Monday, 2017-02-27

« Sunday, 2017-02-26 Index Tuesday, 2017-02-28 »
*** sdake has joined #tripleo00:00
*** sdake_ has quit IRC00:00
*** ansmith_ has joined #tripleo00:10
*** sdake has quit IRC00:13
*** flepied has quit IRC00:18
*** oanson has quit IRC00:21
*** lihi has quit IRC00:22
*** oanson has joined #tripleo00:30
*** lihi has joined #tripleo00:31
*** ansmith_ has quit IRC00:32
*** ansmith has quit IRC00:33
*** flepied has joined #tripleo00:36
*** ansmith_ has joined #tripleo00:46
openstackgerritMerged openstack/tripleo-quickstart-extras master: Remove hardcoded instances of `stack` from roles  https://review.openstack.org/42146300:46
*** ansmith has joined #tripleo00:46
*** flepied has quit IRC01:01
*** egafford has joined #tripleo01:05
*** dmacpher has joined #tripleo01:06
*** dmacpher_ has joined #tripleo01:06
*** lblanchard has joined #tripleo01:11
openstackgerritMerged openstack/puppet-tripleo master: Add httpchk for http services  https://review.openstack.org/39468601:13
*** dmacpher has quit IRC01:14
*** dmacpher_ has quit IRC01:14
*** dmacpher has joined #tripleo01:15
*** rbrady has joined #tripleo01:28
*** ansmith_ has quit IRC01:34
*** ansmith has quit IRC01:34
*** rbrady has quit IRC02:06
*** fzdarsky__ has joined #tripleo02:32
*** fzdarsky_ has quit IRC02:36
*** sdake has joined #tripleo03:01
*** achadha has joined #tripleo03:03
*** sudipto has joined #tripleo03:05
*** sudipto_ has joined #tripleo03:05
*** achadha has quit IRC03:10
*** achadha has joined #tripleo03:11
*** sudipto has quit IRC03:12
*** sudipto_ has quit IRC03:12
*** achadha_ has joined #tripleo03:14
*** achadha has quit IRC03:18
*** bvandenh has quit IRC03:24
*** sdake_ has joined #tripleo03:30
*** achadha_ has quit IRC03:30
*** achadha has joined #tripleo03:31
*** sdake has quit IRC03:34
*** pgadiya has joined #tripleo03:36
*** psahoo has joined #tripleo04:04
*** udesale has joined #tripleo04:14
*** xnodeza has joined #tripleo04:16
*** gkadam has joined #tripleo04:22
*** saneax-_-|AFK is now known as saneax04:22
*** pmannidi has quit IRC04:35
openstackgerritMerged openstack/tripleo-heat-templates stable/ocata: Deploy CI scenarios with Pacemaker  https://review.openstack.org/43829404:40
*** sdake_ has quit IRC04:42
*** sudipto_ has joined #tripleo04:57
*** sudipto has joined #tripleo04:57
*** cmyster has quit IRC05:03
*** prateek has joined #tripleo05:14
*** diana_clarke has quit IRC05:16
*** limao has joined #tripleo05:24
*** diana_clarke has joined #tripleo05:29
*** masco has joined #tripleo05:31
*** fragatina has joined #tripleo05:39
openstackgerritJeffrey Zhang proposed openstack/diskimage-builder master: Add qemu-guest-agent element  https://review.openstack.org/43833405:40
*** fragatina has quit IRC05:43
*** fragatina has joined #tripleo05:44
*** pmannidi has joined #tripleo05:45
*** jaganathan has joined #tripleo05:46
*** skramaja has joined #tripleo05:53
*** nyechiel has joined #tripleo06:12
*** florianf has joined #tripleo06:14
*** dparkes has quit IRC06:19
*** nmathew has joined #tripleo06:21
*** twi99y has joined #tripleo06:22
openstackgerritChandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow  https://review.openstack.org/43191606:30
*** jtomasek_ has joined #tripleo06:37
*** fragatina has quit IRC06:39
openstackgerritItzik Brown proposed openstack/tripleo-heat-templates stable/newton: Adding the ip_conntrack_proto_sctp kernel module  https://review.openstack.org/43834306:41
*** iranzo has joined #tripleo06:51
*** iranzo has joined #tripleo06:51
*** hjensas has quit IRC06:52
*** leanderthal|afk is now known as leanderthal06:58
*** flepied has joined #tripleo06:59
*** fragatina has joined #tripleo07:01
*** fragatina has quit IRC07:06
*** ramishra has joined #tripleo07:07
*** cylopez has joined #tripleo07:09
*** dmacpher has quit IRC07:11
bandinimorning07:14
*** tobias_fiberdata has joined #tripleo07:17
*** pcaruana has joined #tripleo07:20
*** rcernin has joined #tripleo07:21
*** rcernin has quit IRC07:23
*** rcernin has joined #tripleo07:23
*** vwker has joined #tripleo07:24
*** abehl has joined #tripleo07:25
*** aufi has joined #tripleo07:26
vwkerhi, i am following the [doc](https://docs.openstack.org/developer/tripleo-docs/environments/virtual.html), but when i execute `instack-virt-setup` it end up with put 'libvirtError: invalid argument: could not find capabilities for arch=x86_64 domaintype=kvm', can anyone help?07:34
*** jtomasek_ has quit IRC07:38
*** sdake has joined #tripleo07:38
*** jtomasek has joined #tripleo07:38
*** sdake_ has joined #tripleo07:40
*** hjensas has joined #tripleo07:40
*** hjensas has quit IRC07:40
*** hjensas has joined #tripleo07:40
*** achadha_ has joined #tripleo07:43
*** sdake has quit IRC07:44
*** saibarspeis has joined #tripleo07:45
*** achadha has quit IRC07:47
*** pmannidi has quit IRC07:56
*** dparkes has joined #tripleo07:56
*** jprovazn has joined #tripleo07:58
openstackgerritAttila Darazs proposed openstack/tripleo-quickstart-extras master: GATE CHECK  https://review.openstack.org/43027708:00
*** brault has joined #tripleo08:02
*** xnodeza_ has joined #tripleo08:04
*** xnodeza_ has quit IRC08:04
*** xnodeza has quit IRC08:06
*** yamahata has joined #tripleo08:11
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci master: Support deploy args in OVB jobs  https://review.openstack.org/43753108:13
*** mcornea has joined #tripleo08:14
openstackgerritSteven Hardy proposed openstack/tripleo-heat-templates master: Add "UPGRADE" to StackAction  https://review.openstack.org/43804408:14
*** Guest68960 has quit IRC08:14
*** iranzo has quit IRC08:16
*** shardy has joined #tripleo08:17
*** ccamacho has joined #tripleo08:18
*** iranzo has joined #tripleo08:18
*** mgagne has joined #tripleo08:20
*** mgagne is now known as Guest1144308:20
*** yprokule has joined #tripleo08:23
*** sdake has joined #tripleo08:24
*** sdake_ has quit IRC08:27
*** flepied has quit IRC08:31
*** achadha_ has quit IRC08:31
openstackgerritThomas Herve proposed openstack/instack-undercloud master: Use Swift as a Zaqar backend.  https://review.openstack.org/43472708:31
*** fzdarsky__ has quit IRC08:32
*** matbu|pto is now known as matbu08:33
*** fzdarsky has joined #tripleo08:33
*** suuuper has joined #tripleo08:36
*** salmankhan has joined #tripleo08:40
*** vwker has quit IRC08:43
*** salmankhan has quit IRC08:44
openstackgerritSagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation  https://review.openstack.org/43743708:44
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci master: Support deploy args in OVB jobs  https://review.openstack.org/43753108:46
*** jaosorior has joined #tripleo08:47
*** karimb has joined #tripleo08:47
*** flepied has joined #tripleo08:48
*** dbecker has joined #tripleo08:49
*** amoralej|off is now known as amoralej08:49
*** gfidente has joined #tripleo08:51
*** agurenko has joined #tripleo08:51
*** shardy has quit IRC08:53
*** shardy has joined #tripleo08:53
*** jpich has joined #tripleo08:56
*** sdake has quit IRC08:59
*** pgadiya has quit IRC08:59
*** zzzeek has quit IRC09:00
*** zzzeek has joined #tripleo09:00
yolandahi, question... how could we enable networking-bgpvpn on tripleo ci? could that be done on some scenario?09:04
openstackgerritJuan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job with RabbitMQ  https://review.openstack.org/42630709:06
shardyyolanda: Hi!  I'm not familiar with bgpvpn or it's requirements, but if it has a driver we can use it CI it could potentially be enabled via a scenario09:07
*** jpena|away is now known as jpena09:07
* shardy looks for supported drivers09:08
*** tosky has joined #tripleo09:08
shardyhttps://docs.openstack.org/developer/networking-bgpvpn/overview.html#drivers09:09
shardyLooks like most of the drivers require a third-party SDN backend?09:09
shardyHmm, perhaps the BaGPipe driver would work09:10
jaosoriorbandini: hey dude, I need a bit of help with the mysqlclient configuration file09:10
*** edim has joined #tripleo09:12
edimHi, I am using RHOSP 10 and adding custom networks for network isolation09:13
*** bogdando has joined #tripleo09:14
shardyedim: Hi, TripleO doesn't yet support custom networks, but we've discussed adding support for that during Pike09:14
*** b00tcat has quit IRC09:14
shardyhttps://etherpad.openstack.org/p/tripleo-ptg-pike-composable-networking has some notes09:15
shardyedim: ^^09:15
*** pgadiya has joined #tripleo09:15
*** edim_ has joined #tripleo09:16
openstackgerritChandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow  https://review.openstack.org/43191609:16
*** sshnaidm has quit IRC09:16
edim_@shardy, i am creating new network in the same manner as existing networks09:17
*** Vijayendra has joined #tripleo09:17
edim_modifying the role.role.j2.yaml and overcloud.j2.yaml file with new parameters09:17
shardyedim: Ok, so you're running modified templates09:18
shardyedim: the discussion above is about avoiding that via some additional j2 templating09:18
shardyin a future release09:18
*** zoli is now known as zoliXXL09:18
bandinijaosorior: fire away (am totally jet-lagged, so I give no guarantees about sensible responses ;)09:18
*** edim has quit IRC09:18
openstackgerritMerged openstack/tripleo-heat-templates master: Don't recalculate EndpointMap to get outputs  https://review.openstack.org/42783609:19
jaosoriorbandini: lol, no biggie09:19
*** yprokule_ has joined #tripleo09:20
*** yprokule has quit IRC09:20
*** yprokule_ is now known as yprokule09:20
jaosoriorbandini: well, I tried doing the SSL bits there, but it keeps failing, as if it wasn't trying to establish the client connection via SSL :/09:21
jaosoriorbandini: here's the log http://logs.openstack.org/95/425295/6/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/12b26f9/09:21
* bandini looks09:22
bandinijaosorior: which logs are you looking at specifically?09:23
bandinijust so we're looking at the same things09:23
jaosoriorbandini: so, the postci logs show the error: http://logs.openstack.org/95/425295/6/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/12b26f9/logs/postci.txt.gz#_2017-02-24_11_42_32_00009:25
jaosoriorbandini: and if you get the logs from the node http://logs.openstack.org/95/425295/6/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/12b26f9/logs/overcloud-controller-0.tar.xz you can see that the clients should be trying to use SSL (at least according to the tripleo.cnf file)09:25
*** pgadiya has quit IRC09:26
bandiniI see. let me check09:26
*** owalsh has joined #tripleo09:27
*** hewbrocca_afk is now known as hewbrocca09:29
*** lucas-afk is now known as lucasagomes09:30
*** gkadam_ has joined #tripleo09:35
bandinijaosorior: do we know if the issue is a) clients not even trying to do ssl or b) something failing in the ssl connection (be it haproxy or galera)09:35
bandinithe db string seems correct09:35
*** athomas has joined #tripleo09:36
hjensasedim_: shardy: It can be done, with custom templates. I have been experimenting with the same to deploy on routed networks using composable roles + modified templates. patch example here: https://paste.fedoraproject.org/paste/xmm5SI0klbXXh-qydcyqdV5M1UNdIGYhyRLivL9gydE=09:36
*** gkadam has quit IRC09:36
*** psahoo has quit IRC09:36
jaosoriorbandini: so, at least I know that the galera server is serving SSL; I connect manually09:37
bandinijaosorior: ack. how about haproxy, you reckon we're good there as well?09:37
*** ansiwen has quit IRC09:37
shardyhjensas: Hi!  Yeah I'm aware it can be done, but it's not something we directly support yet09:37
shardyhjensas: at the PTG last week we agreed to make progress on auto-generating more things via jinja2 to hopefully make it much easier in future09:38
bandiniguess so, can't think of a reason why haproxy could get in the way09:38
*** pgadiya has joined #tripleo09:38
*** mrunge has quit IRC09:38
shardyhjensas: dsneddon is also working on full support for routed networks https://etherpad.openstack.org/p/tripleo-routed-networks09:39
hjensasedim_: shardy: I think however for the routed-networks it is overkill to create additional networks. We should have subnets only...09:39
shardywhich is related to the work to enable custom networks09:39
*** salmankhan has joined #tripleo09:39
shardyhjensas: it would be very helpful if you could review the specs linked from that etherpad09:39
bandinijaosorior: so in mysqld.log I see bunch of "2017-02-24 11:19:15 139741327099648 [Warning] 'user' entry 'aodh@overcloud.internalapi.tripleodomain' ignored in --skip-name-resolve mode."09:39
shardyin particular https://review.openstack.org/#/c/421009/09:39
*** ansiwen has joined #tripleo09:40
shardyhjensas: so we can compare your experiences with the approach proposed by dsneddon09:40
bandinijaosorior: I wonder if that is our culprit09:40
bandinidciabrin: ever seen the messages above ^ ? (this is on an ssl-enabled galera)09:40
shardyI think it aligns with your comments, but it'd be good to get your feedback directly on the specs09:40
hjensasshardy: yes, I did. It includes use of the new network segments in neutron that was just completed. A feuture I am currently reading up on to understand fully. (So far I feel it is a bit overkill for the tripleo use case, but I need to study it more to fully understand it.)09:41
jaosorioruhm... those are funky09:41
dciabrinbandini, jaosorior not sure but I would have said it's just a logging concerns, i.e. whether to log full names or ip09:41
openstackgerritMerged openstack/tripleo-ui stable/ocata: Imported Translations from Zanata  https://review.openstack.org/43815409:42
*** achadha has joined #tripleo09:42
openstackgerritMerged openstack/tripleo-ui master: Imported Translations from Zanata  https://review.openstack.org/43782509:42
*** achadha has quit IRC09:42
*** achadha has joined #tripleo09:42
shardyhjensas: ack, thanks09:43
shardyhjensas: I think in https://review.openstack.org/#/c/421009/9/specs/pike/tripleo-routed-networks-deployment.rst09:43
shardyhjensas: that the "multiple subnets per neutron network" may describe your preferred approach?09:43
jaosoriorbandini: uhm... it might be that there are issues with the mysql providers in puppet, that are not actually using SSL. then it wouldn't be able to create the users databases, and thus, the db-manages would fail, since the credentials wouldn't be valid09:43
hjensasshardy: yes, same idea.09:44
bandinijaosorior: that could be indeed. should be simple to test on an ssl-enabled setup?09:45
jaosoriorbandini: deploying one right now09:46
*** ckyriakidou has joined #tripleo09:46
bandinijaosorior: ack, let us know how it goes09:46
hjensasshardy: I started some work here: https://review.openstack.org/#/c/437544/ - focus on inspector, and instack-undercloud bits first. dansneddon is aware, and providing feedback.09:46
shardyhjensas: ah, excellent, thanks!  Will check out the review09:47
*** yprokule_ has joined #tripleo09:47
*** yprokule has quit IRC09:48
*** yprokule_ is now known as yprokule09:48
hjensasshardy: on the topic, but further down. If we use multiple subnets, how will that affect upgrades? If we in templates separate the networks from the subnets, we will get resources trying to create a subnet already created by previous version templas. How does Heat deal with that? Conflict, and error?09:49
hjensass/separate subnets from networks/09:49
shardyhjensas: that's a good question - right now we override the default heat update behavior for a few resources, including subnets09:50
shardyhttps://github.com/openstack/tripleo-common/blob/master/undercloud_heat_plugins/immutable_resources.py#L4809:51
*** bogdando has quit IRC09:51
shardyI think that means any changes will be ignored on update, but we probably would delete the subnet if it moves to a different template during the refactoring09:51
*** zoliXXL is now known as zoli09:52
*** zoli is now known as zoliXXL09:52
*** zoliXXL is now known as zoli09:53
hjensasshardy: hm, if we delete it and create again a resource with the same settings. But actually ignore doing the change with override plugin, that is probably what we want?09:54
*** zoli is now known as zoliXXL09:54
*** florianf has quit IRC09:55
shardyhjensas: possibly - I guess it's something we'll have to test pretty carefully09:59
hjensasshardy: ack.10:00
*** b00tcat has joined #tripleo10:02
jaosoriorbandini: so, the users were created.10:02
jaosoriorbandini: and the databases too10:03
bandinijaosorior: ok so those messages can be ignored I guess?10:04
*** derekh has joined #tripleo10:04
bandinijaosorior: I wonder if this issues has to do with the db sync commands maybe ignoreing the tripleo.cnf file even though it is in the db string?10:04
jaosoriornot sure :/10:05
jaosoriorbandini: http://paste.openstack.org/show/600559/10:06
bandinijaosorior: if you run one of the db sync commands what do we get?10:07
jaosoriortherve: hey dude, by the way, I could abandon my patch, that's no biggie. I never got my stuff to work on ubuntu.10:07
jaosoriorbandini: lets see10:07
thervejaosorior, No your patch is great :)10:08
therveI believe we can skip the feature on10:08
therveI believe we can skip the feature on ubuntu?10:08
therveBut we still have the selinux issue10:08
jaosoriortherve: the puppet folks were very keen on keeping that on ubuntu (even though it seems to me they didn't package the wsgi files)10:08
jaosoriorso I never got that merged10:08
jaosoriortherve: I think mwhahaha knew more about the selinux issue.10:08
therveI mean it's cool that they want ubuntu, but then they fix the package?10:09
jaosoriortherve: I never got anybody to answer that :/10:10
therveThough I may have a workaround if we get to that10:10
*** florianf has joined #tripleo10:10
jaosoriortherve: do you?10:10
thervejaosorior, Yeah I think we can refer to heat python module directly10:11
jaosoriortherve: anyway, if you have time to take that commit. And we can ping mwhahaha to ask him again about the selinux issue, at least he might have a reference to the bz10:12
therveYeah the bz is in the review comments10:12
thervehttps://bugzilla.redhat.com/show_bug.cgi?id=13965510:12
openstackbugzilla.redhat.com bug 139655 in up2date "up2date fails with ImportError: /usr/lib/python2.2/site-packages/libxml2mod.so: undefined symbol: xmlNewTextReaderFilename" [Medium,Closed: notabug] - Assigned to bretm10:12
jaosoriorah, didn't notice10:13
thervehttps://bugzilla.redhat.com/show_bug.cgi?id=139655310:13
openstackbugzilla.redhat.com bug 1396553 in openstack-selinux "apache unable to listen on port 8000 to front heat-api-cfn" [Unspecified,New] - Assigned to lhh10:13
therveThat looks better :)10:13
therveI'm not sure why we can't simply make a semanage call10:14
therveOr why we don't need it for other ports, actually10:14
jaosoriortherve: well, ideally we should get permission from the openstack-selinux policy file10:15
jaosoriorbrb10:15
*** jaosorior is now known as jaosorior_lunch10:15
edim_hjensas: shardy: thank you for the clarification.10:17
edim_Can we have a node deployed with multiple roles?10:17
*** jaosorior_lunch is now known as jaosorior10:17
jaosoriorfalse alarm10:17
*** dtantsur has joined #tripleo10:19
thervejaosorior, Yeah there doesn't seem to be any specific heat stuff in openstack-selinux10:21
therveMaybe there needs to be?10:21
* social annoys with https://review.openstack.org/#/c/436535/10:21
therveThough I don't get why 8003/8004 works, and not 800010:21
shardyedim_: a role is a group of nodes, running some services10:22
shardyedim_: so no, a node can only be defined as exactly one role10:22
openstackgerritKarthik S proposed openstack/os-net-config master: Restart Openvswitch after adding DPDK ports  https://review.openstack.org/43425510:22
shardyedim_: but you can adjust the services on any role, via either the ServicesDefault value in roles_data.yaml, or the $RoleServices parameter in an environment file10:23
shardyedim_: that means that you can, for example, run the compute services normally assigned to the Compute role on some other role10:23
openstackgerritGael Chamoulaud proposed openstack/tripleo-quickstart master: Ensure that polkit packages are installed on virthost  https://review.openstack.org/43841410:24
*** achadha_ has joined #tripleo10:24
hjensasedim_: you can create a custom role, including the "sum of services2 from multiple roles to achive what you want. See: http://hardysteven.blogspot.se/2016/10/tripleo-composablecustom-roles.html and http://hardysteven.blogspot.se/2016/08/tripleo-composable-services-101.html10:24
shardywhich I think may actually be what you want ref "node deployed with multiple roles" ?10:24
openstackgerritMerged openstack/tripleo-heat-templates master: Install openstack-heat-agents on upgrade  https://review.openstack.org/43820410:25
*** limao has quit IRC10:26
*** achadha has quit IRC10:26
openstackgerritFlavio Percoco proposed openstack/tripleo-docs master: Show dev warning for the containers section  https://review.openstack.org/43841810:28
*** sshnaidm has joined #tripleo10:28
*** bogdando has joined #tripleo10:30
openstackgerritKarthik S proposed openstack/os-net-config master: Restart Openvswitch after adding DPDK ports  https://review.openstack.org/43425510:30
*** bogdando has quit IRC10:30
*** mrunge has joined #tripleo10:31
*** gkadam_ has quit IRC10:41
*** yamahata has quit IRC10:44
*** akrivoka has joined #tripleo10:45
*** gkadam_ has joined #tripleo10:53
*** agurenko has quit IRC10:54
jaosoriortherve: I have no idea :/10:57
therveOK, we'll see when Alex is around10:57
*** bvandenh has joined #tripleo10:58
*** udesale has quit IRC10:58
*** akrivoka has quit IRC10:58
*** zoliXXL is now known as zoli|lunch10:58
*** lmasiero has joined #tripleo11:00
edim_hjensas: shardy: thank you for the clarification :)11:00
gfidenteshardy was looking at the post deployment part to plug the workflow11:01
*** achadha_ has quit IRC11:01
gfidenteI remember you mentioned the post steps11:01
gfidentebut I was thinking to trigger it only once, not at each step11:01
lmasierogood morning guys.. i just deployed overcloud and exited with signal 0, says CREATE_COMPLETE but it doesn't generate the overcloudrc file.. any thoughts?11:01
lmasieroi'm using newton upstream release11:01
lmasieroany help would be appreciated11:02
lmasiero:)11:02
*** mhenkel has quit IRC11:03
*** akrivoka has joined #tripleo11:03
shardygfidente: yeah I was thinking the same about the requirements around parameter generation, we need some way to trigger workflow execution after the role resource groups are created11:05
shardynot yet looked at exactly where we might add it11:05
*** mhenkel has joined #tripleo11:06
*** nmathew has quit IRC11:06
openstackgerritMerged openstack/tripleo-common stable/ocata: Updated from global requirements  https://review.openstack.org/43625511:08
openstackgerritMerged openstack/tripleo-common master: Add TripleO template overrides file  https://review.openstack.org/43798711:08
openstackgerritMerged openstack/tripleo-heat-templates master: Install openstack-panko-api on upgrade  https://review.openstack.org/43820111:09
openstackgerritMerged openstack/instack-undercloud stable/ocata: Return 1 when an error occurs  https://review.openstack.org/43818111:09
*** saibarspeis has quit IRC11:09
*** cylopez has quit IRC11:11
*** mrunge has quit IRC11:12
*** mrunge has joined #tripleo11:14
*** yolanda has quit IRC11:17
*** yolanda has joined #tripleo11:19
openstackgerritHonza Pokorny proposed openstack/tripleo-ui master: Add i18n strings for deployment status messages  https://review.openstack.org/43803711:22
*** dciabrin has quit IRC11:27
*** dciabrin has joined #tripleo11:27
*** edim_ has quit IRC11:31
*** cmyster has joined #tripleo11:33
*** cmyster has quit IRC11:33
*** cmyster has joined #tripleo11:33
*** abregman has joined #tripleo11:34
*** lmasiero has quit IRC11:34
*** jkilpatr has quit IRC11:34
*** bvandenh has quit IRC11:36
*** psahoo has joined #tripleo11:36
openstackgerritYair Fried proposed openstack/instack-undercloud master: VIPs should not be part of PXE boot subnet  https://review.openstack.org/43845211:37
openstackgerritChandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow  https://review.openstack.org/43191611:45
openstackgerritGiulio Fidente proposed openstack/tripleo-validations master: Check for Ceph status in post-deployment  https://review.openstack.org/43619411:48
openstackgerritTom Barron proposed openstack/tripleo-heat-templates master: Default Manila CephFS snapshot enablement to false  https://review.openstack.org/43846011:55
*** pgadiya has quit IRC11:57
*** milan has joined #tripleo11:58
*** sudipto_ has quit IRC11:58
*** sudipto has quit IRC11:58
*** yamahata has joined #tripleo12:00
EmilienMhello12:00
*** yamahata_ has joined #tripleo12:01
*** jkilpatr has joined #tripleo12:02
*** yamahata has quit IRC12:02
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates stable/ocata: Install openstack-panko-api on upgrade  https://review.openstack.org/43846212:02
*** adarazs is now known as adarazs_lunch12:03
*** karimb has quit IRC12:03
*** abregman has quit IRC12:04
*** jaganathan has quit IRC12:05
*** agurenko has joined #tripleo12:06
*** jpena is now known as jpena|lunch12:07
openstackgerritArx Cruz proposed openstack/tripleo-quickstart-extras master: Updating skip file mechanism  https://review.openstack.org/43846512:09
jaosoriorbandini: well, so far it seems to me that the ssl parameters in the configuration are not being read12:09
jaosoriorbandini: one could specify the connection args via python, but not really via the connection URL12:09
jaosoriorbandini: any idea where these end up?  It's not sqlalchemy12:09
openstackgerritArx Cruz proposed openstack/tripleo-quickstart-extras master: Updating skip file mechanism  https://review.openstack.org/43846512:09
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci master: Support deploy args in OVB jobs  https://review.openstack.org/43753112:10
bandinijaosorior: so you observed that they are not being read when running the db sync commands?12:11
bandinijaosorior: they are read by pymysql if the config file + config section are set in the db connection uri12:11
jaosoriorbandini: it appears so. And by appears, I mean that I've been trying to get as much traces as I can get, but db_sync doesn't cooperate12:12
jaosoriorbut it is the same error as when I try to log in that user without ssl12:12
EmilienMshardy: hey! fyi I'm enabling Pacemaker in all scenarios jobs, including upgrades one, so we increase testing coverage. All good for you?12:12
jaosoriorbandini: this https://github.com/PyMySQL/PyMySQL ?12:12
bandinijaosorior: yes12:14
bandinijaosorior: the reason for having to use a completely separate file is https://github.com/PyMySQL/PyMySQL/issues/54812:14
bandini(ftr)12:14
EmilienMcan someone review this backport please? https://review.openstack.org/#/c/438297/12:14
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates stable/ocata: Install openstack-heat-agents on upgrade  https://review.openstack.org/43846612:15
EmilienMshardy: I also think https://review.openstack.org/#/c/427836/ would be great to backport, wdyt?12:17
jaosoriorbandini: now I see from the code that the SSL parameters are being ignored12:17
jaosoriorneed to add support for that in PyMSQL12:17
bandinijaosorior: good catch! although meh :/12:17
bandinidciabrin: ^ fyi12:18
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates stable/ocata: Don't recalculate EndpointMap to get outputs  https://review.openstack.org/43846812:18
shardyEmilienM: Hi!, sure enabling pacemaker sounds good, I think it's been quite well tested for upgrades locally12:21
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM: test scenario001 upgrade from newton to ocata  https://review.openstack.org/43573812:21
EmilienMshardy: ack12:22
*** zoli|lunch is now known as zoli12:23
*** pkovar has joined #tripleo12:23
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM: test scenario001 upgrade from newton to ocata  https://review.openstack.org/43573812:23
shardyEmilienM: yes to https://review.openstack.org/#/c/427836 backport, I'll just check that https://review.openstack.org/#/c/413625/ also exists on stable/ocata12:23
shardyOk it does so we're good12:24
EmilienMshardy: https://review.openstack.org/#/c/413625/ is in stable/ocata since we branched recently12:24
shardyEmilienM: Yeah I just wanted to check as I couldn't remember when it landed exactly12:24
shardy(both patches were waiting for review for a while)12:24
EmilienMshardy: excellent. We now have scenarios in place, I did some testing this weekend and scenario001 was passing12:25
shardyEmilienM: FYI snecklifter was asking for that same backport to stable/newton12:25
shardyEmilienM: I got the dependent heat patch landed to stable/newton, so will propose both t-h-t patches after we get this one onto ocata12:26
EmilienMshardy: i'm fine to backport it!12:26
*** thrash|travel is now known as thrash12:26
shardyEmilienM: Ok thanks, I can propose it after your ocata patch lands12:26
EmilienMshardy: also pingtest seems to work fine on regular upgrade job https://review.openstack.org/#/c/438055/12:27
*** saneax is now known as saneax-_-|AFK12:27
EmilienMI propose to enable it once we have all 3 scenarios passing the upgrade job12:27
shardyEmilienM: Ok, that looks good as a first step, then we can consider moving the pingtest to before the upgrade in future12:28
*** fultonj has joined #tripleo12:28
EmilienMshardy: yeah exactly, step by step12:28
EmilienMshardy: I don't want make big changes at this stage12:29
shardyEmilienM: ack, agreed12:29
*** rhallisey has joined #tripleo12:30
*** dougbtv has joined #tripleo12:30
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM - test scenario upgrades with pingtest  https://review.openstack.org/43847112:31
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: Deploy CI scenarios with Pacemaker  https://review.openstack.org/43828712:31
arxcruzchandankumar: http://logs.openstack.org/16/431916/11/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha-oooq-nv/f520ef7/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz12:37
chandankumararxcruz: i have updated the patch https://review.openstack.org/43191612:39
arxcruzchandankumar: oh, okay, I just saw this result on my email :)12:40
fpanEmilienM: could you +workflow on https://review.openstack.org/#/c/435074/12:43
*** masco has quit IRC12:43
EmilienMfpan: done12:44
fpanEmilienM: thanks!12:44
jaosoriorbandini: what do you think https://github.com/PyMySQL/PyMySQL/pull/552 ?12:48
*** sudipto has joined #tripleo12:50
*** sudipto_ has joined #tripleo12:50
bandinijaosorior: looks sane to me, thanks12:51
*** jcoufal has joined #tripleo12:51
jaosoriordciabrin: can you check this out https://github.com/PyMySQL/PyMySQL/pull/552 ?12:51
openstackgerritMerged openstack/tripleo-quickstart-extras master: Reintroduce undercloud_enable_novajoin for undercloud.conf  https://review.openstack.org/43790812:55
*** lucasagomes is now known as lucas-hungry12:57
*** psahoo has quit IRC12:58
*** Guest49819 is now known as rook12:58
EmilienMjaosorior: fyi https://review.openstack.org/#/c/426274/ you should not use depends-on for a patch in the same project, but rather use git rebase.12:59
*** jayg|g0n3 is now known as jayg13:00
jaosoriorEmilienM: thought depends-On worked as well13:00
*** jaganathan has joined #tripleo13:00
jaosoriorEmilienM: anyway, the MySQL SSL work is currently blocked by this https://github.com/PyMySQL/PyMySQL/pull/55213:00
EmilienMjaosorior: not 100% sure, moreover when you have another depends-on from another project in the second depends-on13:00
EmilienMit makes things complex to debug :)13:00
jaosoriorwell, until that merges13:00
*** jaganathan has quit IRC13:01
*** owalsh is now known as owalsh-afk13:01
EmilienMmerges + released + bumped + packaged13:01
jaosoriorright13:01
jaosoriorso, it'll be a while13:01
*** jaganathan has joined #tripleo13:01
jaosoriorEmilienM: it took me a while to get to that commit haha13:01
*** rbrady has joined #tripleo13:01
jaosoriorEmilienM: either way, at least the depends-on's will block those commits from getting merged, until the beginning of the chain merges13:03
*** jpena|lunch is now known as jpena13:04
*** karimb has joined #tripleo13:04
*** dprince has joined #tripleo13:05
*** beagles is now known as beagles_mtg13:05
*** suuuper has quit IRC13:09
openstackgerritAthlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Make sure we don't stop service when pacemaker is handling them.  https://review.openstack.org/43809113:09
*** michapma has joined #tripleo13:11
*** amoralej is now known as amoralej|off13:11
openstackgerritYair Fried proposed openstack/instack-undercloud master: VIPs should not be part of PXE boot subnet  https://review.openstack.org/43845213:15
*** gkadam_ has quit IRC13:15
*** dsariel has quit IRC13:16
*** limao has joined #tripleo13:16
*** liverpooler has joined #tripleo13:18
*** eck` is now known as eck`gone13:19
EmilienMshardy: can you please review https://review.openstack.org/#/c/438090/ ?13:19
*** limao_ has joined #tripleo13:19
*** trown|ATL is now known as trown13:20
*** limao has quit IRC13:21
*** adarazs_lunch is now known as adarazs13:21
*** prateek_ has joined #tripleo13:22
jaosoriorbandini, dciabrin: Is there a freenode channel where I can poke the pymysql folks?13:24
shardyEmilienM: sure, looking now13:25
shardythe only thing I notice is the upgrades job took 2h11m13:25
dciabrinjaosorior, don't think so. I usually interact via github, and he's in japan13:25
shardyhopefully that's a one-off13:25
EmilienMshardy: we can do a recheck13:25
EmilienMlet me do it, you're right13:25
shardyEmilienM: adding a step shouldn't have much impact unless we've got tasks getting stuck already13:26
openstackgerritMerged openstack/tripleo-heat-templates stable/ocata: Don't assume default network names in net_ip*map  https://review.openstack.org/43829713:26
shardybut yeah lets recheck and I'll hold off +A until we see the results13:26
openstackgerritMerged openstack/tripleo-quickstart-extras master: Ansible role for built packages install  https://review.openstack.org/42686913:28
*** cwolferh has joined #tripleo13:29
jaosoriordciabrin: holy crap he answers fast13:29
*** DrBacchus has joined #tripleo13:31
*** ccamacho is now known as ccamacho|lunch13:34
openstackgerritHarry Rybacki proposed openstack/tripleo-quickstart master: Add option to pass a config file with full path to collect-logs.sh  https://review.openstack.org/43771113:34
*** panda|bah is now known as panda13:35
radezEmilienM: could you take a look at https://review.openstack.org/#/c/438343 when you get a chance?13:37
*** limao_ has quit IRC13:38
*** limao has joined #tripleo13:38
EmilienMradez: done13:39
radezEmilienM: thx!13:39
*** bkopilov has joined #tripleo13:39
*** lblanchard has quit IRC13:40
*** rlandy has joined #tripleo13:42
*** tbonds has quit IRC13:43
openstackgerritMerged openstack/tripleo-quickstart master: Make it possible to customize the list of extradisks to be created  https://review.openstack.org/43742713:47
*** bfournie has joined #tripleo13:49
shardyflaper87: Hey is there a blueprint for the containerized overcloud work?13:50
shardyI see https://blueprints.launchpad.net/tripleo/+spec/containerize-tripleo linked from https://github.com/openstack/tripleo-specs/blob/master/specs/ocata/containerize-tripleo-overcloud.rst13:50
shardybut it's marked obsolete13:50
shardyand AFAICT none of the patches we're landing reference any blueprint13:50
shardyit'd be good to have one or more blueprints to track progress in a transparent way13:51
openstackgerritSaravanan KR proposed openstack/tripleo-heat-templates master: Disable exit on error for pacemaker commands for update flow  https://review.openstack.org/43849513:51
flaper87shardy: wait, that should not be marked as obsolete13:51
flaper87it must be a mistake, let me talk to EmilienM13:51
shardyflaper87: Ok, we can fix that and assign it to you I guess13:52
flaper87shardy: https://github.com/openstack/tripleo-specs/blob/master/specs/ocata/containerize-tripleo-overcloud.rst is the spec for the docker phase and the containerized overcloud13:52
*** zaneb has joined #tripleo13:52
flaper87shardy: that sounds perfect to me13:52
flaper87shardy: thanks for noticing13:52
shardyflaper87: Ok I'll fix the blueprint and target it to pike as a high priority13:52
EmilienMshardy: pike-3?13:52
*** bogdando has joined #tripleo13:52
*** cylopez has joined #tripleo13:52
shardyif we can divide the tasks it'd be good to break out some pieces so we don't have one mega-blueprint13:52
EmilienM++13:53
flaper87EmilienM: shardy pike-3 sounds good13:53
flaper87and also breaking up the remaining tasks13:53
shardyEmilienM: yeah but it'd be good to have some smaller blueprints we can mark complete before then13:53
flaper87EmilienM: I just mentioned to shardy that I'll be working tomorrow on breaking this work in smaller tasks13:53
*** ansmith_ has joined #tripleo13:54
shardyEmilienM: I started https://etherpad.openstack.org/p/tripleo-pike-container-tasks as a possible way to help track the tasks against the schedule13:54
shardynot sure if flaper87 wants to do it that way yet tho, happy to see other approaches13:54
*** myoung|alt is now known as myoung13:54
flaper87shardy: mmh, not sure about smaller blueprints, though. there are quite a few things. I'm happy to add more work items13:54
*** ansmith has joined #tripleo13:54
openstackgerritChandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow  https://review.openstack.org/43191613:54
EmilienMshardy: +1 for etherpad on this case13:54
shardyflaper87: sure, but we need some way to track things so we don't get to pike-3 having landed 200 patches against the same BP and realize it's going to slip :)13:55
shardywe've done that a few times now and it doesn't work that well13:55
shardymaybe just an etherpad is enough though, your call13:55
flaper87shardy: EmilienM yep, agreed. Just saying I've been working on this. Will come up with something soon13:56
shardyflaper87: ack, sure thanks!13:56
*** jmelvin has joined #tripleo13:56
flaper87I just don't want to spend too much time writing N blueprints for items that really belong to the containers spec13:56
flaper87hope it makes sense13:56
*** dsavineau has quit IRC13:56
shardyflaper87: yup, sure fair enough13:56
*** dsavineau has joined #tripleo13:56
EmilienMccamacho|lunch: for https://review.openstack.org/#/c/437545/ - I did like keystone13:57
EmilienMccamacho|lunch: feel free to take over if I missed something13:57
*** chlong has joined #tripleo13:59
fpanshardy: could you take a look at https://review.openstack.org/#/c/427437/ and https://review.openstack.org/#/c/424426/ when you get a chance13:59
*** limao has quit IRC13:59
*** limao has joined #tripleo14:00
*** fultonj has quit IRC14:00
*** jaganathan has quit IRC14:00
*** owalsh-afk is now known as owalsh14:01
skramajaEmilienM: raised https://bugs.launchpad.net/tripleo/+bug/1668266, this is needed for this release. please let me know if it is ok.14:01
openstackLaunchpad bug 1668266 in tripleo "Package update fails in the non-ha cluster" [Undecided,In progress] - Assigned to Saravanan KR (skramaja)14:01
*** jprovazn has quit IRC14:03
*** chlong has quit IRC14:04
*** fultonj has joined #tripleo14:05
jtomasekakrivoka: hi, can you please point me to the code which creates default plan in swift?14:08
*** bogdando has quit IRC14:09
shardyfpan: ack will do, thanks for the patches14:09
*** tzumainn has joined #tripleo14:09
fpanshardy: thanks! sorry the os-net-config is big, but should be mostly contained to vpp interface itself and not other parts...14:10
*** ukalifon has joined #tripleo14:10
*** limao has quit IRC14:11
ukalifonjtomasek: hi14:11
*** lucas-hungry is now known as lucasagomes14:11
*** limao has joined #tripleo14:12
*** toure is now known as toure_ptg14:12
jtomasekukalifon: I am trying to find the code which creates default plan in swift14:12
ukalifonjtomasek: didn't Florian write that code? He should know14:12
jtomasekukalifon: I think it was akrivoka. I asked here already14:13
*** Vijayendra has quit IRC14:14
openstackgerritFlorian Fuchs proposed openstack/tripleo-ui master: Add documentation for adding new languages  https://review.openstack.org/43152714:16
*** gbarros has joined #tripleo14:17
*** jaganathan has joined #tripleo14:17
EmilienMskramaja: sure14:18
*** sdake has joined #tripleo14:21
*** ccamacho|lunch is now known as ccamacho14:22
*** amoralej|off is now known as amoralej14:23
*** sai is now known as smalleni14:24
*** Vijayendra has joined #tripleo14:24
*** achadha has joined #tripleo14:25
akrivokajtomasek: here's the workflow that does that https://github.com/openstack/tripleo-common/blob/master/workbooks/plan_management.yaml#L13314:26
akrivokajtomasek: does that answer your q?14:26
jtomasekakrivoka: thanks, yeah, is that run by tripleo-puppet?14:27
akrivokajtomasek: I'm not sure, probably d0ugal would know?14:27
openstackgerritRonelle Landy proposed openstack/tripleo-quickstart master: Add composable deployment config files and mixed release  https://review.openstack.org/41083114:28
jtomasekakrivoka: ok, thanks14:28
d0ugaljtomasek: instack-undercloud14:29
akrivokajtomasek: about plan import/export - I thought a bit more about it14:29
d0ugaljtomasek: https://github.com/openstack/instack-undercloud/blob/master/instack_undercloud/undercloud.py#L134914:29
jtomasekd0ugal: ack, is it possible that that workflow creates user-environment.yaml?14:29
*** achadha has quit IRC14:29
akrivokajtomasek: all the patches are up for review and went through several rounds of review comments14:29
d0ugaljtomasek: I don't think so.14:30
*** abehl has quit IRC14:30
akrivokajtomasek: and I feel they might be ready to merge or very close to it14:30
d0ugaljtomasek: it doesn't use tripleoclient, but calls the workflow directly.14:30
akrivokajtomasek: what do you think about merging the work that is up now and then refactoring it later to make it use the new approach?14:30
jtomasekd0ugal: yeah, ukalifon is getting the default plan with this environment present in swift for some reason14:30
akrivokajtomasek: I volunteer to do the follow up work as well :)14:31
jtomasekukalifon: what setup tool do you use to get undercloud up and running?14:31
*** sdake has quit IRC14:31
jtomasekakrivoka: I don't have objections to do the change in subsequent patches. I plan to look at your patches tomorrow hopefully14:31
ukalifonjtomasek: I have a bash script for that14:31
jtomasekakrivoka: although it would be nice to have the changes done as soon as possible, obviously we need to create spec for it first and then implement everything in tripleo-common14:32
*** limao has quit IRC14:32
jtomasekakrivoka: I plan to start on the spec today and then I'd love to get input from you and dougal on it14:33
akrivokajtomasek: sounds good14:33
*** limao has joined #tripleo14:33
*** v1k0d3n has joined #tripleo14:33
jtomasekukalifon: it would be great if you could install new undercloud and see how the default plan looks without doing anything in GUI or CLI14:34
v1k0d3nflaper87: hey man, are you around by chance?14:34
jtomasekukalifon: so we can isolate whether the problem is caused when default plan is created14:34
d0ugaljtomasek: \o/14:34
*** lblanchard has joined #tripleo14:34
jtomasekakrivoka, d0ugal: cool!14:35
akrivoka\o/14:35
d0ugaljtomasek: is the goal of the spec to define the structure of the file in swift? or?14:35
ukalifonjtomasek: OK, I will see what the default plan looks like right after installation. I think maybe that env got into the default plan when I deployed a custom plan that had IPv6, so I'll pay attention to that too14:36
*** sdake has joined #tripleo14:36
jtomasekd0ugal: yeah, I'd like to outline the change from mistral environment to file in swift, and we than identify blueprints that do it14:36
jtomasekd0ugal: if you think spec is not required I am all in haha14:36
*** dmacpher has joined #tripleo14:36
d0ugaljtomasek: I don't really feel like it is, but a short one would be good14:36
v1k0d3nfor those of you who went to the PTG, as a representative of the #openstack-helm group...i really want to give a big  "thank you" for inviting us to your meetings, and for including/collaborating w/us and the other official openstack deployment groups.14:36
v1k0d3nit was a very productive PTG (i had doubts, and i was proven wrong). :)14:37
*** eck`gone is now known as eck`14:37
masterjcoolHi guys. Following the tripleo docs, I have installed the undercloud, build and uploaded images. I now want to register nodes but when I run `openstack overcloud node import instackenv.json` I get "openstack: 'overcloud' is not an openstack command.", and it then lists several options for overcloud (including `overcloud image upload`). Am I missing something here? Running as user stack, with the stackrc14:38
masterjcoolsourced.14:38
d0ugaljtomasek: I think the most useful part will be defining what the file structure. So far it has just evolved and never really been designed :)14:38
jtomasekd0ugal: maybe the spec could be called "Deployment Plan management updates in tripleo-common" and it would also describe the requirements to manage roles and networks through tripleo-common. moving from mistral environment is sort of initial step of this initiative14:38
*** trozet has joined #tripleo14:38
jtomasekd0ugal: good point14:38
d0ugaljtomasek: Yeah, the migration is going to be a pain :)14:38
*** noslzzp has quit IRC14:39
jtomasekd0ugal:  we can keep current files as a first step (roles-data.yaml and network-data.yaml, capabilities-map.yaml...) and later decide whether we want to move to having everything in single meta file14:39
*** noslzzp has joined #tripleo14:39
jtomasekukalifon: thanks!14:40
openstackgerritMerged openstack/puppet-tripleo master: Remove todo comment  https://review.openstack.org/43829014:40
openstackgerritJuan Antonio Osorio Robles proposed openstack/instack-undercloud master: Add a notifications topic for novajoin  https://review.openstack.org/43851514:40
openstackgerritMerged openstack/tripleo-ui stable/ocata: Fix the deletePlan action  https://review.openstack.org/43793514:40
openstackgerritMerged openstack/tripleo-heat-templates stable/newton: Adding the ip_conntrack_proto_sctp kernel module  https://review.openstack.org/43834314:40
openstackgerritMerged openstack/tripleo-ui master: Cleanup Nodes selectors  https://review.openstack.org/43151514:41
openstackgerritMerged openstack/tripleo-ui stable/ocata: Add i18n to Nodes table  https://review.openstack.org/43753914:41
mwhahahatherve, jaosorior: https://bugzilla.redhat.com/show_bug.cgi?id=139655314:41
openstackbugzilla.redhat.com bug 1396553 in openstack-selinux "apache unable to listen on port 8000 to front heat-api-cfn" [Unspecified,New] - Assigned to lhh14:41
masterjcoolAh sorry, my bad. The documentation shows I need to use `openstack baremetal` for this.14:42
EmilienMmcornea: have you seen this one? http://logs.openstack.org/71/438471/1/check/gate-tripleo-ci-centos-7-scenario001-multinode-upgrades-nv/fbd12af/console.html#_2017-02-27_14_26_51_48339914:42
EmilienMmcornea: when doing your testing14:42
mwhahahatherve, jaosorior: when i looked at it, 8000 is assigned to something else in selinux so apache never works. i forget the specifics but 8000 is associated with something else and i don't think we have a proper rule in openstack-selinux14:42
jaosoriormwhahaha: any chance someone will pick up the openstack-selinux bit?14:43
mwhahahano idea14:43
shardymwhahaha: tcp/8000 runs heat-api-cfn by default14:44
mwhahahashardy: right but not httpd14:44
jaosoriorshardy: the context is that we try to run heat over httpd14:44
jaosoriorand running heat-cfn on httpd fails because of a selinux issue.14:44
shardyah, Ok thanks for the context14:45
fultonjmandre the latest containerized deploy had an issue for me via quickstart. are you interested in looking further like last time?14:46
*** lblanchard has quit IRC14:47
fultonjmandre: i can file a bug if you like14:47
mcorneaEmilienM: I haven't seen it so far14:47
*** sshnaidm has quit IRC14:47
EmilienMmcornea: digging14:47
*** lblanchard has joined #tripleo14:48
*** owalsh has quit IRC14:49
*** owalsh has joined #tripleo14:49
EmilienMmcornea: same for http://logs.openstack.org/71/438471/1/check/gate-tripleo-ci-centos-7-scenario002-multinode-upgrades-nv/8cc08e7/console.html#_2017-02-27_14_10_03_80203214:50
EmilienMmcornea: I think I found why14:50
EmilienMmcornea: do we run puppet at step3 during upgrade process?14:50
mcorneaEmilienM: no, we run puppet after the upgrade14:51
EmilienMmcornea: at which step?14:51
mcorneaEmilienM: it could be that the service is misconfigured during the upgrade14:51
EmilienMI don't think it's a service, it's an endpoint missing14:51
EmilienMmcornea: after the upgrade, do we run puppet for every step?14:52
flaper87v1k0d3n: here14:52
mcorneaEmilienM: I believe so - https://github.com/openstack/tripleo-heat-templates/blob/master/environments/major-upgrade-converge.yaml#L414:53
v1k0d3nhey flaper87...so going to start the process on the infra side to get openstack-helm in openstack proper.14:53
v1k0d3nwe're going to discuss this in greater detail at a community meeting tomorrow. would you like to attend?14:54
*** limao has quit IRC14:54
flaper87v1k0d3n: yes, please14:54
flaper87v1k0d3n: lemme join the os-helm channel14:54
EmilienMmcornea: I'll dig into logs14:54
v1k0d3nawesome. see you there.14:54
*** limao has joined #tripleo14:54
*** jprovazn has joined #tripleo14:55
mandrefultonj: yeah! tell me more about the failure14:55
fultonjmandre: like last time it timed out but this time it doesn't have any containers14:55
EmilienMmcornea: trying scenario-upgrades, they all fail on something14:55
EmilienMmcornea: on pingtest14:55
fultonjmandre: also, it seems to have failed while attmepting to containerize the controller14:56
fultonjmandre: this was a two-node deploy (compute + control) with environments/docker.yaml + environments/docker-network.yaml14:56
EmilienMjaosorior: can you look https://review.openstack.org/#/c/438055/ please?14:56
fultonjmandre: built right after you landed the heat json size change14:56
mandrefultonj: ok, and that was like last time using the old arch, or you used master this time?14:56
EmilienMmcornea: look http://logs.openstack.org/71/438471/1/check/gate-tripleo-ci-centos-7-scenario003-multinode-upgrades-nv/63331f0/console.html#_2017-02-27_14_03_31_20030514:56
fultonjmandre: yes14:56
jtomasekflorianf: what do you mean by breaking websocket connection? On logout the connection should be terminated14:57
EmilienMmcornea: it all fails on finding the endpoints that were added in Ocata14:57
fultonjlatest master from friday afternoon14:57
fultonjmandre: your ssh key is still on the hypervisor if you want to look14:57
fultonjmandre: i can file but too if you like14:57
EmilienMjistr: I'm confused by your review on https://review.openstack.org/#/c/438044/, you -1 or +2 ?14:59
florianfjtomasek: I haven't figured out what the exact circumstances are, but in one SSL setup the websocket connection breaks on logout (Invalid frame header) and cannot be re-established without a page reload14:59
jtomasekflorianf: ack14:59
mandrefultonj: is it possible you're hitting https://review.openstack.org/#/c/438031/?15:00
EmilienMmcornea: I confirm, puppet didn't run step 3 after the upgrade15:01
EmilienMmcornea: I checked in the logs ...15:01
EmilienMshardy: any idea? it seems like puppet is not executed through all the steps after upgrade, probably only the last one?15:01
fultonjmandre: yes, because I don't have that file in tht :)15:01
fultonjmandre: i will redeploy using latest, ok?15:02
openstackgerritJohn Trowbridge proposed openstack/tripleo-heat-templates master: WIP Temporarily disable nova server containers  https://review.openstack.org/43853615:02
openstackgerritJohn Trowbridge proposed openstack/tripleo-heat-templates master: WIP Temporarily disable neutron server containers  https://review.openstack.org/43853715:02
*** sshnaidm has joined #tripleo15:02
jistrEmilienM: neither. There's no clear good choice. I'd like to at least make sure that you and shardy are aware of the peculiarity of UPDATE vs. MINOR_UPDATE, should we ever need to distinguish between the two. I suggested an alternative approach, but it's not without issues either, mainly it would need accompanying code in tripleo-common to take care of resetting StackUpdateType back to NORMAL15:02
mcorneaEmilienM: hm, checking my environment I can see the new endpoints created but I could miss some upstream bits15:03
mandrefultonj: btw, I'm also trying to deploy the containerized overcloud on my side15:03
fultonjmandre: to clarify, mine still has Compute not NodeUserData. i will git pull to get latest THT and redeploy15:03
fultonjmandre: ok, thanks15:03
EmilienMmcornea: what is your version of puppet-tripleo?15:04
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart-extras master: Stop pulling Atomic image  https://review.openstack.org/42274215:05
mcorneaEmilienM: puppet-tripleo.noarch 0:6.2.0-0.20170222195517.7f91deb.el7ost15:05
*** udesale has joined #tripleo15:05
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart-extras master: Stop pulling Atomic image  https://review.openstack.org/42274215:05
mcorneaEmilienM: where can I see the upgrade_roles_data.yaml ?15:07
*** trown is now known as trown|brb15:07
*** jaganathan has quit IRC15:07
EmilienMmcornea: https://github.com/openstack-infra/tripleo-ci/blob/master/test-environments/upgrade_roles_data.yaml15:07
shardyEmilienM: Hmm, not sure, checking15:10
*** trown|brb is now known as trown15:11
*** trown is now known as trown|mtg15:11
*** trown|mtg is now known as trown15:11
shardyEmilienM. mcornea: I see all the steps rendered in the puppet/post-upgrade.yaml and locally I'm seeing all steps run15:11
shardymcornea: do you have an environment with this reproduced locally?15:12
trownmandre: I got it working with the couple quickstart patches and disabling nova and neutron server containers15:12
EmilienMmcornea, shardy: https://bugs.launchpad.net/tripleo/+bug/166829415:12
openstackLaunchpad bug 1668294 in tripleo "puppet is not run at step 3 after upgrade from newton to ocata" [Critical,Triaged]15:12
trownmandre: I will file bugs for the nova and neutron issues15:12
mandredprince, jistr, flaper87: one of you wants to +A https://review.openstack.org/#/c/438031/?15:12
mandrenot trying to influence your judgement :)15:12
mcorneashardy: no, I haven't seen it on my environment15:13
mcorneashardy: shouldn't https://github.com/openstack-infra/tripleo-ci/blob/master/test-environments/upgrade_roles_data.yaml contain some services?15:13
mandretrown: great! that already represents quite a few containerized services15:13
*** skramaja has quit IRC15:13
shardymcornea: No because we set the services via t-h-t environments, so upgrades can work with scenarios15:13
EmilienMmcornea: here i think https://github.com/openstack/tripleo-heat-templates/blob/master/ci/environments/scenario001-multinode.yaml15:14
dprincemandre: +A. you are apparently very infuential15:14
*** limao has quit IRC15:15
mandredprince: thx15:15
*** limao has joined #tripleo15:15
shardyhttp://logs.openstack.org/71/438471/1/check/gate-tripleo-ci-centos-7-scenario002-multinode-upgrades-nv/8cc08e7/logs/subnode-2/var/log/messages15:16
shardyEmilienM: you can see the puppet apply for step3 at Feb 27 14:02:12 (Feb 27 13:15:23 for the initial deploy)15:16
shardyI wonder if one of the ansible running puppet tasks breaks things by writing a hard-coded step15:17
EmilienMshardy: right but not after the upgrade15:17
openstackgerritBrent Eagles proposed openstack/puppet-tripleo master: Default neutron dhcp_agents_per_network to number of agents  https://review.openstack.org/43854515:17
* shardy looks15:17
shardyEmilienM: Ok so we run it on deploy, then during the upgrade, but not during the converge15:17
EmilienMshardy: during the upgrade? like all the 5 puppet steps?15:17
shardyEmilienM: yes we run them all after the ansible tasks15:18
shardythen we should run them all again during the converge15:18
EmilienMmhh, then I wonder why endpoints are not created at step3 after ansible tasks15:18
shardyso you should see step3 three times15:19
*** dsariel has joined #tripleo15:19
openstackgerritBrent Eagles proposed openstack/tripleo-heat-templates master: Make neutron dhcp agents per network conditional  https://review.openstack.org/43854915:19
*** athomas has quit IRC15:20
*** sdake_ has joined #tripleo15:20
*** mcornea has quit IRC15:21
*** mcornea has joined #tripleo15:21
*** cylopez has quit IRC15:21
EmilienMshardy: weird, i see ControllerDeployment_Step3 3 times in /var/log/messages15:22
openstackgerritOpenStack Proposal Bot proposed openstack/tripleo-heat-templates stable/ocata: Updated from global requirements  https://review.openstack.org/43625615:22
*** prateek_ has quit IRC15:23
EmilienMshardy: yeah you're right, timestamps say puppet ran at step 3 during the upgrade15:23
*** sdake has quit IRC15:23
EmilienMshardy: I just don't see the Keystone_endpoint resources15:24
*** gbarros has quit IRC15:24
shardyEmilienM: yeah - http://paste.openstack.org/show/600639/15:25
*** gbarros has joined #tripleo15:25
EmilienMoh wait15:25
*** chlong has joined #tripleo15:25
EmilienMI found why15:25
*** athomas has joined #tripleo15:25
EmilienMshardy: pingtest is taken from master while services were deployed from stable/newton15:26
EmilienMshardy: in the case of zaqar, zaqar was not deployed during the upgrade process: http://logs.openstack.org/71/438471/1/check/gate-tripleo-ci-centos-7-scenario002-multinode-upgrades-nv/8cc08e7/logs/subnode-2/var/log/ (no logs for zaqar)15:26
*** chlong_ has joined #tripleo15:26
EmilienMshardy: so I guess it's a bug in tripleo-ci, if we consider that we don't want to deploy the services that are in master (versus newton); do you see what I mean?15:28
*** chlong_ has quit IRC15:28
*** prateek has quit IRC15:28
EmilienMbug is here: https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/tripleo.sh#L104215:29
EmilienMit should try to load the stable/newton pingtest I think15:29
shardyEmilienM: Yeah, initially I wasn't expecting to support adding services during upgrades but perhaps it's best if we do, so the scenarios align15:29
shardyEmilienM: yeah, or pass the updated environments to the upgrade15:30
EmilienMshardy: I don't know how to do what you said, but I think that would be the way to go.15:30
*** twi99y has quit IRC15:30
EmilienMI can look15:30
*** twi99y has joined #tripleo15:30
thervemwhahaha, Do you know who can help moving that selinux problem forward?15:31
mwhahahatherve: i poked rhallisey and he said lon was the best person15:31
jschlueterdo you guys have a good log in permissive captured and a good way to re-produce it?15:32
EmilienMshardy: somewhere around https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L452 ?15:32
thervejschlueter, https://review.openstack.org/#/c/394835/ has the failing builds15:33
jschluetermwhahaha, therve: lon is over in the the #rdo channel ...15:33
therveCool let's move there15:34
jschlueterbest approach is to capture the /var/log/audit/audit.log in permissive for the whole thing and identify an easy re-producer if possible ...15:34
shardyhttps://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L31015:34
jschlueterthat's what's needed to debug/figure out selinux issues15:34
*** sshnaidm has quit IRC15:34
shardyEmilienM: yeah, I think we need to remove ^^ and rewrite to use the /usr/share path for upgrade15:35
shardyhttps://github.com/openstack-infra/tripleo-ci/blob/master/scripts/deploy.sh#L45215:35
shardyEmilienM: Hmm, ^^ is supposed to already do that15:35
*** dmacpher is now known as dmacpher-afk15:36
*** sshnaidm has joined #tripleo15:36
*** limao_ has joined #tripleo15:37
*** limao has quit IRC15:39
*** jaganathan has joined #tripleo15:40
openstackgerritBrent Eagles proposed openstack/tripleo-heat-templates master: Make neutron dhcp agents per network conditional  https://review.openstack.org/43854915:40
openstackgerritBrent Eagles proposed openstack/python-tripleoclient master: Remove calculation for NeutronDhcpAgentsPerNetwork  https://review.openstack.org/43856415:41
EmilienMshardy: maybe in the meantime I can run pingtest from stable...15:42
*** liverpooler has quit IRC15:43
EmilienMshardy: UPGRADE_ENV is only set for multinode-upgrades and not for scenarios, FYI15:43
EmilienMthat's maybe why15:43
EmilienMI'm trying something15:44
shardyaha15:46
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Export UPGRADE_ENV for scenario-upgrades  https://review.openstack.org/43856715:46
EmilienMshardy: that? ^15:46
*** michapma has quit IRC15:46
*** artom has quit IRC15:47
openstackgerritMerged openstack/tripleo-heat-templates master: Update the Nova Endpoint information for Cinder  https://review.openstack.org/43715315:47
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates stable/ocata: Update the Nova Endpoint information for Cinder  https://review.openstack.org/43856815:48
*** hewbrocca is now known as hewbrocca_afk15:48
*** liverpooler has joined #tripleo15:48
flaper87dprince: EmilienM https://review.openstack.org/#/c/438418/ quick docs patch to mark the docs as 'under dev'15:48
*** liverpooler has quit IRC15:49
EmilienMyolanda: so for tripleo scenarios15:49
*** liverpooler has joined #tripleo15:49
EmilienMyolanda: yes, we can pick one scenario and deploy the new services in https://github.com/openstack/tripleo-heat-templates/blob/master/ci/environments/15:50
yolandaso do you think we can reuse an existing one?15:50
EmilienMyolanda: and document in https://github.com/openstack/tripleo-heat-templates/blob/master/README.rst15:50
EmilienMyolanda: the question is how to test that?15:50
*** artom has joined #tripleo15:50
*** artom has quit IRC15:51
yolandaEmilienM, so currently there are changes on puppet-neutron for that, already landed. And there should be changes in tripleo as well, that i'm not 100% sure of the status15:51
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Export UPGRADE_ENV for scenario-upgrades  https://review.openstack.org/43856715:51
yolandabut it may be a matter of adding some config var to enable it, and then check about the status of the service, config files created, etc15:51
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM - test scenario upgrades with pingtest  https://review.openstack.org/43847115:52
EmilienMmcornea, shardy ^ testing again with depends-on I0ac0dac42286994241700f803159764fa3a6aee615:52
*** karthiks has quit IRC15:52
EmilienMyolanda: scenario are live tested, so as soon as the services are in TripleO, you can start testing them15:52
openstackgerritMerged openstack/tripleo-docs master: Show dev warning for the containers section  https://review.openstack.org/43841815:53
*** artom has joined #tripleo15:53
yolandaEmilienM, and what about testing that on puppet directly, without tripleo? is that possible?15:53
yolandaso as this is created on puppet-neutron, it can be independent of tripleo, and used by another projects as well15:54
EmilienMyolanda: ah !!! I know understand15:54
EmilienMyolanda: big +115:54
yolandaEmilienM, https://github.com/openstack/puppet-neutron/blob/master/manifests/services/bgpvpn.pp15:55
EmilienMyolanda: so you need to patch puppet-openstack-integration (pick a scenario), patch neutron.pp (with a conditional like $test_bgp) and set it to true in fixtures. Also patch run_tests.sh with the Tempest tests you want to run to actually test the feature15:55
yolandawhat are the criterias for picking an scenario?15:56
*** karimb has quit IRC15:56
EmilienMyolanda: in that case, zero. But we try to keep tripleo/puppet scenarios consistent in term of coverage. Easy to understand15:56
openstackgerritAttila Darazs proposed openstack/tripleo-quickstart-extras master: collect-logs: use rsync to gather logs  https://review.openstack.org/43787815:56
EmilienMyolanda: you might want to see https://github.com/openstack/puppet-openstack-integration#description15:56
yolandaso you will do parallel testing in puppet and tripleo?15:56
EmilienMyolanda: first puppet15:57
EmilienMyolanda: keep in mind puppet/scenario003 test LB and not OVS15:57
EmilienMyolanda: I would pick 004 if you don't mind15:57
yolandasure15:57
*** limao has joined #tripleo15:58
*** tobias_fiberdata has quit IRC15:58
EmilienMfpan: need backport to ocata? https://review.openstack.org/#/c/435074/15:58
yolandathis is actually a joint work between our team and tmorin, that created the bgpvpn project. So i'll share that information with him and we'll come up with something15:58
arxcruzEmilienM: regarding https://bugs.launchpad.net/tripleo/+bug/1667679 it's also affecting overcloud15:59
openstackLaunchpad bug 1667679 in OpenStack Compute (nova) "Setting quota fails saying admin project is not a valid project" [Low,Confirmed]15:59
arxcruzEmilienM: as you can see in http://logs.openstack.org/periodic/periodic-tripleo-ci-centos-7-ovb-nonha-tempest-oooq/065132f/logs/oooq/stackviz/#/stdin/test-details/tempest.api.compute.admin.test_quotas.QuotasAdminTestJSON.test_delete_quota15:59
openstackgerritAttila Darazs proposed openstack/tripleo-quickstart-extras master: GATE CHECK  https://review.openstack.org/43027716:00
*** beagles_mtg is now known as beagles16:00
fpanEmilienM: yeah16:00
EmilienMfpan: ok, please do it when it's merged16:00
*** limao_ has quit IRC16:00
fpanEmilienM: will do, thanks for the reminder16:01
*** sdake_ is now known as sdake16:01
openstackgerritMerged openstack-infra/tripleo-ci master: upgrades: run pingtest at the very end of the CI job  https://review.openstack.org/43805516:01
*** gbarros has quit IRC16:03
*** zoli is now known as zoli|brb16:05
*** karimb has joined #tripleo16:06
EmilienMI have some backports, ready for review https://review.openstack.org/#/q/owner:%22Emilien+Macchi+%253Cemilien%2540redhat.com%253E%22+branch:stable/ocata+status:open+project:openstack/tripleo-heat-templates16:06
openstackgerritAlex Schultz proposed openstack/puppet-tripleo master: Add release note for httpchk  https://review.openstack.org/43857616:08
*** artom has quit IRC16:09
*** artom has joined #tripleo16:10
*** flepied has quit IRC16:10
*** Fenuks has joined #tripleo16:11
FenuksHello! Anyone can help me with narrowing down the source of the problem with os-collect-config element?16:13
*** rcernin has quit IRC16:15
*** gkadam_ has joined #tripleo16:19
*** gkadam_ has quit IRC16:20
*** limao has quit IRC16:21
*** udesale has quit IRC16:21
*** flepied has joined #tripleo16:23
*** milan has quit IRC16:23
openstackgerritRonelle Landy proposed openstack/tripleo-quickstart master: Update repo and image links for master release config  https://review.openstack.org/43858216:24
smalleni/ #browbeat-bros16:27
openstackgerritBrad P. Crochet proposed openstack/tripleo-common master: Fix wrong args to update manager  https://review.openstack.org/43858316:27
*** hewbrocca_afk is now known as hewbrocca16:30
*** artom has quit IRC16:31
*** jaganathan has quit IRC16:32
*** artom has joined #tripleo16:33
openstackgerritCarlos Camacho proposed openstack/tripleo-heat-templates master: Add upgrade support for Horizon  https://review.openstack.org/43754516:33
*** mcornea has quit IRC16:33
thrashd0ugal: can you look at https://review.openstack.org/#/c/438583/16:34
*** karthiks has joined #tripleo16:35
thrashd0ugal: and https://review.openstack.org/#/c/436534/16:35
thrashd0ugal: and https://review.openstack.org/#/c/436536/1 :)16:36
d0ugalthrash: on it16:37
thrashd0ugal: thanks...16:38
thrashd0ugal: sorry for the barage...16:38
d0ugalnp :)16:38
*** dciabrin has quit IRC16:39
*** dciabrin has joined #tripleo16:39
*** limao has joined #tripleo16:39
openstackgerritTim Rozet proposed openstack/tripleo-puppet-elements master: Adds puppet-systemd  https://review.openstack.org/43860016:39
*** matbu is now known as matbu-bbl16:40
openstackgerritTim Rozet proposed openstack/puppet-tripleo master: mariadb: Move generation of systemd drop-in to puppet-tripleo  https://review.openstack.org/40825016:41
*** gatuus has joined #tripleo16:41
*** dparkes has quit IRC16:41
*** karthiks has quit IRC16:43
openstackgerritMerged openstack/tripleo-heat-templates master: Enable docker for all roles  https://review.openstack.org/43803116:44
openstackgerritMerged openstack/tripleo-heat-templates stable/newton: Deploy CI scenarios with Pacemaker  https://review.openstack.org/43829516:44
*** Fenuks has quit IRC16:45
*** zoli|brb is now known as zoli16:45
arxcruzjaosorior: hey, around?16:46
jaosoriorarxcruz: what's up?16:46
arxcruzjaosorior: do you know where in tripleo it sets the publicURL for identity for overcloud ?16:46
arxcruzjaosorior: I'm asking because your comment in https://bugs.launchpad.net/tripleo/+bug/166767916:47
openstackLaunchpad bug 1667679 in OpenStack Compute (nova) "Setting quota fails saying admin project is not a valid project" [Low,Confirmed]16:47
arxcruzit also affects overcloud, and it seems to be fixed only in undercloud16:47
jaosoriorarxcruz: well, that was because of the stackrc16:47
*** dmarlin has joined #tripleo16:47
jaosoriorarxcruz: this is how it was fixed in the undercloud https://review.openstack.org/#/c/438017/16:47
jaosoriorarxcruz: you need to make a fix in tripleo-common or python-tripleoclient (I don't exactly remember which), cause that's where the overcloudrc is generated16:48
openstackgerritMerged openstack/tripleo-heat-templates master: Add OpenDaylightConnectionProtocol parameter to opendaylight-api service  https://review.openstack.org/43507416:48
jaosorioroh16:48
jaosoriorwait a minute16:48
jaosoriorthat fix was not for the stackrc16:48
jaosoriorit was for the public keystone endpoint16:48
arxcruzjaosorior: exactly16:49
jaosoriorarxcruz: let me verify16:49
arxcruzI set manually the public endpoint on overcloud to versionless and it worked16:49
arxcruzat least the quota tests16:49
arxcruzhowever I'm running full tempest again just to ensure that it doesn't break other tests16:49
*** skramaja has joined #tripleo16:49
jaosoriorarxcruz: it's set in puppet, via puppet-tripleo, and you should be able to set it via t-h-t16:50
trozetjaosorior: do you mind reviewing this please? https://review.openstack.org/#/c/427437/16:50
*** leanderthal is now known as leanderthal|afk16:50
openstackgerritMerged openstack-infra/tripleo-ci master: update promotion scripts for stable branches  https://review.openstack.org/42738516:51
jaosoriortrozet: what's your timezone?16:51
openstackgerritMerged openstack/puppet-tripleo master: Add release note for httpchk  https://review.openstack.org/43857616:51
trozetjaosorior: EST16:51
jaosoriortrozet: mind reminding me tomorrow?16:52
trozetjaosorior: sure np16:52
trozetjaosorior: ty16:52
openstackgerritJuan Antonio Osorio Robles proposed openstack/tripleo-heat-templates master: Deploy versionless keystone endpoints (for keystone only)  https://review.openstack.org/43862016:55
jaosoriorarxcruz: ^^16:55
*** funzo_ is now known as funzo16:56
*** limao_ has joined #tripleo16:56
skramajacan anyone review this qucker - https://review.openstack.org/#/c/438495/?16:57
openstackgerritCarlos Camacho proposed openstack/tripleo-heat-templates master: Add upgrade support for Horizon  https://review.openstack.org/43754516:57
jaosoriortrozet: had some time for it; +A16:59
*** limao has quit IRC16:59
trozetjaosorior: oh sweet, thanks!17:00
*** Fenuks has joined #tripleo17:00
EmilienMjaosorior: nice about https://review.openstack.org/#/c/438620/1 - I was about to fix it in puppet-keystone: https://review.openstack.org/#/c/438061/ - but THT is also fine for now17:01
jaosoriorEmilienM: probably it should also be fixed in puppet-keystone17:01
EmilienMjaosorior: I'll take care of it, see my patch ^17:01
*** agurenko has quit IRC17:02
EmilienMjaosorior: please do not backport https://review.openstack.org/#/c/438620/ though17:02
EmilienMjaosorior: the patch in nova that broke us was on pike only AFIK17:02
*** twi99y_ has joined #tripleo17:03
jaosoriorEmilienM: good to know, won't backport it17:03
*** zoli is now known as zoli|gone17:04
*** twi99y has quit IRC17:04
*** rajinir_ has joined #tripleo17:05
openstackgerritGael Chamoulaud proposed openstack/tripleo-quickstart-extras master: Fix hosts file generation  https://review.openstack.org/43862517:08
openstackgerritFeng Pan proposed openstack/tripleo-heat-templates stable/ocata: Add OpenDaylightConnectionProtocol parameter to opendaylight-api service  https://review.openstack.org/43862617:08
*** jmelvin has quit IRC17:08
*** Fenuks has quit IRC17:10
*** yprokule has quit IRC17:12
*** yfried has joined #tripleo17:13
yfriedslagle: ping re https://review.openstack.org/#/c/438452/2. can you understand why it failed gate?17:14
*** twi99y_ has quit IRC17:14
*** twi99y has joined #tripleo17:15
*** sudipto_ has quit IRC17:15
*** sudipto has quit IRC17:15
*** trown is now known as trown|lunch17:15
openstackgerritBen Nemec proposed openstack/tripleo-docs master: Enable warning-is-error for docs job  https://review.openstack.org/43047617:16
*** twi99y_ has joined #tripleo17:16
*** bnemec has joined #tripleo17:17
jristjtomasek: ugh not quite done, looks like I need to rebase or something17:18
slagleyfried: did you check the console log?17:19
EmilienMshardy: I found out that adding services during the upgrade CI job, makes the upgrade failings: http://logs.openstack.org/71/438471/2/check/gate-tripleo-ci-centos-7-scenario001-multinode-upgrades-nv/1c9b160/console.html#_2017-02-27_16_52_00_738631 - I'm investigating17:19
*** twi99y has quit IRC17:19
* bnemec just noticed he hasn't been on IRC all morning17:20
*** abehl has joined #tripleo17:20
*** achadha has joined #tripleo17:20
*** matbu-bbl is now known as matbu17:20
openstackgerritJason E. Rist proposed openstack/tripleo-ui master: Adds tooltips to the deployment plan page  https://review.openstack.org/43772717:21
jristjtomasek: ^17:21
shardyEmilienM: Hmm, did cinder-api get added to the scenario between newton and ocata?17:22
shardyit failed because we have a step0 validation that checks things are running17:22
EmilienMshardy: no, it shouldn't. I saw the trace in validation17:23
EmilienMFeb 27 16:51:56 centos-7-2-node-osic-cloud1-s3700-7497998-443517 os-collect-config: fatal: [localhost]: FAILED! => {"changed": true, "cmd": "/usr/bin/systemctl show 'openstack-cinder-api' --property ActiveState | grep '\\bactive\\b'", "delta": "0:00:00.008762", "end": "2017-02-27 16:51:56.191418", "failed": true, "rc": 1, "start": "2017-02-27 16:51:56.182656", "stderr": "", "stdout": "", "stdout_lines": [],17:23
EmilienM"warnings": []}17:23
EmilienMshardy: oh wait17:23
*** twi99y_ has quit IRC17:23
EmilienMshardy: yes, cinder was added in Newton17:23
*** twi99y has joined #tripleo17:23
chandankumarsshnaidm: arxcruz hello17:23
EmilienMshardy: ok so it means everytime we'll add a service in scenarios, upgrade validation will fail?17:24
shardyEmilienM: the problem is if it's not in newton but we added it in ocata17:24
*** dujelly has joined #tripleo17:24
shardyEmilienM: it means we should only add new services where there aren't any upgrade_tasks17:24
*** karimb has quit IRC17:24
*** limao_ has quit IRC17:25
shardyfor services which exist on newton and ocata, we should backport the addition to the scenario environment, because upgrade_tasks exists17:25
openstackgerritGael Chamoulaud proposed openstack/tripleo-quickstart-extras master: Fix hosts file generation  https://review.openstack.org/43862517:25
shardyEmilienM: the other option is to set SkipUpgradeConfigTags: validation17:26
EmilienMshardy: ok, but it won't fix the case where we add a new service in pike, upgrades will be broken between scenario on ocata & pike, right?17:26
shardybut then we lose some useful coverage17:26
shardyEmilienM: I think new services for pike should probably have the upgrade_tasks (or at least the validation) disabled17:26
arxcruzchandankumar: hey, I'll check your patch in a moment17:26
shardyas we know they won't exist prior to the upgrade (e.g when upgrade_tasks runs)17:26
arxcruzjaosorior: thanks, I'm running tempest right now, once I confirm that there's no failures, I'll add my +1 :)17:26
EmilienMshardy: mhh ok. I'll try something today then17:27
shardyEmilienM: either that or the validations need to be smarter17:27
shardye.g figure out if the service should be running (look for the config file perhaps?)17:27
yfriedslagle: yes. seems like yum install failure. which seems unrelated17:27
yfriedslagle: also, it passed before and this time I've only modifed unittests17:28
slagleyfried: ok, so probably a network issue then17:28
slagleyfried: i'd just try a recheck17:28
*** achadha has quit IRC17:29
*** jpich has quit IRC17:29
*** achadha has joined #tripleo17:30
shardyEmilienM, marios: https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/cinder-api.yaml#L15417:30
slaglebnemec: can you review https://review.openstack.org/#/c/438452/, it seems reasonable to me, but i wasnt sure if that validator was added for a reason17:30
*** prateek has joined #tripleo17:30
shardywe should probably (for all services) only do the ActiveState check if the service is actually enabled17:31
EmilienMshardy: oh, indeed17:31
shardysystemctl can probably tell us that17:31
EmilienMshardy: yes, I'll investigate that way17:31
*** hewbrocca is now known as hewbrocca_afk17:32
bnemecslagle: That was added because of https://bugzilla.redhat.com/show_bug.cgi?id=126845117:33
openstackbugzilla.redhat.com bug 1268451 in instack-undercloud "[RFE] Improve unspecified value for admin_vip in undercloud.conf" [Unspecified,Verified] - Assigned to bnemec17:33
sshnaidmchandankumar, hi17:33
*** owalsh is now known as owalsh-afk17:34
bnemecCrud, no dtrainor today.17:34
bnemecI thought this had been discussed among the ui team.17:35
chandankumarsshnaidm: http://logs.openstack.org/23/436923/3/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha-mitaka/c45553a/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz17:35
bnemecjtomasek: Do you know what the current strategy for accessing the ui remotely is?17:35
chandankumarsshnaidm: is failing due to missing dependency17:35
bnemecjtomasek: In regards to the conflict between https://review.openstack.org/#/c/438452/ and https://bugzilla.redhat.com/show_bug.cgi?id=126845117:35
openstackbugzilla.redhat.com bug 1268451 in instack-undercloud "[RFE] Improve unspecified value for admin_vip in undercloud.conf" [Unspecified,Verified] - Assigned to bnemec17:35
chandankumarsshnaidm: http://logs.openstack.org/23/436923/3/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha-newton/c7c0a0e/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz due to some config issue17:36
chandankumarsshnaidm: http://logs.openstack.org/16/431916/13/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha-oooq-nv/81acadf/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz17:36
openstackgerritAna Krivokapic proposed openstack/tripleo-common master: Enhance plan creation and update with plan-environment  https://review.openstack.org/41416917:36
openstackgerritAna Krivokapic proposed openstack/tripleo-common master: Add plan export action and workflow  https://review.openstack.org/42278917:36
chandankumarsshnaidm: tests are not able to run17:36
sshnaidmchandankumar, so, what is dependency?17:36
EmilienMshardy: on the very short term, I'm going to run pingtest from stable/newton after the upgrade - so we can make progress17:37
slaglebnemec: i'd think that we want to validate that admin_vip is on pxe cidr, but public_vip doesnt necessarily need to be17:37
EmilienMshardy: meanwhile I'm working on the thing you said, checking service is actually enabled during ansible validation17:37
sshnaidmchandankumar, we need to solve everything there, both config issues and dependencies17:37
chandankumararxcruz: ^^17:37
sshnaidmchandankumar, I can provide you a dev env that is exactly copy of upstream CI to try your patches if it will help17:38
sshnaidmchandankumar, if this works locally and fails only on ci17:38
chandankumarsshnaidm: that will work17:38
sshnaidmchandankumar, does it work for you with libvirt? does tempest run?17:39
jtomasekbnemec: I am not entirely sure, we access the UI by configuring undercloud endpoints to use public ip, configured by undercloud_public_vip in undercloud.conf17:39
yfriedslagle: oh. I typed "retest" and couldn't understand why it failed...17:39
*** owalsh-afk has quit IRC17:39
yfriedslagle: I'm rusty17:39
chandankumarsshnaidm: for mitaka failure ionic tempest pluig and neuton tempest plugin is mossong17:39
mandreEmilienM: I dunno if this can help, but this is how we're checking if a service is enabled on the overcloud nodes within tripleo-validations, https://review.openstack.org/#/c/436194/6/validations/ceph-health.yaml@1317:39
chandankumarsshnaidm: i will take a look again tomorrow to fix some issue17:40
sshnaidmchandankumar, I think it's point to raise on #rdo, maybe tosky knows..17:40
EmilienMmandre: oh wow17:40
EmilienMmandre: that is exactly what I needed man17:40
EmilienMmandre: can I shamelessly copy/paste?17:40
toskysshnaidm: chandankumar: I thought that they were included in the image like the others17:41
mandreEmilienM: sure, credit goes to gfidente :)17:41
toskyundercloud image17:41
chandankumartosky: no idea17:41
chandankumartosky: http://logs.openstack.org/23/436923/3/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha-mitaka/c45553a/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz17:41
jtomasekbnemec: in general, public undercloud endpoints should be set to a publicly accessible ip since GUI is downloaded to client browser and makes requests to these ips17:42
shardyEmilienM, mandre: that won't work, because we've set cinder_api to enabled_services, but not started/enabled it yet17:42
*** gchamoul is now known as gchamoul|afk17:42
shardybecause we run the ansible upgrade tasks, then run puppet17:42
*** jmelvin has joined #tripleo17:42
toskysshnaidm, chandankumar: is that tempest from git?17:43
shardyI think we need something which looks at systemctl enabled directly17:43
shardymandre: thanks though :)17:43
EmilienMshardy: I'll re-use some bits, and use systemctl17:43
chandankumartosky: nope from rpm17:43
skramajamarios: shardy - ptal, https://review.openstack.org/#/c/43849517:44
sshnaidmchandankumar, where is install part here? http://logs.openstack.org/23/436923/3/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha-mitaka/c45553a/logs/oooq/undercloud/home/jenkins/tempest-setup.sh.txt.gz17:44
mandreshardy: I see... yeah in that case, probably best to call servicectl directly17:44
mandreor a combination of the two17:44
bnemecslagle: jtomasek: Okay, then someone needs to verify that it's only the admin vip we need to validate then.  I guess that's the one specifically called out in the bz.17:45
openstackgerritMerged openstack/tripleo-heat-templates master: Make heat max_json_body_size configurable  https://review.openstack.org/43793217:45
bnemecMaybe the public_vip validation should be changed to just ping the public vip and make sure it's accessible.17:45
*** david-lyle has joined #tripleo17:45
*** david-lyle has quit IRC17:45
bnemecI suspect in ci it happens to work because we forward ctlplane to the external network.17:45
bnemecThat's not necessarily the case in the real world though.17:45
chandankumarsshnaidm: it has the openstack-tempest installed17:45
chandankumarfrom undercloud image i think17:46
*** david-lyle has joined #tripleo17:46
sshnaidmchandankumar, there is no undercloud image in upstream ci17:46
chandankumarsshnaidm: http://logs.openstack.org/23/436923/3/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha-mitaka/c45553a/logs/rpm-qa.txt.gz17:47
*** derekh has quit IRC17:47
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: upgrade jobs: execute stable pingtest (and not master)  https://review.openstack.org/43864217:47
EmilienMshardy: ^17:47
*** cylopez has joined #tripleo17:47
sshnaidmchandankumar, ok, somewhere it was installed, I suppose as dependency of some openstack-* package17:48
chandankumarsshnaidm: so i need to add for install openstack-tempest in configure -tempest template17:49
toskystill, the error comes from some of the plugins17:49
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM - test upgrades with stable pingtest  https://review.openstack.org/43864417:49
chandankumarsshnaidm: let me dig one more time tomorrow17:50
*** fragatina has joined #tripleo17:50
toskyI would recheck if some backport was done to the mitaka branches of those repositories17:50
sshnaidmchandankumar, yeah, it'd be better to make sure it's installed17:50
sshnaidmchandankumar, sure17:50
sshnaidmtosky, do we know that it works with mitaka at all?17:50
toskysshnaidm: elsewhere, yes17:50
toskysshnaidm: or... maybe17:51
toskyI can't guarantee 100% about it right now17:51
*** dsariel has quit IRC17:51
sshnaidmtosky, just wonder if somebody notice if it's broken,,17:51
toskysshnaidm: but anyway, I don't think it's a blocker for the change that chandankumar was writing17:51
*** fzdarsky is now known as fzdarsky|afk17:52
sshnaidmtosky, if it doesn't work for mitaka? we need to run tempest on mitaka17:52
*** owalsh-afk has joined #tripleo17:52
toskysshnaidm: and did it work without the patch, with whatever tests was there before?17:52
toskywere*17:52
chandankumartosky: http://logs.openstack.org/16/431916/13/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha-oooq-nv/81acadf/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz17:53
sshnaidmtosky, we didn't run with packages, only venv17:53
chandankumartosky: i am not sure what i  am doing wrong  here17:53
toskysshnaidm: venv with plugins from master, or from their respective branches?17:53
sshnaidmtosky, venv from redhat-openstack/tempest repo I mean17:54
*** mburned is now known as mburned_out17:54
toskysshnaidm: isolated venv, or with --system-site-packages?17:54
sshnaidmtosky, like it's configured now17:54
EmilienMmandre, shardy: i'm preparing a patch that uses 'systemctl is-enabled XXX'17:55
sshnaidmtosky, isolated17:55
*** mburned_out is now known as mburned17:55
toskysshnaidm: and the tempest plugins?17:55
toskywhere they installed?17:55
*** chlong has quit IRC17:55
sshnaidmtosky, packages?17:55
openstackgerritDan Radez proposed openstack/tripleo-heat-templates master: Adding keystone parameters for Congress  https://review.openstack.org/43864617:55
toskychandankumar: no idea, try to add some more debugging information17:55
toskysshnaidm: no, if you install in an isolated venv, you need to install the plugins in the same venv iirc17:55
sshnaidmtosky, venv script installs everything it needs17:56
toskysshnaidm: so all plugins from git?17:56
sshnaidmtosky, yep17:56
toskysshnaidm: do you have a working run?17:56
toskysshnaidm: git from master, or from their respective branches?17:56
sshnaidmtosky, mitaka in particular?17:56
toskymitaka in that case17:56
toskyyep17:56
sshnaidmtosky, master17:56
*** aufi has quit IRC17:57
openstackgerritDan Radez proposed openstack/tripleo-heat-templates master: Adding keystone parameters for Tacker  https://review.openstack.org/43864817:57
toskysshnaidm: so never tried the branched plugins, which are in the packages17:57
*** nyechiel has quit IRC17:57
sshnaidmtosky, there is option for packages in file as you can see, but not sure if somebody used it and it worked17:57
*** fragatina has quit IRC17:58
toskysshnaidm: no, I mean: if plugins were installed from master, the scenario "plugins from $wherever branch" (mitaka in this case) was not tested, if I understand it correctly17:58
*** david-lyle has quit IRC17:58
*** dtantsur is now known as dtantsur|afk17:58
*** fragatina has joined #tripleo17:58
toskythe rpm contains the code of the plugins from the respective branch17:58
*** cylopez has quit IRC17:58
sshnaidmtosky, seems like this.. I'm not even sure rdo tempest had branches at all17:59
toskysshnaidm: it had17:59
toskyredhat-openstack/tempest is (was) branched regularly17:59
*** yfried has quit IRC17:59
toskyso the proposed change is just (maybe) uncovering an existing issue18:00
sshnaidmtosky, definitely may be18:00
*** DrBacchus is now known as rbowen18:00
*** achadha has quit IRC18:00
*** david-lyle has joined #tripleo18:01
*** vpickard has quit IRC18:01
openstackgerritMerged openstack/tripleo-heat-templates master: Add VPP composable service  https://review.openstack.org/42743718:01
chandankumartosky: sshnaidm i will try to make sure that patch works with ocata by tomorrow18:02
sshnaidmchandankumar, ok, thansk18:02
chandankumarthen later on we can fix previous release18:02
*** sshnaidm is now known as sshnaidm|off18:02
chandankumartosky: what you say?18:02
beaglesEmilienM: are we requiring launchpad bugs for backports to the Ocata branch?18:02
beaglesor does that happen after release?18:03
toskychandankumar: fine18:03
EmilienMbeagles: it's better18:03
beaglesEmilienM: k18:03
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist  https://review.openstack.org/43865718:04
EmilienMshardy, mandre: wdyt? (I just did glance api)18:04
*** jaosorior has quit IRC18:05
EmilienMmandre: I wonder if I correctly used ignore_errors18:06
shardyI think it should be true?18:06
*** sshnaidm|off has quit IRC18:07
EmilienMshardy: right, done18:07
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist  https://review.openstack.org/43865718:07
EmilienMshardy: I just hope the shell doesn't return True even if glance api is not here18:08
EmilienMbut I don't think that's the behavior18:08
shardyEmilienM: it doesn't, I just tried it18:08
*** prateek_ has joined #tripleo18:08
EmilienMshardy: ok, so can I do all other services?18:08
shardyit returns 1 when a service is disabled or missing which I think is what we want18:08
*** athomas has quit IRC18:09
shardyEmilienM: Yes I think so18:09
EmilienMexcellent18:09
shardysed challenge ;)18:09
EmilienMor 32 files to change :P18:09
*** lucasagomes is now known as lucas-afk18:09
*** yfried has joined #tripleo18:10
*** chlong has joined #tripleo18:10
*** prateek has quit IRC18:11
*** gbarros has joined #tripleo18:13
*** gbarros has quit IRC18:15
*** salmankhan has quit IRC18:16
*** user` has joined #tripleo18:17
*** user` is now known as chem18:17
*** chem has quit IRC18:18
*** chem has joined #tripleo18:18
*** achadha has joined #tripleo18:18
*** yfried has quit IRC18:20
*** gbarros has joined #tripleo18:20
*** achadha_ has joined #tripleo18:20
*** dciabrin has quit IRC18:20
*** dciabrin has joined #tripleo18:21
*** tosky has quit IRC18:22
*** achadha_ has quit IRC18:23
*** achadha has quit IRC18:23
*** achadha has joined #tripleo18:23
*** ckyriakidou has quit IRC18:24
*** achadha has quit IRC18:25
*** gbarros has quit IRC18:26
jristlblanchard: so the tooltips are temporary - we're going to do better tooltips in pike with 'react-joyride'18:26
jristlblanchard: and implement the patternfly styling more specifically18:26
jristlblanchard: this is a temporary solution18:26
*** jpena is now known as jpena|away18:26
skramajashardy: somehow i find the list of enabled services as same across controller and compute - in osp10, any known issues?18:27
skramajashardy: hiera enabled_services18:27
*** ansmith_ has quit IRC18:27
*** achadha_ has joined #tripleo18:28
shardyskramaja: No, the list should match the service_names for each service defined in the roles_data ServicesDefault lists18:31
shardyskramaja: Oh wait18:32
*** yamahata_ has quit IRC18:32
shardywe calculate the list for all roles18:32
* shardy looks for alternative18:32
skramajashardy: can't use systemctl status command, as it also returns error..18:33
skramaja[root@overcloud-compute-0 ~]# hiera pacemaker_enabled18:34
skramajatrue18:34
skramajashardy: ^18:34
*** yfried has joined #tripleo18:34
*** sohail has joined #tripleo18:35
openstackgerritMerged openstack/instack-undercloud master: Add a notifications topic for novajoin  https://review.openstack.org/43851518:35
yfriedbnemec: hi, I think your comment on https://review.openstack.org/#/c/438452 should go on the LP page18:35
skramajabeagles: can you take a look at - https://review.openstack.org/#/c/434255/?18:36
beaglesskramaja: yup18:36
skramajathanks18:36
*** sshnaidm|off has joined #tripleo18:37
shardyskramaja: try hiera service_names18:37
shardythat should be set per-role18:37
shardysorry for the bad info on the review18:37
yfriedbnemec: looking at the bz you mentioned, I don't understand how the current state solves that18:37
skramajashardy: yes it is. thanks, i will make the changes accordingly.18:38
*** gbarros has joined #tripleo18:38
lblanchardjrist: okay that makes sense then!!18:39
*** sudipto_ has joined #tripleo18:40
*** sudipto has joined #tripleo18:40
jristlblanchard: so does that help part of it?18:41
jristlblanchard: I can make the icon blue18:41
rookshardy: deploying with a separated out networker, could I reuse the control flavor?18:42
rookie OvercloudNetworkerFlavor: control18:42
*** achadha_ has quit IRC18:43
*** achadha has joined #tripleo18:43
*** sshnaidm|off is now known as sshnaidm18:43
bnemecyfried: It catches the case where a user changes the default cidr but doesn't change the vips to match.18:44
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist  https://review.openstack.org/43865718:45
*** jkilpatr has quit IRC18:45
*** prateek_ has quit IRC18:46
shardyrook: yes, provided you don't care which nodes from the control tagged nodes are used18:46
rookI don't18:46
rookshardy: but I am seeing :18:47
rook2017-02-27 18:29:49Z [overcloud]: CREATE_FAILED  Resource CREATE failed: ResourceInError: resources.Controller.resources[1].reso18:47
rookurces.Controller: Went to status ERROR due to "Message: No valid host was found. There are not enough hosts available., Code: 5018:47
rook0"18:47
rookeven though I have nodes tagged with control18:47
rook6 nodes to be exact. and I am requesting 518:48
*** gatuus has quit IRC18:50
shardyrook: Ok and all the nodes are the same, e.g memory/disk etc?18:50
rookyup18:50
shardyrook: I'd check the nova scheduler log to see which scheduler rule failed to find a host, and double check the node states and tagging18:51
EmilienMmandre: '|succeeded' means rc == 0?18:51
shardyrook: are any custom roles defaulting their flavor to "baremetal"?18:51
shardyrook: they can use up some of the control tagged nodes if so18:51
rookshardy: yes18:52
rookshardy: how can I avoid that?18:52
mandreEmilienM: I believe they're equivalent, yes... not 100% sure, though18:52
*** pkovar has quit IRC18:52
shardyrook: you'll have to pass Overcloud$RoleFlavor for all roles18:52
shardyto avoid using baremetal anywhere18:52
EmilienMmandre: ok for now I think that's fine. I'll just replace shell by command.18:52
*** gfidente is now known as gfidente|afk18:52
*** abregman has joined #tripleo18:53
shardyrook: we should probably add a validation for this, as mixing the baremetal flavor with anything else generally won't work well18:53
mandreEmilienM: okay, '.rc == 0' is also a pretty common pattern18:53
rookweird shardy18:54
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist  https://review.openstack.org/43865718:54
rook2017-02-27 13:29:47.910 2180 DEBUG nova.scheduler.filters.ram_filter [req-ccf70f06-4c3f-4564-b7e3-3369ebd6157d b53efb7ccbdc4a47ab11c85ad37fde48 f1ec65886fd948cfbeebe23790ccbe9a - - -] (undercloud, 9ba12aa9-0b77-41d8-b34b-db1336404e9c) ram: 0MB disk: 0MB io_ops: 0 instances: 0 does not have 4096 MB usable ram before overcommit, it only has 0 MB. host_passes18:54
rook/usr/lib/python2.7/site-packages/nova/scheduler/filters/ram_filter.py:4518:54
rookhowever shardy , this node has:18:54
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: Deploy CI scenarios with Pacemaker  https://review.openstack.org/43828718:54
rookshardy {u'memory_mb': u'262144', u'cpu_arch': u'x86_64', u'local_gb': u'37118:55
EmilienMmandre: right18:55
rookshardy: so, something between nova and ironic fubar?18:55
mandreEmilienM: I'm going afk now, I'll review your change in the morning if it's still there18:55
shardyrook: hrm, yeah that looks wrong18:55
* shardy looks around for non-afk nova/ironic folks18:56
honzaakrivoka: why does the plan-environment.yaml file need the `template` and `environments` keys?  (n00b question)18:56
rookyeah shardy all my nodes are returning zero ram18:57
rookhowever, in ironic node-show, I see that it is populated right18:57
*** yamahata_ has joined #tripleo18:57
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM: test scenario001 upgrade from newton to ocata  https://review.openstack.org/43573818:59
openstackgerritEmilien Macchi proposed openstack/tripleo-heat-templates master: DNM - test scenario upgrades with pingtest  https://review.openstack.org/43847118:59
*** jkilpatr has joined #tripleo18:59
bfournierook, shardy: there is an Ironic troubleshooting guide here which may be useful - https://docs.openstack.org/developer/ironic/deploy/troubleshooting.html19:00
openstackgerritSaravanan KR proposed openstack/tripleo-heat-templates master: Disable exit on error for pacemaker commands for update flow  https://review.openstack.org/43849519:00
skramajashardy: ^19:00
rooklooking19:01
honzaakrivoka: is that just the default environment?19:01
openstackgerritEmilien Macchi proposed openstack-infra/tripleo-ci master: Add gate-tripleo-ci-centos-7-ovb-updates into promotion pipeline  https://review.openstack.org/43379619:01
rookbfournie: so, all my nodes mention > 0 GB Ram (actually close to 200+) however nova-scheduler states the RAMFilter shows zero hosts with > 0GB of ram19:04
rookI tried kicking the services and running ironic node-list --provision-state available --maintenance false --associated false ; which gave me all 107 nodes I have.19:05
*** shardy is now known as shardy_afk19:07
beaglesEmilienM: in https://review.openstack.org/#/c/430444/ you asked pradk to add a relnote to the top patch. What top patch was that?19:11
EmilienMbeagles: yeah top patch means to me a patch on top of this one19:11
bfournierook, so when you run "ironic node-show <IRONIC NODE> --fields properties" you're see valid memory?19:11
rookyes bfournie19:11
EmilienMbeagles: I won't block the patch, but if we can do it afterward19:11
beaglesEmilienM: in the same repo though right..19:11
EmilienMbeagles: yeah but a different patch, so we can land this one19:12
EmilienMbeagles: release note patches don't run tripleo jobs19:12
beaglesoooooooohhhh19:12
beaglespfft19:12
EmilienMbeagles: just reno :)19:12
beaglesokay19:12
EmilienMI should remove "top patch" from my vocabulary19:12
akrivokahonza: yes, that's what's needed to create a minimal environment19:12
beaglesEmilienM: no that's what I call it too19:13
beaglesEmilienM: where there didn't seem to be one in the series, I was a bit confused19:13
EmilienMbeagles: top sep patch :P19:13
*** trown|lunch is now known as trown19:13
akrivokaakrivoka: those were previously contained in the capabilities-map.yaml but now we want to have them in this new file19:13
akrivokahonza: ^19:13
bfournierook, did this show memory_mb_used? "openstack hypervisor stats show"19:13
akrivokaI'm talking to myself now \o/19:13
beaglesEmilienM: add a top or follow up all the same :)19:13
rookhttps://gist.github.com/jtaleric/8b53ddf3976fa99446e7c7da70169f2e bfournie19:14
honzaakrivoka: ah, cool, i was a bit perplexed, thanks19:15
*** florianf has quit IRC19:15
arxcruzcan someone review and merge https://review.openstack.org/#/c/438620 before tomorrow? So next periodic job can ran without this failures ?19:15
openstackgerritSaravanan KR proposed openstack/os-net-config master: Restart Openvswitch after adding DPDK ports  https://review.openstack.org/43425519:15
arxcruzEmilienM ^19:15
openstackgerritAna Krivokapic proposed openstack/python-tripleoclient master: Add plan export command  https://review.openstack.org/42585819:16
EmilienMarxcruz: I think I already +2-ed, isn't?19:16
*** sudipto_ has quit IRC19:17
*** sudipto has quit IRC19:17
*** skramaja has quit IRC19:17
yfriedbnemec: could you plz explain to me why tripleo needs to enforce that vip is in the cidr?19:17
yfriedbnemec: of the PXE subnet?19:17
arxcruzEmilienM: I don't know another core who +2 and merge19:18
*** gbarros has quit IRC19:18
arxcruzEmilienM: I've tested, and with this patch, I reduce to 2 failures in full tempest on periodic job, and I would like to see it happening tomorrow :D19:18
yfriedbnemec: As I've mentioned on LP - that's a very unlikely scenario. the PXE subnet is the last place that should be reachable for an undercloud user19:18
*** gatuus has joined #tripleo19:19
bnemecyfried: I already did.  It's for the case when a user changes the cidr but not the vips.  This is reported in the bz to cause undercloud installation failure.19:22
akrivokahonza: thanks for reviewing! I'll address those comments tomorrow19:22
lblanchardjrist: yep just change the icon color…should I update my comment?19:22
yfriedbnemec: I'm challenging the initial assumption that the vip should be part of the pxe cidr in first place19:24
bnemecyfried: And I'm pointing you at the bz that says it breaks if it's not.19:25
yfriedbnemec: I read that19:25
honzaakrivoka: np!19:25
yfriedbnemec: if putting a bad address at the vip is wrong, why are we enforcing the bad assumption instead of doing some other validation?19:26
*** alop has joined #tripleo19:26
*** sohail has quit IRC19:26
*** sudipto_ has joined #tripleo19:27
*** sudipto has joined #tripleo19:27
yfriedbnemec: I'm new to this issue, so I wonder if you have a reason for this initial value19:28
bnemecyfried: Those are not mutually exclusive things.  Historically it was safest to put the vip on the provisioning network because it was reported to cause problems if it wasn't.19:28
thrashrbrady: if you have a moment: https://review.openstack.org/#/c/438583/19:28
bnemecyfried: The UI presents a new use case where that is no longer true.19:28
yfriedbnemec: let's put a pin in the UI19:28
bnemecyfried: Thus we need a more flexible validation.19:28
yfriedbnemec: consider a regular CLI user19:28
*** pcaruana has quit IRC19:29
*** akrivoka has quit IRC19:29
yfriedbnemec: they can't access the API from outside the undercloud, because it's listening on the internal PXE subnet19:29
bnemecyfried: Most cli use is on the undercloud itself.  In the past if you wanted to contact undercloud apis externally then you needed a routable provisioning network.19:30
bnemecBut in the past that mostly didn't work for a lot of things because the cli assumed you were on the undercloud.19:30
bnemec(it still might for a few functions)19:30
yfriedbnemec: which is a bad assumption that should be solved by the VIPs, shouldn't it?19:31
yfriedbnemec: like any regular openstack cloud API19:31
bfournierook, are you running quickstart?19:31
*** sshnaidm is now known as sshnaidm|afk19:31
*** sudipto_ has quit IRC19:32
*** sudipto has quit IRC19:32
*** yamahata_ has quit IRC19:33
bfournierook, wondering if you are hitting this https://bugs.launchpad.net/tripleo/+bug/1651704, which is just the manifestation of this bug with quickstart and virtualbmc which has recently been resolved: https://bugs.launchpad.net/tripleo/+bug/165171919:34
openstackLaunchpad bug 1651704 in tripleo "Errors when starting introspection are silently ignored" [High,Fix released] - Assigned to Dougal Matthews (d0ugal)19:34
openstackLaunchpad bug 1651719 in tripleo "Introspection fails on nodes with the same IPMI address but different IPMI ports" [High,Triaged]19:34
yfriedbnemec: I have a different idea19:34
yfriedbnemec: instead of having actual address in VIPs, set default to "None"19:35
yfriedbnemec: in that case, pull arbitrary addresses from the given CIDR (say - first 2) if you must have VIPs19:36
*** sdake has quit IRC19:36
rookbfournie: nope, no qs19:36
bfournierook; k19:36
yfriedbnemec: that way, you keep current functionality without the weird rule enforcing this when users actually wish to use the API as documented19:36
yfriedbnemec: what do you think?19:37
*** vpickard_afk has joined #tripleo19:37
rookfultonj: you happen to be around... got a custom role question...19:37
fultonjrook: ok19:37
rookhttps://gist.github.com/jtaleric/e2c4d1c310af04305f5d3effe04b8f7c fultonj19:37
rookso, I created the CNCFNetworker, and CNCFBaremetal roles19:38
rooki see the networker/controller/ceph node19:38
rookbut no compute or baremetal node deployed19:38
rookam I just being impatient?19:38
rook(deployment is still going)19:38
*** vpickard_afk is now known as vpickard19:38
openstackgerritJohn Trowbridge proposed openstack/tripleo-quickstart-extras master: Add build-images role  https://review.openstack.org/41433619:38
*** sudipto has joined #tripleo19:39
*** sudipto_ has joined #tripleo19:39
fultonji'd expect to see it in the nova list too. too early to call if something is wrong19:42
fultonjrook: 619:43
fultonjrook: s/6/^/19:43
rookfultonj: yeah, no messages sent to stdout yet19:43
rookI do see a reference to 2017-02-27 19:19:50Z [overcloud.CNCFBaremetal]: CREATE_IN_PROGRESS  state changed19:43
fultonjyou'll either get the nodes or a scheduler error19:44
bnemecyfried: That could be useful.  A few gotchas to watch out for though:19:44
bnemec1) It can't just be the first two because usually the first IP is used for local_ip (if for no other reason than because it's what we default to)19:44
bnemec2) You would also need to be careful not to overlap either DHCP range.  Basically 1 and 2 mean the vip selection needs to be intelligent based on the other settings, which is doable.19:44
bnemec3) Whatever problem caused https://bugzilla.redhat.com/show_bug.cgi?id=1268451 is not necessarily addressed, although it's mitigated by the fact that the user would have to explicitly configure a broken vip.19:44
openstackbugzilla.redhat.com bug 1268451 in instack-undercloud "[RFE] Improve unspecified value for admin_vip in undercloud.conf" [Unspecified,Verified] - Assigned to bnemec19:44
fultonjrook: i guess we'll see what you get over time19:44
yfriedbnemec: I understand 1 and 219:45
EmilienMchem: why did you recheck https://review.openstack.org/#/c/438091 ? i think the error is valid19:45
EmilienMhttp://logs.openstack.org/91/438091/2/check/gate-tripleo-ci-centos-7-multinode-upgrades-nv/8ad40a5/console.html#_2017-02-27_18_57_13_86025719:45
yfriedbnemec: but do you understand that the solution to the original bug was to actually render VIPs unusable for explicit addresses? they only work for FQDN now19:45
*** sudipto has quit IRC19:46
*** sudipto_ has quit IRC19:46
bnemecyfried: Yes, but that was later fixed.19:47
bnemecFQDN should never have worked, but it happened to due to a bug in the validation.19:47
bnemecSo we ended up with a feature that depended on a bug.19:47
bnemecBut like I said, that's been addressed.  FQDN works fine now, which is why the opts were renamed _host19:47
openstackgerritJoe Talerico proposed openstack/puppet-tripleo master: Tuned should be configured properly  https://review.openstack.org/43769819:47
openstackgerritRajini Ram proposed openstack/tripleo-heat-templates stable/newton: Composable services support for Cinder Dell EMC PS Series  https://review.openstack.org/43558319:48
yfriedbnemec: yes, but addresses don't :)19:48
bnemecyfried: I'm fine with pursuing the automatic vip assignment.  It still doesn't address why https://bugzilla.redhat.com/show_bug.cgi?id=1268451 failed, but at least that specific case isn't regressed because we'll DTRT.19:50
openstackbugzilla.redhat.com bug 1268451 in instack-undercloud "[RFE] Improve unspecified value for admin_vip in undercloud.conf" [Unspecified,Verified] - Assigned to bnemec19:50
yfriedbnemec: ok19:52
yfriedbnemec: I'll try to do that later19:52
*** trown is now known as trown|brb19:56
*** ukalifon has quit IRC19:56
openstackgerritRajini Ram proposed openstack/puppet-tripleo stable/newton: Composable services for Dell EMC PS Series  https://review.openstack.org/43558719:56
*** trown|brb is now known as trown20:08
*** yamahata_ has joined #tripleo20:14
*** ansmith_ has joined #tripleo20:14
*** ansmith__ has joined #tripleo20:14
*** ansmith_ has quit IRC20:14
*** ansmith__ has quit IRC20:14
*** toure_ptg is now known as toure20:15
*** bogdando has joined #tripleo20:15
*** bogdando has quit IRC20:17
*** toure is now known as toure_ptg20:17
*** toure_ptg is now known as toure20:18
openstackgerritTim Rozet proposed openstack/tripleo-heat-templates master: Adds upgrade tasks for OpenDaylight services  https://review.openstack.org/43868920:21
openstackgerritTim Rozet proposed openstack/tripleo-heat-templates master: Adds upgrade tasks for OpenDaylight services  https://review.openstack.org/43868920:25
*** dbecker has quit IRC20:27
*** limao has joined #tripleo20:30
*** gbarros has joined #tripleo20:30
*** ccamacho has quit IRC20:33
openstackgerritTim Rozet proposed openstack/tripleo-heat-templates master: Adds upgrade tasks for OpenDaylight services  https://review.openstack.org/43868920:33
openstackgerritDan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh  https://review.openstack.org/43741120:34
*** limao has quit IRC20:35
openstackgerritTim Rozet proposed openstack/tripleo-puppet-elements master: Adds puppet-systemd  https://review.openstack.org/43860020:38
*** salmankhan has joined #tripleo20:38
openstackgerritwes hayutin proposed openstack/tripleo-quickstart master: break out general_config into config and topology  https://review.openstack.org/42545120:39
*** jprovazn has quit IRC20:41
*** eggmaster has quit IRC20:41
openstackgerritwes hayutin proposed openstack/tripleo-quickstart master: break out general_config into config and topology  https://review.openstack.org/42545120:42
openstackgerritJason E. Rist proposed openstack/tripleo-ui master: Adds tooltips to the deployment plan page  https://review.openstack.org/43772720:43
*** eck` is now known as eck`gone20:44
openstackgerritBen Kero proposed openstack-infra/tripleo-ci master: Switch quickstart config from scenario to multinode-settings  https://review.openstack.org/43869520:45
yfriedbnemec: could you plz point me to where the logic for the addresses is?20:49
*** abehl has quit IRC20:49
*** ccamacho has joined #tripleo20:50
*** liverpooler has quit IRC20:50
openstackgerritDan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh  https://review.openstack.org/43741120:51
*** dparkes has joined #tripleo20:52
*** jkilpatr has quit IRC20:52
*** rcernin has joined #tripleo20:52
*** Guest11443 is now known as mgagne20:57
*** mgagne has quit IRC20:57
*** mgagne has joined #tripleo20:57
bnemecyfried: There isn't any.  Currently they're read directly from the conf file.  However, we did something similar to this in https://review.openstack.org/#/c/320072/3/instack_undercloud/undercloud.py20:58
yfriedbnemec: tnx20:59
*** jkilpatr has joined #tripleo21:01
openstackgerritMerged openstack/tripleo-common master: Fix wrong args to update manager  https://review.openstack.org/43858321:01
openstackgerritMerged openstack/puppet-tripleo master: Add ceilometer polling agent profile  https://review.openstack.org/43044121:01
openstackgerritMerged openstack/tripleo-heat-templates stable/ocata: Install openstack-heat-agents on upgrade  https://review.openstack.org/43846621:02
openstackgerritMerged openstack/tripleo-heat-templates stable/ocata: Don't recalculate EndpointMap to get outputs  https://review.openstack.org/43846821:02
openstackgerritMerged openstack/tripleo-heat-templates stable/ocata: Update the Nova Endpoint information for Cinder  https://review.openstack.org/43856821:02
EmilienMbnemec: can you please create a blueprint for https://etherpad.openstack.org/p/tripleo-environment-generator ?21:08
*** lblanchard has quit IRC21:09
*** paramite has joined #tripleo21:12
openstackgerritBen Kero proposed openstack-infra/tripleo-ci master: Switch quickstart config from scenario to multinode-settings  https://review.openstack.org/43869521:16
*** jayg is now known as jayg|g0n321:18
*** eck`gone is now known as eck`21:22
*** sdake has joined #tripleo21:29
*** sdake_ has joined #tripleo21:33
openstackgerritDan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh  https://review.openstack.org/43741121:34
*** sdake has quit IRC21:34
*** dprince has quit IRC21:35
*** paramite has quit IRC21:39
*** paramite has joined #tripleo21:42
*** ansmith has quit IRC21:46
*** chlong has quit IRC21:50
*** dparkes has quit IRC21:51
*** trown is now known as trown|outtypewww21:53
*** mburned is now known as mburned_out22:03
*** sdake_ has quit IRC22:05
*** bfournie has quit IRC22:05
*** paramite has quit IRC22:08
*** ccamacho has quit IRC22:10
*** jmelvin has quit IRC22:11
*** shardy_afk has quit IRC22:23
openstackgerritBen Kero proposed openstack-infra/tripleo-ci master: Switch quickstart config from scenario to multinode-settings  https://review.openstack.org/43869522:23
*** thrash is now known as thrash|g0ne22:29
*** owalsh-afk has quit IRC22:31
*** panda is now known as panda|Zz22:32
*** sdake has joined #tripleo22:33
openstackgerritBrad P. Crochet proposed openstack/tripleo-common stable/ocata: Fix wrong args to update manager  https://review.openstack.org/43872522:33
*** toure is now known as toure|gone22:34
*** ckyriakidou has joined #tripleo22:39
openstackgerritRonelle Landy proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation  https://review.openstack.org/43743722:51
*** mburned_out is now known as mburned22:52
*** achadha has quit IRC22:55
*** jkilpatr has quit IRC22:56
*** achadha has joined #tripleo22:56
*** gfidente|afk has quit IRC22:59
*** rhallisey has quit IRC23:01
*** owalsh-afk has joined #tripleo23:01
openstackgerritMerged openstack/tripleo-heat-templates master: Deploy CI scenarios with Pacemaker  https://review.openstack.org/43828723:04
*** ntpttr_ has joined #tripleo23:05
*** owalsh has joined #tripleo23:06
*** owalsh-afk has quit IRC23:06
*** abregman has quit IRC23:08
openstackgerritMatthew Flusche proposed openstack/tripleo-heat-templates master: yum_update.sh - Use the yum parameter: check-update  https://review.openstack.org/43873023:09
openstackgerritwes hayutin proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation  https://review.openstack.org/43743723:11
*** saneax-_-|AFK is now known as saneax23:16
*** ckyriakidou has quit IRC23:22
*** owalsh_ has joined #tripleo23:22
*** owalsh has quit IRC23:23
*** gbarros has quit IRC23:28
openstackgerritBen Kero proposed openstack-infra/tripleo-ci master: Switch quickstart config from scenario to multinode-settings  https://review.openstack.org/43869523:30
*** dmarlin has quit IRC23:34
*** jkilpatr has joined #tripleo23:34
*** rcernin has quit IRC23:41
*** ansmith has joined #tripleo23:41
*** dougbtv_ has joined #tripleo23:41
*** dougbtv has quit IRC23:42
*** jcoufal has quit IRC23:45
*** pmannidi has joined #tripleo23:49
*** jcoufal has joined #tripleo23:51
openstackgerritMerged openstack/python-tripleoclient stable/ocata: Pass empty stack_fields to update manager  https://review.openstack.org/43653423:53
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Add support for $DO_BOOTSTRAP_SUBNODES  https://review.openstack.org/43743423:54
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Use bootstrap subnodes environment in CI  https://review.openstack.org/42588823:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Copy nodepool private key to ~/.ssh  https://review.openstack.org/43774323:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Add support for $OOOQ_TAGS  https://review.openstack.org/43743223:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Add generated nodepool key to authorized_keys  https://review.openstack.org/43774223:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Fix setting of node and node_private  https://review.openstack.org/43743323:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Pass $MTU env var to toci_vxlan_networking.sh  https://review.openstack.org/43774423:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Improved initial package handling  https://review.openstack.org/43743023:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Reinstall python-requests  https://review.openstack.org/43743123:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Don't hardcode jenkins user  https://review.openstack.org/43742923:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Don't output os-collect-config log to console  https://review.openstack.org/43810523:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Don't hardcode /home/jenkins  https://review.openstack.org/43671923:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Stop dstat at the end of a job  https://review.openstack.org/43874523:55
openstackgerritJames Slagle proposed openstack-infra/tripleo-ci master: Update pip in DLRN venv  https://review.openstack.org/43874623:55
*** yfried has quit IRC23:56
« Sunday, 2017-02-26 Index Tuesday, 2017-02-28 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!