Friday, 2019-09-20

*** mrda has quit IRC00:28
*** jcoufal has joined #tripleo00:31
*** mmethot_ has quit IRC00:40
*** mmethot has joined #tripleo00:40
*** mmethot has quit IRC00:45
*** mmethot has joined #tripleo00:45
*** mmethot has quit IRC00:46
*** mmethot has joined #tripleo00:47
*** mmethot has quit IRC00:55
*** mmethot_ has joined #tripleo00:55
*** jaganathan has quit IRC01:01
*** mschuppert has quit IRC01:19
openstackgerritMerged openstack/tripleo-puppet-elements stable/rocky: Add proper permissions on script  https://review.opendev.org/68253501:29
*** mrda has joined #tripleo01:35
openstackgerritMerged openstack/tripleo-puppet-elements stable/stein: Add proper permissions on script  https://review.opendev.org/68253401:40
openstackgerritMerged openstack/tripleo-upgrade stable/stein: Explicitly make a list  https://review.opendev.org/68267401:40
openstackgerritMerged openstack/tripleo-common master: mistral-executor image install novajoin package  https://review.opendev.org/68252701:52
*** pierrepr1netti has joined #tripleo01:53
*** mcornea has joined #tripleo01:55
*** ricolin has joined #tripleo01:55
*** pierreprinetti has quit IRC01:56
*** jcoufal has quit IRC01:56
openstackgerritMerged openstack/paunch stable/stein: Add --cpuset-cpus support for both Docker and Podman  https://review.opendev.org/68248401:58
openstackgerritMerged openstack/ansible-role-tripleo-modify-image master: Update README examples on yum update/install  https://review.opendev.org/68237001:58
*** aakarsh has joined #tripleo02:00
*** mcornea has quit IRC02:04
*** dsneddon has quit IRC02:04
openstackgerritMerged openstack/paunch master: Cleanup useless warnings for podman  https://review.opendev.org/68265102:14
openstackgerritMerged openstack/puppet-tripleo master: Add basic unit tests for tripleo::profile::base::keystone  https://review.opendev.org/68251902:14
openstackgerritMerged openstack/tripleo-upgrade master: Use Python3 client for OSP15  https://review.opendev.org/68097002:14
openstackgerritMerged openstack/paunch master: Bump the openstackdocstheme extension to 1.20  https://review.opendev.org/67376302:14
openstackgerritMerged openstack/paunch master: Blacklist sphinx 2.1.0 (autodoc bug)  https://review.opendev.org/67376202:14
*** dsneddon has joined #tripleo02:17
*** dsneddon has quit IRC02:22
*** rlandy|bbl is now known as rlandy02:37
*** HardCase has quit IRC02:41
*** dsneddon has joined #tripleo02:45
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move nested bash script into a real file  https://review.opendev.org/68327002:51
*** ricolin_ has joined #tripleo03:04
*** rfolco has quit IRC03:05
*** ricolin has quit IRC03:06
mwhahahacloudnull, EmilienM: so I think one issue with dropping the pool numbers is that we communicate with too many services with the pool so it's constantly thrashing new connections03:08
mwhahahacloudnull, EmilienM: Looking in my logs we talk to docker.io, the local registry, cloudflare. so each time we talk to a different location, it's closing the pool to the previous place03:08
mwhahahaI think we need a large connection pool but smaller threads03:09
weshayeveryone should have a large pool, beer, bbq03:14
* mwhahaha is currently consuming canadian beer03:14
* mwhahaha raises one for EmilienM 03:14
weshaythe nicest beer in the world03:14
mwhahahait appologizes for getting you hungover03:14
weshaylolz03:14
openstackgerritLogan V proposed openstack/diskimage-builder master: Ensure machine-id is not included in images  https://review.opendev.org/68327203:15
cloudnullmwhahaha that makes a lot of sense03:16
cloudnullup the pool to something like 32 (maybe 64?), but lower worker thread max ?03:16
mwhahahawe could try that03:17
openstackgerritLogan V proposed openstack/diskimage-builder master: Ensure machine-id is not included in images  https://review.opendev.org/68327203:17
mwhahahai'm currently running a prep to gather logs, i'll run that tomorrow03:17
mwhahahais too late to be dealing with such shenanigans03:17
openstackgerritLogan V proposed openstack/diskimage-builder master: Ensure machine-id is not included in images  https://review.opendev.org/68327203:17
cloudnull++03:18
cloudnullwe had https://review.opendev.org/#/c/682731/9/tripleo_common/image/image_uploader.py going03:18
cloudnullmaybe a good start ?03:18
cloudnullits still running, only had to wait ~14 hours, so should be done soon03:19
weshayya man.. it gets ugly when this stuff happens03:19
cloudnullindeed03:19
cloudnullbut good seeing some things merge03:19
weshaywe had a good run :)03:19
mwhahahacloudnull: yea that's the patch i pulled actually03:19
weshayya..03:19
cloudnullthere was a bit of a merge party earlier03:20
cloudnullwhich was good03:20
mwhahahacloudnull: oh did we ever revert the session close stuff?03:22
mwhahahathat might also be contributing03:22
mwhahahano we did not03:23
mwhahahawill need to try that as well03:23
cloudnullwe did not03:23
cloudnullhttps://review.opendev.org/#/c/682719/03:24
cloudnullwe have that one here. we could propose that on its own03:24
cloudnullright now its stacked03:24
mwhahahayikes it took 31 mins for my run w/ the lower counts03:31
openstackgerritPranali Deore proposed openstack/puppet-tripleo master: WIP, Add glance multiple backend support  https://review.opendev.org/68327503:32
*** psachin has joined #tripleo03:36
*** rlandy has quit IRC03:38
*** ykarel|away has joined #tripleo03:39
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-ansible master: Use async to ensure faster task execution  https://review.opendev.org/68327703:45
*** ramishra has joined #tripleo03:47
*** gkadam has joined #tripleo03:58
*** gkadam has quit IRC03:58
*** skramaja has joined #tripleo03:59
*** dsneddon has quit IRC03:59
*** dsneddon has joined #tripleo04:01
*** dsneddon has quit IRC04:05
*** ykarel|away is now known as ykarel04:11
*** karthiks has joined #tripleo04:12
*** soniya29 has joined #tripleo04:24
*** rh-jelabarre has quit IRC04:32
*** karthiks has quit IRC04:33
openstackgerritwes hayutin proposed openstack/tripleo-quickstart master: DNM - test CI against quay.io  https://review.opendev.org/68321204:37
*** karthiks has joined #tripleo04:39
*** ratailor has joined #tripleo04:40
*** _mmethot_ has joined #tripleo04:41
*** mmethot_ has quit IRC04:41
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move nested bash script into a real file  https://review.opendev.org/68327004:42
*** surpatil has joined #tripleo04:43
*** weshay has quit IRC04:52
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-ansible master: Use async to ensure faster task execution  https://review.opendev.org/68327704:56
*** ade_lee_ has joined #tripleo04:57
*** ade_lee__ has quit IRC04:57
*** ade_lee_ has quit IRC05:04
*** ade_lee_ has joined #tripleo05:04
*** holser has joined #tripleo05:08
*** holser has quit IRC05:12
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move basic global functions into ansible tasks  https://review.opendev.org/68328605:15
*** pcaruana has joined #tripleo05:28
*** dsneddon has joined #tripleo05:34
*** dsneddon has quit IRC05:40
*** ykarel has quit IRC05:43
*** pcaruana has quit IRC05:53
*** ricolin_ is now known as ricolin05:54
openstackgerritSagi Shnaidman proposed openstack/tripleo-heat-templates master: Don't use pacemaker in ironic sa job  https://review.opendev.org/68241005:55
*** ksambor has joined #tripleo05:57
openstackgerritSagi Shnaidman proposed openstack/tripleo-heat-templates master: Remove docker resource from scen004  https://review.opendev.org/68328905:57
*** lmiccini has joined #tripleo05:59
*** ykarel has joined #tripleo06:03
*** gfidente has joined #tripleo06:06
openstackgerritMerged openstack/ansible-role-collect-logs master: Don't gzip ARA when generating HTML  https://review.opendev.org/68303906:07
*** apetrich has quit IRC06:08
TenguEmilienM: hey! I didn't follow all the thing about the CI issue with inflight validations, but there's already a way to deactivate them in the CLI directly, at least in master. iirc you even reviewed the change, at least mwhahaha did. So maybe it's "just" a matter of changing the defaults in there instead of adding a new param in t-h-t?06:10
TenguEmilienM: also..... the spec emphasis multiple times the inflight should NOT be invasive nor too long nor too expensive in term of resources. Maybe the real thing is to understand why the nova inflight takes so long, and put it back on the bench instead of dropping all the existing inflight checks?06:11
Tenguanyway. we can talk about that on Monday once I'm really back ;).06:11
openstackgerritMerged openstack/paunch master: Improve volume validation  https://review.opendev.org/68214106:11
*** Tengu has quit IRC06:16
*** Tengu has joined #tripleo06:17
*** jfrancoa has joined #tripleo06:19
*** slaweq has joined #tripleo06:19
*** Tengu has quit IRC06:20
*** Tengu has joined #tripleo06:21
*** jfrancoa has quit IRC06:23
*** psachin has quit IRC06:23
*** Tengu has quit IRC06:28
*** mschuppert has joined #tripleo06:28
openstackgerritMerged openstack/python-tripleoclient master: Add ovn_dbs_vip to export_data  https://review.opendev.org/68229406:32
openstackgerritMerged openstack/tripleo-heat-templates stable/queens: Revert "Move nova cell v2 discovery to deploy_steps_tasks"  https://review.opendev.org/68264406:32
*** kopecmartin|off is now known as kopecmartin06:36
openstackgerritNatal Ngétal proposed openstack/python-tripleoclient master: Add test for the function check_deprecated_parameters.  https://review.opendev.org/67453106:36
*** xek has joined #tripleo06:36
*** psachin has joined #tripleo06:39
*** dciabrin has joined #tripleo06:40
openstackgerritKamil Sambor proposed openstack/tripleo-heat-templates master: Add posibilities to set ovn_openflow_probe_interval for controller  https://review.opendev.org/68329406:40
*** holser has joined #tripleo06:41
*** Tengu has joined #tripleo06:42
openstackgerritSagi Shnaidman proposed openstack/tripleo-heat-templates master: Remove docker resource from scen004  https://review.opendev.org/68328906:43
*** Tengu has quit IRC06:44
openstackgerritIan Wienand proposed openstack/diskimage-builder master: Install Python 3 libselinux packages for Fedora  https://review.opendev.org/68329506:45
*** zbr is now known as zbr|ruck06:48
*** Tengu has joined #tripleo06:50
*** bogdando has joined #tripleo06:57
*** hberaud|gone is now known as hberaud06:59
*** gfidente has quit IRC07:04
*** ccamacho has joined #tripleo07:13
*** marios has joined #tripleo07:15
*** tosky has joined #tripleo07:18
*** udesale has joined #tripleo07:20
*** yolanda has quit IRC07:21
*** amoralej|off is now known as amoralej07:23
*** yolanda has joined #tripleo07:23
*** rcernin has quit IRC07:29
*** rpittau|afk is now known as rpittau07:31
*** yolanda has quit IRC07:32
openstackgerritIan Wienand proposed openstack/diskimage-builder master: Install Python 3 libselinux packages for Fedora  https://review.opendev.org/68329507:33
openstackgerritMartin Schuppert proposed openstack/tripleo-heat-templates stable/queens: [WIP] Queens-only: Run cellv2 host discovery via workflow or deploy_steps_tasks  https://review.opendev.org/68306907:35
*** dsneddon has joined #tripleo07:36
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/ansible-role-tripleo-modify-image master: Add support for yum caching when buildah updates  https://review.opendev.org/68310007:41
*** apetrich has joined #tripleo07:41
*** jpena|off is now known as jpena07:42
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/ansible-role-tripleo-modify-image master: Add support for yum caching when buildah updates  https://review.opendev.org/68310007:44
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/ansible-role-tripleo-modify-image master: Add support for yum caching when buildah updates  https://review.opendev.org/68310007:46
*** gfidente has joined #tripleo07:47
*** yolanda has joined #tripleo07:47
*** iurygregory has joined #tripleo07:51
*** ykarel is now known as ykarel|lunch07:52
*** soniya29 has quit IRC07:54
*** jpich has joined #tripleo07:54
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/ansible-role-tripleo-modify-image master: Add support for yum caching when buildah updates  https://review.opendev.org/68310007:55
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/ansible-role-tripleo-modify-image master: Add support for yum caching when buildah updates  https://review.opendev.org/68310007:59
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/tripleo-ci master: DNM test CI containers update with yum caching  https://review.opendev.org/68310107:59
openstackgerritAlfredo Moralejo proposed openstack/tripleo-quickstart-extras master: Move from ocata to queens for the initial packages installation  https://review.opendev.org/68331308:02
*** paramite|clone has joined #tripleo08:05
*** paramite has quit IRC08:07
*** dsneddon has quit IRC08:07
*** apetrich has quit IRC08:07
*** tkajinam has quit IRC08:09
*** alexmcleod has joined #tripleo08:14
*** suuuper has joined #tripleo08:32
*** dsneddon has joined #tripleo08:34
openstackgerritKamil Sambor proposed openstack/tripleo-heat-templates master: Add posibilities to set ovn_openflow_probe_interval for controller  https://review.opendev.org/68329408:39
*** cylopez has joined #tripleo08:39
openstackgerritMartin Schuppert proposed openstack/tripleo-heat-templates master: Depreace EC2 API  https://review.opendev.org/68331908:43
odyssey4mebogdando: Any chance for a re-review for https://review.opendev.org/682855 now that the pep8 issues are fixed?08:47
*** ykarel|lunch is now known as ykarel08:50
*** rcernin has joined #tripleo08:56
*** florianf has joined #tripleo08:59
*** dsneddon has quit IRC09:01
*** dsneddon has joined #tripleo09:06
*** dsneddon has quit IRC09:10
*** pcaruana has joined #tripleo09:12
bogdandoodyssey4me: done, thanks for this improvement!09:14
bogdandoodyssey4me: please take a look in return https://review.opendev.org/#/c/683100/ :)09:15
bogdandodoing local benchmarking09:15
bogdando10 images , now 36 (all I got for some env deployed)09:15
*** paramite|clone has quit IRC09:22
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/tripleo-ci master: DNM test CI containers update with yum caching  https://review.opendev.org/68310109:23
*** derekh has joined #tripleo09:27
*** dtantsur|afk is now known as dtantsur09:27
*** paramite has joined #tripleo09:29
zbr|ruckanyone that is also core on kolla, i want to merge https://review.opendev.org/#/c/683200/ - thanks.09:30
*** apetrich has joined #tripleo09:41
*** apetrich has quit IRC09:45
openstackgerritSandeep Yadav proposed openstack/tripleo-quickstart-extras master: Replace StandaloneLocalMtu with InterfaceLocalMtu  https://review.opendev.org/68334309:46
odyssey4mebogdando: review done, some comments/questions there09:48
*** rcernin has quit IRC09:49
*** hberaud is now known as hberaud|lunch09:54
*** paramite has quit IRC09:56
*** apetrich has joined #tripleo09:59
openstackgerritMerged openstack/puppet-tripleo master: Correct wrong name in certmonger_certificate for ovn_controller  https://review.opendev.org/68258610:01
*** jaosorior has quit IRC10:03
*** gfidente has quit IRC10:05
openstackgerritArx Cruz proposed openstack/tripleo-quickstart-extras master: Adding tests to skip list in Rocky  https://review.opendev.org/68335010:08
DamjanekHmm. My RHOSP13 deployment dies ~10min after running openstack overcloud deploy. It did not yet managed to start stack creation. The message I'm seeing is: "Timed out waiting for messages from Execution (ID: f9852847-e5a3-4ecc-af0b-ec96a7737f0f, State: RUNNING). The WebSocket timed out before the Workflow completed." I can't find anything suspicious in mistral logs or anywhere.10:08
DamjanekAny pointers, where I should look for issues?10:08
slaweqarxcruz|rover: chandankumar: hi, do You remember my patch for tobiko to add there tripleo based job?10:09
openstackgerritArx Cruz proposed openstack/tripleo-quickstart-extras master: Adding tests to skip list in Rocky  https://review.opendev.org/68335010:09
arxcruz|roverslaweq: yes...10:09
slaweqarxcruz|rover: chandankumar: it was https://review.opendev.org/#/c/651828/ but it seems that I will need multinode environment to run tobiko10:09
chandankumarslaweq: ok10:09
slaweqmostly because tobiko spawns resources for tests and don't clean them - it's intentional10:09
arxcruz|roverhmmmmm10:10
chandankumarslaweq: then you need to switch to fs01010:10
slaweqand on single node we can be out of e.g. memory10:10
openstackgerritSandeep Yadav proposed openstack/tripleo-docs master: Replace StandaloneLocalMtu with InterfaceLocalMtu  https://review.opendev.org/68335110:10
slaweqchandankumar: arxcruz|rover so I started new patch https://review.opendev.org/#/c/683126/10:10
chandankumarslaweq: tripleo-ci-centos-7-containers-multinode10:10
openstackgerritMerged openstack/tripleo-ansible master: Let ReaR uses different backup strategy  https://review.opendev.org/68258310:10
slaweqbut it fails for me for now on multinode-overcloud.yaml10:10
slaweqhttps://storage.bhs1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_cc9/683126/2/check/tobiko-tripleo-containers-update/cc9e3ca/job-output.txt10:11
slaweqcan You take a look on it? Maybe You will quickly know what I'm missing again10:11
chandankumarzbr|ruck: https://storage.bhs1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_cc9/683126/2/check/tobiko-tripleo-containers-update/cc9e3ca/logs/undercloud/home/zuul/overcloud_deploy.log.txt.gz10:12
slaweqchandankumar: fs010 will have heat? I don't remember now why by I choose fs016 - will it be good also or not?10:12
chandankumarzbr|ruck: resources.ServiceChain: Property error: resources[59].properties: Property DockerRsyslogConfigImage not assigned issue known?10:12
arxcruz|roverslaweq: 2019-09-19 22:58:22 |     resources.ServiceChain: Property error: resources[59].properties: Property DockerRsyslogConfigImage not assigned10:12
arxcruz|roverchandankumar: nope, not known10:12
chandankumarslaweq: fs016 is triggered in rdo thirdparty or periodic job in rdo side10:12
slaweqchandankumar: ok, so featureset '010' than would be better, right?10:14
chandankumarslaweq: yes,10:15
slaweqchandankumar: now, another question, should I add ci/environments/scenario002-multinode-containers.yaml to be used on this job? to have e.g. heat installed10:15
slaweqchandankumar: and if yes, how I should add it to job definition?10:16
slaweqchandankumar: should it be something like:10:16
slaweqcomposable_scenario: scenario002-multinode-containers.yaml10:16
DamjanekOk. Seems that this is an issue:10:16
Damjanekhttps://bugzilla.redhat.com/show_bug.cgi?id=166405510:16
openstackbugzilla.redhat.com bug 1664055 in openstack-tripleo-common "Message collection size is too large for Zaqar" [High,Post] - Assigned to apetrich10:16
slaweqin feature_override section?10:17
openstackgerritMartin Schuppert proposed openstack/tripleo-common stable/queens: [WIP] Add workflow to do cellv2 host discovery  https://review.opendev.org/68306710:18
chandankumarslaweq: let me check that10:18
slaweqchandankumar: thx a lot10:18
chandankumarslaweq: Here is the job defined https://opendev.org/openstack/tripleo-ci/src/branch/master/zuul.d/multinode-jobs.yaml#L18310:21
chandankumarslaweq: here is the multinode.yaml https://github.com/openstack/tripleo-quickstart-extras/blob/master/playbooks/multinode.yml10:25
chandankumarslaweq: you can reorder the playbook in same and it will work10:25
openstackgerritMerged openstack/puppet-tripleo master: Use memcached for token caching in barbican authtoken  https://review.opendev.org/68137110:27
*** artom has quit IRC10:32
slaweqchandankumar: thx a lot10:34
openstackgerritSlawek Kaplonski proposed openstack/tripleo-heat-templates master: Enable "port_forwarding" feature in neutron ML2 ovs environment  https://review.opendev.org/68335710:35
EmilienMTengu: they were enabled by default10:36
slaweqchandankumar: but in my job I have those playbooks in same order10:40
slaweqexcept multinode-overcloud-upgrade.yml but job is failing before that even10:40
*** paramite has joined #tripleo10:41
*** paramite|clone has joined #tripleo10:45
*** paramite has quit IRC10:46
*** brault has joined #tripleo10:47
chandankumarEmilienM: need some help on this failure http://storage.bhs1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_cc9/683126/2/check/tobiko-tripleo-containers-update/cc9e3ca/logs/undercloud/home/zuul/overcloud_deploy.log.txt.gz10:52
chandankumarEmilienM:  resources.ServiceChain: Property error: resources[59].properties: Property DockerRsyslogConfigImage not assigned10:52
openstackgerritMerged openstack/tripleo-ci master: Run build containers on rocky branch  https://review.opendev.org/68233510:53
*** pbandark has joined #tripleo10:54
*** pcaruana has quit IRC10:56
*** cylopez has quit IRC10:56
EmilienMchandankumar: I'm on pto today10:56
chandankumarEmilienM: ok10:56
*** jpena is now known as jpena|lunch11:01
*** panda is now known as panda|lunch11:03
*** artom has joined #tripleo11:05
*** dsneddon has joined #tripleo11:06
*** apetrich has quit IRC11:07
*** raildo has joined #tripleo11:07
openstackgerritMerged openstack/tripleo-upgrade stable/stein: Fix image validation when AP services get several images.  https://review.opendev.org/68306611:08
openstackgerritMerged openstack/tripleo-upgrade master: Fix passing tags through infrared to the tripleo-upgrade role.  https://review.opendev.org/68126811:17
*** lucasagomes has joined #tripleo11:19
*** ccamacho has quit IRC11:21
*** hberaud|lunch is now known as hberaud11:26
*** jaosorior has joined #tripleo11:27
*** pcaruana has joined #tripleo11:29
*** dsneddon has quit IRC11:31
*** rh-jelabarre has joined #tripleo11:32
*** chandankumar is now known as raukadah11:33
*** pcaruana has quit IRC11:42
*** pcaruana has joined #tripleo11:42
*** dsneddon has joined #tripleo11:42
*** dmellado has quit IRC11:45
TenguEmilienM: I know. not sure if it's a good thing to deactivate, since a feature won't be tested as it should :/.11:47
*** jaosorior has quit IRC11:47
Tengui.e. if a validation takes too long, maybe it's an issue to correct in order to make it faster, instead of dropping them all.11:47
*** dsneddon has quit IRC11:47
*** dmellado has joined #tripleo11:48
*** apetrich has joined #tripleo11:51
*** udesale has quit IRC11:52
*** two_oes has joined #tripleo11:52
*** jpena|lunch is now known as jpena11:53
*** udesale has joined #tripleo11:54
openstackgerritMerged openstack/puppet-tripleo stable/stein: Safeguard against 'Undef' length() checks.  https://review.opendev.org/68238511:55
apetrichDamjanek, where is it the message too large issue happening? Can I help with something?11:55
*** morazi has joined #tripleo11:57
*** weshay has joined #tripleo12:00
*** dsneddon has joined #tripleo12:01
Damjanekapetrich: I'm observing that in mistral engine log. When applied a workaround with bumping sizes for both zaqar and mistral and then rebooted undercloud (because for some reason zaqar and mistral services restart was not enough), it went thru12:02
DamjanekBtw, is this normal behavour when using idrac pm driver, that every single time node is being rebooted by ironic, it reconfigures bios?12:03
DamjanekOr I have some wrong config or something?12:03
*** artom has quit IRC12:04
apetrichDamjanek, weird just rekicking the services is not enough. but good that it passed12:04
apetrichDamjanek, anyway feel free to ping if you have another mistral issue :)12:04
Damjanekapetrich: Will keep that in mind. Thanks!12:05
*** dsneddon has quit IRC12:06
openstackgerritLukas Bezdicka proposed openstack/tripleo-heat-templates master: Workaround to avoid puppet managing internal files.  https://review.opendev.org/65440712:07
*** panda|lunch is now known as panda12:16
*** amoralej is now known as amoralej|lunch12:19
openstackgerritMerged openstack/tripleo-common master: Add missing allocations for active nodes  https://review.opendev.org/67869812:23
openstackgerritMerged openstack/tripleo-upgrade master: Ensure pacemaker bootstrap controller is upgraded first.  https://review.opendev.org/68059512:23
openstackgerritBrent Eagles proposed openstack/tripleo-quickstart master: Add support for running tempest against Octavia featuresets  https://review.opendev.org/66966612:23
openstackgerritBrent Eagles proposed openstack/tripleo-quickstart master: Set the amphora image to stein branch  https://review.opendev.org/67889912:23
*** ratailor has quit IRC12:25
*** psachin has quit IRC12:28
*** skramaja has quit IRC12:37
*** jaosorior has joined #tripleo12:38
*** rlandy has joined #tripleo12:38
*** cylopez has joined #tripleo12:39
*** apetrich has quit IRC12:42
*** rfolco has joined #tripleo12:46
*** aakarsh has quit IRC12:47
*** mcornea has joined #tripleo12:50
*** rcernin has joined #tripleo12:57
openstackgerritSlawek Kaplonski proposed openstack/tripleo-heat-templates master: Enable "port_forwarding" feature in neutron ML2 ovs environment  https://review.opendev.org/68335713:01
openstackgerritKamil Sambor proposed openstack/tripleo-heat-templates master: WIP: Support connecting OVN DB over SSL  https://review.opendev.org/68034513:01
openstackgerritSergii Golovatiuk proposed openstack/tripleo-upgrade stable/stein: Fix passing tags through infrared to the tripleo-upgrade role.  https://review.opendev.org/68337913:05
openstackgerritSlawek Kaplonski proposed openstack/tripleo-heat-templates master: Remove leftovers after NeutronExternalNetworkBridge option  https://review.opendev.org/68338013:05
openstackgerritMerged openstack/os-net-config master: Add linkdelay option  https://review.opendev.org/68056913:06
*** dsneddon has joined #tripleo13:08
*** paramite|clone has quit IRC13:13
*** brault has quit IRC13:13
*** dsneddon has quit IRC13:13
*** aakarsh has joined #tripleo13:15
openstackgerritBrent Eagles proposed openstack/tripleo-heat-templates master: Add Octavia driver agent service  https://review.opendev.org/65811813:22
openstackgerritMartin Schuppert proposed openstack/tripleo-common stable/queens: [Queens] Add workflow to do cellv2 host discovery  https://review.opendev.org/68306713:25
openstackgerritMartin Schuppert proposed openstack/tripleo-heat-templates stable/queens: [Queens] Run cellv2 host discovery via workflow or deploy_steps_tasks  https://review.opendev.org/68306913:25
*** munimeha1 has joined #tripleo13:26
*** amoralej|lunch is now known as amoralej13:27
openstackgerritGael Chamoulaud proposed openstack/tripleo-validations master: Add new healthcheck-service-status validation  https://review.opendev.org/68237713:30
*** gfidente has joined #tripleo13:31
*** gfidente has quit IRC13:33
*** devep has joined #tripleo13:35
ykarelslaweq, regarding the issue u faced, DockerRsyslogConfigImage, chandankumar asked me to look13:37
ykarelslaweq, https://review.opendev.org/#/c/624330/ missed changing docker image names to container like https://review.opendev.org/#/c/651550/1313:37
ykarelhttps://review.opendev.org/#/c/624330/14/deployment/logging/rsyslog-container-puppet.yaml13:37
ykarelDockerRsyslogImage and DockerRsyslogConfigImage are outdated13:38
ykarelcorrect names are https://opendev.org/openstack/tripleo-common/src/branch/master/container-images/overcloud_containers.yaml.j2#L924-L92513:38
*** apetrich has joined #tripleo13:39
openstackgerritBogdan Dobrelya (bogdando) proposed openstack/tripleo-specs master: Add a spec for K8s Metal3 baremetal operator  https://review.opendev.org/65928813:39
slaweqykarel: thx for looking but how I can/should fix it?13:40
slaweqI'm not playing with those variables in my patch at all13:40
ykarelslaweq, so u using fs016 multinode13:41
openstackgerritAlex Schultz proposed openstack/paunch master: Add --cpuset-cpus support for both Docker and Podman  https://review.opendev.org/68125713:41
ykarelslaweq, https://opendev.org/openstack/tripleo-heat-templates/src/branch/master/ci/environments/scenario001-multinode-containers.yaml#L1713:41
*** devep has quit IRC13:41
ykareland deployment/logging/rsyslog-container-puppet.yaml has wrong parameters13:41
ykarelso you hit it13:41
mwhahahawait does that mean scenario001 doesn't have rsyslog but the multinode job did?13:42
slaweqykarel: ok, so if I changed for fs 010 I should not have this issue anymore, right?13:43
ykarelslaweq, yes atlesat this issue will not be hit13:43
ykarelmwhahaha, you comparing standalone vs multinode?13:44
mwhahahayea that's what i mean13:44
ykarelif yes multinode have rsyslog but standalone don't13:44
mwhahahabecause if you're saying multinode is broken, then we should have found this via standalone13:45
mwhahahathe fix is to fix the Docker* vars on rsyslog tho13:45
mwhahahaand we should have the service enabled for a scenario13:45
ykarelmwhahaha, yes right13:45
ykarel001-multinode are not running in master13:45
ykarelso standalone should cover it13:46
openstackgerritAlex Schultz proposed openstack/paunch stable/stein: Improve volume validation  https://review.opendev.org/68338613:46
*** surpatil has quit IRC13:47
*** weshay is now known as weshay|ruck13:47
ykarelcloudnull, mwhahaha can u check https://review.opendev.org/#/c/681876/13:50
ykareldumb-init has broken jobs in queens/rocky13:50
mwhahaha2019-09-20 05:56:13 |         "/usr/bin/mysqld_safe: line 755: ulimit: -1: invalid option",13:50
ykarelnope13:50
mwhahaha?13:50
ykareldumb-init not found13:50
mwhahaha2019-09-20 05:56:13 |         "stderr: container_linux.go:235: starting container process caused \"exec: \\\"dumb-init\\\": executable file not found in $PATH\"",13:50
ykarelyes13:51
openstackgerritLuca Miccini proposed openstack/puppet-tripleo stable/stein: Enable deep_compare for fencing resource  https://review.opendev.org/68338813:51
mwhahahaso dumb init is missing from haproxy13:51
openstackgerritLuca Miccini proposed openstack/puppet-tripleo stable/rocky: Enable deep_compare for fencing resource  https://review.opendev.org/68339013:51
*** apetrich has quit IRC13:51
mwhahahahaproxy/rabbit/mysql13:51
openstackgerritLuca Miccini proposed openstack/puppet-tripleo stable/queens: Enable deep_compare for fencing resource  https://review.opendev.org/68339113:51
mwhahahabandini: did we have to do something to fix dumb-init for haproxy/rabbit/mysql containers13:51
ykarelmwhahaha, issue is dumb-init is installed there, but /usr/bin is bind mounted in container13:51
slaweqykarel: thx13:51
mwhahahalol13:51
ykarelso binaries in container are lost13:52
mwhahahabrilliant13:52
ykarel:D13:52
* mwhahaha blames bandini for good measure13:52
mwhahahameh i'll look into the logs for stein to figure out what we did later13:52
* bandini totally guilty as usual13:52
bandiniwhat's broken exactly?13:53
mwhahahawe switched to dumb-init13:54
mwhahahawhich is loaded from /usr/bin13:54
mwhahahait seems the pacemaker containers bindmoutn host /usr/bin13:54
mwhahahamaking dumb-init unavailable13:54
*** artom has joined #tripleo13:54
bandinihohum13:54
mwhahahacan we drop the host bind mount?13:55
mwhahahaseems odd that we do that13:55
mwhahahai would assume it's for the pacemaker binaries maybe?13:55
*** dsneddon has joined #tripleo13:55
bandinido you have a log of the failure and which bind mount are we talking about exactly?13:55
mwhahahahttps://logs.rdoproject.org/openstack-periodic-24hr/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-7-multinode-1ctlr-featureset010-queens/50b2520/logs/undercloud/home/zuul/overcloud_deploy.log.txt.gz#_2019-09-20_05_56_1313:56
mwhahaha2019-09-20 05:56:13 |         "stderr: container_linux.go:235: starting container process caused \"exec: \\\"dumb-init\\\": executable file not found in $PATH\"",13:56
bandiniok let me see13:56
mwhahaha \"/usr/bin:/usr/bin:ro\"13:56
openstackgerritSagi Shnaidman proposed openstack/tripleo-heat-templates master: Don't use pacemaker in ironic sa job  https://review.opendev.org/68241013:57
openstackgerritSagi Shnaidman proposed openstack/tripleo-ci master: Use podman for scenario012  https://review.opendev.org/68234113:57
ykarelmwhahaha, bind mount is there for running docker tag:- https://github.com/openstack/tripleo-heat-templates/blob/stable/queens/docker/services/pacemaker/haproxy.yaml#L213-L21813:59
mwhahahacan we bind mount docker then13:59
ykareland it's not there in train/stein as tripleo-container-tag role is used14:00
mwhahahaand not all of it14:00
mwhahahaor should we just install docker in those containers14:00
ykareland that role directly run docker tag, instead running from containers14:00
ykarelor install dumb-init on host?14:00
mwhahahaactually i thought we installed docker in the containers14:00
*** dsneddon has quit IRC14:00
ykarelthat will be a workaround though14:01
*** gfidente has joined #tripleo14:01
mwhahahawell for minor updates we need to limit it to containers14:01
mwhahahait seems like installing docker in those containers is a better idea14:01
mwhahahawe did that for neutron and stuff14:01
bogdandoactually bind-mount docker binary would be better14:03
*** ayoung has joined #tripleo14:03
bogdandoa statically compiled binary, why installing the whole thing?..14:03
mwhahahaok let's switch to that then14:03
bandiniif that will work, then i think it is the better approach14:03
bandinialthough we need to check if docker binary exist otherwise a podman deploy will fail due to missing bindmount?14:04
bogdandoyes14:04
mwhahahathis is queens/rocky14:04
mwhahahano podman14:04
mwhahahawe don't have this problem stein+14:04
mwhahahaas mentioned we switched to something else14:04
bandiniok so let's tweak just in pre stein to /usr/bin/docker and see if CI does not barf on us14:04
mwhahahathis showed up because we're backporting dumb-init14:04
bandiniright14:05
bogdandomay be we could also backport tagging it on host :)14:05
bogdandonot in container14:05
mwhahahalet's not14:05
bogdando:D14:05
bandiniseems a bit risky to backport14:05
bandinialthough could work with some solid testing14:06
mwhahahalet's start with the docker bind mount :D then we can work on switching if we want14:06
*** openstackgerrit has quit IRC14:06
bandiniagree, want me to propose a rocky review ?14:07
bogdandocautious cautiosness right14:07
mwhahahayes plz14:07
bandiniack14:07
*** openstackgerrit has joined #tripleo14:07
openstackgerritMerged openstack/tripleo-quickstart-extras master: Statically import CI reproducer roles  https://review.opendev.org/68117114:07
*** rascasoft has quit IRC14:10
openstackgerritGael Chamoulaud proposed openstack/tripleo-validations master: Add new healthcheck-service-status validation  https://review.opendev.org/68237714:10
weshay|ruckcloudnull, check this out14:11
openstackgerritLuca Miccini proposed openstack/tripleo-heat-templates stable/stein: Enable deep_compare by default for stonith resources  https://review.opendev.org/68339514:11
weshay|ruck2019-09-20 06:42:33 | RUNNING HANDLER [tripleo-container-image-prepare : Delete param file] **********14:11
weshay|ruck2019-09-20 06:42:33 | task path: /usr/share/ansible/roles/tripleo-container-image-prepare/handlers/main.yml:1814:11
weshay|ruck2019-09-20 06:42:33 | Friday 20 September 2019  06:41:06 +0000 (0:15:06.780)       0:20:53.993 ******14:11
weshay|ruck2019-09-20 06:42:33 | changed: [undercloud] => {14:11
weshay|ruck2019-09-20 06:42:33 |     "changed": true,14:11
weshay|ruck2019-09-20 06:42:33 |     "path": "/tmp/ansible.xiUhDl-prepare-param",14:11
weshay|ruck2019-09-20 06:42:33 |     "state": "absent"14:11
weshay|ruckhttps://storage.bhs1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_9c8/678698/2/check/tripleo-ci-centos-7-containers-multinode/9c86690/logs/undercloud/home/zuul/overcloud_deploy.log.txt.gz14:11
weshay|ruck15min14:11
mwhahahawe didn't do anything?14:11
cloudnullmornings14:12
openstackgerritMichele Baldessari proposed openstack/tripleo-heat-templates stable/rocky: Stop bindmounting /usr/bin  https://review.opendev.org/68339614:12
bandinimwhahaha, dciabrin: ^14:12
bandinilet's see if ci passes14:12
mwhahahak14:12
weshay|ruckcloudnull, we're going to put up a change that removes the handlers here https://review.opendev.org/#/c/674414/24/tripleo_ansible/roles/tripleo-container-image-prepare/handlers/main.yml14:13
weshay|ruck2019-09-19 21:13:31 | RUNNING HANDLER [tripleo-container-image-prepare : Delete param file] **********14:13
weshay|ruck2019-09-19 21:13:31 | Thursday 19 September 2019  21:12:46 +0000 (0:28:32.293)       0:36:39.510 ****14:13
weshay|ruck2019-09-19 21:13:31 | changed: [undercloud] => {14:13
weshay|ruck2019-09-19 21:13:31 |     "changed": true,14:13
weshay|ruck2019-09-19 21:13:31 |     "path": "/tmp/ansible.Q4cd6h-prepare-param",14:13
weshay|ruck2019-09-19 21:13:31 |     "state": "absent"14:13
cloudnullweshay|ruck is it the handler or just that the delete takes too long ?14:13
weshay|ruckdelete takes too long14:14
dciabrinbandini, sec14:14
weshay|ruckarxcruz|rover, is working it ^14:14
openstackgerritGael Chamoulaud proposed openstack/tripleo-validations master: New validation: detect failed containers  https://review.opendev.org/65678514:15
cloudnullmaybe we change those to simply command: rm https://pasted.tech/pastes/50c863cb2f3c45a3511e4366c9393072a154cc3114:15
cloudnullsee if that's in fact faster?14:15
weshay|ruckcloudnull, sounds fine14:16
cloudnullarxcruz|rover ^14:16
weshay|ruckcloudnull, what's being removed there... the containers or yml file?14:16
cloudnulltemp files14:16
cloudnullI suspect the ansible file module is walking the tree, and I guess there's lots of files ?14:17
mwhahahalikely14:17
mwhahahawe can't drop the cleanup, but we can rm -rf14:17
openstackgerritArx Cruz proposed openstack/tripleo-ansible master: Remove notify handler to delete the prepare parameter temp file  https://review.opendev.org/68340214:17
ykarelbandini, we need to run that patch with promotion pipeline where container exists with new entrypoint14:17
*** dsneddon has joined #tripleo14:17
weshay|ruckarxcruz|rover, ^14:17
ykarels/promotion pipeline/promotion jobs14:17
arxcruz|roverweshay|ruck: so rm -rf then ?14:17
weshay|ruckarxcruz|rover, ya14:17
arxcruz|roverokay, abandoning this one14:18
weshay|ruckupdate that review to use shell: rm14:18
arxcruz|roverokay14:18
mwhahahaykarel: no we want to make sur eit doesn't break the current stuff14:18
ykarelmwhahaha, i meant additionally we need to check with promotion jobs14:18
mwhahahaykarel: then we can land it and run it in promotion14:18
mwhahahayea14:18
bandiniyeah let's first see that it passes14:18
ykarelbandini, i can check that locally14:19
cloudnullmake sure to add the ansible lint skip tag there - https://pasted.tech/pastes/58188095b175122dd58f7c69bd8839ad0db1be40 - but other than that the rm method should be faster.14:19
arxcruz|rovercloudnull: ack14:20
ykarelbandini, i got You don't have either docker-client or docker-client-latest installed. Please install either one and retry14:20
ykarelwith /usr/bin/docker bind mount14:20
mwhahahayea that was my concern14:20
mwhahahais that there's more stuff14:20
*** iurygregory_ has joined #tripleo14:21
bandinilet me try here14:21
cloudnullarxcruz|rover you might need to use shell, and make an test to see if the directory exists before running `rm`,14:21
weshay|ruckarxcruz++14:21
cloudnulllet me know how it goes14:21
*** iurygregory has quit IRC14:21
openstackgerritArx Cruz proposed openstack/tripleo-ansible master: Switching the delete param handler to use rm  https://review.opendev.org/68340214:22
*** rcernin has quit IRC14:23
openstackgerritLuca Miccini proposed openstack/tripleo-heat-templates stable/rocky: Enable deep_compare by default for stonith resources  https://review.opendev.org/68340414:23
weshay|ruckarxcruz|rover, doesn't rm .. require shell: and not command?14:23
openstackgerritLuca Miccini proposed openstack/tripleo-heat-templates stable/queens: Enable deep_compare by default for stonith resources  https://review.opendev.org/68340514:23
weshay|ruckI think lint may fail on that14:23
weshay|ruckcloudnull, does that matter? ^14:23
arxcruz|roverweshay|ruck: checking14:23
weshay|ruckthis in theory should bring containers-multinode back to under 3hrs14:24
cloudnullweshay|ruck lint will be ok, arxcruz|rover used the skip_ansible_lint tag so it shouldn't complain14:24
*** hberaud is now known as hberaud|school-r14:25
weshay|ruckah right14:25
openstackgerritArx Cruz proposed openstack/tripleo-ansible master: Switching the delete param handler to use rm  https://review.opendev.org/68340214:25
arxcruz|rovercloudnull: added || true in case the directory doesn't exist14:25
arxcruz|roverso it will not fail the task14:26
*** cylopez has quit IRC14:26
* weshay|ruck notes standalone jobs are ALSO over 3 hours tripleo-ci-centos-7-scenario001-standaloneSUCCESS in 3h 05m 56s14:26
weshay|ruck:(14:26
*** jaosorior has quit IRC14:30
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move basic global functions into ansible tasks  https://review.opendev.org/68328614:30
openstackgerritMichele Baldessari proposed openstack/tripleo-heat-templates stable/rocky: Stop bindmounting /usr/bin  https://review.opendev.org/68339614:30
bandinidciabrin, ykarel, mwhahaha: ^ this approach works for me locally14:30
mwhahahacloudnull: -^14:31
* cloudnull running that review myself 14:31
openstackgerritAlex Schultz proposed openstack/paunch master: Add --cpuset-cpus support for both Docker and Podman  https://review.opendev.org/68125714:32
openstackgerritArx Cruz proposed openstack/tripleo-ansible master: Switching the delete param handler to use rm  https://review.opendev.org/68340214:32
ykarelbandini, yes working for me too14:33
*** hberaud|school-r is now known as hberaud14:33
ykarelarxcruz|rover, fyi ^^14:33
*** dtantsur is now known as dtantsur|afk14:33
arxcruz|roverykarel: sorry man, i was working on another issue, i'm all ears now :)14:34
ykarelarxcruz|rover, np14:34
arxcruz|roverykarel: i'll submit a patch to revert it ok ?14:35
ykarelarxcruz|rover, no14:35
ykarelarxcruz|rover, bandini has pushed a fix14:36
ykarelso let's observe results there14:36
arxcruz|roverykarel: ack14:36
ykarelarxcruz|rover, and regarding master, i reran failed job with https://review.rdoproject.org/r/2216314:37
openstackgerritLuca Miccini proposed openstack/tripleo-heat-templates stable/stein: Enable deep_compare by default for stonith resources  https://review.opendev.org/68340914:37
ykarelmaster promotion14:37
ykarel001,050 passed14:37
ykarel020 still running, 039 failed14:37
ykarelso i think we can consider promotion14:38
arxcruz|roverykarel: okay, let's see14:38
*** rcernin has joined #tripleo14:38
arxcruz|roverabishop: I'll let the job finishe, because we need to ensure that's the issue that is making overcloud timeout14:39
openstackgerritAlex Schultz proposed openstack/paunch master: Add --cpuset-cpus support for both Docker and Podman  https://review.opendev.org/68125714:40
abishoparxcruz|rover: np, good to verify the approach before refining it :)14:41
openstackgerritAlex Schultz proposed openstack/paunch stable/stein: Fixing upper stable url  https://review.opendev.org/68341314:47
*** udesale has quit IRC14:50
*** udesale has joined #tripleo14:51
openstackgerritLuca Miccini proposed openstack/tripleo-heat-templates stable/stein: Enable deep_compare by default for stonith resources  https://review.opendev.org/68340914:53
*** iurygregory_ is now known as iurygregory14:53
arxcruz|roverweshay|ruck: I've update https://bugs.launchpad.net/tripleo/+bug/1844778 with the logs from featureset020 and the Delete param14:54
openstackLaunchpad bug 1844778 in tripleo "Overcloud deployment is timeing out on scenario001 ovb" [Critical,Triaged] - Assigned to Arx Cruz (arxcruz)14:54
*** Vorrtex has joined #tripleo14:56
openstackgerritSlawek Kaplonski proposed openstack/tripleo-heat-templates master: Enable "port_forwarding" feature in neutron ML2 ovs environment  https://review.opendev.org/68335714:59
openstackgerritSergii Golovatiuk proposed openstack/tripleo-upgrade stable/stein: Change docker references to podman.  https://review.opendev.org/68341814:59
*** dsneddon has quit IRC15:00
*** _mmethot_ has quit IRC15:04
*** _mmethot_ has joined #tripleo15:05
*** bnemec is now known as beekneemech15:06
openstackgerritMerged openstack/tripleo-upgrade master: Change docker references to podman.  https://review.opendev.org/68308115:09
weshay|ruckcloudnull, so that handler runs on the undercloud as well15:17
weshay|ruckbut we can't see how long it's running for due to verbosity of ansible-playbook15:17
* weshay|ruck going to put up a change for that15:17
weshay|ruckwe may save 30-50min15:18
mwhahahai've got a change to reduce some of that15:18
mwhahahahttps://review.opendev.org/#/c/682988/15:18
* weshay|ruck looks15:18
weshay|ruckmwhahaha, so.. what I need to help re: perf is a more verbose ansible-playbook call for the undercloud15:19
cloudnullweshay|ruck that would be spectacular15:19
mwhahahaweshay|ruck: you don't need verbose, you need the timing outputs right?15:20
mwhahahabecause there's not really  more verbosity to be had15:20
mwhahahaso we can identify wtf is taking forever consistently?15:20
* weshay|ruck will paste for clarity15:20
openstackgerritSlawek Kaplonski proposed openstack/tripleo-quickstart-extras master: Add playbook and role to run tobiko  https://review.opendev.org/65542315:21
weshay|ruckmwhahaha, http://paste.openstack.org/show/778064/15:21
weshay|ruckthere is no way to know how long that handler took in the undercloud atm15:21
weshay|ruckbecause ansible logging sucks15:21
mwhahahadon't really need that15:22
mwhahahawhat you need is the Thursday 19 September 2019  21:12:46 +0000 (0:28:32.293)       0:36:39.510 ****15:22
mwhahahathat's not a verbosity thing15:22
weshay|ruckya15:22
mwhahahaand that's what i was talking about15:22
weshay|ruckagree15:22
mwhahahaverbosity would be all the changed details (which we don't want)15:22
weshay|ruckok.. so I'm not groking your patch then yet15:22
mwhahahai agree we do want the timing15:22
weshay|ruckk.. ya15:22
mwhahahamy patch clears up all the excessive var details15:22
mwhahahabecuase import_var dumps the vars being imported15:23
mwhahahawhich clutters logs15:23
weshay|ruckya.. getting rid of logs we don't need is ++15:23
weshay|ruckbut I also want the real elapsed time15:23
weshay|rucks/want/need15:24
mwhahahathat's an output plugin thing if i recal15:24
weshay|ruckaye15:24
weshay|ruckwe could update a plugin for sure15:24
mwhahahaprofile_tasks or whatever15:25
weshay|ruckcallback_whitelist = profile_tasks15:25
weshay|ruckhttps://docs.ansible.com/ansible/latest/plugins/callback/profile_tasks.html15:25
*** Vorrtex has quit IRC15:25
mwhahahayea that15:25
mwhahahai don't know where we're setting that in the overcloud15:26
mwhahahaare we hacking that in via ci?15:26
weshay|ruckI'd rather not15:26
mwhahahano i mean is that what we're doing now15:26
mwhahahacause i'm not finding a reference to that in tripleo-common or tripleoclient15:26
weshay|ruckyou are actively working on doing that now? just to be clear?15:26
mwhahahai can, i'm just looking15:27
mwhahahayes we're hacking it in via quickstart atm15:27
mwhahahahttps://opendev.org/openstack/tripleo-quickstart-extras/src/branch/master/roles/overcloud-deploy/ansible.cfg15:27
* mwhahaha throws things are people15:27
mwhahahas/are/at15:27
weshay|ruck3 years ago15:28
weshay|ruckthat is not related to the overcloud / undercloud deployment15:29
* weshay|ruck not clear what you are trying to say atm..15:29
weshay|ruckwe need that set in tripleo-common15:30
mwhahaharight15:30
weshay|ruckit can be in both dude15:30
mwhahahai think it's getting inherieted15:30
mwhahahaanyway15:30
weshay|ruckdoubt it15:30
mwhahahawonder if we should enable full_skip too15:32
mwhahahawill check that out later15:32
mwhahahahttps://docs.ansible.com/ansible/latest/plugins/callback/full_skip.html15:32
weshay|ruckk15:33
mwhahahasending a tripleo-common patch in a sec15:34
*** dpeacock has quit IRC15:34
openstackgerritMerged openstack/tripleo-ansible master: Fix the non-HA case where OVN has its own VIP  https://review.opendev.org/68296815:34
mwhahahaso i think what's happening is that we shell out to ansible for image modify which loads https://opendev.org/openstack/ansible-role-tripleo-modify-image/src/branch/master/ansible.cfg15:34
mwhahahaso thats why timings exist for those15:34
openstackgerritAlex Schultz proposed openstack/tripleo-common master: Add profile_tasks to ansible runs  https://review.opendev.org/68342515:35
*** kopecmartin is now known as kopecmartin|off15:35
*** apetrich has joined #tripleo15:37
openstackgerritAlex Schultz proposed openstack/tripleo-common master: WIP: Add full_skip to default ansible.cfg  https://review.opendev.org/68342815:40
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Convert subprocess to a reusable method  https://review.opendev.org/68342915:40
*** dpeacock has joined #tripleo15:41
weshay|ruckmwhahaha, /me will add the same to tq15:42
openstackgerritAlex Schultz proposed openstack/tripleo-common master: Set display_skipped_host to no  https://review.opendev.org/68342815:42
mwhahahaactually full skip is deprecated15:42
mwhahahathere's a way to do it built in now15:43
mwhahahawant to see what it looks like first15:43
*** rcernin has quit IRC15:44
mwhahahadid gerrit just die?15:47
mwhahahaor is it just me15:47
openstackgerritMerged openstack/tripleo-heat-templates master: Add new parameter options to Octavia service  https://review.opendev.org/68264115:47
mwhahahaweird it's back15:47
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-ansible master: Use shell to ensure faster task execution  https://review.opendev.org/68327715:48
weshay|ruckarxcruz|rover, and look at https://review.opendev.org/#/c/683428/15:50
weshay|ruckarxcruz|rover, I'm going to add the same for tq15:50
*** iurygregory has quit IRC15:52
ykarelweshay|ruck, arxcruz|rover master promoting:- promoter Promoting the container images for dlrn hash 41408527ae88ac42de11ca2818ff405320795a8b on master to current-tripleo15:54
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-ansible master: Switching the delete param handler to use rm  https://review.opendev.org/68340215:54
cloudnullarxcruz|rover updated your review based on some feedback - just to keep that moving - sorry.15:54
weshay|ruckykarel, ?15:54
weshay|ruckykarel, did you change criteria?15:54
ykarelweshay|ruck, nope15:55
*** suuuper has quit IRC15:55
weshay|ruckykarel,  why is there a promotion?15:55
ykarelweshay|ruck, https://trunk-primary.rdoproject.org/api-centos-master-uc/api/civotes_detail.html?commit_hash=41408527ae88ac42de11ca2818ff405320795a8b&distro_hash=9e5988125e88f803ba20743be7aa99079dd275f215:55
ykarelall required job passed15:55
*** cylopez has joined #tripleo15:56
weshay|ruckreally?15:56
mwhahahapffft don't be shocked :D15:56
mwhahahawe write really awesome code!15:56
weshay|rucklolz15:56
ykarelweshay|ruck, run fail jobed in check, those were failed at overcloud deploy timeout15:56
ykarelhttps://review.rdoproject.org/r/#/c/22163/15:57
*** rpittau is now known as rpittau|afk15:57
*** lmiccini has quit IRC15:58
weshay|ruckok.. thanks ykarel15:59
*** ykarel is now known as ykarel|away16:00
*** jpich has quit IRC16:02
*** fultonj has quit IRC16:04
arxcruz|rovercloudnull: no problem as far as the promotion pipeline move forward :D16:04
arxcruz|roverI don't want to have nightmares on friday :P16:04
arxcruz|roverbrb, upgrade firmware16:05
*** two_oes has quit IRC16:05
*** two_oes has joined #tripleo16:06
bogdandoas far as we have here a little optimization party, PTAL https://review.opendev.org/#/c/683100/16:08
bogdandoeach saved minute saves a kitten16:08
*** Vorrtex has joined #tripleo16:09
*** cylopez has quit IRC16:13
weshay|ruckbogdando, ah.. that is a sweet way to express that16:18
*** pcaruana has quit IRC16:19
bogdandowhatever encourages you folks merge that16:19
bogdando:D16:19
bogdandowill fix pep issues after the testing CI job passes16:19
bogdando*completes16:20
odyssey4mebogdando: LGTM other than that pesky pep8 issue16:27
bogdandoodyssey4me: thanks! I'll push an update shortly. Don't want to spoil https://review.opendev.org/#/c/683101/ that checks that patch but with caching ON16:28
*** holser__ has joined #tripleo16:29
*** onmywaytosunset has joined #tripleo16:30
openstackgerritwes hayutin proposed openstack/tripleo-quickstart-extras master: remove unused ansible.cfg file  https://review.opendev.org/68350316:30
*** jpena is now known as jpena|off16:31
arxcruz|roverbogdando: would be faster instead of a cat, save one beer16:31
arxcruz|roverfor me16:31
bogdandoheh16:31
*** holser has quit IRC16:31
*** onmywaytosunset has left #tripleo16:32
*** cylopez has joined #tripleo16:32
*** karthiks has quit IRC16:35
openstackgerritwes hayutin proposed openstack/tripleo-quickstart master: do not log skipped tasks  https://review.opendev.org/68353016:35
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Convert subprocess to a reusable method  https://review.opendev.org/68342916:35
*** bogdando has quit IRC16:36
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move nested bash script into a real file  https://review.opendev.org/68327016:37
*** onmywaytosunset has joined #tripleo16:37
onmywaytosunsetgetting permssion denied /etc/ceilometer/gnocchi-resources inside the container16:38
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move basic global functions into ansible tasks  https://review.opendev.org/68328616:38
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Convert subprocess to a reusable method  https://review.opendev.org/68342916:39
*** marios has quit IRC16:41
*** ykarel|away has quit IRC16:41
*** derekh has quit IRC16:42
*** hberaud is now known as hberaud|gone16:43
*** fultonj has joined #tripleo16:45
*** fultonj has joined #tripleo16:46
openstackgerritwes hayutin proposed openstack/tripleo-ansible master: increase coverage of tripleo-ansible  https://review.opendev.org/68358716:48
weshay|ruckmwhahaha, you cool w/ ^16:48
mwhahahanot really16:48
mwhahahabut whatever16:48
mwhahahaweshay|ruck: why multinode?16:49
mwhahahawe already run scenario00416:49
mwhahahawhat are we missing16:49
weshay|ruckwould help ensure that changes there don't mess w/ perf16:49
weshay|ruckwe're adding a lot of stuff16:49
weshay|ruckfor some amount of time... seems like a good idea to me16:50
mwhahahanot sure if that's really the solution but ok16:50
mwhahahawe can give it a shot16:50
weshay|ruckya.. this is not a have to16:50
weshay|ruckbut I think it makes sense for now16:50
weshay|ruckthe coverage is light16:50
weshay|ruckimho16:50
weshay|ruckcloudnull, thinking it would be a good idea to turn on container-multinode on all tripleo-ansible roles while were in heavy devel16:52
weshay|ruckpita I know.. sorry :)16:53
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move globals into main level 0 case  https://review.opendev.org/68359016:54
*** cylopez has quit IRC16:55
*** ramishra has quit IRC16:55
openstackgerritwes hayutin proposed openstack/tripleo-heat-templates master: Run Manila in scenario004 without pacemaker  https://review.opendev.org/68285316:55
*** udesale has quit IRC16:57
*** gbarros has joined #tripleo16:58
*** poopcat has quit IRC16:58
*** holser__ has quit IRC16:58
*** ykarel|away has joined #tripleo16:58
cloudnullweshay|ruck ++16:58
cloudnullone comment but otherwise lgtm16:59
*** florianf has quit IRC17:00
*** holser has joined #tripleo17:01
*** mcornea has quit IRC17:02
*** mcornea has joined #tripleo17:03
openstackgerritwes hayutin proposed openstack/tripleo-ansible master: increase coverage of tripleo-ansible  https://review.opendev.org/68358717:03
weshay|ruckcloudnull, good point.. updated17:03
*** onmywaytosunset has quit IRC17:05
*** holser has quit IRC17:05
mwhahahacloudnull: oh so lookin in the image uploader code, i think we have some excessive session creation17:09
mwhahahacloudnull: we have a bunch of authenticate calls that don't have session passed with them17:09
cloudnullthat does sound about right :)17:10
mwhahahacloudnull: things like discover_image_tag, filter_images_with_labels, discover_tag_from_inspect, tags_for_image17:10
mwhahahathough it's likely because of the stupid scope stuff17:11
cloudnull++17:11
mwhahahabut i don't think we should be creating a sesssion for each one of these calls17:11
mwhahahaseems like we need to seperate the pooling aspect from the token bits17:12
tbarronweshay|ruck: i'm being dense about what CI we are waiting for on 682853, it keeps passing scenario004 and gets +1 vote from zuul17:12
cloudnullthe way a lot of that code works is a little odd. its authenticating on the scoped image, which we should be able to track with the sha1 ids, but its a bit all over the place17:12
weshay|rucktbarron, we can't do a depends on from upstream to third party.. has to work the other way.. so17:13
cloudnullits all using the same auth regardless of the image, so it makes a lot of sense to try and separate those functions, if we can17:14
weshay|rucktbarron, to see the results upstream we had to get that zuul change merged17:14
weshay|ruckwhich it now is17:14
tbarronweshay|ruck: ah, that's why the depends-on seemed reversed to me.  So where do I look to see the results that we're waiting on?17:15
weshay|rucktbarron, I rebased the change.. the job is running, you'll see results in https://review.opendev.org/#/c/682853/17:15
weshay|rucktbarron, job is running here https://review.rdoproject.org/zuul/status17:16
weshay|ruckfilter on 68285317:16
weshay|ruckhttps://review.rdoproject.org/zuul/stream/986c0953040b4601ab20b6a92b440e52?logfile=console.log17:16
weshay|ruckhopefully the right container_cli command will be passed17:16
tbarronweshay|ruck: ok, cool, and thanks.  I can do the zuul ongoing change thing, just wasn't getting that we're looking at this change set17:16
weshay|ruckshould be17:16
tbarronmakes sense since you rebased though17:17
*** suuuper has joined #tripleo17:23
ykarel|awaymmm Delete param taking 25-40 minutes, really?17:25
ykarel|awaycan't digest17:25
ykarel|awayme looking at https://review.opendev.org/#/c/683402/517:25
ykarel|awaymwhahaha, arxcruz|rover any link for this ^^17:26
ykarel|awaylog link17:26
arxcruz|roverykarel|away: https://bugs.launchpad.net/tripleo/+bug/184477817:27
openstackLaunchpad bug 1844778 in tripleo "Overcloud deployment is timeing out on scenario001 ovb" [Critical,Triaged] - Assigned to Arx Cruz (arxcruz)17:27
* ykarel|away looks17:27
ykarel|awayarxcruz|rover, you mean 36:15.383 taken by for that delete task?17:28
ykarel|awaythat time is for tripleo-container-image prepare task i think17:28
arxcruz|roverykarel|away: don't think so...17:29
arxcruz|roverykarel|away: did you check the log ?17:29
ykarel|awayarxcruz|rover, yes17:29
ykarel|awayhttps://logs.rdoproject.org/openstack-periodic-master/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-7-ovb-1ctlr_2comp-featureset020-master/8e0ca89/logs/undercloud/home/zuul/overcloud_deploy.log.txt.gz#_2019-09-19_22_11_1317:29
arxcruz|roverok, first of all, how can you get the link for the time?17:30
ykarel|awayso in ^^, Thursday 19 September 2019  22:10:28 +0000 (0:36:15.383)       0:43:02.648 ****17:31
ykarel|awayi say, 36:15 is for tripleo-container-image prepare task17:31
ykarel|awaynot handler17:31
ykarel|awayarxcruz|rover, see https://logs.rdoproject.org/openstack-periodic-master/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-7-ovb-1ctlr_2comp-featureset020-master/8e0ca89/logs/ara_oooq_overcloud/ara-report/ for more clearity17:32
ykarel|awaysort it based on Duration17:32
ykarel|awayand u will see 36:15 is for tripleo-container-image preapare command17:32
arxcruz|roverhmmmm17:35
arxcruz|roverweshay|ruck: ^17:35
weshay|ruckya.. just amazing how awful ansible logging is at times17:35
*** amoralej is now known as amoralej|off17:35
weshay|ruckgood way to sell tower17:36
arxcruz|roverstill, i want to see the logs from the patch17:38
arxcruz|roverhow much time the image prepare was before ?17:39
*** tosky_ has joined #tripleo17:48
*** holser has joined #tripleo17:49
ykarel|awayarxcruz|rover, befor which patch17:51
*** tosky is now known as Guest8725917:51
*** tosky_ is now known as tosky17:51
*** Guest87259 has quit IRC17:52
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move config processing into a single method  https://review.opendev.org/68367017:52
*** holser has quit IRC17:54
*** Tengu has quit IRC18:07
*** Tengu has joined #tripleo18:07
*** gbarros has quit IRC18:09
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Consolodate mp_puppet_config  https://review.opendev.org/68368318:17
*** paramite|clone has joined #tripleo18:17
*** dsneddon has joined #tripleo18:22
*** morazi has quit IRC18:26
*** ricolin has quit IRC18:27
*** ricolin has joined #tripleo18:27
*** dsneddon has quit IRC18:27
*** jistr has quit IRC18:28
*** redrobot has quit IRC18:28
*** elGamal has quit IRC18:28
*** jistr has joined #tripleo18:28
*** elGamal has joined #tripleo18:29
*** dansmith has quit IRC18:29
*** dansmith has joined #tripleo18:29
*** Vorrtex has quit IRC18:34
openstackgerritwes hayutin proposed openstack/tripleo-heat-templates master: Run Manila in scenario004 without pacemaker  https://review.opendev.org/68285318:36
*** holser has joined #tripleo18:37
weshay|ruckcloudnull, didn't EmilienM say he did NOT see resets using quay Resetting dropped connection: quay.io18:40
weshay|ruckhttps://88c78217ea88b68a79a3-c409a01604f6f3a7ce618899a7a4cd0d.ssl.cf1.rackcdn.com/683212/2/check/tripleo-ci-centos-7-containers-multinode/2f599e9/logs/undercloud/var/log/tripleo-container-image-prepare.log.txt.gz18:40
cloudnullcorrect.18:40
cloudnullno resets no retries18:40
cloudnullerr , maybe it was no retries, but resets.18:40
weshay|ruckhrm.. there are a bunch in that log18:41
* cloudnull doesn't remember. 18:41
*** holser has quit IRC18:41
*** poopcat has joined #tripleo18:41
mwhahahait was retries18:41
mwhahaharesets happen18:41
mwhahahanot sure why18:41
mwhahahawas wondering if they have connection-close or something18:41
cloudnulllooking at that log I dont see any "re-auth"18:41
weshay|ruckok.. so still worth trying?18:42
mwhahahai'm looking at the session code and i'm not sure we can move it because auth'd sessions are scoped to a specific image18:42
mwhahahayes it is18:42
mwhahahait's way less calls18:42
mwhahahathe problem is that docker.io auth calls are expensive18:42
mwhahaha2x18:42
weshay|ruck24min task18:42
mwhahaha24 no mirroring right?18:42
weshay|rucknot really improving time, but we can still move18:42
weshay|ruckya.. I don't think it was mirrored yet18:43
* weshay|ruck checks cloudnull's patch18:43
weshay|ruckhttps://review.opendev.org/#/c/683240/18:43
cloudnullwe need our infra core :)18:43
* weshay|ruck just asked in channel18:45
weshay|ruckcloudnull, saw something on reddit.. re: everyone gets to merge what ever they want for 12 hours once a year18:45
weshay|ruck12 hour window18:45
mwhahahasounds like a terrible idea18:46
cloudnullthe purge18:46
weshay|ruckthe name was funny18:46
cloudnull's/purge/merge/'18:46
weshay|ruckhttps://www.reddit.com/r/ProgrammerHumor/comments/d6d947/may_the_merge_begin/18:46
weshay|ruckha ha..18:46
weshay|ruckthe sequal... "the rebase"18:46
mwhahahahttps://www.youtube.com/watch?v=VU0GYSA1POs18:47
weshay|rucktru18:47
weshay|ruckthat is how you get ants18:47
weshay|ruckthe ovb job was quick18:49
weshay|ruckhttp://logs.rdoproject.org/12/683212/2/openstack-check/tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001/1725a7a/logs/undercloud/home/zuul/install-undercloud.log.txt.gz18:49
weshay|ruck14min18:49
weshay|rucktripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001SUCCESS in 2h 59m 10s18:49
weshay|ruckkind of points back to the infra18:49
weshay|ruckwhen ovb is 41 min faster than container-multinode18:50
*** Vorrtex has joined #tripleo18:50
weshay|rucktotally need a an infra core18:50
mwhahahathe wide range of times totally points to environmental things18:51
weshay|ruckya18:51
weshay|ruck\0/18:51
weshay|rucktoo much C02 in the atmosphere for containers to work18:52
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move config processing into consolodated methods  https://review.opendev.org/68367019:06
*** morazi has joined #tripleo19:07
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Move config processing into consolodated methods  https://review.opendev.org/68367019:10
*** alexmcleod has quit IRC19:12
*** ricolin has quit IRC19:16
*** ykarel|away has quit IRC19:17
openstackgerritMerged openstack/tripleo-docs master: Fix container configuration file path  https://review.opendev.org/68242419:18
*** lucasagomes has quit IRC19:25
*** dsneddon has joined #tripleo19:31
*** slaweq has quit IRC19:32
weshay|rucktbarron, fyi.. got results back19:32
tbarronweshay|ruck: you sound like an oncologist19:33
tbarronweshay|ruck: i'll look and see what's happening with ceph now that wasn't happening before you  rebased19:34
tbarronweshay|ruck: i bet running *with* pacemaker would fail with your rebase too :)19:34
weshay|rucktbarron, please call the hotline at the ci center for confidential results19:34
weshay|ruckthere will be  survey at the end of the call for quality assurance19:34
tbarronweshay|ruck: that's why i'm confused about why we hold up the normal tripleo fix for dependencies elsewhere even though I'm happy to work on thodse too19:35
tbarronthose19:35
tbarronbut I'm always confused so don't worry19:35
weshay|rucktbarron, this one is confusing.. because of pacemaker19:35
tbarroni'm gonna want to backport this fix to stable though, maybe as is?  it worked just fine ...19:36
weshay|ruckI think scen01/02 are using podman on both centos and rhel19:36
weshay|rucktbarron, backport to what exactly?19:36
tbarronall the way to stable/queens, the non pcs path should work even if we didn't support it downstream19:37
tbarronand it never did19:37
weshay|ruckah19:37
tbarronso that's what i was fixing, that's what the bug and the @#$#@$ nag on irc and the trello card say19:38
*** morazi has quit IRC19:38
tbarronnow I understand it's for a higher cause :)19:38
weshay|ruckwe are saving lives here19:38
tbarronand am perfectly willing to help get done what needs to be done, but i'm foggy on our goal19:38
tbarronbut that doesn't matter, i'm just a soldier19:38
tbarronanyways, i'll look at the new logs, something went wrong deploying ceph cause those pools didn't get created19:39
*** suuuper has quit IRC19:45
*** morazi has joined #tripleo19:52
*** morazi has quit IRC20:01
*** morazi has joined #tripleo20:02
*** pbandark has quit IRC20:05
*** pbandark has joined #tripleo20:06
*** artom has quit IRC20:10
cloudnullmwhahaha any chance you could give https://review.opendev.org/#/c/683277/ a nudge20:15
mwhahahak20:16
openstackgerritMartin Mágr proposed openstack/puppet-tripleo master: Add collectd-sensubility configuration  https://review.opendev.org/67886320:18
*** two_oes has quit IRC20:19
*** two_oes has joined #tripleo20:19
cloudnullty20:20
*** Vorrtex has quit IRC20:29
*** jtomasek has quit IRC20:32
*** redrobot has joined #tripleo20:34
openstackgerritDamien Ciabrini proposed openstack/tripleo-heat-templates master: WIP HA: reorder init_bundle and restart_bundle for improved updates  https://review.opendev.org/67599320:36
openstackgerritMichael Johnson proposed openstack/tripleo-heat-templates master: Added the connection_logging parameter for Octavia  https://review.opendev.org/68372820:44
openstackgerritAlex Schultz proposed openstack/tripleo-common master: Cleanup override file  https://review.opendev.org/68245521:02
openstackgerritAlex Schultz proposed openstack/tripleo-common master: Remove tacker container code  https://review.opendev.org/68245721:03
openstackgerritAlex Schultz proposed openstack/tripleo-common master: Remove Trove password constant  https://review.opendev.org/68247921:03
*** _mmethot_ has quit IRC21:05
*** mmethot_ has joined #tripleo21:05
*** rfolco has quit IRC21:08
openstackgerritMichael Johnson proposed openstack/tripleo-heat-templates master: Add the Octavia log offload parameters  https://review.opendev.org/68373121:12
*** gfidente has quit IRC21:16
*** paramite|clone has quit IRC21:17
*** apetrich has quit IRC21:19
*** aakarsh has quit IRC21:20
*** rlandy has quit IRC21:25
*** pierrepr1netti is now known as pierreprinetti21:26
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-heat-templates master: Consolodate file operations  https://review.opendev.org/68373421:35
openstackgerritAlex Schultz proposed openstack/tripleo-heat-templates master: Remove python executable print  https://review.opendev.org/68373521:35
mwhahahacloudnull: for blocks like https://opendev.org/openstack/tripleo-ansible/src/branch/master/tripleo_ansible/roles/tripleo-ceph-common/tasks/main.yml#L21-L33 should we no_log it like we did for https://review.opendev.org/#/c/682988/21:41
mwhahahai'm looking at the deploy logs and we have a bunch of extra logs/fake fatals because of those env var lookup things21:42
mwhahahahttps://logs.rdoproject.org/25/683425/1/openstack-check/tripleo-ci-rhel-8-scenario001-standalone-rdo/0015380/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz#_2019-09-20_16_16_5021:42
*** rh-jelabarre has quit IRC21:43
cloudnullI think some of those load blocks can just be removed.21:43
mwhahahawe call these tripleo-ceph things a lot21:43
cloudnullthe role template adds that block by default but I suspect some of our roles done have any vars files so there no reason to have it21:44
mwhahahawith https://review.opendev.org/#/c/683428/ they'll stop showing up (at least the skipped things) though it seems excessive21:44
*** nkinder has quit IRC21:44
*** raildo has quit IRC21:44
*** rh-jelabarre has joined #tripleo21:44
mwhahahahttps://logs.rdoproject.org/28/683428/2/openstack-check/tripleo-ci-rhel-8-scenario001-standalone-rdo/cbe88d0/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz#_2019-09-20_16_32_0121:45
mwhahahastill a bit chatty21:45
openstackgerritMerged openstack/tripleo-ansible master: Switching the delete param handler to use rm  https://review.opendev.org/68340221:47
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-ansible master: remove unessisary blocks  https://review.opendev.org/68373621:48
cloudnull^ mwhahaha quick scan of the roles, those were the two we had that were using that block without any vars , so i remvoed it21:49
mwhahahai mean we can probably leave it if we ever want them, but include_vars dumps the contents to the script when we run with -vv21:49
mwhahahawould be nice if this was just inherient to ansible21:50
mwhahahaperhaps a module?21:50
cloudnullwe might be able to write a vars module21:50
cloudnullbut no, ansible doesn't have a built in var file loader, other than for the file `vars/main.yml`21:51
mwhahahathe logic we have for the files seems backwards from other examples i've seen anyway21:51
mwhahahaso maybe just a tripleo_load_os_vars21:51
cloudnullmost folks include those types of vars in the playbook, sadly we cant run that at the playbook level, so we're using that task to scan for and load OS specific var files when they're encountered.21:54
mwhahahayea that's why i was wondering if a module could work21:54
* mwhahaha shrugs21:54
cloudnullnot every role in tripleo-ansible has that block , so i think it makes sense to remove it where its not needed.21:54
cloudnullhowevere, I can look into making a module .21:54
cloudnullthat's something we could probably do?21:55
* cloudnull reading https://docs.ansible.com/ansible/latest/plugins/vars.html21:55
mwhahahaseems like for repeated tasks it would be beneficial21:55
cloudnull++21:55
*** mcornea has quit IRC21:59
mwhahahawe seem to be including tripleo-ceph stuff multiple times int eh same step22:01
mwhahahaweird22:01
mwhahahacloudnull: does the meta/main.yaml do anythhing at run time or is it just a galaxy thing22:02
cloudnullit load dependencies if there are any.22:03
cloudnullin most cases its just a galaxy thing22:03
mwhahahaso we run tripleo-ceph-common repeatedly then22:03
mwhahahabecause of teh dependencies22:03
cloudnullprobably22:03
mwhahahai wonder if we need to drop those22:03
mwhahahahttps://logs.rdoproject.org/28/683428/2/openstack-check/tripleo-ci-rhel-8-scenario001-standalone-rdo/cbe88d0/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz#_2019-09-20_16_32_0122:04
mwhahahacheck all the repeats of tripleo-ceph-common22:04
mwhahahaactually https://logs.rdoproject.org/25/683425/1/openstack-check/tripleo-ci-rhel-8-scenario001-standalone-rdo/0015380/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz#_2019-09-20_16_16_51 is a better example22:04
cloudnullgood question for fultonj - maybe we can figure out a way to run once ?22:04
mwhahahawe're including it in our playbook itself22:04
mwhahahaso can we just not have that?22:04
* mwhahaha doesn't like dependencies in ansible because of this22:05
cloudnullif we can control the playbook, we can control the load ordering22:05
mwhahahawe do22:06
mwhahahacause these are controlled via deploy tasks22:06
cloudnullwe can also set run_once on  https://github.com/openstack/tripleo-ansible/blob/master/tripleo_ansible/roles/tripleo-ceph-common/tasks/main.yml#L35-L4922:07
cloudnullwhich will make it a little more quite22:07
mwhahahathat works too22:07
mwhahahaceph-{work-dir,fetch-dir,common} run repeatedly22:08
*** xek has quit IRC22:12
mwhahahaactually i don't think we want run_once22:15
mwhahahabecause that shares it across the deploy22:15
mwhahahaif i'm reading https://docs.ansible.com/ansible/latest/user_guide/playbooks_delegation.html#run-once correctly22:15
mwhahaharun_once is one host, not one instance per host22:16
*** stevebaker has joined #tripleo22:20
*** rh-jelabarre has quit IRC22:22
mwhahahafultonj: https://bugs.launchpad.net/tripleo/+bug/184483722:25
openstackLaunchpad bug 1844837 in tripleo "ceph runs the tripleo-ceph-{work-dir,fetch-dir,common} roles more than once in the deployment steps" [High,Triaged]22:25
mwhahahacloudnull: -^22:25
*** gbarros has joined #tripleo22:29
* mwhahaha gives up for the week22:29
openstackgerritKevin Carter (cloudnull) proposed openstack/tripleo-ansible master: Add conditions to ceph roles for dependencies  https://review.opendev.org/68373922:30
cloudnullmwhahaha ^22:31
cloudnullI think i worked out whats needed, but we'll see in the gate22:31
mwhahahak22:31
mwhahahawill check on monday when the results might be ready :D22:31
cloudnull++22:31
* cloudnull was going to break away myself 22:31
cloudnulltalk to you monday22:32
cloudnullhave a good weekend22:32
*** gbarros has quit IRC22:40
*** nkinder has joined #tripleo23:01
*** gbarros has joined #tripleo23:08
*** nkinder has quit IRC23:26
openstackgerritMerged openstack/tripleo-ansible master: Use shell to ensure faster task execution  https://review.opendev.org/68327723:27
*** tosky has quit IRC23:28
*** dsneddon has quit IRC23:29
*** munimeha1 has quit IRC23:35
openstackgerritMerged openstack/puppet-tripleo master: Disable keystone token_flush by default  https://review.opendev.org/68251223:56
openstackgerritMerged openstack/tripleo-heat-templates master: Don't run keystone_cron container if fernet token is used  https://review.opendev.org/68226523:56
*** rfolco has joined #tripleo23:57
*** gbarros has quit IRC23:58

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