does it fail on that same task each time?13:49
andymccrmhayden: ive seen the master one more than pike13:49
andymccrwe could check periodics going back a bit13:49
hwoarangodyssey4me: stable/pike on os_neutron has 'master' in tests/a-r-r13:49
evrardjp_looks to me they are13:49
andymccri'll build a pike one to try - but master always fials on the master fail13:49
mhaydenwhat's weird is that the test run installs from EPEL just fine for other pkgs13:49
evrardjp_hwoarang: it seems ok to me13:49
odyssey4mehwoarang ah, tests/a-r-r isn't used by anything13:49
*** hk135 has quit IRC13:49
mhaydenbut it says the baseurl is invalid13:49
hwoarangis it not used for fetching the role dependencies?13:50
hwoarangwhen running the gate tests?13:50
odyssey4mehttps://github.com/openstack/openstack-ansible-tests/blob/master/ansible-role-requirements-gen.sh and https://github.com/openstack/openstack-ansible-tests/blob/master/ansible-role-requirements.yaml are completely unused13:50
*** marst has joined #openstack-ansible13:50
evrardjp_andymccr: mhayden I just pasted the periodic13:50
evrardjp_http://logs.openstack.org/periodic/periodic-openstack-ansible-deploy-ceph-pike-centos-7/03f89c7/console.html#_2017-09-05_07_20_58_475266 and http://logs.openstack.org/periodic/periodic-openstack-ansible-upgrade-ceph-pike-centos-7/dad6760/console.html#_2017-09-05_06_58_52_30804613:50
hwoarangodyssey4me: yeah but what about the tests/a-r-r in the individual role files?13:50
odyssey4mehwoarang we could perhaps start to use it instead of managing individual repo a-r-r's, but that is not what's going on now13:50
*** yifei has quit IRC13:51
andymccryou can pretty much check any PR since the ceph centos7 job fails out before timeout so you can get logs.13:51
odyssey4mehwoarang yes, each repo has its own a-r-r and that *should be done right13:51
hwoarangok that's what i am talking about13:51
*** madorn has quit IRC13:51
mhaydeni'm going to see if i can add the /etc stuff to the integrated build13:51
hwoarangsince it appears all of them pull down 'master' dependencies on stable branches13:51
odyssey4meyep, this is wrong13:52
odyssey4me:https://github.com/openstack/openstack-ansible-os_neutron/blob/stable/pike/tests/ansible-role-requirements.yml13:52
*** shardy has quit IRC13:52
hwoarangcrap13:52
odyssey4meI must have missed it, unless my patch hasn't merged yet13:52
hwoarangaha13:52
hwoarangit's not broken everywhere13:52
evrardjp_nova is fine13:52
hwoarangnova is good for example13:53
hwoaranginteresting13:53
odyssey4mehwoarang ah, I esse it https://review.openstack.org/49662213:53
odyssey4me*see13:53
odyssey4methe problem is that while the patch has been proposed to fix it, it hasn't merged13:53
hwoarangoh we still have these around13:53
odyssey4mehttps://review.openstack.org/#/q/topic:create-pike+project:%22%255E(openstack%257Copenstack-infra)/(designate%257Cmonitorstack%257Copenstack-ansible%257Cansible-%257Cdevstack-gate).*%22+status:open13:54
*** madorn has joined #openstack-ansible13:54
hwoarangright13:54
hwoarangmhayden: noted :)13:54
*** yifei has joined #openstack-ansible13:54
*** shardy has joined #openstack-ansible13:54
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible-rabbitmq_server master: Upgrade to RabbitMQ 3.6.11  https://review.openstack.org/49680113:55
evrardjp_logan-: are you there?13:56
*** jmorgan1 has joined #openstack-ansible13:57
*** galstrom is now known as galstrom_zzz14:01
openstackgerritMarc Gariépy (mgariepy) proposed openstack/openstack-ansible-ops master: Add a check for galera repo vars.  https://review.openstack.org/50083914:04
mgariepycloudnull, when you have a minutes i would like your comment on https://review.openstack.org/#/c/500839/14:07
openstackgerritMajor Hayden proposed openstack/openstack-ansible master: Preserve /etc in job results  https://review.openstack.org/50084214:08
mhaydenodyssey4me / andymccr: ^^14:08
odyssey4memhayden - the os_neutron role for ocata is struggling to pass consistently for the centos job14:08
* mhayden adds it to his growing list ;)14:08
odyssey4mecan you please take a look see to figure out what's made things better because that's holding a very important patch back14:09
odyssey4meright now neutron db migrations are broken for ocata14:09
logan-o/14:09
logan-evrardjp_: hello14:10
evrardjp_We have this:  http://logs.openstack.org/22/496622/4/check/gate-openstack-ansible-os_neutron-ansible-calico-ubuntu-xenial/7f2d17d/console.html.gz#_2017-08-29_11_35_15_31812514:10
evrardjp_from https://review.openstack.org/#/c/496622/414:10
evrardjp_could you have a look at it when you have time?14:10
evrardjp_or at least give an insight?14:11
logan-yup looking14:11
logan-for the development sha id leave networking_calico at master14:12
evrardjp_including for stable branch?14:13
logan-for stable branch I think we were pinning based on master sha snapshots not by tag14:13
logan-they rarely tag it, its just a rolling release meant to maintain compatibility from icehouse to latest openstack14:13
evrardjp_ok14:14
logan-yeah confirmed https://github.com/openstack/openstack-ansible/blob/stable/ocata/playbooks/defaults/repo_packages/openstack_services.yml#L16614:14
logan-we dont use the tags14:14
evrardjp_that's for the integrated14:15
evrardjp_I meant for role testing14:15
logan-right, for role testing its master14:16
*** vnogin has quit IRC14:18
odyssey4meit might be best to just make it the current sha for the stable branch14:19
odyssey4mebut meh, logan-'s the SME here :p14:19
logan-thats fine too, just not the tag14:19
logan-its a minimally updated repo and most of the commits are adjusting to shifts in neutron internals to maintain the driver14:20
*** dxiri_ has joined #openstack-ansible14:20
evrardjp_I'd rather keep it at master if we plan to roll14:21
logan-++ i think its ideal14:21
*** DanyC has quit IRC14:22
neiththe task "Gather software version" list always fails in /opt/openstack-ansible/playbooks/os-nova-install.yml with:14:22
neithhttps://www.irccloud.com/pastebin/qYTqliCv/14:22
neithany idea? seems like the var is undefined but why14:23
evrardjp_logan is where felix info version stored for os_neutron test? It's just https://github.com/openstack/openstack-ansible-os_neutron/blob/stable/pike/defaults/main.yml#L352 ?14:23
evrardjp_logan-: in other words, calico seems fine in stable/pike os_neutron test14:24
logan-thats it, the L350-352 determine the version installed in role test + integrated14:26
openstackgerritJean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Update vars and test tooling for Pike  https://review.openstack.org/49662214:26
evrardjp_logan-: ^ this should be better then14:26
logan-it is a go based agent so we just download the binary in the URL and use it14:26
openstackgerritMajor Hayden proposed openstack/ansible-hardening stable/pike: Add release note for F26 support  https://review.openstack.org/50005514:27
openstackgerritJean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron stable/pike: Update vars and test tooling for Pike  https://review.openstack.org/49662214:31
*** vikrant has joined #openstack-ansible14:32
*** galstrom_zzz is now known as galstrom14:32
openstackgerritMajor Hayden proposed openstack/openstack-ansible-os_neutron stable/ocata: Simplify the string check for offline db migrations  https://review.openstack.org/49461414:35
openstackgerritJean-Philippe Evrard proposed openstack/openstack-ansible stable/pike: Bump etcd role  https://review.openstack.org/50085014:36
*** kristian__ has joined #openstack-ansible14:36
openstackgerritMajor Hayden proposed openstack/openstack-ansible-os_neutron stable/ocata: Improve the execution time  https://review.openstack.org/46629914:37
*** vnogin has joined #openstack-ansible14:37
mhaydenodyssey4me: on ocata os_neutron, we may need to merge the bindep + db migration patches to get it through the gate14:39
mhaydentesting that scenario now14:39
odyssey4memhayden the db migration patch is not required for the other one14:39
odyssey4methe issue relates to bad repo issues in centos... nothing to do with the db migrations14:39
odyssey4methe db migrations only affect integrated deployments, and only in real life14:39
mhaydenwell the pyOpenSSL patch is stuck in os_neutron14:39
mhaydentrying to figure out why14:39
odyssey4meyes, so the db migration needs to be based on the bindep change - which it has been, except cloudnull keeps removing the dep14:40
mhaydenoh damn14:40
odyssey4meso we need to bindep change in, that'll unblock the rest14:40
mhaydenbindep patch is failing on http://logs.openstack.org/00/497900/1/gate/gate-openstack-ansible-os_neutron-ansible-func-centos-7/8f321e6/console.html#_2017-09-04_15_01_09_18760614:40
*** mrch has quit IRC14:40
mhaydeni thought i saw a patch for that problem14:40
evrardjp_dear cloudnull, DimGR, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, spotz, vdo, jmccrory, alextricity25, jasondotstar, admin0, michaelgugino, ametts, v1k0d3n, severion, bgmccollum, darrenc, J14:41
evrardjp_Robinson__, asettle, colinmcnamara, thorst, adreznec, eil397, qwang,nishpatwa_, cathrichardson, drifterza, hwoarang, cshen:14:41
evrardjp_the osa bug triage will start in 1h20’. Please have a look at the bug list before starting: https://etherpad.openstack.org/p/osa-bugtriage14:41
evrardjp_Thank you14:41
cloudnullmornings14:42
cloudnullwhat dep did i remove ?14:44
*** udesale has joined #openstack-ansible14:44
mhaydencloudnull: that depends14:45
cloudnullhahahaa14:45
cloudnullrather, what dep do I "keep" removing?14:45
mhaydencloudnull: apparently something in os_neutron's ocata branch14:45
odyssey4mecloudnull I keep basing https://review.openstack.org/494614 on the bindep change required for it to pass, and you keep removing that dep14:46
odyssey4mewe're having a little rebase war ;)14:46
mhaydensounds like someone's off base14:46
* mhayden will be here all week14:47
cloudnullthe conditional ?14:47
cloudnullthat pr has never passed gate in ocata.14:47
*** yifei has quit IRC14:47
odyssey4mecloudnull https://review.openstack.org/#/c/494614/ is based on https://review.openstack.org/#/c/497900/1 - please leave it like that :)14:47
*** udesale has quit IRC14:47
odyssey4mecloudnull it has14:47
*** jheroux has joined #openstack-ansible14:47
odyssey4metwice in ps214:48
*** udesale has joined #openstack-ansible14:48
odyssey4methrice actually14:48
cloudnullI'm lost... how is rebasing that patch bad?14:49
odyssey4mecloudnull you removed the base of the one on the other14:49
*** udesale has quit IRC14:49
odyssey4methat caused confusion14:49
odyssey4mebecause then people waste time trying to see why it's failing, but it's failing because it needs the other patch14:50
odyssey4mewhich had already been worked out, which is why it was based on the patch14:50
odyssey4meie, if a patch is based on another patch - why rebase it... I just don't understand why you changed it?14:51
cloudnullto pull in the head of the stable branch.14:52
odyssey4mebut it's based on another patch14:52
odyssey4meand zuul rebases for you - we never need to rebase14:52
cloudnullthats not true14:52
cloudnullrecheck will not rebase14:52
odyssey4meit does, and if it doesn't infra would like to know because that's a fundamental function of zuul14:53
cloudnullthat doesn't make sense.14:54
odyssey4meback to the point of the discussion - that one patch depends on the other patch, please stop rebasing ot back to the head of the branch14:55
cloudnullhttps://www.mediawiki.org/wiki/Gerrit/Tutorial14:55
cloudnull^ rebase is different than recheck14:55
odyssey4megerrit is not zuul14:55
cloudnullso we're no longer using the gerrit functionality any longer.14:56
odyssey4mealso, don't get me wrong here - I wasn't complaining, I was explaining to mhayden what happened and why the review was not based on the other review14:56
odyssey4mecloudnull how gerrit/zuul works and what the functionality is is something you can discuss with infra - I have been informed previously that zuul rebases every patch it tests14:57
hwoaranginteresting that zuul always rebases without explicitly asking gerrit to do so14:57
cloudnull^14:57
hwoarangthen what's the point to change the base of the patch is zuul will rebase it to the head anyway?14:57
hwoarang*if14:57
* hwoarang is confused14:58
odyssey4methe only time changing th ebase of the patch is useful is if the base you want has not yet merged14:58
odyssey4mebasically zuul checks out, then rebases14:58
odyssey4meat least that's what I've been told14:58
*** agrebennikov has joined #openstack-ansible14:59
hwoaranghmm subtle operation...14:59
cloudnullso yea, infra confirms. zuul does magic on the backside and is always rebasing which is overriding the gerrit functionality.15:03
cloudnullodyssey4me: and to the issue at hand, I'll leave those patches a lone.15:04
odyssey4mecloudnull thanks15:04
openstackgerritLogan V proposed openstack/openstack-ansible master: [DNM] Does it actually rebase  https://review.openstack.org/50087115:05
*** Oku_OS is now known as Oku_OS-away15:07
*** vikrant has quit IRC15:07
*** rodolof has quit IRC15:09
cloudnullodyssey4me: https://www.irccloud.com/pastebin/qYTqliCv/ have you seen this before? Looks like something related to the our setting of local facts?15:12
cloudnulltask path: /opt/openstack-ansible/playbooks/os-nova-install.yml:13015:12
cloudnullusing 14.2.815:12
odyssey4mecloudnull never seen it, but if, for some reason, ansible_local doesn't exist then it would cause that I guess15:13
odyssey4methat would only exist if there was some sort of previous error where a host was removed from the execution list15:13
odyssey4mewe need to use any_error_fatal more I think15:13
odyssey4meand we perhaps need  to be better at caging against missing facts15:14
neithodyssey4me:  the weird thing is that during the first run of the setup-openstack.yml this tasks worked but not after replaying it :(15:15
*** lucasxu has joined #openstack-ansible15:17
*** weezS has joined #openstack-ansible15:17
*** kristian__ has quit IRC15:17
neithodyssey4me: is ansible_local a var defined by OSA or an ansible builtin var?15:18
*** kristian__ has joined #openstack-ansible15:18
odyssey4meneith ansible15:18
neithodyssey4me: alright so its even more weird then15:18
odyssey4mebut it will only be there is a local fact is set - ie in the target host there should be a .fact file in /etc/ansible/facts.d/15:18
odyssey4me*if15:19
*** kristia__ has joined #openstack-ansible15:19
*** kristian__ has quit IRC15:19
*** kristia__ has quit IRC15:20
neithodyssey4me: I don't have this file on the target host15:20
*** kristian__ has joined #openstack-ansible15:20
odyssey4meneith is it a container, or host?15:21
neithodyssey4me: a compute host15:21
odyssey4meok, then the only thing that would put it there is are the nova/neutron plays15:21
odyssey4mehttps://github.com/openstack/openstack-ansible-os_nova/blob/master/tasks/nova_install.yml#L169-L19315:22
odyssey4meif it is not there, then some task before that failed and that host was taken out of the play execution15:22
*** kristian__ has quit IRC15:22
*** gparaskevas has quit IRC15:22
*** kristian__ has joined #openstack-ansible15:23
odyssey4meand because we don't have any_errors_fatal, the next play brought it back in and failed because the assumed fact was not present15:23
openstackgerritMerged openstack/openstack-ansible-os_horizon stable/pike: Updated from OpenStack Ansible Tests  https://review.openstack.org/49750815:24
*** armaan has joined #openstack-ansible15:25
*** lbragstad has joined #openstack-ansible15:25
*** kristian__ has quit IRC15:27
*** vnogin has quit IRC15:27
openstackgerritMerged openstack/openstack-ansible-os_neutron stable/pike: Updated from OpenStack Ansible Tests  https://review.openstack.org/49689815:31
openstackgerritMerged openstack/openstack-ansible-os_neutron master: Fix a typo  https://review.openstack.org/50074415:31
*** vikrant has joined #openstack-ansible15:33
*** b4rti has quit IRC15:33
neithodyssey4me: I just followed this doc https://docs.openstack.org/openstack-ansible/newton/newton/developer-docs/ops-add-computehost.html and I failed in the error cloudnull mentioned15:33
openstackgerritLogan V proposed openstack/openstack-ansible-lxc_container_create master: Fix the bind mount options to match integrated repo  https://review.openstack.org/45690515:33
odyssey4meneith ok, but if you inspect the playbook output, can you see any task failing for that host prior to the one in the pastebin?15:34
neithodyssey4me: let me replay and log all again15:35
mhaydenodyssey4me: so this is what's holding up the neutron gate in ocata -> http://logs.openstack.org/14/494614/4/check/gate-openstack-ansible-os_neutron-ansible-func-centos-7/7ea5865/console.html#_2017-09-05_15_29_29_90769815:36
mhaydeni'm wondering if we're missing a backport15:36
*** kristian__ has joined #openstack-ansible15:36
odyssey4memhayden look earlier in the log, there's another failure which has removed that host from rotation, which is why that task fails15:37
odyssey4mehttp://logs.openstack.org/14/494614/4/check/gate-openstack-ansible-os_neutron-ansible-func-centos-7/7ea5865/console.html#_2017-09-05_15_14_45_14699115:37
odyssey4methe one play doesn't complete properly, resulting in subsequent plays doing stuff which is inconsequential15:38
odyssey4meagain, we should use any_error_fatal more :p15:38
mhaydenah, i see15:38
mhaydeni've seen this come up in other jobs for some unknown reason15:38
mhaydenthe dns record exists15:38
odyssey4meyep, see it come up a fair bit, not sure if it is the DNS or if it's some sort of network wiring or something else15:39
mhaydeni wish i could see what's in /etc15:41
*** kristia__ has joined #openstack-ansible15:41
mhaydeni wonder if we had an issue in /etc/resolv.conf15:42
mhaydenthe yum repo appears to be configured properly15:42
openstackgerritMerged openstack/openstack-ansible-pip_install master: vars: SUSE: Switch openSUSE repository to Pike release  https://review.openstack.org/50049015:44
*** kristian__ has quit IRC15:45
openstackgerritJean-Philippe Evrard proposed openstack/openstack-ansible stable/pike: Add Galera/RabbitMQ bump timeframe  https://review.openstack.org/50089915:45
openstackgerritMerged openstack/openstack-ansible-os_octavia master: Add opendaylight role to tests a-r-r  https://review.openstack.org/50054315:45
openstackgerritMerged openstack/openstack-ansible-os_octavia master: Updated from OpenStack Ansible Tests  https://review.openstack.org/49727715:45
*** kristian__ has joined #openstack-ansible15:49
openstackgerritMajor Hayden proposed openstack/openstack-ansible-tests stable/ocata: Backport /etc collection from pike  https://review.openstack.org/50090115:50
neithodyssey4me: I can confirm there are no failing tasks prior this one15:52
odyssey4meneith and that task ran after the role executed against all hosts?15:53
*** kristia__ has quit IRC15:53
*** kristian__ has quit IRC15:53
*** drifterza has joined #openstack-ansible15:53
odyssey4meneucan you confirm that the role executed against that host?15:54
neithodyssey4me: I don"t understand your question15:54
odyssey4meneith specifically, did https://github.com/openstack/openstack-ansible-os_nova/blob/master/tasks/nova_install.yml#L169-L193 execute against that host?15:54
odyssey4meie can you see that task name in the log being executed against the compute node that's got the issue?15:54
neithodyssey4me: this task is not executed15:56
mhaydenodyssey4me: spinning up a test vm for those neutron shenanigans15:56
odyssey4meneith not against any hosts, or not against only that host15:57
neith'im limiting the play to this very one host15:57
neith*i'm15:58
odyssey4meneith and are any of the other tasks before it in that task file executing against that hosty?15:58
openstackgerritJean-Philippe Evrard proposed openstack/openstack-ansible master: Connectivity testing  https://review.openstack.org/49549215:58
openstackgerritJean-Philippe Evrard proposed openstack/openstack-ansible master: Add infra healthchecks  https://review.openstack.org/49549215:59
evrardjp_Bug triage cloudnull, DimGR, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, spotz, vdo, jmccrory, alextricity25, jasondotstar, admin0, michaelgugino, ametts, v1k0d3n, severion, bgmccollum, darr16:01
evrardjp_enc, JRobinson__, asettle, colinmcnamara, thorst, adreznec, eil397, qwang,nishpatwa_, cathrichardson, drifterza, hwoarang, cshen16:01
evrardjp_Here is our bug list for today https://etherpad.openstack.org/p/osa-bugtriage16:01
evrardjp_#startmeeting openstack_ansible_meeting16:01
openstackMeeting started Tue Sep  5 16:01:21 2017 UTC and is due to finish in 60 minutes.  The chair is evrardjp_. Information about MeetBot at http://wiki.debian.org/MeetBot.16:01
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:01
openstackThe meeting name has been set to 'openstack_ansible_meeting'16:01
andymccro/16:01
evrardjp_#topic this week's bugs16:01
cloudnullo/16:01
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/171459716:01
openstackLaunchpad bug 1714597 in openstack-ansible "multi OS repo-build needs to be more intuitive " [Undecided,New]16:01
neithodyssey4me: yes some tasks are played from that task file16:02
evrardjp_never heard about this problem but I agree we don't have cross repo testing16:03
andymccrhmm16:03
evrardjp_cross distro*16:03
andymccrwe dont really have a way to test that though16:04
evrardjp_manual testing only indeed16:04
evrardjp_let's maybe leave that as is?16:04
evrardjp_until someone confirms?16:04
neithevrardjp_: I'm currently struggling uprading from ubuntu 14 to 16  and there are some possible improvements16:04
*** hamzy has quit IRC16:04
odyssey4meit may make sense for that role to change up the tests to implement 3 qemu vm's via libvirt instead of using a container for the build16:04
neithevrardjp_: mostly wheel package are not synced correctly16:05
odyssey4mebut that's obviously an improvement that someone would need to implement, which is not really the bug subject16:05
*** marst has quit IRC16:05
odyssey4meneith this is the bug triage meeting - please wait until afterwards16:05
*** marst has joined #openstack-ansible16:05
odyssey4me(or register bugs!)16:05
evrardjp_:D16:05
evrardjp_don't! :p16:05
evrardjp_(joking)16:06
evrardjp_that problem should at least have been caught by a few users before16:06
*** sauloaislan has joined #openstack-ansible16:06
sauloaislanMorning16:06
evrardjp_I think we should leave it as is, and if someone has the issue again we'll see16:06
evrardjp_ok for everyone?16:07
evrardjp_let's move on16:07
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/171446216:07
openstackLaunchpad bug 1714462 in openstack-ansible "Ansible security leads to FAIL_DELAY configuration error on ocata" [Undecided,New]16:07
evrardjp_mhayden: ?16:07
evrardjp_have you seen that?16:07
* mhayden ganders16:08
mhaydenassign it to me and i'll dig into it16:08
evrardjp_thanks16:08
evrardjp_next16:08
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/171137616:08
openstackLaunchpad bug 1711376 in openstack-ansible "intermittent AIO error: Timeout (7s) waiting for privilege escalation prompt" [Undecided,New]16:08
hwoarangyeah i haven't seen that^^ ever since I increased the memory requirements on the Vagrantfile16:09
evrardjp_hwoarang: is that done?16:09
*** fellypefca has joined #openstack-ansible16:09
evrardjp_let's mark it as fixed for now, and maybe reopen then?16:09
*** kristian__ has joined #openstack-ansible16:09
hwoarangi guess we can close that since we don't see it on the gates16:10
hwoarangyeah ok16:10
evrardjp_next16:10
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/171087416:10
openstackLaunchpad bug 1710874 in tripleo "tacker requires conductor service in pike" [High,Triaged] - Assigned to Tim Rozet (trozet)16:10
*** marst_ has joined #openstack-ansible16:11
*** marst has quit IRC16:12
evrardjp_do we even have ppl working on mistral?16:14
*** kristian__ has quit IRC16:14
evrardjp_who was working on tacker recently?16:14
andymccrit was mbuil mostly i think - but he's out atm16:14
evrardjp_well let's say confirmed, based on what we can see in other deploy projects16:15
sauloaislanandymccr: I have a problem with ironic-api16:15
evrardjp_sauloaislan: maybe later?16:15
evrardjp_andymccr: how would you triage this?16:16
evrardjp_low?16:16
andymccrevrardjp_: yeah we dont officially spuport tacker in pike anyway16:16
evrardjp_ok let's move on16:16
andymccrsauloaislan: we're doing bug triage right now - happy to try help after16:16
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/170836016:16
openstackLaunchpad bug 1708360 in openstack-ansible "Openstack-Ansible Install Fails with OVS" [Undecided,New]16:16
andymccri mean - probably :P16:17
evrardjp_same thing as last week?16:17
sauloaislanMy problem can be a bug, but ok16:17
andymccrif somebody is keen on OVS we should get some testing etc going16:17
evrardjp_hwoarang: ?16:17
andymccrsauloaislan: if you put it up on launchpad then its probably on the list to go through today :)16:17
evrardjp_list is already define, but I leave time at the end of each bug triage to talk about other bust16:18
evrardjp_bugs*16:18
evrardjp_ok let's move on16:18
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/170715116:18
openstackLaunchpad bug 1707151 in oslo.middleware "Remove method CORS.set_latent" [Undecided,New]16:18
*** marst has joined #openstack-ansible16:18
jmccrorythe glare paste file was removed, i don't know if there's anything else for OSA to do there16:19
*** marst_ has quit IRC16:19
evrardjp_yeah I'd say I don't see anything in hound16:19
evrardjp_I'll remove it16:19
andymccrhmm16:20
evrardjp_next16:20
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/170714216:20
openstackLaunchpad bug 1707142 in openstack-ansible "pip_install: "The following packages cannot be authenticated"" [Undecided,New]16:20
evrardjp_gonna expire soon!16:20
evrardjp_next16:20
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/170537416:21
openstackLaunchpad bug 1705374 in openstack-ansible "realpath() of /etc/uwsgi/nova-placement-uwsgi.ini failed" [Undecided,New] - Assigned to Andy McCrae (andrew-mccrae)16:21
evrardjp_andymccr: I assigned that to you because you said you're gonna check on it because you had to revisit the code16:22
evrardjp_now that it's revisited, should we close that?16:22
andymccrevrardjp_: lemme double check ill close if its been backported i think it was16:22
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server stable/pike: Increase RabbitMQ's ulimit to 65536  https://review.openstack.org/50091316:22
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server stable/ocata: Increase RabbitMQ's ulimit to 65536  https://review.openstack.org/50091416:22
andymccryeh done16:23
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server stable/newton: Increase RabbitMQ's ulimit to 65536  https://review.openstack.org/50091516:23
andymccrnext!16:23
*** marst_ has joined #openstack-ansible16:23
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/170361816:23
openstackLaunchpad bug 1703618 in openstack-ansible "repo_container on arm host contains packages for x86 hosts only" [Undecided,New]16:23
*** marst has quit IRC16:23
evrardjp_should we decide to backport this to older branches?16:24
evrardjp_we are not sure it would solve the issue16:24
evrardjp_if nobody answers, it's gonna expire16:24
evrardjp_we could mark it as incomplete too16:25
evrardjp_incomplete ok for everyone?16:25
evrardjp_316:25
evrardjp_216:25
evrardjp_116:25
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/170361616:26
openstackLaunchpad bug 1703616 in openstack-ansible "Error " This is not a recognized Fernet token" when installing os-trove" [Undecided,New]16:26
evrardjp_incomplete?16:26
andymccrhmm16:27
andymccryeah16:27
andymccrim not sure how to recreate16:27
evrardjp_andymccr: it looks like we are expecting data16:27
andymccryeah16:28
andymccrincomplete is good i think - if we cant recreate its hard to fix :)16:28
evrardjp_sorry16:28
evrardjp_next16:29
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/169887116:29
openstackLaunchpad bug 1698871 in openstack-ansible "[master] [os_gnocchi]Gnocchi role tests failing on installing pip packages" [Undecided,New]16:29
evrardjp_anyone got the chance to look at that?16:29
andymccri mean its unsurprising16:29
evrardjp_yes16:29
andymccrthe whole telemetry stack is pretty much in disrepair since nobody seems to use/maintain it16:29
evrardjp_agreed16:29
evrardjp_We have a talk at the PTG about phasing out roles16:30
evrardjp_feedback gathering session at least16:30
*** kristian__ has joined #openstack-ansible16:30
evrardjp_ok let's continue then16:30
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/169778216:30
openstackLaunchpad bug 1697782 in openstack-ansible "Mounting of ceph-backed cinder volumes is broken after Ocata upgrade" [Undecided,New] - Assigned to Jesse Pretorius (jesse-pretorius)16:30
evrardjp_odyssey4me: I don't recall why I assigned that to you. I can check on the logs16:31
odyssey4meI added a patch to check whether it was true, as far as I can see builds are still working with that patch merged.16:31
openstackgerritMerged openstack/openstack-ansible-rabbitmq_server master: tests: test-vars: Install the RabbitMQ file directly  https://review.openstack.org/50062916:31
*** marst has joined #openstack-ansible16:31
evrardjp_https://review.openstack.org/48354116:32
*** marst_ has quit IRC16:32
odyssey4melogan- appears to be working on it though16:32
logan-i have not done any work on it since my last reply, have not had any luck reproducing it :(16:32
evrardjp_should we care?16:32
logan-im about to dive into ocata upgrade stuff so i'm keeping an eye out for it cropping up in my deploys here16:33
odyssey4meI've removed myself as assignee and added a comment about the testing implemented16:33
evrardjp_thanks odyssey4me!16:34
evrardjp_next16:34
evrardjp_#link https://bugs.launchpad.net/openstack-ansible/+bug/171134716:34
openstackLaunchpad bug 1711347 in openstack-ansible "Config file override mechanism broken" [Wishlist,New]16:34
evrardjp_andymccr: did we revert the python3 yet?16:35
*** kristian__ has quit IRC16:35
evrardjp_sorry I haven't followed that, I just see 1 commit.16:35
andymccrevrardjp_: we did in pike yeah16:35
andymccrwe dont install it as part of the bootstrap and we dont select it so it should be fine16:35
logan-check experimental in plugins will expose a bunch of py3 plugins issues16:35
odyssey4melogan- you had another patch to test the connection.strategy plugins which needs a revision too - it'd be nice to get that done16:36
logan-yup16:37
logan-was looking at that this weekend, trying to wrap up old patches I have open16:37
logan-but ill ping after the mtg about that :)16:37
evrardjp_no I meant these kind of patches: https://review.openstack.org/#/c/471531/16:37
evrardjp_https://review.openstack.org/#/c/471529/16:38
evrardjp_if we are good then let's continue with our lives16:38
logan-yeah16:38
logan-there's a lot more of those needed in plugins before py3 works correctly there16:38
evrardjp_What I mean there, is that there were other bugs to introduce python3 compatibility16:38
evrardjp_sorry let me rephrase16:38
*** sauloaislan has quit IRC16:39
evrardjp_if some of our code is python 2+3 ready, good.16:39
evrardjp_if it's breaking under 3, meh IF we ensure python3 is not used.16:39
evrardjp_but this isn't the case. So should we care now?16:39
evrardjp_Because expectations to run python3 may be in the community already16:40
evrardjp_so should way say somewhere: run python2? (in our shabangs ?)16:40
evrardjp_in the meantime, I'd say  let's leave the bug as is, and just remember it :)16:41
logan-bootstrap-ansible essentially enforces "run python 2" or "run python 3" because it manages the venv build16:41
evrardjp_ok for everyone?16:41
logan-yeah16:41
cloudnull++16:42
evrardjp_logan-: oh yeah that's true. So except if someone is running with it's own bootstrapped ansible we should be fine16:42
evrardjp_ok16:42
evrardjp_done for today16:42
evrardjp_anyone else has a bug to raise?16:43
evrardjp_516:43
evrardjp_416:43
evrardjp_316:43
evrardjp_216:43
evrardjp_116:43
andymccrthanks :)16:43
evrardjp_ty !16:44
evrardjp_#endmeeting16:44
openstackMeeting ended Tue Sep  5 16:44:01 2017 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:44
openstackMinutes:        http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2017/openstack_ansible_meeting.2017-09-05-16.01.html16:44
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2017/openstack_ansible_meeting.2017-09-05-16.01.txt16:44
openstackLog:            http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2017/openstack_ansible_meeting.2017-09-05-16.01.log.html16:44
evrardjp_andymccr: see what you did there!16:44
openstackgerritMerged openstack/openstack-ansible-tests stable/pike: test-prepare-host: Fix typo in postup/postdown options for SUSE  https://review.openstack.org/50050816:44
andymccrso much!16:44
openstackgerritOpenStack Proposal Bot proposed openstack/openstack-ansible-os_swift stable/pike: Updated from OpenStack Ansible Tests  https://review.openstack.org/49667116:45
openstackgerritOpenStack Proposal Bot proposed openstack/openstack-ansible-os_tacker stable/pike: Updated from OpenStack Ansible Tests  https://review.openstack.org/50092316:45
* neith apologizes for being intrusive16:46
cloudnull^ never!16:47
cloudnullif there's a problem then we need to solve it16:47
neithodyssey4me: do you want to get back to my issue or later?16:48
openstackgerritMajor Hayden proposed openstack/openstack-ansible master: Preserve /etc in job results  https://review.openstack.org/50084216:48
odyssey4meneith it's a little difficult to step you through each and every step to troubleshoot it... it'd be a lot easier if you could just pastebin the log output16:49
neithodyssey4me: https://pastebin.com/b8iFMncL16:49
neithodyssey4me: I was expecting you to ask ^^16:50
*** kristian__ has joined #openstack-ansible16:51
odyssey4meneith this is newton, right?16:53
neithodyssey4me: 14.2.816:53
openstackgerritMerged openstack/openstack-ansible-os_octavia master: Add blank line to avoid malformed jinja2 template  https://review.openstack.org/49835416:54
*** kristian__ has quit IRC16:55
openstackgerritMajor Hayden proposed openstack/ansible-hardening stable/ocata: Configure pam_faildelay on Ubuntu  https://review.openstack.org/50092416:55
mhaydenodyssey4me / andymccr: i wonder if the ocata neutron issues are due to the lxc-net bug that andymccr found16:57
mhaydenwhere lxc-net was failing to start temporarily on centos16:57
openstackgerritMerged openstack/openstack-ansible-tests stable/pike: network_interfaces: Remove useless 'break' command from case statement  https://review.openstack.org/50050916:57
andymccrmhayden: if it isnt able to find packages it could be.16:58
andymccri meant to backport it but the initial backport created problems16:58
andymccri could do a combined thing (there are 2 patches)16:58
odyssey4meneith that is bizarre17:00
odyssey4meit does the spice/novnc includes, then simply skips on to the next task file17:00
odyssey4meneith is this only on this one host, or does it do the same on others?17:00
neithit does the same on the freshly upgraded infra node17:01
neithfrom 14.04 to 16.0417:01
*** shardy is now known as shardy_afk17:01
*** a0x3e8 has quit IRC17:02
odyssey4meneith was that a fresh 16.04 build, or has it been an in-place upgrade?17:03
neithodyssey4me: in place17:03
odyssey4mehmm, I wonder if there isn't some kinda of permissions issue or some other python funk that's not working right17:04
odyssey4medoes /etc/ansible/facts.d exist?17:05
odyssey4meon that host17:05
neithodyssey4me: yes on both hosts compute3 and infra117:06
*** tasker has joined #openstack-ansible17:06
taskerhello and good day17:06
odyssey4meok, can you re-run with -vvv and pastebin the results?17:06
neithodyssey4me: ok17:07
openstackgerritFellype Cavalcante proposed openstack/openstack-ansible-os_ironic master: Add libxml2-dev on ironic api distro packages  https://review.openstack.org/50092817:09
odyssey4meneith, maybe also worth a try is to use the ad-hoc command... something like: ansible -m ini_file -a 'dest=/etc/ansible/facts.d/test.fact,section=test1,value=value1' <host>17:09
openstackgerritMerged openstack/openstack-ansible-os_neutron master: Add defaults for nova metadata settings  https://review.openstack.org/50050217:09
openstackgerritJesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/pike: Add defaults for nova metadata settings  https://review.openstack.org/50092917:10
fellypefcaandymccr Hi! If you have some time, please take a look https://review.openstack.org/50092817:10
*** kristian__ has joined #openstack-ansible17:11
fellypefcaOur main problems with osa pike and master are happened because of this issue..17:12
neithodyssey4me: https://paste.ubuntu.com/25473326/17:14
openstackgerritFellype Cavalcante proposed openstack/openstack-ansible-os_ironic master: Add libxml2-dev on ironic api distro packages  https://review.openstack.org/50092817:15
andymccrfellypefca: nice catch - seems right to me. I must've missed that since we had to make the same change on other services17:15
fellypefcaOh, sorry andymccr i just update the commit message :/17:15
*** sauloaislan has joined #openstack-ansible17:16
fellypefcaAnd thank you :) andymccr17:16
*** kristian__ has quit IRC17:16
andymccrfellypefca: no problem. thanks for taking care of the fix!17:16
neithodyssey4me: I think you are right17:18
sauloaislanandymccr: thk17:18
*** pbandark has quit IRC17:18
fellypefcaandymccr I'm a member of Ironic Team. Here at Openstack-Ansible we also need two +2 and a +1 at workflow to be merged?17:18
openstackgerritMerged openstack/openstack-ansible master: Adding serial console support  https://review.openstack.org/49948717:18
odyssey4meneith call me insane, but can you confirm whether https://github.com/openstack/openstack-ansible-os_nova/blob/14.2.8/tasks/nova_install.yml matches your role's task file in /etc/ansible/os_nova ?17:19
andymccrfellypefca: we do yeah. but usually there are a few ppl to review - mhayden evrardjp_ could one of you take a look and merge if youre ok with it (its pretty simple patch)17:19
evrardjp_sure17:21
neithodyssey4me: I think I understood17:21
*** hamzy has joined #openstack-ansible17:21
neithodyssey4me: I checkout the 14.2.8 in openstack-ansible, but it doesnt automatically clone the roles in /etc/ansible right?17:22
evrardjp_fellypefca: did you test with any other distro?17:22
neithI have to rebootstrap17:22
odyssey4meneith nope, you need to run bootstrap-ansible every time you change the integrated repo to ensure that the updated roles as pulled in17:23
neithodyssey4me: damn I totally forgot about that17:23
neithodyssey4me: what a shame17:23
fellypefcaevrardjp_: No :/ only with Ubuntu17:23
neithodyssey4me: I'm sorry I waste your time17:23
openstackgerritMarkos Chandras (hwoarang) proposed openstack/openstack-ansible-pip_install stable/pike: vars: SUSE: Switch openSUSE repository to Pike release  https://review.openstack.org/50093317:24
neithodyssey4me: thats why I have so much mess with upgrading probably17:24
odyssey4meneith that's probably caused quite a few problems yeah :/ sorry17:25
* neith would like to disappear in deep space17:25
spotzawww neith don't do that17:27
fellypefcaevrardjp_: but with the patch, it's everything ok for you?17:29
evrardjp_the patch itself is good. I'm worried for other distros to not be included there.17:29
neithspotz: :)17:29
odyssey4meneith well, at least we're now to the bottom of it :) hopefully updating the roles and re-executing across your cluster will make things go properly17:30
neithodyssey4me: I have been distracted because my main goal was to upgrade from trusty to  xenial and In the middle of the process I was advised to also upgrade from 14.2.1 to 14.2.817:32
*** gouthamr has quit IRC17:32
evrardjp_I voted for inclusion fellypefca17:33
fellypefcaevrardjp_ thank you very munch! :)17:34
sauloaislanevrardjp_: thk :D17:34
*** vikrant has quit IRC17:34
evrardjp_don't forget to bump your roles!17:34
odyssey4meneith well, perhaps it's time we made the process a little more intuitive for our operators :)17:34
evrardjp_fine for me to backport it too17:35
fellypefcaandymccr, evrardjp_ I can open the backport for Ocata now?17:35
fellypefcaOPs17:36
fellypefcaPike**17:36
neithodyssey4me: thks again, I have taken a lot of notes if needed for any guides17:36
evrardjp_yeah pike first, but you'd need to wait for it to bump in the integrated, except if you apply your own patch17:38
*** dmsimard|afk is now known as dmsimard17:38
fellypefcaevrardjp_: Sorry, i didn't' understand what you mean :/ It's wait to pass on jenkins?17:41
*** stuartgr has quit IRC17:42
openstackgerritMerged openstack/openstack-ansible stable/pike: roles: bootstrap-host: Fix typo in postup/postdown options for SUSE  https://review.openstack.org/50070017:46
admin0back .. \o .17:50
admin0anyone uses check_mk for monitoring ?17:50
*** kristian__ has joined #openstack-ansible17:53
openstackgerritMerged openstack/openstack-ansible stable/pike: roles: bootstrap-host: Remove useless 'break' command from case statement  https://review.openstack.org/50070117:55
*** kristia__ has joined #openstack-ansible17:56
hwoarangmhayden: hmm there still seems to be something wrong with apparmor according to https://review.openstack.org/#/c/500055/3 :(17:56
hwoarangdebian seems to randomly fail (same job passed and failed without patch modifications)17:56
*** shardy_afk is now known as shardy17:57
*** kristian__ has quit IRC17:57
*** gouthamr has joined #openstack-ansible17:59
*** armaan has quit IRC18:04
*** exodusftw has quit IRC18:05
*** drifterza has quit IRC18:12
*** shardy is now known as shardy_afk18:14
*** chas has quit IRC18:20
*** eumel8 has joined #openstack-ansible18:33
*** pcaruana has quit IRC18:49
*** fellypefca is now known as fellypefca-afk18:50
*** thorst_afk has quit IRC18:59
*** thorst_afk has joined #openstack-ansible18:59
mhaydenhwoarang: ugh19:02
mhaydenhwoarang: it's this one again http://logs.openstack.org/55/500055/3/check/gate-ansible-hardening-ansible-func-debian-jessie/5508984/console.html#_2017-09-05_18_02_28_22837419:02
*** thorst_afk has quit IRC19:04
*** vnogin has joined #openstack-ansible19:04
*** eumel8 has quit IRC19:06
*** thorst_afk has joined #openstack-ansible19:11
*** hybridpollo has joined #openstack-ansible19:11
mhaydenhwoarang: i don't think checking apparmor in dmesg will work :/19:11
* mhayden looks for another way19:11
mhaydenespecially if the kernel ring buffer has looped around already19:12
*** hybridpolio has joined #openstack-ansible19:13
*** hybridpollo has quit IRC19:15
sauloaislanmhayden: Hi I have this patch https://review.openstack.org/#/c/500928 and broken in CI on the opensuse-42319:15
*** thorst_afk has quit IRC19:15
sauloaislanFailed http://paste.openstack.org/show/620432/19:16
*** hybridpolio is now known as hybridpollo19:19
*** TxGirlGeek has joined #openstack-ansible19:20
*** armaan has joined #openstack-ansible19:24
*** thorst has joined #openstack-ansible19:27
openstackgerritMerged openstack/openstack-ansible master: scripts: scripts-library.sh: Fix tracepath logging  https://review.openstack.org/50031219:29
openstackgerritMerged openstack/openstack-ansible stable/ocata: Bump ceph-ansible to 2.2.11 tag  https://review.openstack.org/48519419:29
*** vnogin has quit IRC19:36
cloudnullsauloaislan: likely related to some other failure in the run ?19:36
openstackgerritMajor Hayden proposed openstack/ansible-hardening master: Check apparmor_status output  https://review.openstack.org/50096119:38
openstackgerritMajor Hayden proposed openstack/openstack-ansible-tests stable/ocata: Backport /etc collection from pike  https://review.openstack.org/50090119:40
jamesdentonAnyone know offhand if nova-manage gets installed anywhere?19:41
openstackgerritKevin Carter (cloudnull) proposed openstack/openstack-ansible-specs master: Hyperconverged Containers  https://review.openstack.org/49988619:42
logan-jamesdenton: should be present in the nova venv on any of the nova hosts19:42
jamesdentonthanks logan-19:43
*** kristia__ has quit IRC19:48
*** vnogin has joined #openstack-ansible19:49
*** esberglu has quit IRC19:50
*** esberglu has joined #openstack-ansible19:50
*** dxiri_ has quit IRC19:51
*** dxiri has joined #openstack-ansible19:51
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server master: Optimize CentOS Erlang installation  https://review.openstack.org/49637619:54
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server master: Upgrade to RabbitMQ 3.6.11  https://review.openstack.org/49680119:54
*** esberglu has quit IRC19:55
*** vnogin has quit IRC20:01
*** armaan has quit IRC20:03
*** esberglu has joined #openstack-ansible20:03
evrardjp_fellypefca-afk: you can propose to pike and others, don't worry :)20:07
*** esberglu has quit IRC20:07
*** smatzek has quit IRC20:14
admin0pinpointed my 404 in glance to this line:  2017-09-05 22:15:38.214 3511 INFO eventlet.wsgi.server [req-29bcf0e6-47b0-40c0-9352-b0a001f52b0b c1e056c28217422887adc7302d565d38 c97e9a0814fe49179735282894bd59f9 - default default] 10.10.0.9,172.29.236.12 - - [05/Sep/2017 22:15:38] "GET /v2/images HTTP/1.1" 404 261 0.583009 — how to inteprete this as why there is a 404 :D20:16
admin0issue:   admin can list images, tenant gets a 404 .. same tenant on using horizon can view the images .. tenant credentials from the utility container also works ..20:17
admin0m stuck for almost a week now20:17
*** eumel8 has joined #openstack-ansible20:18
*** shardy_afk has quit IRC20:25
*** woodard has quit IRC20:30
*** woodard has joined #openstack-ansible20:30
*** marst has quit IRC20:33
openstackgerritMajor Hayden proposed openstack/ansible-hardening master: Check apparmor_status output  https://review.openstack.org/50096120:41
lbragstadhey folks - quick question on https://governance.openstack.org/tc/goals/queens/policy-in-code.html20:43
lbragstadjust double checking - but osa shouldn't have anything on the hook for that in queens, right?20:43
lbragstadif so - i've proposed a patch to governance omitting openstack-ansible from the goal https://review.openstack.org/50097920:47
*** lucasxu has quit IRC20:48
openstackgerritMajor Hayden proposed openstack/openstack-ansible-pip_install master: Use RDO Pike repository  https://review.openstack.org/50098020:48
openstackgerritMajor Hayden proposed openstack/openstack-ansible-openstack_hosts master: Use RDO Pike repository  https://review.openstack.org/50098120:50
openstackgerritMajor Hayden proposed openstack/openstack-ansible-tests master: [TEST] Test with RDO Pike repo  https://review.openstack.org/50098220:52
evrardjp_lbragstad: thanks champion20:53
evrardjp_;)20:53
lbragstadevrardjp_: thanks for the confirmation, expert!20:53
openstackgerritMajor Hayden proposed openstack/ansible-hardening master: Check apparmor_status output  https://review.openstack.org/50096120:54
mhaydenandymccr: CentOS now has a pike repo20:54
evrardjp_lbragstad: I hear everything nowadays!20:54
evrardjp_me expert? :p20:55
mhaydenya;; are both special to me20:55
mhaydeny'all even20:55
lbragstadevrardjp_: mhayden told me you're the expert20:55
evrardjp_"no, you're special"20:55
evrardjp_haha20:55
mhaydenNO YOU ARE20:56
*** oneswig has joined #openstack-ansible20:56
admin0!!! YOU ALL ARE !!!20:57
openstackadmin0: Error: "!!" is not a valid command.20:57
admin0 :D20:58
evrardjp_mhayden: :)20:58
*** smatzek has joined #openstack-ansible21:02
*** jheroux has quit IRC21:04
*** gouthamr has quit IRC21:05
mhaydenhwoarang: this should nip our apparmor idempotency issue https://review.openstack.org/50092421:08
spotzuh oh what did I miss here?!!21:14
mhaydenspotz: evrardjp_ and lbragstad are enjoying each other's company on irc21:16
mhaydensome kinda secret PTL code21:16
spotzhrrrm mhayden.... lbragstad and evrardjp_ you 2 behave!!:)21:16
lbragstadpsh - it's not even next week yet21:17
lbragstadyou just wait until the ptg21:18
*** thorst has quit IRC21:18
spotzI'll be in LA:(21:18
spotzI told them to carry around a picture of a dalmatian to include me21:19
*** smatzek has quit IRC21:19
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server master: Optimize CentOS Erlang installation  https://review.openstack.org/49637621:20
mhaydenandymccr: also, i think we're bringing lxc-net up on centos but we shouldn't be doing so :|21:31
openstackgerritMajor Hayden proposed openstack/openstack-ansible-lxc_hosts master: Disable lxc-net on CentOS  https://review.openstack.org/50099521:37
mhaydenandymccr: ^^21:37
mhaydenjust tested that in my ocata CentOS deployment21:37
mhaydenand it works okay21:37
*** thorst has joined #openstack-ansible21:37
* mhayden apologizes to the group for suggesting we run lxc-net in the first place :(21:38
openstackgerritMajor Hayden proposed openstack/openstack-ansible-rabbitmq_server master: Upgrade to RabbitMQ 3.6.11  https://review.openstack.org/49680121:39
*** thorst has quit IRC21:42
hwoarangmhayden: ok thanks21:47
andymccrmhayden: weird - i didnt have any dns if lxc-net wasnt up properly.21:56
*** galstrom is now known as galstrom_zzz22:01
*** thorst has joined #openstack-ansible22:07
*** oneswig has quit IRC22:08
*** weezS has quit IRC22:08
*** thorst has quit IRC22:10
*** weezS has joined #openstack-ansible22:18
*** dxiri_ has joined #openstack-ansible22:23
*** dxiri has quit IRC22:26
*** zcourts has quit IRC22:40
*** esberglu has joined #openstack-ansible22:50
*** esberglu has quit IRC22:55
*** weezS has quit IRC22:57
*** toddnni has quit IRC22:58
*** toddnni has joined #openstack-ansible22:58
*** thorst has joined #openstack-ansible23:11
*** sauloaislan has quit IRC23:11
*** thorst has quit IRC23:16
*** gouthamr has joined #openstack-ansible23:17
*** TxGirlGeek has quit IRC23:18
*** thorst has joined #openstack-ansible23:21
*** thorst has quit IRC23:25
*** weezS has joined #openstack-ansible23:33
*** weezS has quit IRC23:38
*** markvoelker has quit IRC23:43
*** yifei has joined #openstack-ansible23:59

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