*** patrickeast has quit IRC | 00:02 | |
*** NachoDuck has quit IRC | 00:02 | |
*** melwitt has quit IRC | 00:02 | |
*** Hazelesque has quit IRC | 00:02 | |
*** achadha has quit IRC | 00:03 | |
*** achadha has joined #tripleo | 00:04 | |
*** owalsh_ is now known as owalsh | 00:04 | |
*** Hazelesque has joined #tripleo | 00:04 | |
*** owalsh is now known as owalsh-afk | 00:04 | |
*** patrickeast has joined #tripleo | 00:06 | |
*** NachoDuck has joined #tripleo | 00:06 | |
*** ukalifon has joined #tripleo | 00:08 | |
*** jcoufal has quit IRC | 00:09 | |
*** melwitt has joined #tripleo | 00:09 | |
*** melwitt is now known as Guest68076 | 00:09 | |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Remove puppet-midonet https://review.openstack.org/438749 | 00:10 |
---|---|---|
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Add virtual_packages support to norpm provider https://review.openstack.org/437701 | 00:11 |
*** Guest68076 is now known as melwitt | 00:12 | |
*** salmankhan has quit IRC | 00:19 | |
*** dmacpher-afk has quit IRC | 00:20 | |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Enable object-expirer on Swift proxy profile https://review.openstack.org/437541 | 00:20 |
*** rajinir_ has quit IRC | 00:21 | |
*** rbrady has quit IRC | 00:28 | |
*** achadha_ has joined #tripleo | 00:29 | |
*** achadha has quit IRC | 00:32 | |
*** saneax is now known as saneax-_-|AFK | 00:43 | |
*** limao has joined #tripleo | 00:48 | |
*** gatuus has quit IRC | 00:51 | |
*** ukalifon has quit IRC | 00:54 | |
*** limao has quit IRC | 01:01 | |
*** fragatina has quit IRC | 01:01 | |
*** limao has joined #tripleo | 01:01 | |
*** alop has quit IRC | 01:03 | |
*** dujelly has quit IRC | 01:06 | |
openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh https://review.openstack.org/437411 | 01:10 |
*** fultonj has quit IRC | 01:14 | |
*** fultonj has joined #tripleo | 01:19 | |
*** ukalifon has joined #tripleo | 01:20 | |
*** ukalifon has quit IRC | 01:25 | |
*** ukalifon has joined #tripleo | 01:25 | |
*** fultonj has quit IRC | 01:28 | |
*** sdake_ has joined #tripleo | 01:30 | |
openstackgerrit | Merged openstack/diskimage-builder master: pip-and-virtualenv: also handle rhel distros https://review.openstack.org/433693 | 01:31 |
openstackgerrit | Merged openstack/diskimage-builder master: gentoo: do not manually clean /tmp https://review.openstack.org/436101 | 01:31 |
openstackgerrit | Merged openstack/diskimage-builder master: [suse] remove --no-confirm from zypper invocation https://review.openstack.org/436167 | 01:31 |
*** sdake has quit IRC | 01:33 | |
*** fragatina has joined #tripleo | 01:36 | |
*** sdake_ has quit IRC | 01:40 | |
*** rdopiera has quit IRC | 02:03 | |
*** rdopiera has joined #tripleo | 02:05 | |
*** bugzy has quit IRC | 02:07 | |
*** bfournie has joined #tripleo | 02:11 | |
*** ukalifon has quit IRC | 02:11 | |
*** achadha_ has quit IRC | 02:12 | |
*** achadha has joined #tripleo | 02:13 | |
*** achadha has quit IRC | 02:20 | |
*** achadha has joined #tripleo | 02:21 | |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates master: Adding congress datasource to pingtest scn001 https://review.openstack.org/438769 | 02:24 |
*** rlandy has quit IRC | 02:24 | |
*** bkopilov has quit IRC | 02:27 | |
*** fzdarsky_ has joined #tripleo | 02:33 | |
*** fzdarsky|afk has quit IRC | 02:34 | |
*** dougbtv has joined #tripleo | 02:37 | |
*** larstobi has quit IRC | 02:40 | |
*** larstobi has joined #tripleo | 02:42 | |
*** achadha has quit IRC | 02:44 | |
*** achadha has joined #tripleo | 02:46 | |
*** gbarros has joined #tripleo | 02:50 | |
*** achadha has quit IRC | 02:50 | |
*** iranzo has quit IRC | 02:53 | |
openstackgerrit | Merged openstack-infra/tripleo-ci master: Remove rdo-qemu-ev workaround https://review.openstack.org/425753 | 02:55 |
openstackgerrit | Merged openstack-infra/tripleo-ci master: Drop ceph workaround for mitaka https://review.openstack.org/425754 | 02:56 |
*** dmacpher has joined #tripleo | 03:04 | |
*** achadha has joined #tripleo | 03:12 | |
*** achadha has quit IRC | 03:17 | |
*** achadha_ has joined #tripleo | 03:17 | |
*** achadha_ has quit IRC | 03:22 | |
*** limao has quit IRC | 03:23 | |
*** tzumainn has quit IRC | 03:28 | |
*** dougbtv has quit IRC | 03:31 | |
*** achadha has joined #tripleo | 03:37 | |
*** achadha has quit IRC | 03:37 | |
*** limao has joined #tripleo | 03:37 | |
*** achadha has joined #tripleo | 03:37 | |
*** achadha_ has joined #tripleo | 03:38 | |
*** achadha has quit IRC | 03:39 | |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates master: Adding congress datasource to pingtest scn001 https://review.openstack.org/438769 | 03:42 |
*** limao has quit IRC | 03:42 | |
*** gkadam has joined #tripleo | 03:45 | |
*** gbarros has quit IRC | 03:48 | |
*** bkopilov has joined #tripleo | 03:48 | |
*** nyechiel has joined #tripleo | 03:52 | |
*** psahoo has joined #tripleo | 03:56 | |
*** nmathew has joined #tripleo | 03:56 | |
*** yamahata_ has quit IRC | 03:57 | |
*** fragatin_ has joined #tripleo | 03:59 | |
*** fragati__ has joined #tripleo | 04:01 | |
*** limao has joined #tripleo | 04:01 | |
*** fragatina has quit IRC | 04:03 | |
*** fragatin_ has quit IRC | 04:04 | |
*** fragati__ has quit IRC | 04:05 | |
*** limao has quit IRC | 04:16 | |
*** limao_ has joined #tripleo | 04:16 | |
*** saneax-_-|AFK is now known as saneax | 04:22 | |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist https://review.openstack.org/438657 | 04:25 |
*** dtrainor has joined #tripleo | 04:31 | |
EmilienM | it seems like https://github.com/openstack/swift/commit/1f36b5dd1617eb971327893635402529becd584d broke us but not 100% sure | 04:37 |
EmilienM | sshnaidm|afk: we need to add scenarios jobs in promotion pipeline asap imho | 04:38 |
*** Fenuks has joined #tripleo | 04:40 | |
*** sudipto has joined #tripleo | 04:40 | |
*** sudipto_ has joined #tripleo | 04:40 | |
*** sudipto_ has quit IRC | 04:41 | |
*** sudipto has quit IRC | 04:41 | |
*** sudipto has joined #tripleo | 04:43 | |
*** sudipto_ has joined #tripleo | 04:43 | |
openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: temprevert Swift Global EC Cluster Support https://review.openstack.org/438797 | 04:45 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: DNM - temprevert swift testing https://review.openstack.org/438798 | 04:47 |
*** jaganathan has joined #tripleo | 04:52 | |
*** nyechiel has quit IRC | 04:58 | |
*** sudipto_ has quit IRC | 04:58 | |
*** sudipto has quit IRC | 04:59 | |
EmilienM | https://bugs.launchpad.net/tripleo/+bug/1668493 | 04:59 |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 04:59 |
*** sdake has joined #tripleo | 05:03 | |
openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo master: Revert "Add httpchk for http services" https://review.openstack.org/438801 | 05:06 |
openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo master: Revert "Add httpchk for http services" https://review.openstack.org/438801 | 05:06 |
*** rickflare has joined #tripleo | 05:07 | |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: DNM - test a revert to fix scenario004 https://review.openstack.org/438798 | 05:07 |
*** sdake has quit IRC | 05:08 | |
*** sdake_ has joined #tripleo | 05:08 | |
*** ooolpbot has joined #tripleo | 05:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 05:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 05:10 |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 05:10 |
*** ooolpbot has quit IRC | 05:10 | |
*** bugzy has joined #tripleo | 05:17 | |
*** udesale has joined #tripleo | 05:19 | |
*** karthiks has joined #tripleo | 05:20 | |
*** fragatina has joined #tripleo | 05:20 | |
*** fragatina has quit IRC | 05:21 | |
*** fragatina has joined #tripleo | 05:21 | |
*** skramaja has joined #tripleo | 05:24 | |
*** suuuper has joined #tripleo | 05:30 | |
*** udesale__ has joined #tripleo | 05:33 | |
*** prateek has joined #tripleo | 05:34 | |
*** agurenko has joined #tripleo | 05:34 | |
*** udesale has quit IRC | 05:35 | |
*** agurenko has quit IRC | 05:36 | |
*** agurenko has joined #tripleo | 05:39 | |
*** jaosorior has joined #tripleo | 05:41 | |
*** bugzy has quit IRC | 05:43 | |
*** masco has joined #tripleo | 05:43 | |
*** sdake_ has quit IRC | 05:47 | |
*** dsariel has joined #tripleo | 05:47 | |
*** sdake has joined #tripleo | 05:47 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-tripleoclient master: Updated from global requirements https://review.openstack.org/438809 | 05:48 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates master: Configure SSL connection for MySQL client via client config file https://review.openstack.org/437364 | 05:49 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: Configure MySQL client SSL connections via the config file https://review.openstack.org/437365 | 05:49 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: Force MySQL users to use SSL if internal TLS is enabled https://review.openstack.org/426274 | 05:50 |
*** twi99y has quit IRC | 05:50 | |
*** bugzy has joined #tripleo | 05:55 | |
*** bugzy has quit IRC | 05:59 | |
*** pgadiya has joined #tripleo | 06:05 | |
openstackgerrit | Merged openstack/tripleo-quickstart-extras master: Stop building heat-agents image https://review.openstack.org/437546 | 06:08 |
*** sdake_ has joined #tripleo | 06:10 | |
*** abregman has joined #tripleo | 06:10 | |
*** fragatina has quit IRC | 06:12 | |
*** sdake has quit IRC | 06:13 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: Enable TLS in the internal network for RabbitMQ https://review.openstack.org/327482 | 06:19 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: Enable TLS in the internal network for RabbitMQ https://review.openstack.org/327482 | 06:19 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates master: Pass hieradata for internal TLS for RabbitMQ https://review.openstack.org/409143 | 06:20 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job with RabbitMQ https://review.openstack.org/426307 | 06:20 |
openstackgerrit | Saravanan KR proposed openstack/os-net-config master: Restart Openvswitch after adding DPDK ports https://review.openstack.org/434255 | 06:20 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job with RabbitMQ https://review.openstack.org/426307 | 06:21 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-heat-templates stable/ocata: Updated from global requirements https://review.openstack.org/436256 | 06:24 |
*** artom has quit IRC | 06:28 | |
*** fzdarsky_ is now known as fzdarsky | 06:35 | |
*** udesale__ has quit IRC | 06:35 | |
*** yamahata_ has joined #tripleo | 06:35 | |
*** ukalifon has joined #tripleo | 06:39 | |
*** artom has joined #tripleo | 06:42 | |
*** bugzy has joined #tripleo | 06:45 | |
*** paramite has joined #tripleo | 06:46 | |
*** agurenko has quit IRC | 06:48 | |
*** sudipto_ has joined #tripleo | 06:48 | |
*** agurenko has joined #tripleo | 06:48 | |
*** sudipto has joined #tripleo | 06:48 | |
*** iranzo has joined #tripleo | 06:50 | |
*** iranzo has joined #tripleo | 06:50 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates master: Switch keystone default provider to fernet https://review.openstack.org/432420 | 06:50 |
*** nyechiel has joined #tripleo | 06:52 | |
openstackgerrit | Merged openstack/puppet-tripleo master: Ironic inspector support https://review.openstack.org/436143 | 06:52 |
openstackgerrit | Merged openstack/puppet-tripleo master: Default neutron dhcp_agents_per_network to number of agents https://review.openstack.org/438545 | 06:53 |
*** jprovazn has joined #tripleo | 06:53 | |
openstackgerrit | Merged openstack/python-tripleoclient stable/ocata: Make overcloud image upload honor --image-path https://review.openstack.org/434975 | 06:53 |
*** saneax is now known as saneax-_-|AFK | 06:53 | |
*** apetrich has quit IRC | 06:55 | |
*** saneax-_-|AFK is now known as saneax | 06:56 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: Configure MySQL client SSL connections via the config file https://review.openstack.org/437365 | 06:57 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: Force MySQL users to use SSL if internal TLS is enabled https://review.openstack.org/426274 | 06:57 |
*** achadha_ has quit IRC | 06:58 | |
*** achadha has joined #tripleo | 06:58 | |
*** bugzy has quit IRC | 06:59 | |
*** bugzy has joined #tripleo | 06:59 | |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 06:59 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart master: Set network isolation type to single-nic by default https://review.openstack.org/437816 | 07:02 |
openstackgerrit | Merged openstack/instack-undercloud master: Purge /var/lib/os-collect-config https://review.openstack.org/432830 | 07:03 |
openstackgerrit | Merged openstack/tripleo-common stable/ocata: Fix wrong args to update manager https://review.openstack.org/438725 | 07:03 |
openstackgerrit | Merged openstack/tripleo-common stable/ocata: Just set the breakpoints on args passed to heat update https://review.openstack.org/436536 | 07:03 |
*** dparkes has joined #tripleo | 07:04 | |
openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: Support deploy args in OVB jobs https://review.openstack.org/437531 | 07:04 |
*** dmacpher has quit IRC | 07:05 | |
*** karthiks has quit IRC | 07:05 | |
*** artom has quit IRC | 07:09 | |
*** ooolpbot has joined #tripleo | 07:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 07:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 07:10 |
*** ooolpbot has quit IRC | 07:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 07:10 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 07:11 |
*** rcernin has joined #tripleo | 07:11 | |
*** apetrich has joined #tripleo | 07:16 | |
*** karthiks has joined #tripleo | 07:18 | |
bandini | morning | 07:24 |
*** yfried has joined #tripleo | 07:27 | |
*** tesseract has joined #tripleo | 07:34 | |
*** gmann has quit IRC | 07:35 | |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Adds http proxy support for registering RHEL overcloud nodes https://review.openstack.org/437247 | 07:36 |
*** gmann has joined #tripleo | 07:37 | |
*** jpena|away is now known as jpena | 07:41 | |
*** fzdarsky has quit IRC | 07:43 | |
*** fzdarsky has joined #tripleo | 07:44 | |
*** pgadiya has quit IRC | 07:48 | |
*** abehl has joined #tripleo | 07:48 | |
*** yprokule has joined #tripleo | 07:53 | |
*** jpena is now known as jpena|away | 07:54 | |
*** shardy has joined #tripleo | 07:56 | |
*** sdake_ is now known as sdake | 07:56 | |
openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart-extras master: GATE CHECK https://review.openstack.org/430277 | 08:00 |
chem | matbu: marios do we have the openvswitch nopostun trick ? I cannot find it from https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/tripleo-packages.yaml | 08:00 |
marios | chem: o/ no right now we don't have it in the controller (ansible) upgrade steps at all | 08:00 |
marios | chem: but matbu wanted to add it... though it was a while ago and it could be that latest fixes the issue for him too... (see https://review.openstack.org/#/c/434346/ ) | 08:01 |
chem | matbu mariso I'm trying to answer the comment from beagles https://review.openstack.org/#/c/438090/3/puppet/services/neutron-ovs-agent.yaml | 08:01 |
chem | marios: ^ | 08:01 |
*** jpena|away is now known as jpena | 08:01 | |
marios | chem: and in https://review.openstack.org/#/c/436990/ i want to remove it from the tripleo_upgrade_node.sh alltogether too (looking) | 08:01 |
*** pgadiya has joined #tripleo | 08:02 | |
*** gchamoul|afk is now known as gchamoul | 08:02 | |
marios | chem: ah so i think beagles is referring to the _other_ openvswitch issue we had for 9-10 i mean | 08:02 |
marios | chem: the no postun was one issue, the secure mode on bridegs was another | 08:02 |
*** ramishra has quit IRC | 08:03 | |
marios | chem: lemme comment there | 08:03 |
chem | marios: aren't they somehow related ... waiting for comment :) | 08:03 |
*** ramishra has joined #tripleo | 08:05 | |
matbu | chem: you hit the ovs issue ? | 08:06 |
*** aufi has joined #tripleo | 08:06 | |
matbu | chem: marios i answer to your comment about that | 08:06 |
matbu | marios: chem in the latest newton package, ovs is 2.6, but i'm not if it would be the case in downstream | 08:07 |
matbu | in all cases, this special upgrade case would be safe and usefull i think | 08:07 |
*** ccamacho has joined #tripleo | 08:07 | |
chem | matbu: yep it will be downstream as well; nope I don't hit the issue because I didn't run it | 08:08 |
*** stendulker has joined #tripleo | 08:09 | |
*** ooolpbot has joined #tripleo | 08:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 08:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 08:10 |
*** ooolpbot has quit IRC | 08:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 08:10 |
*** pgadiya has quit IRC | 08:10 | |
marios | matbu: no unfortunately it is not safe | 08:14 |
marios | chem: ^^ | 08:14 |
marios | matbu: at least when i tried it on the compute it killed the network | 08:14 |
marios | matbu: see comments at https://review.openstack.org/#/c/436990/ | 08:15 |
marios | matbu: and i'll leave a new acomment at your https://review.openstack.org/#/c/434346/6/puppet/services/tripleo-packages.yaml maybe wasn't clear there but it caused problems for me | 08:15 |
marios | matbu: for controllers i don't do anything special and have the ansible yum update on step3 to pull in the latest openvswitch and there are no explosions | 08:16 |
marios | 17:42 < marios> beagles: it used to be 'use workaround or die' but today was 'use workaround and die' (from comment at https://review.openstack.org/#/c/436990/ ) | 08:17 |
*** pcaruana has joined #tripleo | 08:17 | |
matbu | marios: you mean, with my review, you loose the network on the compute ? | 08:18 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add upgrade support for Horizon https://review.openstack.org/437545 | 08:18 |
marios | matbu: no your review doesn't affect the compute | 08:18 |
matbu | marios: yep right | 08:19 |
marios | matbu: i did *not* have your review applied and the yum update went fine (and other ansible and puppet tasks for controller/ceph) | 08:19 |
* matbu loose his brain during pto :) | 08:19 | |
marios | matbu: and i got openvswitch-2.6.1-8.git20161206.el7fdb.x86_64 | 08:19 |
marios | matbu: on my controller ^ (osp11 puddle) | 08:19 |
matbu | marios: and you are working on downstream | 08:19 |
matbu | yep ack | 08:19 |
marios | matbu: then when i tried to upgrade compute which DOES have the workaround , explosions | 08:19 |
*** florianf has joined #tripleo | 08:19 | |
marios | matbu: well actually more like radio silence not explosions | 08:19 |
marios | matbu: so i posted https://review.openstack.org/#/c/436990/ to remove | 08:20 |
marios | matbu: and i added a comment on your one at https://review.openstack.org/#/c/434346/6 to suggest that it could in fact have th eopposite effect and we should consider nuking it from orbit | 08:20 |
matbu | marios: when i was testing it, i had to upgrade python-ovs in the same way, if not the network was shuting down during the upgrade of ovs package | 08:20 |
*** athomas has joined #tripleo | 08:21 | |
marios | matbu: ack, so i know the ovs guys were working on some new package version we *could* upgrade to (i.e. it was broken before and we had to versionlock) | 08:21 |
marios | matbu: so perhaps the latest bothup and d/stream is all good now? | 08:21 |
matbu | it looks like, i didn't hit it yesterday | 08:22 |
matbu | on upstream | 08:22 |
*** flepied has quit IRC | 08:24 | |
*** yfried has quit IRC | 08:25 | |
*** shardy has quit IRC | 08:25 | |
*** pgadiya has joined #tripleo | 08:26 | |
*** cylopez has joined #tripleo | 08:33 | |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add auditd upgrate steps https://review.openstack.org/438289 | 08:42 |
*** bvandenh has joined #tripleo | 08:43 | |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add etcd composable upgrade steps https://review.openstack.org/438288 | 08:44 |
*** karimb has joined #tripleo | 08:48 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-ui stable/ocata: Imported Translations from Zanata https://review.openstack.org/438855 | 08:48 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add zaqar service for composable upgrade https://review.openstack.org/435906 | 08:48 |
*** dj_ has joined #tripleo | 08:51 | |
*** leanderthal|afk is now known as leanderthal | 08:53 | |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-heat-templates master: DON'T MERGE: testin oooq scenario https://review.openstack.org/438858 | 08:53 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add support for upgrading ec2-api https://review.openstack.org/425285 | 08:53 |
*** tosky has joined #tripleo | 08:53 | |
*** owalsh-afk is now known as owalsh | 08:57 | |
*** jpich has joined #tripleo | 08:58 | |
openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: Add option for scenarios jobs promotion https://review.openstack.org/425819 | 08:59 |
*** zzzeek has quit IRC | 09:00 | |
*** pgadiya has quit IRC | 09:01 | |
*** jlinkes has joined #tripleo | 09:02 | |
*** akrivoka has joined #tripleo | 09:04 | |
*** aufi has quit IRC | 09:04 | |
*** flepied has joined #tripleo | 09:04 | |
*** achadha has quit IRC | 09:05 | |
openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: DONT REVIEW: test ceilometer patch with periodic https://review.openstack.org/437008 | 09:06 |
*** achadha has joined #tripleo | 09:08 | |
*** pgadiya has joined #tripleo | 09:08 | |
*** ooolpbot has joined #tripleo | 09:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 09:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 09:10 |
*** ooolpbot has quit IRC | 09:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 09:10 |
*** zzzeek has joined #tripleo | 09:12 | |
*** bvandenh has quit IRC | 09:13 | |
*** achadha has quit IRC | 09:13 | |
*** sdake has quit IRC | 09:13 | |
*** cylopez has quit IRC | 09:14 | |
*** dbecker has joined #tripleo | 09:15 | |
*** udesale has joined #tripleo | 09:17 | |
*** dsariel has quit IRC | 09:18 | |
*** aufi has joined #tripleo | 09:19 | |
*** ramishra has quit IRC | 09:23 | |
*** ramishra has joined #tripleo | 09:24 | |
*** zoli|gone is now known as zoli | 09:26 | |
openstackgerrit | Chandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow https://review.openstack.org/431916 | 09:27 |
*** pgadiya has left #tripleo | 09:27 | |
*** gfidente has joined #tripleo | 09:27 | |
*** gfidente has quit IRC | 09:27 | |
*** gfidente has joined #tripleo | 09:27 | |
*** aufi has quit IRC | 09:27 | |
*** apetrich has quit IRC | 09:29 | |
*** bvandenh has joined #tripleo | 09:30 | |
*** gkadam is now known as gkadam-afk | 09:33 | |
chandankumar | sshnaidm|afk: \o/ | 09:35 |
sshnaidm|afk | chandankumar, passes? | 09:35 |
*** bogdando has joined #tripleo | 09:37 | |
*** derekh has joined #tripleo | 09:37 | |
chandankumar | sshnaidm|afk: no i added testr to get more information if it fails | 09:37 |
chandankumar | sshnaidm|afk: https://review.openstack.org/#/c/431916/ | 09:37 |
chandankumar | with some more modification | 09:37 |
*** apetrich has joined #tripleo | 09:38 | |
*** skramaja has quit IRC | 09:39 | |
*** jaosorior has quit IRC | 09:39 | |
*** aufi has joined #tripleo | 09:39 | |
*** jaosorior has joined #tripleo | 09:40 | |
*** skramaja has joined #tripleo | 09:41 | |
*** sshnaidm|afk is now known as sshnaidm | 09:41 | |
openstackgerrit | Arx Cruz proposed openstack/tripleo-quickstart-extras master: Updating skip file mechanism https://review.openstack.org/438465 | 09:41 |
*** mcornea has joined #tripleo | 09:42 | |
sshnaidm | chandankumar, ok, let's see.. | 09:43 |
*** florianf has quit IRC | 09:44 | |
*** achadha has joined #tripleo | 09:44 | |
*** yamahata_ has quit IRC | 09:45 | |
*** cylopez has joined #tripleo | 09:48 | |
*** lucas-afk is now known as lucasagomes | 09:50 | |
*** achadha has quit IRC | 09:53 | |
*** skramaja has quit IRC | 09:57 | |
*** bogdando has quit IRC | 09:58 | |
*** artom has joined #tripleo | 10:01 | |
*** ramishra has quit IRC | 10:01 | |
*** artom has quit IRC | 10:01 | |
*** artom has joined #tripleo | 10:02 | |
*** ckyriakidou has joined #tripleo | 10:02 | |
*** ramishra has joined #tripleo | 10:03 | |
*** bvandenh has quit IRC | 10:04 | |
*** hewbrocca_afk is now known as hewbrocca | 10:05 | |
*** sdake has joined #tripleo | 10:05 | |
*** skramaja has joined #tripleo | 10:06 | |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add support for upgrading ec2-api https://review.openstack.org/425285 | 10:07 |
*** pmannidi has quit IRC | 10:07 | |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add zaqar service for composable upgrade https://review.openstack.org/435906 | 10:08 |
*** ooolpbot has joined #tripleo | 10:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 10:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 10:10 |
*** ooolpbot has quit IRC | 10:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 10:10 |
b00tcat | Hi, if I want to deploy an overcloud with just "openstack overcloud deploy --templates", what compute.yaml and what controller.yaml files are used? I couldn't figure it out from the tripleo-heat-templates repo | 10:16 |
jaosorior | b00tcat: puppet/compute-role.yaml and puppet/controller-role.yaml | 10:17 |
jaosorior | b00tcat: or, do you mean where are the list of services defined? | 10:17 |
jaosorior | b00tcat: cause the defaults are in role_data.yaml | 10:17 |
jaosorior | * roles_data.yaml | 10:18 |
b00tcat | jaosorior: no I meant that thanks :-) | 10:18 |
b00tcat | jaosorior: oh right, I forgot | 10:18 |
*** tosky has quit IRC | 10:21 | |
*** tosky has joined #tripleo | 10:22 | |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-heat-templates master: DON'T MERGE: testin oooq scenario https://review.openstack.org/438858 | 10:24 |
openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart master: Add composable deployment config files and mixed release https://review.openstack.org/410831 | 10:25 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-docs master: Update docs for HA by default. https://review.openstack.org/418756 | 10:28 |
*** nyechiel has quit IRC | 10:28 | |
*** nyechiel has joined #tripleo | 10:31 | |
b00tcat | also, does anybody know which network config at https://github.com/openstack/tripleo-heat-templates/tree/master/network/ is picked up by default? | 10:32 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart master: Set network isolation type to single-nic by default https://review.openstack.org/437816 | 10:32 |
marios | b00tcat: i think single nic vlans is the pretty standard dev one at least is what i use https://github.com/openstack/tripleo-heat-templates/tree/master/network/config/single-nic-vlans | 10:33 |
marios | b00tcat: (that config ^^^ is applied/referred to from the corresponding env file @ https://github.com/openstack/tripleo-heat-templates/blob/master/environments/net-single-nic-with-vlans.yaml ) hope it helps | 10:35 |
b00tcat | marios: yeah it's just that I never explicitly used one or the other, and I was wondering which one is used by default (or if it's used dynamically depending on the ironic introspection or something) | 10:37 |
b00tcat | thanks marios | 10:37 |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Deploy versionless keystone endpoints (for keystone only) https://review.openstack.org/438620 | 10:39 |
*** sdake has quit IRC | 10:42 | |
*** udesale has quit IRC | 10:43 | |
*** shardy has joined #tripleo | 10:45 | |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add upgrade task for panko api https://review.openstack.org/438298 | 10:46 |
openstackgerrit | Merged openstack/instack-undercloud master: Use Swift as a Zaqar backend. https://review.openstack.org/434727 | 10:46 |
openstackgerrit | Merged openstack/tripleo-ui stable/ocata: Imported Translations from Zanata https://review.openstack.org/438855 | 10:46 |
*** dtantsur|afk is now known as dtantsur | 10:47 | |
jaosorior | bandini, ccamacho, marios: if you have time could you review these? | 10:49 |
jaosorior | https://review.openstack.org/#/q/status:open+topic:mysql-ssl | 10:49 |
jaosorior | bandini: I go around the configparser issue you reported on github by using ssl=1 instead of just ssl. And recently added support for pymysql for the configuring the ssl attributes via the config file https://github.com/PyMySQL/PyMySQL/commit/557d0d90a5305bb7b3f49aafb83425ef5d1b4490 | 10:51 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Put service stop at step1 and quiesce at step2. https://review.openstack.org/438090 | 10:52 |
bandini | jaosorior: I saw it got merged, nice work ;) | 10:52 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Put service stop at step1 and quiesce at step2. https://review.openstack.org/438090 | 10:52 |
jaosorior | bandini: I'm very impressed by the maintainer, really fast and efficient folk | 10:53 |
bandini | jaosorior: he is very fast (although a little too opinionated at times if you ask me ;) | 10:53 |
jaosorior | bandini: hahaha I noticed | 10:54 |
bandini | jaosorior: we will need to make sure your commit is in the pymysql package in rdo first? | 10:54 |
jaosorior | bandini: well, having those attributes in the configuration doesn't hurt. This one is the one that's blocked until we get the commit in rdo https://review.openstack.org/#/c/426274/ | 10:54 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Make sure we don't stop service when pacemaker is handling them. https://review.openstack.org/438091 | 10:55 |
bandini | jaosorior: ack, so until we have fixed packages the tls stuff just won't work | 10:55 |
jaosorior | bandini: right, it won't really be using SSL for that | 10:55 |
jaosorior | just adding it to the configuration | 10:55 |
jaosorior | then it will finally get read when that commit lands | 10:55 |
bandini | ack got it | 10:56 |
jaosorior | ccamacho: why the -1? | 10:56 |
bandini | jaosorior: will look at it right after this call I am in | 10:56 |
jaosorior | ccamacho: : I really don't think the order makes a difference in this case | 10:56 |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Write out a json file containing container startup info and create tool to use it. https://review.openstack.org/431746 | 10:58 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-tripleoclient stable/mitaka: Updated from global requirements https://review.openstack.org/438912 | 10:59 |
*** limao_ has quit IRC | 10:59 | |
openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart master: break out general_config into config and topology https://review.openstack.org/425451 | 11:01 |
shardy | bandini: Hey I added a question/comment on https://review.openstack.org/#/c/438091 | 11:03 |
shardy | can you clarify why we need pacemaker upgrade logic in the non-pacemaker service templates? | 11:03 |
shardy | I was expecting that to go into e.g puppet/services/pacemaker/cinder-volume.yaml | 11:03 |
*** yfried has joined #tripleo | 11:03 | |
openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart master: break out general_config into config and topology https://review.openstack.org/425451 | 11:04 |
bandini | shardy: I am not sure tbh, I thought it was due to the way we include stuff | 11:05 |
bandini | I'll need to look at it a bit more in detail | 11:05 |
shardy | bandini: yeah puppet/services/pacemaker/cinder-volume.yaml references ../cinder-volume.yaml but it doesn't inherit any upgrade_tasks | 11:06 |
shardy | chem: ^^ perhaps you can clarify if you've been testing this | 11:06 |
openstackgerrit | Jiri Tomasek proposed openstack/tripleo-specs master: Deployment plan management changes spec https://review.openstack.org/438918 | 11:06 |
bandini | shardy: ack, then you are correct I believe | 11:06 |
jtomasek | d0ugal: ^^ | 11:07 |
shardy | environments/puppet-pacemaker.yaml still references puppet/services/pacemaker so I think we probably need to copy the upgrade tasks (or inherit from the base service template and append the pacemaker specific tasks e.g via a list_join | 11:07 |
shardy | unless I'm missing something :) | 11:07 |
*** prateek_ has joined #tripleo | 11:07 | |
jtomasek | d0ugal: if you're ok with the spec name, I'll create a general blueprint which should be parent of subsequent ones which we define | 11:08 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 11:08 |
*** gkadam-afk is now known as gkadam | 11:08 | |
chem | shardy: bandini nope, it's not tested, and I may be missing something, it was more of a reminder to make sure we look into it | 11:10 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 11:10 |
*** ooolpbot has joined #tripleo | 11:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 11:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 11:10 |
*** ooolpbot has quit IRC | 11:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 11:10 |
ccamacho | hey jaosorior, ack, I thought base on that here https://review.openstack.org/#/c/437364/3/puppet/services/database/mysql-client.yaml you are setting "tripleo::profile::base::database::mysql::client::enable_ssl" which is a new parameter introduced in https://review.openstack.org/#/c/437365/5/manifests/profile/base/database/mysql/client.pp the THT patch should depends on the puppet-tripleo one | 11:10 |
marios | jaosorior: will do but adding to list man have a few to look at before i get there | 11:10 |
openstackgerrit | Ana Krivokapic proposed openstack/tripleo-ui master: Add plan export https://review.openstack.org/437676 | 11:11 |
shardy | jtomasek: hey, can you provide me with a paste of the output from the UI parameter flattening code? | 11:11 |
shardy | or point me at any tests I can replicate? | 11:11 |
shardy | I was going to try adding a mistral workflow/action to do the same as we discussed last week | 11:11 |
d0ugal | jtomasek: yeah, the same seems fine - I don't have strong opinions about spec names :) | 11:11 |
d0ugal | jtomasek: I'll read through it all shortly. | 11:11 |
jtomasek | d0ugal: cool, thanks | 11:11 |
shardy | chem: Ok np, I think we may have to move the tasks around then, let me know if I can help with that (I can push another revision to that review if you like showing how I think it needs to work) | 11:11 |
jtomasek | shardy: I think I sent it to you when you asked at PTG, did you get the emails from me? | 11:12 |
jtomasek | shardy: resources: http://paste.openstack.org/show/600292/ | 11:13 |
jtomasek | parameters: http://paste.openstack.org/show/600293/ | 11:13 |
shardy | jtomasek: Oh, sorry yes you did and I missed it, apologies | 11:13 |
shardy | thanks! | 11:13 |
*** prateek has quit IRC | 11:14 | |
chem | shardy: that may be simpler in order to get what you mean. But currently the pacemaker/*.yaml get their upgrade from the non-pacemaker service conterpart right ? so the check is needed. If I understand well you want to add upgrade task to the pacemaker resources. But then it makes things messy, no ? | 11:15 |
*** prateek has joined #tripleo | 11:15 | |
shardy | chem: currently we don't expose any upgrade_tasks at all from the pacemaker service templates: | 11:15 |
*** gchamoul is now known as gchamoul|afk | 11:16 | |
shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/pacemaker/cinder-volume.yaml#L46 | 11:16 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 11:16 |
shardy | chem: I'm suggesting we "inherit" from the base template (similar to the config_settings on line 41) | 11:16 |
shardy | then just add the pacemaker specific tasks | 11:16 |
openstackgerrit | Merged openstack/python-tripleoclient master: Updated from global requirements https://review.openstack.org/438809 | 11:16 |
chem | shardy: this https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/pacemaker/cinder-volume.yaml#L25 doesn't make it happen ? | 11:17 |
shardy | chem: AFAICT the templates you modified aren't used for upgrade in the pacemaker case, but I may be missing something | 11:17 |
shardy | chem: No, because we don't reference that resource in the outputs (like we do for config_settings and some other keys) | 11:17 |
*** stendulker has quit IRC | 11:18 | |
*** zoli is now known as zoli|lunch | 11:19 | |
*** jaganathan is now known as jaganathan_afk | 11:19 | |
jaosorior | ccamacho: well, I could do it that way, I honestly don't think it makes a difference. But if you really think I should change the dependencies on the commits, I could | 11:19 |
chem | shardy: oki, I think I get, we need to be explicit for all outputs ... let me retry my patch then | 11:20 |
shardy | chem: you can check on any pacemaker enabled deployment by doing something like: | 11:20 |
shardy | openstack stack resource show overcloud ControllerServiceChain --format json | jq ".attributes.role_data.upgrade_tasks" | 11:21 |
lhinds | do we have any need for Tftp Server on the o'cloud. I know we need it on the undercloud for ironic | 11:21 |
shardy | then you can see for sure which tasks are included | 11:21 |
ccamacho | jaosorior, no worries just take it into account and if by any reason you need to push a new patchset you can do it :) | 11:21 |
*** tobias_fiberdata has joined #tripleo | 11:21 | |
jaosorior | ccamacho: ok, will do | 11:22 |
chem | shardy: hum .. in a upgrade_task I would need the stackrc available on the node to do that | 11:22 |
chem | shardy: but I think the patch is not needed at all if the upgrade task is not "inherited" automatically | 11:22 |
shardy | chem: No I mean for debugging, on the undercloud | 11:22 |
openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart-extras master: pass the topology of the overcloud w/ topology vs extra-args https://review.openstack.org/425453 | 11:23 |
shardy | then you can grep for your changes and prove what I'm saying (or show that I'm wrong ;) | 11:23 |
shardy | chem: Yeah if I'm correct then we do nothing at all for those services when the puppet-pacemaker environment is enabled | 11:23 |
*** tobias-fiberdata has joined #tripleo | 11:24 | |
chem | shardy: ah, yeah correct which is not expected :) | 11:25 |
chem | shardy: oki I'll do another pass on this | 11:25 |
marios | shardy: chem not following all the conversation but for tha tcase the ansible-pacemaker module is in play (from https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/pacemaker.yaml#L141 ) | 11:25 |
shardy | chem: ack, thanks! | 11:25 |
marios | shardy: chem that case being the pacemaker env | 11:26 |
shardy | marios: yup, the discussion was about https://review.openstack.org/#/c/438091/ which I think changes templates which aren't used in the pacemaker case | 11:26 |
shardy | think we have it figured out now :) | 11:26 |
*** tobias_fiberdata has quit IRC | 11:27 | |
marios | shardy: yeah i caught the start of that and was looking at the other reviews in the topic branch (i like the check for enabled first... i guess it covers the case for things that weren't default before but now are for example so you'd get the upgrade tasks but not necessarily already have the service) | 11:27 |
marios | shardy: i was away yesterday (I guess friday most people were travelling, i was away then too) | 11:27 |
chem | marios: (shardy) the problem is that if one of the pacemaker controlled resource need something else beside restart we won't have it | 11:28 |
shardy | marios: cool - yeah the check before validate thing is due to CI - on upgrade some scenarios add services, so we have to handle that | 11:28 |
chem | marios: that make things inconsistent | 11:28 |
openstackgerrit | Saravanan KR proposed openstack/tripleo-heat-templates master: Disable exit on error for pacemaker commands for update flow https://review.openstack.org/438495 | 11:28 |
marios | chem: yeah we dn't have any changes to the cluster resource definitions... don't think we needed them here (is that what you mean?0 | 11:28 |
*** adarazs is now known as adarazs_lunch | 11:29 | |
marios | shardy: some docs landed :) https://docs.openstack.org/developer/tripleo-docs/post_deployment/upgrade.html#upgrading-the-overcloud-to-ocata-and-beyond | 11:29 |
shardy | marios: nice! :) | 11:29 |
marios | shardy: yeah spent a couple hours last week when it was a bit quieter around here to get something out as a first pass we can improve but at least there is a base | 11:30 |
*** salmankhan has joined #tripleo | 11:30 | |
openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart master: break out general_config into config and topology https://review.openstack.org/425451 | 11:31 |
*** sudipto has quit IRC | 11:32 | |
*** sudipto_ has quit IRC | 11:32 | |
openstackgerrit | Saravanan KR proposed openstack/os-net-config master: Restart Openvswitch after adding DPDK ports https://review.openstack.org/434255 | 11:32 |
skramaja | shardy: can you approve this backport - https://review.openstack.org/#/c/433496/? | 11:35 |
jaosorior | therve: saw this in the submission http://logs.openstack.org/35/394835/17/check/gate-puppet-heat-puppet-beaker-rspec-centos-7/410cee6/console.html#_2017-02-27_15_56_49_738363 any idea if there's anything else tahn heat-api-cfn attempting to use port 8000? | 11:36 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 11:37 |
jaosorior | therve: also, one thing is that in puppet-openstack-integration, keystone is using 127.0.0.1 while the heat services attempt to use 0.0.0.0 http://logs.openstack.org/35/394835/17/check/gate-puppet-heat-puppet-beaker-rspec-centos-7/410cee6/logs/apache_config/ | 11:38 |
*** nyechiel has quit IRC | 11:39 | |
*** ckyriakidou has quit IRC | 11:40 | |
*** nmathew has quit IRC | 11:41 | |
*** jkilpatr has quit IRC | 11:41 | |
*** jaganathan_afk is now known as jaganathan | 11:43 | |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud master: Deploy heat APIs over httpd https://review.openstack.org/394837 | 11:46 |
*** karthiks has quit IRC | 11:47 | |
*** suuuper has quit IRC | 11:48 | |
*** tobias-fiberdata has quit IRC | 11:48 | |
*** cylopez has quit IRC | 11:52 | |
*** athomas has quit IRC | 11:53 | |
*** tosky_ has joined #tripleo | 11:57 | |
*** artom has quit IRC | 11:58 | |
therve | jaosorior, the permission denied is selinux ? | 11:58 |
marios | gerrit a bit slow for others too? | 11:59 |
*** yfried has quit IRC | 11:59 | |
*** tosky has quit IRC | 11:59 | |
*** artom has joined #tripleo | 11:59 | |
therve | jaosorior, I don't understand that other comment | 11:59 |
chem | marios: yep | 11:59 |
*** athomas has joined #tripleo | 12:00 | |
marios | thanks chem | 12:00 |
*** igorbelikov_ has quit IRC | 12:01 | |
*** pkovar has joined #tripleo | 12:02 | |
*** igorbelikov_ has joined #tripleo | 12:03 | |
*** akrivoka has quit IRC | 12:03 | |
chandankumar | sshnaidm: Hello | 12:04 |
*** karthiks has joined #tripleo | 12:04 | |
chandankumar | shardy: mitaka ironic and neutron tempest plugin has some isues http://logs.openstack.org/23/436923/3/check-tripleo/gate-tripleo-ci-centos-7-ovb-nonha-mitaka/653e262/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz | 12:04 |
chandankumar | sorry shardy | 12:05 |
chandankumar | sshnaidm: ^^ | 12:05 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 12:06 |
*** tosky_ is now known as tosky | 12:06 | |
*** pblaho has quit IRC | 12:07 | |
*** ansmith has quit IRC | 12:07 | |
*** ckyriakidou has joined #tripleo | 12:07 | |
*** bkopilov has quit IRC | 12:07 | |
sshnaidm | chandankumar, I see.. | 12:08 |
sshnaidm | chandankumar, it's with packages, right? | 12:08 |
chandankumar | sshnaidm: yes | 12:08 |
chandankumar | sshnaidm: the plugin is installed through package by it got failed to import | 12:09 |
sshnaidm | chandankumar, I'd ask dmellado on #rdo, maybe he knows | 12:09 |
sshnaidm | chandankumar, maybe another dependencies should be installed | 12:09 |
*** dj_ has quit IRC | 12:09 | |
*** ooolpbot has joined #tripleo | 12:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 12:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 12:10 |
*** ooolpbot has quit IRC | 12:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 12:10 |
sshnaidm | chandankumar, if it never worked, we won't support tempest with packages for mitaka | 12:10 |
chandankumar | sshnaidm: i need to look in to package, Daniel is not longer looking into tempest stuff or we can poke ironic and neutron guys | 12:10 |
chandankumar | i am waiting for other jobs to get completed | 12:11 |
sshnaidm | chandankumar, I'm almost sure it's some packaging issue.. and who is tempest folks now? | 12:11 |
*** fungi has quit IRC | 12:11 | |
bandini | matbu, chem: appreciate your thoughts on https://review.gerrithub.io/350823 | 12:11 |
sshnaidm | chandankumar, if it doesn't go anywhere, just disable packages option for mitaka, we'll continue to use venv for it | 12:11 |
chandankumar | sshnaidm: me and mkopec with the help of tosky will be looking into tempest and respective tempest plugin will be looked by their dfg | 12:12 |
openstackgerrit | Flavio Percoco proposed openstack/tripleo-quickstart-extras master: Use openstack deploy to install the undercloud https://review.openstack.org/419040 | 12:12 |
*** jkilpatr has joined #tripleo | 12:13 | |
sshnaidm | chandankumar, oh, then maybe it's worth to check with these dgfs | 12:13 |
chandankumar | yup | 12:13 |
jaosorior | therve: the second comment was about the possibility of it having to do with the specific IP. But I tried it out and it didn't help | 12:13 |
therve | jaosorior, Ah, ok, for the vhost | 12:13 |
jaosorior | therve: no, I'm not sure if it's the selinux issue. It's been months since I worked on this. and was doing so in my free time | 12:13 |
therve | jaosorior, https://bugzilla.redhat.com/show_bug.cgi?id=1396553 seems pretty convincing :) | 12:13 |
openstack | bugzilla.redhat.com bug 1396553 in openstack-selinux "apache unable to listen on port 8000 to front heat-api-cfn" [Unspecified,New] - Assigned to lhh | 12:13 |
jaosorior | anyway, I'm back to being confused about the error :/ | 12:14 |
tosky | chandankumar, sshnaidm: the suggestions of setting venv for mitaka for now it's not bad; we can have the main change (which brings benefit for the more recent versions) now and fix it later | 12:14 |
tosky | (or forget about it when mitaka is EOL :P ) | 12:14 |
sshnaidm | chandankumar, tosky agree, anyway we get rid off it soon | 12:15 |
jaosorior | therve: oh, nice, from the looks of it seems they figured out the overlap | 12:15 |
therve | jaosorior, Yeah. Not that there is any solution, but that's good :) | 12:15 |
therve | jaosorior, I wonder if we could make it so that we don't change the deployment model on upgrades? | 12:16 |
therve | And change the default port to something else | 12:16 |
jaosorior | what do you mean about not changing the deployment model on upgrades? | 12:17 |
*** fungi has joined #tripleo | 12:17 | |
therve | Not moving from current deployment to apache mod_wsgi | 12:18 |
jaosorior | therve: that could certainly work for getting CI to pass on the puppet patch. And was the way I had it in early commits. But I was told to move it so it would be tested :/ | 12:19 |
*** akrivoka has joined #tripleo | 12:20 | |
openstackgerrit | Chandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow https://review.openstack.org/431916 | 12:21 |
therve | jaosorior, Does the move in puppet-heat matter? I thought the impact would be in puppet-openstackintegration, or tripleo? | 12:22 |
chandankumar | arxcruz: i have fixed your chnages, | 12:22 |
therve | It doesn't seem we change the default in puppet-heat, just the tests default | 12:22 |
*** Fenuks has quit IRC | 12:22 | |
*** zoli|lunch is now known as zoli | 12:22 | |
jaosorior | therve: I'll re-attempt to remove it, lets see if someone mentions anything about it | 12:23 |
arxcruz | chandankumar: ;) | 12:23 |
therve | jaosorior, What do you want to remove? | 12:23 |
*** rhallisey has joined #tripleo | 12:23 | |
*** pblaho has joined #tripleo | 12:23 | |
*** achadha has joined #tripleo | 12:23 | |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 12:24 |
jaosorior | therve: deploying heat_cfn via httpd in that commit | 12:24 |
therve | jaosorior, Ah so do the regular api, and leave cfn behind? | 12:25 |
jaosorior | therve: yeah | 12:25 |
therve | Hum | 12:25 |
jaosorior | therve: for the testing of that commit. And I left a note saying we can't do that yet due to the selinux issue | 12:26 |
therve | jaosorior, Ah yeah just the tests, maybe. | 12:26 |
jaosorior | therve: yeah | 12:26 |
therve | Now that we identified the conflict, it's possible we can move forward | 12:27 |
ansiwen | flaper87: how far are you with the data container approach? if it is not done yet, I would go for that now. | 12:28 |
jaosorior | therve: here I'm adding it to the undercloud though https://review.openstack.org/#/c/394837/11 it had passed at some point. Lets see how it does nowadays | 12:28 |
flaper87 | ansiwen: not done yet, working on it on the tht side | 12:29 |
openstackgerrit | Michele Baldessari proposed openstack/tripleo-heat-templates master: Upgrades: wait for galera to be settled https://review.openstack.org/438947 | 12:29 |
therve | jaosorior, Yeah unless we disabled selinux, it shouldn't work | 12:29 |
ansiwen | flaper87: shall we synchronize somehow? | 12:30 |
*** pblaho has quit IRC | 12:30 | |
*** limao has joined #tripleo | 12:30 | |
*** thrash|g0ne is now known as thrash | 12:31 | |
flaper87 | ansiwen: yeah, I'll ping you in a bit because I'm stuck in calls for a bit | 12:31 |
ansiwen | flaper87: ok | 12:31 |
*** skramaja is now known as skramaja|afk | 12:31 | |
chem | shardy: marios matbu Am I missing something or there is nothing to load the pacemaker/{haproxy,rabbitmq ....} service in https://github.com/openstack/tripleo-heat-templates/blob/master/overcloud-resource-registry-puppet.j2.yaml ? | 12:32 |
EmilienM | hello | 12:33 |
*** achadha has quit IRC | 12:33 | |
jaosorior | chem: that's the pacemaker.yaml environment | 12:33 |
*** ccamacho has quit IRC | 12:34 | |
jaosorior | Any +A's for this? https://review.openstack.org/#/c/438801/2 will fix the scenario004 | 12:34 |
chem | jaosorior: hum, oki but what is the use of https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/pacemaker/rabbitmq.yaml for instance ? | 12:34 |
jaosorior | chem: that's loaded when the deployer uses -e environments/pacemaker.yaml | 12:35 |
jaosorior | chem: sorry, environments/puppet-pacemaker.yaml | 12:35 |
chem | jaosorior: oki, thanks. I see it now. | 12:36 |
jaosorior | EmilienM: the revert worked, I ran it again after your commit to project-config merged, so now it ran all the scenarios. but still needs another +A | 12:36 |
*** limao_ has joined #tripleo | 12:38 | |
chandankumar | tosky: sshnaidm https://bugzilla.redhat.com/show_bug.cgi?id=1356092 the above error is linked with this bug | 12:38 |
openstack | bugzilla.redhat.com bug 1356092 in openstack-ironic "Tempest test failing for cinder, sahara and neutron in RHOS-9" [High,Closed: errata] - Assigned to lmartins | 12:38 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 12:39 |
*** saneax is now known as saneax-_-|AFK | 12:39 | |
tosky | chandankumar: sure? | 12:39 |
*** limao has quit IRC | 12:39 | |
*** bfournie has quit IRC | 12:39 | |
tosky | chandankumar: 2 of the components there were fixed | 12:39 |
*** achadha has joined #tripleo | 12:42 | |
gfidente | jaosorior I don't see manu references | 12:42 |
gfidente | can I switch the defaults in puppet-tripleo ? | 12:43 |
*** pblaho has joined #tripleo | 12:43 | |
jaosorior | gfidente: which references? | 12:43 |
*** dmacpher has joined #tripleo | 12:43 | |
jaosorior | gfidente: this https://github.com/openstack/puppet-tripleo/blob/master/manifests/profile/base/database/mysql/client.pp#L41 ? | 12:44 |
gfidente | yeah | 12:44 |
gfidente | and the ones above can just be turned into defaults no? | 12:44 |
gfidente | we need to land the tht change together with it | 12:44 |
jaosorior | gfidente: something we could do is to change the t-h-t values to fill those, then we could remove those hiera calls | 12:45 |
EmilienM | jaosorior: ok | 12:45 |
EmilienM | folks, I need a quick review on https://review.openstack.org/#/c/438801/ please - it's blocking CI | 12:46 |
jaosorior | gerrit is particularly slow today, anybody else experiencing this? | 12:46 |
openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates master: Add upgrade task for panko api https://review.openstack.org/438298 | 12:47 |
gfidente | /me too | 12:47 |
EmilienM | chem, bandini, marios and all: very quick FYI: I enabled pacemaker in scenarios jobs, including scenario-upgrades | 12:47 |
*** masco has quit IRC | 12:47 | |
mandre | jaosorior: yep, experiencing the same gerrit slowness here | 12:48 |
bandini | EmilienM: ah perfect. thanks! | 12:48 |
chem | EmilienM: cool thank. | 12:48 |
gfidente | EmilienM did you add the services to the multinode jobs as I did for scenario4 or set it globally in tripleo.sh ? | 12:48 |
EmilienM | gfidente: I added them to multinode jobs | 12:48 |
gfidente | ack, cool | 12:49 |
EmilienM | gfidente: maybe you can verify in THT please, to be sure I didn't do wrong | 12:49 |
marios | thanks EmilienM | 12:49 |
gfidente | EmilienM I think it's fine just wanted to have them all behaving in the same way | 12:49 |
EmilienM | marios, chem, bandini: I probably broke the CI job or something but that's on purpose, so we can iterate on making the job passing | 12:49 |
EmilienM | gate-tripleo-ci-centos-7-multinode-upgrades-nv is passing well this morning | 12:49 |
EmilienM | though scenario upgrades are broken - I'll work on it | 12:49 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates master: mysqlclient: Use actual parameter in puppet to set bind-address https://review.openstack.org/438950 | 12:50 |
jaosorior | gfidente: ^^ | 12:50 |
EmilienM | jaosorior, marios, gfidente: I need you core review on https://review.openstack.org/#/c/438567/ please | 12:50 |
gfidente | jaosorior ok nice let me do the puppet-tripleo part | 12:50 |
jaosorior | gfidente: once we have that, we can remove the hiera calls | 12:50 |
jaosorior | gfidente: wait up | 12:50 |
EmilienM | marios, gfidente: and someone to approve https://review.openstack.org/#/c/438801/ | 12:50 |
*** ccamacho has joined #tripleo | 12:50 | |
jaosorior | gfidente: I can do it, on top of this one https://review.openstack.org/#/c/437365/ | 12:51 |
*** adarazs_lunch is now known as adarazs | 12:51 | |
*** achadha has quit IRC | 12:51 | |
EmilienM | mandre: please do not recheck https://review.openstack.org/#/c/438657/ until the https://review.openstack.org/#/c/438801/ is merged... | 12:52 |
jaosorior | gfidente: which, by the way, could you review this https://review.openstack.org/#/c/437365/ ? | 12:52 |
marios | EmilienM: ack... unfortunate about https://review.openstack.org/#/c/438801/2 i mean is a big change but a nice tidy-up/refactor | 12:53 |
*** limao_ has quit IRC | 12:53 | |
marios | EmilienM: the thing we are reverting i mean was a nice tidy-up | 12:53 |
paramite | Hey guys, can somebody else from core please check following patch? https://review.openstack.org/#/c/419460/ | 12:53 |
*** psahoo has quit IRC | 12:54 | |
gfidente | jaosorior ack update that comment too! :) | 12:54 |
jaosorior | gfidente: aaah yeah, will do! | 12:54 |
gfidente | (in the one fixing puppet-tripleo) | 12:54 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: mysqlclient: Drop hiera calls in favor of getting these via t-h-t https://review.openstack.org/438952 | 12:54 |
EmilienM | marios: any other suggestion for now? It's blocking some patches to merge | 12:55 |
*** limao has joined #tripleo | 12:55 | |
EmilienM | marios: I spent time last night to investigate and saw more services that could be affected. I suggest to postpone this refactor | 12:55 |
marios | EmilienM: no just saying we should get it back asap if it breaks something then it should be reverted | 12:55 |
EmilienM | marios: I'm pretty sure we'll get it soon again | 12:55 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo master: mysqlclient: Drop hiera calls in favor of getting these via t-h-t https://review.openstack.org/438952 | 12:56 |
jaosorior | gfidente: done ^^ | 12:56 |
*** chem has quit IRC | 12:57 | |
gfidente | cool we got things rolling | 12:57 |
rook | gfidente: hey - question about using the device-name for root-disk hints | 12:58 |
rook | gfidente: do we know if using /dev/sdX for the root disk hints works w/ OOO newton? | 12:58 |
rook | dtantsur ^ | 12:59 |
rook | I am doing a deployment in CNCF (going to be 900-1000 baremetal nodes) | 12:59 |
rook | and I am trying to deploy Ceph on a subset of nodes, the networking works for all of the nodes but the ceph machines - I cannot easily get console :/ | 12:59 |
gfidente | rook so it doesn't because cloud-init extends the partition to the full disk size | 13:00 |
rook | I am also curious the output in /var/log/ironic/deploy -- is this just during the install? | 13:00 |
gfidente | rook but if you deploy a 'whole disk' image | 13:00 |
gfidente | and that is gpt labeled and prepartitioned | 13:00 |
gfidente | then you can use sdXN | 13:00 |
*** suuuper has joined #tripleo | 13:00 | |
rook | ah - so maybe it is a issue with the root-disk hints | 13:01 |
*** achadha has joined #tripleo | 13:01 | |
rook | gfidente "whole disk image", can you explain that more? | 13:01 |
dtantsur | rook, should be, but it may be just "sdX", I don't remember | 13:01 |
rook | so dtantsur gfidente is_whole_disk_image': False ; u'root_device': {u'name': u'/dev/sdc'} | 13:02 |
*** dprince has joined #tripleo | 13:02 | |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 13:02 |
rook | i would want to flip the whole_disk_image bit, and change the name to just sdc (ugh does that mean I have to re-introspect again?) | 13:03 |
openstackgerrit | Feng Pan proposed openstack/tripleo-heat-templates master: Add validation for VPP upgrade tasks https://review.openstack.org/438959 | 13:05 |
gfidente | rook yeah it has to be a partitioned image to use whole_disk though | 13:05 |
gfidente | or you could disable cloud-init partgrow | 13:05 |
gfidente | and do the additional partitions after the deployment | 13:06 |
*** achadha has quit IRC | 13:06 | |
gfidente | but that'd be manual, sharing the root disk is not automated yet | 13:06 |
*** jayg|g0n3 is now known as jayg | 13:06 | |
rook | gfidente: roger, thanks! | 13:07 |
*** jpena is now known as jpena|lunch | 13:09 | |
-openstackstatus- NOTICE: restarting gerrit to address performance problems | 13:10 | |
*** ChanServ changes topic to "restarting gerrit to address performance problems" | 13:10 | |
*** ooolpbot has joined #tripleo | 13:10 | |
ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 13:10 |
ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1668493 | 13:10 |
*** ooolpbot has quit IRC | 13:10 | |
openstack | Launchpad bug 1668493 in tripleo "ci/scenario004 fails to create an image in glance/swift" [Critical,Triaged] | 13:10 |
dprince | mandre: pushed a fix to build-kolla.conf's location | 13:12 |
*** limao_ has joined #tripleo | 13:12 | |
jaosorior | dprince: is undercloud_containers working at the moment? Tried it just now and it failed like this http://paste.openstack.org/show/600769/ | 13:15 |
*** limao has quit IRC | 13:16 | |
dprince | jaosorior: I'm going to rebase things today and try it again post-merge (all the patches merged last week) | 13:16 |
*** apetrich has quit IRC | 13:16 | |
*** bfournie has joined #tripleo | 13:16 | |
dprince | jaosorior: in your case I think the actual error is before that though | 13:16 |
jaosorior | dprince: ok, let me know how that goes | 13:16 |
dprince | jaosorior: there is probably some os-collect-config output that has useful information in it | 13:16 |
jaosorior | dprince: where could I fetch that? | 13:17 |
*** yfried has joined #tripleo | 13:17 | |
dprince | jaosorior: it might only be on your console at this point | 13:17 |
ccamacho | hey marios, mind to check the comment in https://review.openstack.org/#/c/438090/5/puppet/services/keystone.yaml ? related to Sofer's patch to move stopping the services to step1 | 13:17 |
*** dsariel has joined #tripleo | 13:17 | |
*** sudipto_ has joined #tripleo | 13:17 | |
*** sudipto has joined #tripleo | 13:17 | |
jaosorior | dprince: | 13:17 |
jaosorior | oh | 13:17 |
jaosorior | found it I think | 13:18 |
jaosorior | oh wait | 13:18 |
jaosorior | there are way more errors than I thought | 13:18 |
dprince | jaosorior: I'm curios, the CI job for this is also failing on step 1 ATM | 13:19 |
dprince | jaosorior: https://review.openstack.org/#/c/437411/ | 13:19 |
EmilienM | tripleo weekly meeting in 40 min: https://etherpad.openstack.org/p/tripleo-meeting-items | 13:19 |
jaosorior | dprince: http://paste.openstack.org/show/600772/ | 13:20 |
*** achadha has joined #tripleo | 13:20 | |
*** trown|outtypewww is now known as trown | 13:20 | |
jaosorior | dprince: that's a long log, but looking for the word error shows a bunch of different stuff failing there | 13:20 |
*** lucasagomes is now known as lucas-hungry | 13:21 | |
dprince | jaosorior: its like your puppet modules are all very old or something | 13:21 |
jtomasek | d0ugal, rbrady: by looking at https://github.com/openstack/tripleo-common/blob/master/tripleo_common/actions/templates.py#L263 we don't use root_environment.yaml when passing the environments to heatclient | 13:22 |
jaosorior | dprince: thought that worked like the regular undercloud deploy, pulling the latest packages | 13:22 |
jaosorior | I'm just using the latest image provided in rh1 | 13:22 |
jaosorior | bnemec: ^ | 13:22 |
dprince | jaosorior: I think you need to update your packages on the host | 13:23 |
jtomasek | d0ugal: how do we ensure, that overcloud-resource-registry-puppet.yaml is overriden by contents of other environments? | 13:23 |
jaosorior | do the run.sh idempotent? | 13:23 |
dprince | jaosorior: we are bind mounting in puppet modules from the host for this | 13:23 |
jaosorior | I see | 13:23 |
*** agurenko has quit IRC | 13:23 | |
jtomasek | d0ugal: ah, I think I see now, env_paths is array... nvm:) | 13:24 |
*** apetrich has joined #tripleo | 13:26 | |
EmilienM | ccamacho: I think panko-api needs the validation bits | 13:27 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/ocata: Adds http proxy support for registering RHEL overcloud nodes https://review.openstack.org/438961 | 13:28 |
ccamacho | hey EmilienM we already have the validation bits there, tthe thing is about stopping apache at step2 instead of step1 | 13:28 |
EmilienM | slagle: if you ok, we can backport this one: https://review.openstack.org/438961 (no launchpad but at this stage i think that's acceptable) | 13:28 |
ccamacho | as is a service used by several services | 13:28 |
EmilienM | ccamacho: when should we stop httpd? | 13:28 |
slagle | EmilienM: i think it violates stables:follows-policy, but when i've pointed that out in the past, we don't seem concerned about it | 13:29 |
ccamacho | I think in step2, for example a ceilometer agent tries to communicate with panko and apache is not working we will have errors right?, so, I think we need to stop the ceilometer services first (i.e. https://review.openstack.org/#/c/438090/5/puppet/services/ceilometer-agent-compute.yaml) and then stop httpd. | 13:29 |
*** pblaho has quit IRC | 13:29 | |
ccamacho | EmilienM^ | 13:30 |
slagle | EmilienM: so it's fine i suppose in that regard | 13:30 |
*** pblaho has joined #tripleo | 13:30 | |
jaosorior | dprince: am I supposed to run doit.sh as sudo? | 13:30 |
*** agurenko has joined #tripleo | 13:30 | |
EmilienM | slagle: I can file a launchpad bug, a min | 13:30 |
slagle | EmilienM: why would you do that? | 13:31 |
EmilienM | slagle: to stop violate rules :) | 13:31 |
dprince | jaosorior: it runs sudo for you I think | 13:31 |
slagle | EmilienM: but we don't care about violating rules, based on how we review backports | 13:31 |
jaosorior | dprince: anyway, ran tripleo.sh --repo-setup, updated everything I could, and am running the stuff again, lets see | 13:31 |
EmilienM | ccamacho: can you please make sure it's consistent in our upgrade_tasks? making sure httpd is never stopped at step1 | 13:32 |
EmilienM | slagle: ok | 13:32 |
openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart-extras master: collect-logs: small README fix https://review.openstack.org/438962 | 13:32 |
slagle | EmilienM: besides, this is a feature anyway | 13:32 |
*** amoralej is now known as amoralej|lunch | 13:32 | |
slagle | EmilienM: my point is that, i don't think it really matters. we don't follow the rules, we can backport what we want | 13:33 |
slagle | e.g., i don't know what context you were asking me if i was ok with it | 13:33 |
*** limao has joined #tripleo | 13:34 | |
openstackgerrit | Merged openstack/tripleo-quickstart-extras master: collect-logs: use rsync to gather logs https://review.openstack.org/437878 | 13:34 |
EmilienM | slagle: I don't think we backport what we want. We're trying to be careful I guess | 13:34 |
EmilienM | slagle: I asked you because you were part of the review | 13:35 |
slagle | EmilienM: careful to do what, follow stable:follows-policy? | 13:35 |
EmilienM | slagle: no, careful when backporting things | 13:35 |
*** gchamoul|afk is now known as gchamoul | 13:36 | |
*** sudipto_ has quit IRC | 13:36 | |
*** sudipto has quit IRC | 13:36 | |
*** limao_ has quit IRC | 13:36 | |
jaosorior | dprince: just to be clear, first is the json deployment and then the puppetdeployment takes place, right? | 13:37 |
jaosorior | dprince: and the json deployment dumps the hieradata? or what does it do in the undercloud-containers context? | 13:37 |
slagle | EmilienM: ok. "careful" is pretty subjective, for me, i think stable:follows-policy is the most reasonable and agreed upon definition | 13:38 |
*** tbonds has joined #tripleo | 13:38 | |
slagle | EmilienM: that is what i personally use | 13:38 |
-openstackstatus- NOTICE: ok gerrit is back to normal | 13:39 | |
*** ChanServ changes topic to "ok gerrit is back to normal" | 13:39 | |
*** achadha has quit IRC | 13:39 | |
*** ansmith has joined #tripleo | 13:39 | |
*** ansmith has quit IRC | 13:39 | |
EmilienM | slagle: I don't think this patch violates stable:follows-policy if I file a bug | 13:39 |
EmilienM | slagle: we are fixing the user experience here | 13:40 |
jaosorior | dprince: by the way, it seems it's going forward now :D good catch about the outdated manifests | 13:40 |
EmilienM | if I'm reading https://docs.openstack.org/project-team-guide/stable-branches.html - I see nothing that prevents us to backport this one | 13:40 |
jaosorior | dprince: can you point me to where the bind-mount of the puppet manifests is done? | 13:41 |
dprince | jaosorior: its in docker-puppet.py | 13:42 |
jaosorior | dprince: aaww crap, now it failed in undercloud.AllNodesDeploySteps.UndercloudContainersDeployment_Step3 | 13:42 |
*** ansmith has joined #tripleo | 13:42 | |
openstackgerrit | Martin André proposed openstack/tripleo-common master: Add pull-puppet-modules script to binaries https://review.openstack.org/422772 | 13:43 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/ocata: Adds http proxy support for registering RHEL overcloud nodes https://review.openstack.org/438961 | 13:43 |
EmilienM | slagle: now with a bug report^ | 13:43 |
jaosorior | dprince: http://paste.openstack.org/show/600779/ | 13:44 |
openstackgerrit | Gael Chamoulaud proposed openstack/tripleo-quickstart master: Update accessing overcloud documentation https://review.openstack.org/438965 | 13:44 |
*** ChanServ changes topic to "TripleO : http://tripleo.org/ | https://wiki.openstack.org/wiki/TripleO | Meetings On Tuesdays at 14:00 UTC in #openstack-meeting-alt" | 13:45 | |
-openstackstatus- NOTICE: gerrit is back to normal and I don't know how to use the openstackstaus bot | 13:45 | |
*** rlandy has joined #tripleo | 13:45 | |
*** shardy has quit IRC | 13:46 | |
*** shardy has joined #tripleo | 13:47 | |
*** lblanchard has joined #tripleo | 13:47 | |
EmilienM | ccamacho: for https://review.openstack.org/#/c/438090/ - aren't we stopping mysql at step 2? | 13:47 |
EmilienM | ccamacho: I think stopping httpd at step 1 would be ok, no? | 13:47 |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Configure SSL connection for MySQL client via client config file https://review.openstack.org/437364 | 13:47 |
mandre | dprince, EmilienM, shardy: can I get a quick review on https://review.openstack.org/#/c/422772/? Please note the CI will fail until https://review.rdoproject.org/r/#/c/4391/7 lands | 13:48 |
*** lblanchard has quit IRC | 13:48 | |
*** dciabrin has quit IRC | 13:49 | |
EmilienM | mandre: we need https://review.rdoproject.org/r/#/c/4391/7 merged first and built in delorean | 13:49 |
EmilienM | or it will probably break something | 13:50 |
*** dciabrin has joined #tripleo | 13:50 | |
mandre | EmilienM: exactly, I'm trying to gather +2 so we can quickly merge the patch once https://review.rdoproject.org/r/#/c/4391/ lands | 13:50 |
*** ansmith has quit IRC | 13:51 | |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 13:52 |
*** pradk has joined #tripleo | 13:52 | |
*** limao has quit IRC | 13:55 | |
*** limao has joined #tripleo | 13:55 | |
*** cylopez has joined #tripleo | 13:57 | |
*** jmelvin has joined #tripleo | 13:57 | |
ccamacho | EmilienM yeahp agreed on that, I think is safe to stop http on step1 but we will be consistent and in all services stopping httpd do it on same step | 13:58 |
EmilienM | ccamacho: chem is offline, we need to fix this issue asap | 13:58 |
EmilienM | weekly meeting is about to start | 13:59 |
openstackgerrit | Flavio Percoco proposed openstack/tripleo-quickstart-extras master: Use openstack deploy to install the undercloud https://review.openstack.org/419040 | 14:01 |
*** achadha has joined #tripleo | 14:02 | |
*** openstackgerrit has quit IRC | 14:03 | |
*** bkopilov has joined #tripleo | 14:03 | |
*** openstackgerrit has joined #tripleo | 14:03 | |
*** ansmith has joined #tripleo | 14:03 | |
openstackgerrit | Honza Pokorny proposed openstack/tripleo-specs master: Add gui logging spec https://review.openstack.org/395138 | 14:03 |
*** fultonj has joined #tripleo | 14:03 | |
honza | akrivoka: so, ... I was reviewing your plan export patches yesterday, and now I finally have an idea on how to do GUI logging: coincidence? I think not. | 14:04 |
akrivoka | honza: you're welcome :D | 14:05 |
honza | akrivoka: the temp url for a file in swift was the missing link in my spec :) yay | 14:05 |
honza | akrivoka: yes, a big thank you :) | 14:05 |
akrivoka | \o/ | 14:05 |
*** achadha has quit IRC | 14:06 | |
*** user` has joined #tripleo | 14:07 | |
jpich | \o/ | 14:08 |
openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 14:10 |
*** gbarros has joined #tripleo | 14:10 | |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates master: Adding keystone parameters for Congress https://review.openstack.org/438646 | 14:10 |
*** jpena|lunch is now known as jpena | 14:10 | |
b00tcat | if I wanted to know what network config is picked up by default with tripleo, would you look at the tripleoclient repo or at the tripleo-common repo? I'm trying to figure out what's in each | 14:11 |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates master: Adding congress datasource to pingtest scn001 https://review.openstack.org/438769 | 14:11 |
*** achadha has joined #tripleo | 14:11 | |
*** user` is now known as chem | 14:12 | |
*** chem has quit IRC | 14:12 | |
*** chem has joined #tripleo | 14:12 | |
*** salmankhan has quit IRC | 14:14 | |
openstackgerrit | Merged openstack/tripleo-quickstart master: Set network isolation type to single-nic by default https://review.openstack.org/437816 | 14:15 |
*** lucas-hungry is now known as lucasagomes | 14:15 | |
*** ckyriakidou has quit IRC | 14:15 | |
*** achadha has quit IRC | 14:16 | |
*** toure|gone is now known as toure | 14:16 | |
*** jaganathan is now known as jaganathan_afk | 14:16 | |
*** tzumainn has joined #tripleo | 14:16 | |
*** limao_ has joined #tripleo | 14:17 | |
*** limao has quit IRC | 14:19 | |
*** gkadam has quit IRC | 14:20 | |
*** michapma has joined #tripleo | 14:20 | |
*** eck` is now known as eck`gone | 14:20 | |
*** michapma has quit IRC | 14:20 | |
*** milan has joined #tripleo | 14:20 | |
*** michapma has joined #tripleo | 14:21 | |
*** eck`gone is now known as eck` | 14:21 | |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 14:22 |
*** sudipto_ has joined #tripleo | 14:22 | |
*** sudipto has joined #tripleo | 14:22 | |
openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 14:23 |
openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh https://review.openstack.org/437411 | 14:23 |
*** nyechiel has joined #tripleo | 14:24 | |
*** rbrady has joined #tripleo | 14:24 | |
*** rbrady has joined #tripleo | 14:24 | |
*** sdake has joined #tripleo | 14:25 | |
*** rhallisey has quit IRC | 14:26 | |
d0ugal | jtomasek: sorry, I was away for a bit - looks like you got all your answers? | 14:27 |
jtomasek | d0ugal: yes, no worries | 14:29 |
openstackgerrit | Honza Pokorny proposed openstack/tripleo-specs master: Add gui logging spec https://review.openstack.org/395138 | 14:30 |
*** chlong has joined #tripleo | 14:30 | |
*** ckyriakidou has joined #tripleo | 14:30 | |
*** pblaho has quit IRC | 14:32 | |
openstackgerrit | Merged openstack/puppet-tripleo master: Revert "Add httpchk for http services" https://review.openstack.org/438801 | 14:32 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist https://review.openstack.org/438657 | 14:32 |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Add support for upgrading ec2-api https://review.openstack.org/425285 | 14:32 |
EmilienM | mwhahaha: you probably want to revert https://review.openstack.org/438801 (scenario004 should fail) | 14:33 |
openstackgerrit | Attila Darazs proposed openstack-infra/tripleo-ci master: quickstart settings: fix collect-logs settings https://review.openstack.org/438978 | 14:33 |
openstackgerrit | Merged openstack-infra/tripleo-ci master: Export UPGRADE_ENV for scenario-upgrades https://review.openstack.org/438567 | 14:33 |
mwhahaha | K | 14:33 |
mwhahaha | I'm going to look at it in a biy | 14:33 |
mwhahaha | Bit | 14:33 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 14:35 |
openstackgerrit | Honza Pokorny proposed openstack/tripleo-specs master: Add gui logging spec https://review.openstack.org/395138 | 14:36 |
openstackgerrit | Ana Krivokapic proposed openstack/tripleo-heat-templates master: Add plan-environment.yaml https://review.openstack.org/417385 | 14:37 |
openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Enable composable upgrades for docker service templates https://review.openstack.org/438979 | 14:37 |
shardy | flaper87, jistr: ^^ first pass, not yet tested | 14:38 |
*** amoralej|lunch is now known as amoralej | 14:38 | |
flaper87 | shardy: thanks a bunch | 14:39 |
flaper87 | will take a look | 14:39 |
openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart master: break out general_config into config and topology https://review.openstack.org/425451 | 14:39 |
openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Enable composable upgrades for docker service templates https://review.openstack.org/438979 | 14:40 |
shardy | flaper87: thanks, it'll probably need some fixes, making coffee then will test it out | 14:40 |
jrist | fastest meeting ever? | 14:41 |
*** limao_ has quit IRC | 14:41 | |
jrist | I suppose you guys covered most stuff last week pretty thoroughly | 14:41 |
jistr | shardy: great, thanks! | 14:41 |
*** limao has joined #tripleo | 14:42 | |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Put service stop at step1 and quiesce at step2. https://review.openstack.org/438090 | 14:44 |
EmilienM | shardy: https://review.openstack.org/#/c/438979 is for transition from packaging deployment to containers, right? | 14:45 |
shardy | EmilienM: yes, it's an example of how we might do it after a discussion with flaper87 and jistr | 14:45 |
jrist | akrivoka: did you see jtomasek's review on https://review.openstack.org/#/c/417385/ ? | 14:46 |
EmilienM | shardy: nice. It would be cool to have an experimental ci job | 14:46 |
shardy | basically we stop and disable the services on the host, then start the services in containers like in a normal deploy | 14:46 |
shardy | EmilienM: yeah I was thinking the same, but I'll get it working locally first | 14:46 |
flaper87 | EmilienM: yup, we literally just started working on that | 14:46 |
akrivoka | jrist: yeah, I updated the patch | 14:46 |
shardy | probably we need green containers CI, then we can think about container upgrade CI | 14:46 |
EmilienM | flaper87: cool :D | 14:47 |
EmilienM | shardy: +1 | 14:47 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 14:47 |
EmilienM | shardy: I'm worried about the data, we need a way to migrate it somehow | 14:47 |
flaper87 | EmilienM: I think we'll just bind mount it | 14:48 |
shardy | EmilienM: yeah that's still to be worked out - I think we're likely to just bind mount it for now | 14:48 |
flaper87 | EmilienM: again, everything is under discussion | 14:48 |
openstackgerrit | Ana Krivokapic proposed openstack/tripleo-common master: Enhance plan creation and update with plan-environment https://review.openstack.org/414169 | 14:48 |
jrist | akrivoka: orly | 14:48 |
jrist | akrivoka: haha omg | 14:48 |
jrist | akrivoka: that timing | 14:48 |
EmilienM | ccamacho, chem: let's discuss about https://review.openstack.org/#/c/438090/ | 14:49 |
EmilienM | shardy, flaper87: yes sorry. I'm just giving random thoughts :-P please ignore me | 14:49 |
ccamacho | Emilien, for we would be ideal to stop httpd between the openstack services and myslq | 14:49 |
shardy | EmilienM: sure, it's good to get feedback, feel free to comment on the review | 14:49 |
flaper87 | EmilienM: oh by all means, keep giving them | 14:49 |
ccamacho | EmilienM s/we/me/ | 14:50 |
shardy | the first problem we'll face with testing it is migrating the DB via a bind mount or whatever | 14:50 |
shardy | I'll start testing all-in-one with only keystone+deps like for the first upgrades work | 14:50 |
EmilienM | ccamacho, chem: I think we should stop all openstack services including httpd (that runs APIs) at step 1 | 14:50 |
EmilienM | ccamacho, chem: stopping httpd at step 2 in the same time as mysql is not good, it will through a lot of garbage in logs | 14:51 |
chem | ccamacho: EmilienM oki ... yep but at step1 I maybe service need to get token to discuss their last wishes before shutdown ? | 14:51 |
EmilienM | chem: IMHO, no big deal | 14:51 |
EmilienM | chem: let me explain: | 14:52 |
ccamacho | EmilienM chem cool, so we have a few submissions being merged, so lets rebase chem submission to HEAD after they are merged, and Ill check that all services under httpd are stopping apache on step 1 | 14:52 |
EmilienM | chem: I prefer a service (let's say cinder) not able to get a token and stop functionning, rather than (your proposal) cinder-api able to talk in Keystone but DB is unreachable so data is lost | 14:52 |
ccamacho | yeahp | 14:53 |
ccamacho | EmilienM is right there | 14:53 |
chem | ccamacho: EmilienM, oki so the proper way is really to have this step1.5 | 14:53 |
ccamacho | chem yeah | 14:53 |
ccamacho | step 1.5 :P | 14:53 |
chem | ccamacho: EmilienM where we stop keystone only | 14:53 |
*** pkovar has quit IRC | 14:53 | |
EmilienM | chem: what is 1.5 (/me sorry to ask) | 14:54 |
chem | ccamacho: EmilienM it's just a way of saying between current step1 and step2 | 14:54 |
EmilienM | TIL | 14:54 |
chem | EmilienM TIL ? | 14:54 |
EmilienM | chem: today I learnt | 14:54 |
EmilienM | chem: do we use that already in THT? | 14:54 |
rook | shardy: going through a deployment as we speak and keep seeing : Feb 28 09:54:27 host-192-0-2-24 os-collect-config: 404 Client Error: Not Found for url: http://192.0.2.1:8080/v1/AUTH_f1ec65886fd948cfbeebe23790ccbe9a/ov-uha3rgpxkn-2-tezeq2rgr4a2-CephStorage-o7ou27zvorhy/b6304bf2-843c-414f-a989-867f64443863?temp_url_sig=2c205e2af639ce12ae01b271937bd348e8f5966a&temp_url_expires=2147483586 | 14:55 |
rook | should this be a red-flag for this deployment? | 14:55 |
chem | EmilienM: nope to do that we have to move everything one step up from step2 and add it (step3 becomes step4, step2 becomes step3 ...) | 14:56 |
chem | EmilienM ccamacho emacs refactoring at the rescue | 14:56 |
EmilienM | ah ok, gotcha now | 14:56 |
pradk | can someone +A this please https://review.openstack.org/#/c/430444/ | 14:56 |
matbu | chem: why using a stepX at all ? | 14:56 |
chem | EmilienM ccamacho so everyone agree that the cleanest is to add a new step ? | 14:56 |
EmilienM | chem: I don't | 14:57 |
EmilienM | it's adding complexity and time | 14:57 |
chem | matbu: I'm under age to have stepX | 14:57 |
EmilienM | I would sotp httpd at step1 | 14:57 |
matbu | chem: using tags with ansible doesn't mean incremental tags | 14:57 |
chem | matbu: dependency ? | 14:57 |
matbu | chem: hehe | 14:57 |
EmilienM | matbu: I know but we use it that way | 14:57 |
matbu | chem: we can handle dependency in another way | 14:57 |
chem | matbu: dependency between node ? | 14:58 |
EmilienM | pradk: how upgrade is handled? | 14:58 |
rook | gfidente ever see that 404 with Ceph before? | 14:58 |
chem | EmilienM: oki, so I revert back my change :) (sorry ccamacho, his use case is stronger) | 14:58 |
ccamacho | EmilienM chem matbu, for sanity I would leave it as it is (httpd stop on 1), we can run tests and see the behavior then we can decide | 14:58 |
pradk | EmilienM, the service name is the same | 14:59 |
chandankumar | sshnaidm: hello | 14:59 |
EmilienM | pradk: I commented | 14:59 |
pradk | EmilienM, its just how its invoked changes so upgradses should work without impact | 14:59 |
chem | ccamacho: I already push the step2 stuff, let me put it back at step1 and call it done | 14:59 |
chandankumar | sshnaidm: http://logs.openstack.org/16/431916/15/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha-oooq-nv/8f0d950/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz | 14:59 |
ccamacho | yeah | 14:59 |
EmilienM | pradk: in the upgrade process, we need to make sure the agents are stopped | 14:59 |
EmilienM | pradk: but I think that's already the case | 14:59 |
chandankumar | sshnaidm: why it is taking + /usr/bin/tempest run --regex '.*smoke' --blacklist-file /home/jenkins/skip_file --concurrency=2 | 14:59 |
* ccamacho sorry for the noise there on the step stuff chem EmilienM | 14:59 | |
*** limao_ has joined #tripleo | 15:00 | |
EmilienM | pradk: so for now, I would just wait that scenario001-upgrades job pass (I'm working on it today) and then recheck. do you mind waiting? | 15:00 |
pradk | EmilienM, ok upto you.. i dont see the relation to both.. i dont see the point holding this patch | 15:00 |
chandankumar | sshnaidm: i have passed only smoke as args from where it is taking *.smoke ? | 15:00 |
*** bvandenh has joined #tripleo | 15:00 | |
chem | ccamacho: nope, it's a valide use case, we just sacrifice it on the hostel of performance and complexity sanity | 15:00 |
EmilienM | pradk: the point is that your patch is not testing the transition between Ocata and Pike yet | 15:01 |
pradk | EmilienM, scenario jobs have been having issues for a while.. why block features for that | 15:01 |
chandankumar | sshnaidm: that's why tests are not running | 15:01 |
EmilienM | pradk: for a while? we have the job since yesterday :D | 15:01 |
chem | ccamacho: even if performance may be overrated here ... | 15:01 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Put service stop at step1 and quiesce at step2. https://review.openstack.org/438090 | 15:01 |
pradk | EmilienM, right have not been tested for a while.. why block a patch when we have it since yesterday :) | 15:01 |
pradk | EmilienM, k anyway its ready, when you feel like please merge | 15:02 |
EmilienM | pradk: I'm ok to merge it. I just find it odd that we don't test in CI the transition. | 15:02 |
sshnaidm | chandankumar, it should be defined in test_regex: https://github.com/openstack/tripleo-quickstart-extras/blob/443ade64f54adc6e16ab6a0c2e164d7ae0e7f9a0/roles/validate-tempest/defaults/main.yml#L9 | 15:02 |
*** limao has quit IRC | 15:02 | |
chandankumar | sshnaidm: i have changed it https://review.openstack.org/#/c/431916/15/roles/validate-tempest/defaults/main.yml | 15:02 |
openstackgerrit | Tim Rozet proposed openstack/tripleo-puppet-elements master: Adds puppet-systemd https://review.openstack.org/438600 | 15:02 |
chandankumar | sshnaidm: i have no clue why it is not picking | 15:03 |
EmilienM | pradk: approved; | 15:03 |
*** salmankhan has joined #tripleo | 15:03 | |
sshnaidm | chandankumar, I think I know.. | 15:03 |
chandankumar | sshnaidm: the tests are listing fine | 15:03 |
*** pblaho has joined #tripleo | 15:03 | |
chandankumar | sshnaidm: check for tempest plugins and testr list-tests | 15:03 |
sshnaidm | chandankumar, but are you sure it matters? right now it runs with *smoke and everything works | 15:03 |
chandankumar | sshnaidm: tempest run --regex "smoke" will work | 15:04 |
chandankumar | sshnaidm: but tempest run --regex "*.smoke" will not work i have tested it on my setup then i changed | 15:04 |
sshnaidm | chandankumar, 1 sec, will fix it | 15:05 |
chandankumar | sshnaidm: feel free to test it | 15:05 |
pradk | EmilienM, oh hehe, thx .. i was fine with holding off until you were ready. But yea if upgrades from ocata -> pike are ready for testing we can handle it. It will chnage a bit anyway with container stuff. | 15:05 |
chandankumar | sshnaidm: all the changes from my side is done | 15:05 |
sshnaidm | chandankumar, I change this line: https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/quickstart/ovb-settings.yml#L10 | 15:05 |
* EmilienM it's all flaper87's fault anyway | 15:05 | |
*** abregman has quit IRC | 15:05 | |
* flaper87 agrees with EmilienM | 15:06 | |
flaper87 | wait, what? | 15:06 |
openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: DON'T MERGE: test tempest all branches https://review.openstack.org/436923 | 15:06 |
sshnaidm | chandankumar, ^^ | 15:06 |
honza | mwhahaha: I used your ansible-ovb thingie the other day, and it worked perfectly, A++ would use again | 15:07 |
chandankumar | sshnaidm: check newton tests 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 | 15:07 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 15:07 |
mwhahaha | honza: cool :D | 15:07 |
EmilienM | pradk: also, please submit a release note on top of https://review.openstack.org/#/c/430444/5 (in a separated patch) | 15:07 |
*** gbarros has quit IRC | 15:07 | |
sshnaidm | chandankumar, I see 7 failed there | 15:07 |
sshnaidm | oh, no, 21 | 15:08 |
chandankumar | sshnaidm: that is related to configuraiton not related to my patch | 15:08 |
*** saibarspeis has joined #tripleo | 15:08 | |
*** gbarros has joined #tripleo | 15:08 | |
pradk | EmilienM, i did on the depends patch as you suggested | 15:08 |
sshnaidm | chandankumar, which configuration? | 15:08 |
chandankumar | arxcruz: 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 | 15:08 |
EmilienM | trozet: why do you need puppet-systemd? | 15:08 |
EmilienM | trozet: only for mariadb? | 15:08 |
sshnaidm | chandankumar, right now we have everything passing in all branches, we can't merge something that breaks us | 15:08 |
trozet | EmilienM: that is the first use case | 15:08 |
pradk | EmilienM, also when you have time please review https://review.openstack.org/#/c/430447/ | 15:09 |
trozet | EmilienM: it may be used down the road for uniform systemd modifications | 15:09 |
sshnaidm | chandankumar, if some branch doesn't work like mitaka - let's not change for it anything | 15:09 |
EmilienM | pradk: in THT? | 15:09 |
EmilienM | trozet: ok | 15:09 |
EmilienM | trozet: is https://github.com/camptocamp/puppet-systemd good? | 15:09 |
pradk | EmilienM, in puppet-tripleo i think.. I can do it in tht too if you prefer | 15:09 |
EmilienM | pradk: in both actually, please | 15:09 |
pradk | kk sure | 15:09 |
trozet | EmilienM: well I tested the part I needed to and it worked | 15:10 |
trozet | EmilienM: but I didn't examine the entire repo code, Alex told me about this module | 15:10 |
openstackgerrit | Merged openstack/tripleo-quickstart-extras master: Add disk root device hints https://review.openstack.org/433691 | 15:10 |
EmilienM | pradk: commented https://review.openstack.org/#/c/430447/ | 15:10 |
*** chlong has quit IRC | 15:11 | |
EmilienM | trozet: excellent. Second thing is: do we have it in RDO? | 15:11 |
ansiwen | flaper87, dprince: ok, so I just checked, if you have a data-container with volumes, the volumes are just normal directories in /var/lib/docker/volumes... nothing special... no loopback device. so I can't imagine any drawbacks there. and: anyway we would not need to make the host dir mounts to it's original places. we could have our undercloud namespace on the host, like -v /var/lib/undercloud/tmp:/tmp | 15:11 |
pradk | EmilienM, thx will do | 15:11 |
trozet | EmilienM: yeah I added it there https://review.rdoproject.org/r/#/c/4619/, https://review.rdoproject.org/r/#/c/4159/ | 15:12 |
trozet | EmilienM: newton patch: https://review.rdoproject.org/r/#/c/4620/ | 15:12 |
EmilienM | trozet: good | 15:12 |
chandankumar | arxcruz: can you look newton telemetry failure | 15:12 |
EmilienM | trozet: can you please add a release note to https://review.openstack.org/#/c/438600/3 and I'll +2 right away | 15:13 |
dprince | ansiwen: perhaps we haven't put our finger on exactly cause, but something causes my undercloud to run slower with devicemapper on a loopback | 15:13 |
*** pkovar has joined #tripleo | 15:13 | |
trozet | EmilienM: one question I had about this whole process. So after it is in RDO, it will exist in the RDO repo, but it wont be pulled into the overcloud OOO promote until it is in tripleo-image-elements right? | 15:13 |
dprince | ansiwen: significantly slower I think | 15:13 |
trozet | EmilienM: that is the only way to get DIB to pick it up? | 15:13 |
dprince | ansiwen: unclear if we can get by with doing this for the overcloud without us costing us performance there as well | 15:14 |
trozet | EmilienM: ok will do | 15:14 |
*** suuuper has quit IRC | 15:14 | |
ansiwen | dprince: sure, but I say: we don't need loopbacks. the question is what exactly you did. I'm talking about replacing the host dir mounts with a --volumes-from, and as far as I can see we could make that as performant as a host-dir mount, because it is technically the same | 15:14 |
EmilienM | trozet: we already have puppet-systemd-0.4.0-0.20170117092318.a032136.el7.centos.noarch in the images | 15:15 |
EmilienM | trozet: so we're good | 15:15 |
ansiwen | dprince: and is there an argument against having our own directory at least for the host dir mounts? | 15:15 |
EmilienM | trozet: see http://logs.openstack.org/72/422772/3/check/gate-tripleo-ci-centos-7-nonha-multinode/a8c4198/logs/syslog.txt.gz#_Feb_28_14_12_36 | 15:16 |
trozet | EmilienM: so how does that work? what pulls it from RDO into the overcloud image if the puppet elements ar emissing? | 15:16 |
flaper87 | ansiwen: the problem are not the bind mounts but the named volumes, I think | 15:16 |
EmilienM | trozet: let me show you, give me a sec | 15:16 |
EmilienM | trozet: so we pull openstack-puppet-modules which pulls all packages | 15:16 |
dprince | ansiwen: perhaps not, I'd be fine if we have our own directory | 15:16 |
EmilienM | trozet: https://github.com/rdo-packages/openstack-puppet-modules-distgit/blob/rpm-master/openstack-puppet-modules.spec#L76 | 15:17 |
flaper87 | dprince: speaking of which, how would you feel if we just move everything under `/var/lib/tripleo` ? | 15:17 |
ansiwen | flaper87: yeah, that might be, but I'm not talking about named volumes (although they should also be as performant, if they use the "local" driver as well) | 15:17 |
flaper87 | I was talking to Slower about this at the PTG | 15:17 |
EmilienM | trozet: though it's not working for newton FYI, see https://github.com/rdo-packages/openstack-puppet-modules-distgit/blob/newton-rdo/openstack-puppet-modules.spec#L71 | 15:17 |
trozet | EmilienM: ah ok, bc I dont see puppet-systemd in my newton overcloud image right now | 15:18 |
*** abehl has quit IRC | 15:18 | |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Add upgrade support for Horizon https://review.openstack.org/437545 | 15:18 |
trozet | EmilienM: so is the point of including this tripleo-puppet-elements patch, so that you could build images without RDO/OPM? | 15:18 |
dprince | flaper87: what specifically? Just config file stuff? Or are you talking about service data too? | 15:19 |
flaper87 | dprince: sorry, I was unclear. I meant the config file stuff | 15:19 |
dprince | flaper87: I suppose I'd like to leave service data in-place, to make upgrades easier | 15:19 |
EmilienM | trozet: tbh, I'm not sure we still use these elements for puppet modules. | 15:19 |
ansiwen | dprince, flaper87: if he talks about the same as me: everything you would ever host-dir-mount into a container | 15:19 |
*** dmarlin has joined #tripleo | 15:19 | |
EmilienM | dprince: is there still a point of doing that? https://review.openstack.org/#/c/438600/ | 15:20 |
dprince | flaper87: you already moved the config-file stuff once. So I guess moving the directory again to /var/lib/tripleo is fine. But I though you were making a volume | 15:20 |
trozet | EmilienM: ok slagle is the one that commented to include that patch. slagle do you know why we need tripleo-puppet-elements if OPM already contains it? | 15:20 |
*** limao_ has quit IRC | 15:20 | |
flaper87 | dprince: yup, working on both fronts, fwiw. If we have 1 dir, it'd just be easier :) | 15:21 |
dprince | EmilienM: I'm not aware of a need for it | 15:21 |
*** rhallisey has joined #tripleo | 15:21 | |
EmilienM | trozet: I think it's here for historical reasons when we were installing opm from source in CI | 15:21 |
EmilienM | but since we moved it to delorean, i don't see why we whould need it | 15:22 |
chandankumar | sshnaidm: are we configuring mistral with oooq in ci? | 15:22 |
sshnaidm | chandankumar, hmm.. not sure we do. | 15:22 |
ansiwen | dprince, flaper87: maybe our confusion here is that "host" can be two different things here? I'm talking about the scenario, where the undercloud completely runs in containers, so where you basically have no host involvement. | 15:22 |
openstackgerrit | Merged openstack/tripleo-heat-templates stable/ocata: Install openstack-panko-api on upgrade https://review.openstack.org/438462 | 15:23 |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Add zaqar service for composable upgrade https://review.openstack.org/435906 | 15:23 |
sshnaidm | chandankumar, what kind of configuring? | 15:23 |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Add auditd upgrate steps https://review.openstack.org/438289 | 15:23 |
trozet | EmilienM: so what pulls OPM in, is that during DIB as well? | 15:24 |
dprince | mandre: it is working :) http://logs.openstack.org/11/437411/14/experimental/gate-tripleo-ci-centos-7-undercloud-containers/1ab6f58/console.html. Just need a stackrc for the host now | 15:24 |
mandre | dprince: awesome! | 15:25 |
EmilienM | sshnaidm, adarazs ^ can you make sure gate-tripleo-ci-centos-7-undercloud-containers is also in our list for transition | 15:25 |
EmilienM | dprince: nice work. It would be nice to have a parity in term of logging (grab logs from containers, so we have the same logs as classic jobs, if possible) | 15:25 |
dprince | EmilienM: I think Ian is working on logging. | 15:26 |
*** trown is now known as trown|brb | 15:26 | |
dprince | EmilienM: step at a time | 15:26 |
arxcruz | chandankumar: what exactly you want me to check? the failure is clear there, need to use gnocchi api for metrics | 15:26 |
EmilienM | dprince: right. good to see it mostly working | 15:27 |
adarazs | EmilienM: hm, I'm not sure we have capacity to add that by 10th of March, or how far away we are from making it work on multinode/OVB versus libvirt. weshay, do you think it's likely we can get a container job upstream? | 15:27 |
dprince | EmilienM: cool, thanks | 15:27 |
*** trown|brb is now known as trown | 15:28 | |
ansiwen | dprince, flaper87: to avoid confusion: named volumes and data containers are _not_ the same. | 15:28 |
weshay | adarazs, we have a container job upstream | 15:28 |
adarazs | weshay: quickstart based? what is the name for it? | 15:29 |
flaper87 | ansiwen: I know | 15:29 |
trown | adarazs: ya and it is "working"-ish on libvirt | 15:29 |
trown | adarazs: and we have ovb one too | 15:29 |
adarazs | trown: I know it's kind of working on libvirt :) | 15:29 |
ansiwen | flaper87: ok, just wanted to make sure, because the terms are both used here | 15:29 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add mistral service support for composable upgrades https://review.openstack.org/405246 | 15:29 |
weshay | adarazs, gate-tripleo-ci-centos-7-ovb-containers-oooq | 15:29 |
adarazs | weshay: thanks. what repo is it running against? tripleo-ci? | 15:30 |
openstackgerrit | Arx Cruz proposed openstack/tripleo-quickstart-extras master: Updating skip file mechanism https://review.openstack.org/438465 | 15:30 |
trown | adarazs: ya master-tripleo-ci | 15:31 |
adarazs | okay :) | 15:31 |
*** karthiks has quit IRC | 15:31 | |
adarazs | so I guess we can transition it :) | 15:31 |
*** jcoufal has joined #tripleo | 15:31 | |
weshay | adarazs, I'll send some reviews.. currently tht and some subdirs.. but flaper87 has some reviews to make it a check gate on all | 15:31 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add mistral service support for composable upgrades https://review.openstack.org/405246 | 15:32 |
trown | adarazs: what we dont have is containerized (or not) undercloud deploy via heat job | 15:32 |
*** owalsh has quit IRC | 15:33 | |
*** owalsh_ has joined #tripleo | 15:33 | |
chandankumar | arxcruz: what about mistral failures | 15:33 |
weshay | adarazs, https://review.openstack.org/#/c/437945/ | 15:34 |
arxcruz | chandankumar: it's saying endpoint not found, I'm not familiar with mistral tests, but did you set the endpoint correctly ? | 15:34 |
EmilienM | ccamacho, chem: quick comment on https://review.openstack.org/#/c/438090/5/puppet/services/snmp.yaml | 15:35 |
EmilienM | ccamacho, chem: but other than that, +2 | 15:35 |
flaper87 | weshay: hey, if you re-run the ooq thing again somewhere that I can ssh into, let me know | 15:35 |
weshay | flaper87, I am now | 15:35 |
arxcruz | chandankumar: also, I'm seeing that the endpoint for publicURL is set to https://192.168.24.6:13777 and not https://192.168.24.6:13777/v2.0 | 15:35 |
flaper87 | weshay: awesome! | 15:35 |
*** b00tcat_ has joined #tripleo | 15:35 | |
*** b00tcat has quit IRC | 15:35 | |
arxcruz | chandankumar: this will work in pike, but not in newton ;) | 15:36 |
flaper87 | weshay: EmilienM btw, can you +1 https://review.openstack.org/#/c/437945/ ? (assuming it looks good to you) | 15:36 |
openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: Use a more appropriate domain name for FreeIPA https://review.openstack.org/439023 | 15:36 |
rook | gfidente happen to be around? fultonj maybe you have seen this : Feb 28 10:36:29 overcloud-cephstorage-0 os-collect-config: 404 Client Error: Not Found for url: http://192.0.2.1:8080/v1/AUTH_f1ec65886fd948cfbeebe23790ccbe9a/ov-uha3rgpxkn-0-ocorp65ezdhs-CephStorage-epb3wmvobnj7/8fbb61be-ecc2-42e7-a982-22d156b26c51?temp_url_sig=b35246d11535b7c9fd8748d6fd68d4c001cd48da&temp_url_expires=2147483586 | 15:36 |
chandankumar | sshnaidm: one more thing http://logs.openstack.org/16/431916/15/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha-oooq-nv/8f0d950/logs/oooq/undercloud/home/jenkins/tempest-setup.sh.txt.gz | 15:38 |
EmilienM | flaper87: done | 15:38 |
EmilienM | flaper87: so you do yaml now? | 15:38 |
flaper87 | EmilienM: man, I've no idea what I'm doing | 15:38 |
openstackgerrit | Gael Chamoulaud proposed openstack/tripleo-quickstart-extras master: [overcloud-deploy] Update deprecated references to heat https://review.openstack.org/409178 | 15:38 |
*** jmelvin has quit IRC | 15:38 | |
rook | gfidente fultonj I am curious what the client is trying to get, which service should be serving something? swift? | 15:38 |
chandankumar | sshnaidm: test_regex: 'smoke' but in here it is also taking *.smoke | 15:38 |
sshnaidm | chandankumar, it should be ok now | 15:39 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 15:39 |
*** skramaja|afk is now known as skramaja | 15:39 | |
weshay | adarazs, looks like we're collecting +1's https://review.openstack.org/#/c/437945/ | 15:39 |
openstackgerrit | Michele Baldessari proposed openstack/tripleo-heat-templates master: WIP Upgrades: fix up the rabbitmq HA mode like in new ocata deployments https://review.openstack.org/439025 | 15:40 |
openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh https://review.openstack.org/437411 | 15:41 |
*** yamahata_ has joined #tripleo | 15:41 | |
rook | gfidente fultonj here is some additional debugging https://gist.github.com/jtaleric/5a5d5b2606930517c7cf1d8a65bb2da6 | 15:41 |
openstackgerrit | Brent Eagles proposed openstack/tripleo-heat-templates master: Make neutron dhcp agents per network conditional https://review.openstack.org/438549 | 15:42 |
*** chlong has joined #tripleo | 15:43 | |
*** yfried has quit IRC | 15:44 | |
chandankumar | sshnaidm: in current tripleo-ci jobs where we are installing tempest plugins? | 15:44 |
jaosorior | radez: hey, clarified in the congress patch | 15:44 |
jaosorior | radez: sorry for the confusion | 15:44 |
openstackgerrit | Brent Eagles proposed openstack/tripleo-heat-templates master: Use conditionals for neutron and glance worker defaults https://review.openstack.org/425423 | 15:44 |
*** karthiks has joined #tripleo | 15:44 | |
*** fultonj has quit IRC | 15:44 | |
sshnaidm | chandankumar, yep | 15:45 |
*** fultonj has joined #tripleo | 15:46 | |
*** bvandenh has quit IRC | 15:46 | |
chandankumar | sshnaidm: i grepped in openstack-infra/tripleo-ci but not found | 15:47 |
sshnaidm | chandankumar, https://review.openstack.org/#/c/436923/4/scripts/quickstart/ovb-settings.yml | 15:47 |
sshnaidm | chandankumar, line 10 | 15:47 |
chandankumar | sshnaidm: sorry | 15:48 |
chandankumar | sshnaidm: for running tempest we need to install openstack services tempest plugin like for cinder python-cinder-tests where in the code of tripleo-ci we are installing? | 15:49 |
sshnaidm | chandankumar, I don't think we do, we need to do this in tempest role | 15:49 |
openstackgerrit | Merged openstack/tripleo-quickstart master: Adds a Bashate target to tox.ini. https://review.openstack.org/429703 | 15:51 |
chandankumar | sshnaidm: in tempest role i didnot find any step where it installs tempest plugin | 15:51 |
sshnaidm | chandankumar, ok, we used mostly venv there | 15:52 |
chandankumar | sshnaidm: https://review.openstack.org/#/c/431916/15/roles/validate-tempest/templates/configure-tempest.sh.j2 like 104 line | 15:52 |
chandankumar | sshnaidm: but for venv we also need tempest plugins to be installed on system under tests | 15:53 |
sshnaidm | chandankumar, why? | 15:53 |
radez | jaosorior: no worries, thx for the reply | 15:53 |
sshnaidm | chandankumar, venv install runs isolated, installs everything inside this venv | 15:54 |
ansiwen | flaper87, dprince: ok, and now I tried it with named volumes with the "local" driver, which is default in my installation, and also here: exactly the same as a host-dir-mount, no loopback device involved. | 15:54 |
chandankumar | sshnaidm: tempest is a testing framework which provides interface to discover tempest plugin | 15:54 |
sshnaidm | chandankumar, I don't think it was used | 15:54 |
radez | jaosorior: ah, yes that's as I expected, I'll get it fixed up | 15:55 |
chandankumar | sshnaidm: all the component tests except core services resides in their tempest plugin | 15:55 |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates master: Adding keystone parameters for Congress https://review.openstack.org/438646 | 15:55 |
sshnaidm | chandankumar, better to look at installation part as it runs now, I don't remember all details to be honest | 15:56 |
*** rajinir_ has joined #tripleo | 15:56 | |
chandankumar | sshnaidm: in RDO from newton we have moved all tempest tests of their respective services to tehir -test packages provided that tempest plugin is intree | 15:57 |
*** owalsh_ is now known as owalsh | 15:57 | |
gfidente | rook where is that you see it? the new client is uploading all templates into swift and these are downloaded by mistral on the node where mistral is installed before launching the deployment | 15:58 |
rook | gfidente: omfg i think this is what I am hitting | 15:58 |
rook | https://access.redhat.com/solutions/2924181 | 15:58 |
*** sdake has quit IRC | 15:58 | |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates master: Adding keystone parameters for Tacker https://review.openstack.org/438648 | 15:59 |
openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart-extras master: GATE CHECK https://review.openstack.org/430277 | 16:00 |
chandankumar | sshnaidm: installation part in tripleo-ci of tempest? | 16:00 |
radez | jaosorior: got the tacker one updated too: https://review.openstack.org/#/c/438648/ thx! | 16:00 |
jaosorior | radez: done | 16:00 |
rook | gfidente: maybe not... i see that error on the ceph node, in os-collect-config gfidente | 16:02 |
gfidente | so I havent' ever seen that myself | 16:03 |
chandankumar | sshnaidm: feel free to add your comments on the patch, | 16:03 |
*** jaosorior has quit IRC | 16:03 | |
gfidente | I think the problem is that the temporary url expired during the deployment? | 16:04 |
EmilienM | ccamacho: nice, the pre-validation works :D | 16:04 |
*** dsariel has quit IRC | 16:04 | |
EmilienM | ccamacho: see http://logs.openstack.org/57/438657/6/check/gate-tripleo-ci-centos-7-scenario002-multinode-upgrades-nv/1b35a9c/logs/subnode-2/var/log/messages | 16:05 |
EmilienM | ccamacho: grep "Check if ec2-api is deployed" | 16:05 |
*** jlinkes has quit IRC | 16:05 | |
gfidente | but no the change in os-collect-config is permanent | 16:05 |
sshnaidm | chandankumar, no, on validate-tempest role | 16:06 |
sshnaidm | chandankumar, ok, will review, sorry - on mtg now | 16:06 |
EmilienM | ccamacho: though I wonder why playbook failed | 16:06 |
*** jmelvin has joined #tripleo | 16:07 | |
EmilienM | ccamacho: The conditional check 'ec2-api_enabled.rc == 0' failed. The error was: error while evaluating conditional (ec2-api_enabled.rc == 0): 'api_enabled' is undefined"} | 16:08 |
*** rcernin has quit IRC | 16:08 | |
*** sdake has joined #tripleo | 16:08 | |
openstackgerrit | Merged openstack/tripleo-heat-templates master: Switch central and compute templates to use polling agent https://review.openstack.org/430444 | 16:09 |
matbu | EmilienM: it's probably the '-' | 16:10 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades: fix ec2api conditional https://review.openstack.org/439037 | 16:10 |
matbu | EmilienM: sorry , you got it :) | 16:11 |
EmilienM | matbu: that ^? | 16:11 |
EmilienM | :) | 16:11 |
openstackgerrit | Tim Rozet proposed openstack/tripleo-puppet-elements master: Adds puppet-systemd https://review.openstack.org/438600 | 16:11 |
ccamacho | yeah :( is the - | 16:12 |
EmilienM | ccamacho: I wanted to -1 in the initial review, but I didn't know if would break something | 16:12 |
ccamacho | buahh.. there are other patches with - | 16:12 |
ccamacho | ill check them | 16:12 |
chandankumar | sshnaidm: did not found the steps to install tempest plugin in validate-tempest | 16:12 |
EmilienM | ccamacho: please | 16:12 |
EmilienM | chandankumar: where in https://review.openstack.org/#/c/431916/ CI jobs can we see the output (in logs)? | 16:14 |
sshnaidm | chandankumar, we didn't, I'm trying to say that we used only venv script, just look how it's configured now. | 16:14 |
openstackgerrit | Tim Rozet proposed openstack/tripleo-puppet-elements master: Adds puppet-systemd https://review.openstack.org/438600 | 16:14 |
matbu | ccamacho: the others looks ok | 16:15 |
chandankumar | EmilienM: https://review.openstack.org/#/c/436923/3 is linked to test with all branches | 16:15 |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates master: Add a parameter to control batching of ResourceGroup create https://review.openstack.org/439039 | 16:15 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add mistral service support for composable upgrades https://review.openstack.org/405246 | 16:15 |
*** dmacpher has quit IRC | 16:16 | |
chandankumar | EmilienM: gate-tripleo-ci-centos-7-ovb-ha-oooq-nv linked to https://review.openstack.org/#/c/431916/ will show teh result | 16:16 |
EmilienM | chandankumar: can you link the logs directly? | 16:16 |
chandankumar | EmilienM: http://logs.openstack.org/16/431916/15/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha-oooq-nv/8f0d950/logs/oooq/undercloud/home/jenkins/tempest_output.log.txt.gz | 16:16 |
ccamacho | EmilienM matbu the mistral services | 16:17 |
EmilienM | ccamacho: "The test run didn't actually run any tests" is it expected? | 16:17 |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates master: Batches roles ResourceGroup creation and update https://review.openstack.org/439039 | 16:17 |
EmilienM | ccamacho: only mistral? (also don't forget to backport everything to stable/ocata) | 16:17 |
ccamacho | yeahp | 16:17 |
EmilienM | ok | 16:18 |
chandankumar | EmilienM: but one more problem in ci /usr/bin/tempest run --regex '.*smoke' should be 'smoke' from my patch but it is picking *.smoke, https://review.openstack.org/436923 mighjt fix the issye | 16:18 |
EmilienM | chandankumar: ok | 16:18 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates stable/ocata: Add support for upgrading ec2-api https://review.openstack.org/439042 | 16:18 |
EmilienM | can someone approve https://review.openstack.org/#/c/439037/ please? | 16:19 |
EmilienM | ccamacho: I found another problem on scenario001, I'm digging | 16:20 |
*** aufi has quit IRC | 16:21 | |
EmilienM | fatal: [localhost]: FAILED! => {"changed": false, "failed": true, "msg": "Could not find the requested service collectd: cannot check nor set state"} | 16:21 |
EmilienM | ccamacho: I need to make the prevalidation on collectd too :) I'll do it | 16:21 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist https://review.openstack.org/438657 | 16:21 |
EmilienM | ccamacho: weird, I did it already https://review.openstack.org/#/c/438657/7/puppet/services/metrics/collectd.yaml | 16:23 |
gfidente | EmilienM merge this guy https://review.openstack.org/#/c/438036 pls? | 16:23 |
shardy | EmilienM: do you want to wait for CI on https://review.openstack.org/#/c/439037? | 16:23 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 16:24 |
EmilienM | shardy: no need, I cought the issue in CI job, this patch will fix it. | 16:24 |
EmilienM | gfidente: ok will look in a few | 16:24 |
shardy | EmilienM: Ok, approved | 16:24 |
EmilienM | thx | 16:25 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates stable/ocata: Add zaqar service for composable upgrade https://review.openstack.org/439046 | 16:25 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates stable/ocata: Add auditd upgrate steps https://review.openstack.org/439047 | 16:25 |
*** stendulker has joined #tripleo | 16:26 | |
EmilienM | shardy: it would be pretty nice to have the ansible output generated by Ara (cc dmsimard) | 16:26 |
jtomasek | akrivoka: I haven't checked properly yet, but we'll need to check that all references to root_template and root_environment are removed or root_template is changed to 'template' from plan-environment | 16:26 |
jtomasek | akrivoka: https://github.com/openstack/tripleo-common/search?utf8=%E2%9C%93&q=root_template | 16:26 |
EmilienM | shardy: but I guess it's not easily doable now, since we run ansible from heat | 16:26 |
jtomasek | akrivoka: https://github.com/openstack/tripleo-common/search?utf8=%E2%9C%93&q=root_environment&type=Code | 16:27 |
*** iranzo has quit IRC | 16:27 | |
shardy | EmilienM: well we store the full stdout/stderr for the playbook run inside heat, so it may be possible | 16:28 |
akrivoka | jtomasek: the patch does that | 16:28 |
EmilienM | ccamacho, shardy: ok I found out something new, if scenarioXXX has a new service in ocata, the upgrade fails because it tries to stop it (it happens after validation). I'll fix it by using the same validation I did yesterday. I'm on it | 16:28 |
akrivoka | jtomasek: let me know if you find any remaining references to them, but I don't thing there should be any | 16:29 |
*** yprokule has quit IRC | 16:29 | |
jtomasek | akrivoka: ack, thanks | 16:29 |
shardy | EmilienM: which service? I thought the ansible service module took care of that | 16:29 |
ccamacho | EmilienM ack | 16:29 |
*** iranzo has joined #tripleo | 16:29 | |
EmilienM | dmsimard: can I steal your brain 5 min? I'm just wondering how we could have ara useful for our upgrade process | 16:29 |
shardy | e.g ignored a stopped service if it's already stopped | 16:29 |
EmilienM | shardy: collectd | 16:29 |
flaper87 | weshay: no pressure but, news? :P | 16:29 |
EmilienM | shardy: see http://logs.openstack.org/57/438657/6/check/gate-tripleo-ci-centos-7-scenario001-multinode-upgrades-nv/5f7740a/logs/subnode-2/var/log/messages | 16:29 |
EmilienM | shardy: grep "ould not find the requested service collectd: cannot che" | 16:29 |
dmsimard | EmilienM: I'm fighting with javascript but soon | 16:30 |
shardy | EmilienM: is the problem that it's not installed? | 16:30 |
EmilienM | dmsimard: ack. Maybe we can chat later in the week if you're too busy today | 16:30 |
shardy | perhaps we need to add it to the UpgradeInitCommand? | 16:30 |
dmsimard | EmilienM: nope, just gimme a few | 16:30 |
EmilienM | shardy: service wasn't installed in newton | 16:30 |
shardy | EmilienM: yeah we have a place to install it, sec | 16:31 |
weshay | flaper87, ya.. it just completed | 16:31 |
flaper87 | weshay: good or bad results ? | 16:31 |
shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/environments/major-upgrade-composable-steps.yaml#L12 | 16:32 |
shardy | EmilienM: I think we can add it here | 16:32 |
*** prateek has quit IRC | 16:32 | |
weshay | flaper87, bad.. I think it failed earlier | 16:33 |
shardy | https://review.openstack.org/#/c/438201/2/puppet/services/panko-api.yaml | 16:33 |
openstackgerrit | Merged openstack/tripleo-common master: Add pull-puppet-modules script to binaries https://review.openstack.org/422772 | 16:33 |
EmilienM | shardy: I'm confused, should we add everything we added in ocata? | 16:33 |
weshay | getting you access sec | 16:33 |
*** ipsecguy_ has joined #tripleo | 16:33 | |
flaper87 | weshay: why are you giving me bad news? jeeeeez | 16:33 |
shardy | EmilienM: Or perhaps a better way would be to install it in the service template | 16:33 |
*** ipsecguy has quit IRC | 16:33 | |
shardy | https://review.openstack.org/#/c/438201/2/puppet/services/panko-api.yaml | 16:33 |
*** panda|Zz is now known as panda|bbl | 16:33 | |
shardy | EmilienM: well, we could - all the services are in the ocata overcloud-full after all | 16:33 |
EmilienM | shardy: wait, can't we tell ansible to not fail if the service doesn't exist? | 16:33 |
shardy | EmilienM: yes, but if the service is enabled, we've got to install the packages | 16:34 |
shardy | at some point | 16:34 |
shardy | or puppet will just fail later | 16:34 |
EmilienM | shardy: yes, we shouldn't add packages like this imho (panko) | 16:34 |
EmilienM | puppet doesn't install packages? | 16:34 |
shardy | EmilienM: no, we noop the package provider | 16:34 |
shardy | by default | 16:34 |
EmilienM | ah right. | 16:34 |
*** artom has quit IRC | 16:34 | |
*** pcaruana has quit IRC | 16:34 | |
shardy | but for services added between releases, we have to install them before running puppet | 16:34 |
shardy | the upgrade_tasks seems like an OK place to do it IMO | 16:35 |
EmilienM | shardy: ok so like panko api? | 16:35 |
shardy | at it avoids installing a bunch of new services you may not have enabled | 16:35 |
EmilienM | shardy: problem is ansible tries to stop the services BEFORE upgrading/installing packages | 16:35 |
shardy | EmilienM: I think that's a reasonable approach, unless you disagree | 16:35 |
EmilienM | no I agree, np | 16:35 |
shardy | EmilienM: sure, just change the step | 16:35 |
EmilienM | ok | 16:35 |
EmilienM | I'll work on it and let you know progress | 16:35 |
shardy | EmilienM: thanks, sounds good | 16:36 |
*** artom has joined #tripleo | 16:36 | |
EmilienM | gfidente: where is it tested ? https://review.openstack.org/#/c/438036 | 16:37 |
*** dparkes has quit IRC | 16:37 | |
*** artom has quit IRC | 16:37 | |
*** prateek_ has quit IRC | 16:38 | |
gfidente | EmilienM well the env file isn't because even where we do hci we provide our own list of services, like in the scenarios | 16:38 |
gfidente | I am updating the validation py to check that with every commit though | 16:38 |
gfidente | sec | 16:38 |
shardy | flaper87, dprince: hey is anyone looking at moving haproxy into a container? | 16:41 |
shardy | to start with a minimal upgrade test, I need keystone, mysql, rabbit and haproxy | 16:41 |
*** iranzo has quit IRC | 16:42 | |
openstackgerrit | Flavio Percoco proposed openstack/tripleo-quickstart-extras master: Use openstack deploy to install the undercloud https://review.openstack.org/419040 | 16:43 |
EmilienM | shardy: Packaging things happen at step 3 while services stop at step 2. I don't think we can change this thing now. I would propose to 1. stop services only if they were enabled 2. Deploy the packages at step 3. - that way, upgrade will don't fail and will even deploy new services. Wdyt? | 16:43 |
flaper87 | shardy: yes, bandini's team | 16:43 |
*** gatuus has joined #tripleo | 16:44 | |
flaper87 | shardy: although, if you want to take a stab at it, I'm sure they won't mind | 16:44 |
shardy | EmilienM: sure that sounds OK, so you'll add a task that installs the service in step3? | 16:44 |
shardy | EmilienM: I was thinking we could just do it in an earlier step, but either approach works | 16:44 |
shardy | flaper87: ack, thanks - I'll leave it on baremetal for now and chat with bandini tomorrow | 16:45 |
EmilienM | shardy: if we do at an earlier step, it will upgrade everything else I think | 16:45 |
EmilienM | shardy: because of packaging dependencies | 16:45 |
shardy | EmilienM: Hmm, Ok yeah you're right | 16:45 |
*** dspano has joined #tripleo | 16:45 | |
EmilienM | and imho, packaging thing is very sensible, we should do it in one single step | 16:45 |
shardy | yeah, good point | 16:46 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo master: Add httpchk for http services https://review.openstack.org/439052 | 16:46 |
*** alop has joined #tripleo | 16:47 | |
EmilienM | and we didn't catch this use case before (for nova placement for example) because nova placement was running in httpd, and httpd was already there in newton (and even before) | 16:48 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo master: Add httpchk for http services https://review.openstack.org/439052 | 16:48 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Put service stop at step1 and quiesce at step2. https://review.openstack.org/438090 | 16:50 |
EmilienM | chem: rebasing so we don't have conflicts. I'm working on the same bits (not same topic but same lines) | 16:50 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist https://review.openstack.org/438657 | 16:50 |
flaper87 | weshay: not sure if you want to re-run it with the latest ps | 16:51 |
weshay | flaper87, aye | 16:52 |
weshay | I can | 16:52 |
*** pblaho has quit IRC | 16:52 | |
openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Enable composable upgrades for docker service templates https://review.openstack.org/438979 | 16:53 |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-quickstart master: modify basic test to use non-default non_root_user https://review.openstack.org/439055 | 16:54 |
openstackgerrit | Merged openstack/python-tripleoclient stable/mitaka: Updated from global requirements https://review.openstack.org/438912 | 16:59 |
openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh https://review.openstack.org/437411 | 16:59 |
*** saibarspeis has quit IRC | 17:01 | |
*** achadha has joined #tripleo | 17:01 | |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Update puppet-midonet https://review.openstack.org/438749 | 17:02 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Enable object-expirer on Swift proxy profile https://review.openstack.org/437541 | 17:02 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Add virtual_packages support to norpm provider https://review.openstack.org/437701 | 17:02 |
*** sdake_ has joined #tripleo | 17:03 | |
*** flepied has quit IRC | 17:04 | |
EmilienM | beagles: can you review https://review.openstack.org/#/c/438090/ and the OVS bits please? Is it fine to do it at step 1? | 17:04 |
*** zoli is now known as zoli|gone | 17:04 | |
openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Respect OS_LOG_CAPTURE env var https://review.openstack.org/439060 | 17:04 |
beagles | EmilienM: yup | 17:04 |
beagles | will do | 17:04 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 17:06 |
*** ramishra has quit IRC | 17:06 | |
rook | gfidente fultonj is there a way to add a action during the heat stack-delete to wipe the disks too? | 17:06 |
rook | that way things are cleaned up at that point too? | 17:06 |
*** sdake has quit IRC | 17:07 | |
*** dmacpher has joined #tripleo | 17:08 | |
gfidente | rook in ocata ironic will clean the disks when the node is deleted | 17:08 |
openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates master: Add ceilometer ipmi agent https://review.openstack.org/430447 | 17:08 |
gfidente | for newton we have to stick with the wipe script | 17:08 |
*** tesseract has quit IRC | 17:09 | |
*** cylopez has quit IRC | 17:10 | |
*** agurenko has quit IRC | 17:11 | |
*** sshnaidm is now known as sshnaidm|off | 17:12 | |
EmilienM | larsks: what is the package name for sensu-client? sensu? | 17:15 |
larsks | EmilienM: yes ("sensu") | 17:16 |
EmilienM | larsks: thx! | 17:16 |
*** jaganathan_afk has quit IRC | 17:18 | |
*** funzo has quit IRC | 17:19 | |
*** funzo has joined #tripleo | 17:20 | |
*** yamahata_ has quit IRC | 17:23 | |
*** achadha has quit IRC | 17:24 | |
*** sdake_ is now known as sdake | 17:24 | |
rook | gfidente: i can grow a ceph cluster with newton right?> | 17:24 |
rook | are there any gotchas? | 17:24 |
*** achadha has joined #tripleo | 17:24 | |
gfidente | rook you can yes | 17:25 |
*** stendulker has quit IRC | 17:26 | |
EmilienM | larsks: can I ask the same question for fluentd? | 17:26 |
EmilienM | larsks: (the client) | 17:27 |
larsks | EmilienM: "fluentd" https://github.com/openstack/tripleo-puppet-elements/blob/master/elements/overcloud-opstools/pkg-map | 17:27 |
rook | gfidente: can i start w/o a ceph deployment and add it later. | 17:27 |
EmilienM | larsks: thx for the url! | 17:27 |
gfidente | rook well that could work yes, but you lose data stored in the backends previously used by the services moving to ceph, for example cinder or glance | 17:28 |
rook | what if nothing gets added there | 17:28 |
gfidente | I don't think it's ever been meant to be tested | 17:28 |
rook | oh | 17:28 |
rook | well i much rather avoid that then | 17:28 |
*** karimb has quit IRC | 17:28 | |
*** b00tcat_ has quit IRC | 17:30 | |
*** trown is now known as trown|lunch | 17:30 | |
EmilienM | ccamacho, shardy: https://review.openstack.org/#/c/438657/ is ready for review (any feedback is welcome). I expect this patch fixing all scenarios upgrade jobs (unless I missed something new). Also, I'm not sure pingtest passed, we'll see | 17:30 |
openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Move horizon httpd to be stopped at step1 https://review.openstack.org/439069 | 17:31 |
*** b00tcat_ has joined #tripleo | 17:31 | |
openstackgerrit | Merged openstack-infra/tripleo-ci master: Add gate-tripleo-ci-centos-7-ovb-updates into promotion pipeline https://review.openstack.org/433796 | 17:31 |
*** tosky has quit IRC | 17:31 | |
*** lblanchard has joined #tripleo | 17:32 | |
EmilienM | ccamacho: I fixed it in https://review.openstack.org/438657 | 17:32 |
ccamacho | EmilienM yeah, just saw it :) | 17:32 |
EmilienM | ah :) cool :) | 17:32 |
EmilienM | ccamacho: it will be easier to backport | 17:32 |
ccamacho | agreed | 17:32 |
EmilienM | ccamacho: I think we are close now | 17:32 |
* ccamacho going home | 17:33 | |
*** ccamacho has quit IRC | 17:33 | |
EmilienM | bandini, ccamacho: did we have someone testing real clusters upgrades with pacemaker galera etc? | 17:33 |
chem | EmilienM: yep | 17:33 |
chem | EmilienM: bandini ccamacho me | 17:33 |
EmilienM | chem: how does it work? | 17:34 |
EmilienM | chem: because the upstream CI doesn't test it enough imho (we deploy pacemaker though but in single node) | 17:34 |
chem | EmilienM: I just did a upgrade of the controller and it went ok, I'm manually checking | 17:34 |
openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart-extras master: validate-simple: Use custom THT templates https://review.openstack.org/439071 | 17:35 |
chem | EmilienM: ah, I have also the only one node. I will do the multi node in a few | 17:35 |
chem | EmilienM: ah, I have also the only one node (pacemaker enabled). I will do the multi node in a few | 17:35 |
chem | EmilienM: few minutes | 17:35 |
chem | EmilienM: result in ~2h | 17:36 |
EmilienM | chem: ci jobs are running now. Any feedback on my patch would be welcome | 17:36 |
honza | jrist: i think the latest solution for the log file problem in gui logging is a winner :) https://review.openstack.org/#/c/395138/ | 17:38 |
jrist | sweeeeet | 17:39 |
chem | EmilienM: the "is deployed" patch ? | 17:39 |
jrist | honza: /me reads | 17:39 |
*** jpena is now known as jpena|away | 17:39 | |
honza | jrist: thanks for looking | 17:39 |
*** sudipto_ has quit IRC | 17:39 | |
*** sudipto has quit IRC | 17:39 | |
jrist | honza: tiny nit | 17:40 |
jrist | but worth adding, i think | 17:40 |
honza | all ears | 17:40 |
*** tobias_fiberdata has joined #tripleo | 17:40 | |
EmilienM | chem: https://review.openstack.org/#/c/438657/ | 17:40 |
jrist | honza: I'm about to add it in the spec | 17:41 |
jrist | honza: one question | 17:41 |
chem | EmilienM: I think I don't get how we can get in this state. If the aodh for instance is included then the config is done and it should be up. So I don't get how we get there | 17:41 |
jrist | honza: with your proposal | 17:41 |
honza | jrist: k, good | 17:41 |
jrist | honza: could it be reasonably realtime? | 17:41 |
jrist | honza: like could you go to a page in the gui or a dialog and see the log being generated? | 17:41 |
*** pblaho has joined #tripleo | 17:41 | |
jrist | oh | 17:41 |
chem | EmilienM: (yep the it's the "is deployed patch" :)) | 17:41 |
jrist | I guess it could be | 17:41 |
jrist | log locally but also the zaqar queue | 17:41 |
honza | jrist: yes | 17:41 |
chem | EmilienM: I must be missing something | 17:41 |
honza | jrist: except you'll lose the local data upon reload | 17:42 |
*** flepied has joined #tripleo | 17:42 | |
honza | (page reload) | 17:42 |
*** gatuus has quit IRC | 17:42 | |
jrist | honza: commented | 17:42 |
jrist | but also +1'd | 17:42 |
jrist | not -1 material | 17:42 |
EmilienM | chem: the use case is if someone adds a service in the environment during the upgrade | 17:43 |
EmilienM | chem: my patch will support that case and only try to stop aodh if the service was actually enabled | 17:43 |
EmilienM | chem: that's exactly the problem we hit in CI. Have you looked at scenarios issues we talked with shardy just before? | 17:43 |
openstackgerrit | Ben Kero proposed openstack/tripleo-quickstart-extras master: Add release notes for deprecation of multinic https://review.openstack.org/439066 | 17:43 |
bandini | chem, EmilienM: I am testing with three controllers. I had my first successful upgrade today (with a couple of non-merged yet reviews). will test some more tomorrow | 17:44 |
bandini | I will try some baremetal envs with ceph as well but more towards the end of the week | 17:44 |
EmilienM | bandini: can you link the patches please? | 17:44 |
*** derekh has quit IRC | 17:44 | |
bandini | sure | 17:45 |
chem | EmilienM: well for the existing one not sure it's required then and for "adding a service during an upgrade" then this idiom will be usefull as the step_config isn't "run" before upgrade | 17:46 |
chem | ? | 17:46 |
*** leanderthal is now known as leanderthal|afk | 17:46 | |
bandini | EmilienM: review.openstack.org/438090 review.openstack.org/438947 review.openstack.org/439025 | 17:47 |
EmilienM | chem: can you maybe comment in Gerrit? I'm confused on what you're asking for | 17:47 |
EmilienM | bandini: https://review.openstack.org/#/c/439025/1/puppet/services/pacemaker/rabbitmq.yaml | 17:47 |
EmilienM | bandini: where are steps? | 17:47 |
*** dsneddon has joined #tripleo | 17:47 | |
*** gatuus has joined #tripleo | 17:48 | |
*** sdake_ has joined #tripleo | 17:48 | |
EmilienM | bandini: also see my comment on https://review.openstack.org/#/c/438947/ - CI failed for a valid reason, galera is not started | 17:49 |
bandini | EmilienM: well besides that "conversion" at step0, puppet/services/pacemaker.yaml will take care of stopping and starting everything | 17:49 |
EmilienM | bandini: excellent | 17:49 |
*** sdake has quit IRC | 17:50 | |
EmilienM | bandini: CI job looks passing, I confirm the upgrade of RabbitMQ went well | 17:50 |
bandini | EmilienM: I replied on https://review.openstack.org/#/c/438947/, we first need that ansible-pacemaker change to land | 17:51 |
bandini | EmilienM: unless you tell me I can depends-on on a separate gerrithub instance, in which case I will add a depends-on | 17:51 |
EmilienM | matbu: can you look ^ when you can? | 17:51 |
EmilienM | matbu: also please make sure to release the module for downstream packaging | 17:52 |
*** akrivoka has quit IRC | 17:52 | |
EmilienM | bandini: I don't want to merge it now or it will break upgrade job | 17:53 |
bandini | EmilienM: sounds good. I will do some more testing here and report in the reviews | 17:53 |
bandini | need to feed the kids. bbiab | 17:53 |
*** fragatina has joined #tripleo | 17:55 | |
*** fragatina has quit IRC | 17:57 | |
*** fragatina has joined #tripleo | 17:58 | |
openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates master: Align hyperconverged-ceph.yaml environment and adds some validation https://review.openstack.org/439081 | 17:58 |
chem | bandini: for https://review.openstack.org/#/c/438091/ I first I though I could re-use the non-pacemaker upgrade rule into the pacemaker one using some heat magic, but after seeing your patch for haproxy, I think we should copy/paste what need to be done | 17:58 |
chem | bandini: so I'm going over the whole list and "backport" what need to be backported for pacemaker related service upgrade | 17:59 |
gfidente | EmilienM this adds the validation checks https://review.openstack.org/#/c/439081 | 18:00 |
*** achadha_ has joined #tripleo | 18:00 | |
EmilienM | gfidente: fun | 18:01 |
gfidente | EmilienM yeah it's not but at least we test it, as you pointed | 18:01 |
EmilienM | gfidente: by testing I meant actual CI jobs where we test this arch | 18:02 |
*** sdake_ has quit IRC | 18:02 | |
gfidente | so the arch is tested by the scenarios yes, we have a single node there and we do hci | 18:02 |
gfidente | the environment file wasn't tested before, with the validations it is | 18:02 |
bandini | chem: ack, let's have a chat/call about this tomorrow as well? | 18:03 |
EmilienM | gfidente: ok cool, thanks | 18:03 |
*** achadha has quit IRC | 18:04 | |
*** lucasagomes is now known as lucas-afk | 18:05 | |
*** sdake has joined #tripleo | 18:05 | |
*** jpich has quit IRC | 18:06 | |
*** chlong has quit IRC | 18:06 | |
openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci master: Use custom network-isolation-absolute.yaml in ipv4 net-iso jobs https://review.openstack.org/422297 | 18:07 |
openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci master: Add storage-mgmt network to compute nodes https://review.openstack.org/419049 | 18:07 |
*** achadha_ has quit IRC | 18:12 | |
*** fzdarsky is now known as fzdarsky|afk | 18:13 | |
*** ckyriakidou has quit IRC | 18:13 | |
openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci master: Use custom network-isolation-absolute.yaml in ipv4 net-iso jobs https://review.openstack.org/422297 | 18:14 |
openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci master: Add storage-mgmt network to compute nodes https://review.openstack.org/419049 | 18:14 |
*** ckyriakidou has joined #tripleo | 18:14 | |
mwhahaha | bnemec: fyi https://review.openstack.org/#/c/431185/ caused https://bugs.launchpad.net/tripleo/+bug/1668709 so we might want to revert it | 18:19 |
openstack | Launchpad bug 1668709 in tripleo "openstack undercloud install no longer logs to a file" [High,Confirmed] | 18:19 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist https://review.openstack.org/438657 | 18:19 |
openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: DON'T MERGE: test tempest all branches https://review.openstack.org/436923 | 18:20 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: upgrades: fix ec2api conditional https://review.openstack.org/439037 | 18:20 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Put service stop at step1 and quiesce at step2. https://review.openstack.org/438090 | 18:20 |
*** pkovar has quit IRC | 18:21 | |
*** sdake has quit IRC | 18:21 | |
*** eggmaster has joined #tripleo | 18:22 | |
chem | ah crap, this commit https://review.openstack.org/#/c/439037 got an unexpected rebase, I'm afraid that shardy ccamacho need to +2 again | 18:22 |
*** sdake has joined #tripleo | 18:23 | |
*** gfidente is now known as gfidente|afk | 18:24 | |
skramaja | shardy: CI passed for https://review.openstack.org/#/c/438495/, ptal | 18:24 |
bnemec | mwhahaha: Bleh :-( | 18:24 |
*** pkovar has joined #tripleo | 18:26 | |
shardy | skramaja: lgtm, thanks! | 18:26 |
skramaja | thanks shardy | 18:26 |
skramaja | dsneddon: i have modified ovs restart after ifdown - https://review.openstack.org/#/c/434255/ ptal | 18:27 |
*** milan has quit IRC | 18:27 | |
dsneddon | skramaja, Do you think this will probably apply only to ifcfg-based systems? I'm not sure if this would ever come in handy for a system that uses /etc/network/interfaces (Debian-based). | 18:28 |
*** ckyriakidou has quit IRC | 18:28 | |
*** hewbrocca is now known as hewbrocca_afk | 18:29 | |
dsneddon | skramaja, The other option would be to put the restart method into os_net_config.utils, but I think this is probably fine as-is. | 18:29 |
skramaja | dsneddon: dpdk in debian is still a different package. | 18:29 |
skramaja | dsneddon: yes utils might be an option, shall i take it in follow-up, i got this to be backported for newton. | 18:30 |
dsneddon | skramaja, I'll give it a review, that might not be necessary. | 18:30 |
skramaja | dsneddon: ok.. thanks. | 18:31 |
openstackgerrit | Chandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow https://review.openstack.org/431916 | 18:32 |
EmilienM | chem: arf, I was fixing the horizon stuff in my patch, like I said in the review | 18:32 |
EmilienM | I need to rebase and kill ci jobs now | 18:32 |
chem | EmilienM: ah, I though you wanted to do in that review. | 18:33 |
*** dtantsur is now known as dtantsur|afk | 18:34 | |
EmilienM | nm | 18:34 |
openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Add missing upgrade step for pacemaker controlled resources. https://review.openstack.org/439095 | 18:34 |
EmilienM | np | 18:34 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: upgrades/validation: only run validation when services exist https://review.openstack.org/438657 | 18:35 |
*** karimb has joined #tripleo | 18:35 | |
openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Revert "Turn off propagation for undercloud logger" https://review.openstack.org/439097 | 18:35 |
*** karimb has quit IRC | 18:35 | |
chem | bandini: I've added this on top of your haproxy patch https://review.openstack.org/#/c/439095/ | 18:36 |
openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Revert "Turn off propagation for undercloud logger" https://review.openstack.org/439097 | 18:37 |
bnemec | mwhahaha: ^ | 18:37 |
bnemec | I'll figure out the proper fix when I have time. | 18:37 |
bnemec | I may just switch to oslo.log since that probably handles this correctly. | 18:37 |
chem | bandini: and then I've abandonned https://review.openstack.org/#/c/438091/ | 18:38 |
*** dprince has quit IRC | 18:38 | |
chem | oki, need to leave, bye guy | 18:38 |
*** artom has joined #tripleo | 18:39 | |
dsneddon | skramaja, OK, there is only one small change required, just a fix to the import statement. | 18:41 |
skramaja | ok dsneddon | 18:41 |
dsneddon | skramaja, Once you make that change, I'll +2 and try to get a couple other reviews. | 18:41 |
skramaja | dsneddon: thanks. | 18:42 |
dsneddon | skramaja, I'm glad we were able to meet and chat at the PTG. Let me know if you want me to carve out any time to discuss any remaining tasks to get DPDK working. | 18:44 |
*** milan has joined #tripleo | 18:44 | |
skramaja | sure dsneddon, i will keep you posted. | 18:47 |
*** trown|lunch is now known as trown | 18:47 | |
bandini | chem: ack, let's chat tomorrow | 18:48 |
skramaja | and thanks for all your inputs. makes our work lot easier, dsneddon | 18:48 |
chem | bandini: oki (leaving for real now) | 18:48 |
*** thrash is now known as thrash|biab | 18:48 | |
dsneddon | skramaja, Glad to help, and thank you for putting in all this effort to make this work, this is one of the top priorities for many of the customers and partners I work with. | 18:49 |
*** bkopilov has quit IRC | 18:49 | |
*** yamahata has joined #tripleo | 18:49 | |
*** chem is now known as chem_gone | 18:49 | |
*** milan has quit IRC | 18:50 | |
openstackgerrit | Saravanan KR proposed openstack/os-net-config master: Restart Openvswitch after adding DPDK ports https://review.openstack.org/434255 | 18:51 |
skramaja | dsneddon: ^ | 18:52 |
* dsneddon reviews skramaja's patch... | 18:52 | |
dsneddon | skramaja, no need to do this now, but I think it might be best to do a follow-on patch for OVS 2.7 that actually checks the version of OVS installed and does a restart if the version <= 2.6. | 18:53 |
dsneddon | skramaja, Rather than just removing the restart once OVS 2.7 comes out. That gives us the flexibility to do a backport of a newer os-net-config to an older version of OSP that uses an older OVS. | 18:53 |
dsneddon | skramaja, Did you test this latest patchset on bare metal with a DPDK enabled NIC? | 18:55 |
skramaja | dsneddon: yeah. make sense. i will do that on a follow up along with moving the restart logic to utils. | 18:55 |
skramaja | dsneddon: yes we are testing only in baremetals. | 18:55 |
dsneddon | skramaja, No, after review I no longer think the OVS restart logic would apply to impl_eni, so I think you put it in the right place for now. | 18:55 |
skramaja | ok got it. | 18:56 |
*** nyechiel has quit IRC | 18:56 | |
skramaja | dsneddon: i will just add version check to avoid restart for ovs2.7 onwards | 18:57 |
dsneddon | skramaja, Yes, perfect. I'm going to see if I can get a couple more reviews so you can do the backport ASAP. | 18:57 |
skramaja | thanks. | 18:58 |
dsneddon | EmilienM, bnemec, beagles: If you have a minute, can you look at this simple and semi-urgent review? https://review.openstack.org/#/c/434255 | 18:58 |
EmilienM | I will look after my lunch | 18:59 |
*** achadha has joined #tripleo | 18:59 | |
beagles | dsneddon: done | 19:00 |
dsneddon | beagles, Thanks! | 19:00 |
dsneddon | skramaja, i'll give in an hour or two to hopefully get one more review, otherwise I'll merge this after my lunch. | 19:01 |
skramaja | shardy: in https://review.openstack.org/#/c/438495/, there is a small issue the current code. it is gone to the gate, is it possible to defer it? actually the issue is even grep will also return error "1" when the string stirg is not found. | 19:02 |
skramaja | shardy: it does not affect as the issue will be as is. if it can't, i will add another patch. | 19:03 |
shardy | skramaja: I think you can just push another revision | 19:03 |
*** achadha has quit IRC | 19:03 | |
shardy | grep -q ? | 19:03 |
shardy | and check the return code? | 19:03 |
skramaja | ahh ok. | 19:04 |
shardy | if hiera -c /etc/puppet/hiera.yaml service_names | grep -q pacemaker | 19:04 |
*** achadha has joined #tripleo | 19:04 | |
skramaja | ok | 19:04 |
*** bkopilov has joined #tripleo | 19:05 | |
*** achadha has quit IRC | 19:06 | |
*** achadha has joined #tripleo | 19:06 | |
*** dprince has joined #tripleo | 19:08 | |
shardy | skramaja: Hey FYI I've been looking a little at the workflow for parameter calculation, as we discussed last week | 19:08 |
shardy | skramaja: I've been thinking we could perhaps use your patch https://review.openstack.org/#/c/437956/ to run a mistral workflow in the service template via https://review.openstack.org/#/c/420664/ when it is ready | 19:09 |
shardy | then we can have a workflow which takes a role and some parameters, and retrieves the introspection data and does some calculation per-service per-role | 19:10 |
shardy | it could still use either yaql or a custom action with some pythong | 19:10 |
skramaja | shardy: with this the order cannot be fixed right? | 19:10 |
skramaja | like dpdk should go first and then osd. is it possible? | 19:10 |
*** pkovar has quit IRC | 19:11 | |
shardy | skramaja: Hmm, the order would be defined by the e.g ControllerServices list order I think | 19:11 |
shardy | skramaja: but composable services aren't supposed to care about ordering, that's kind of an implementation detail | 19:11 |
shardy | skramaja: the other option is to explicitly add an interface to run pre-deploy workflows | 19:11 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 19:12 |
shardy | that may fit better with fultonj's requirements around ceph/tendrl, but it'll be more difficult to wire in the modified parameters to e.g the dpdk service | 19:12 |
*** rlandy is now known as rlandy|brb | 19:12 | |
shardy | skramaja: I'll try to share an example later this week then we can discuss further, there may be a better approach, open to ideas :) | 19:13 |
*** amoralej is now known as amoralej|off | 19:13 | |
openstackgerrit | Saravanan KR proposed openstack/tripleo-heat-templates master: Disable exit on error for pacemaker commands for update flow https://review.openstack.org/438495 | 19:14 |
shardy | skramaja: basically I was trying to find a way to define the parameter calculation in the service template, but if that's not possible we'll have to go back to the idea of a standalone pre-deploy mistral workflow, e.g not integrated with t-h-t | 19:14 |
shardy | it'd be nice for integrators if the data was in one place though | 19:15 |
*** Lokesh_Jain_ has joined #tripleo | 19:15 | |
rook | is there a easy way to figure out which node the ping failed to work with? I have the IP that failed, but trying to map that to a specific node so i can check its networking out. | 19:16 |
*** salmankhan has quit IRC | 19:16 | |
*** noslzzp has quit IRC | 19:16 | |
skramaja | shardy: https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/services.yaml#L36 | 19:16 |
Lokesh_Jain_ | Hi.. I am looking to automate SRIOV deployment.. is there a way to automate VF creation when the interfaces are known beforehand? | 19:16 |
shardy | skramaja: Ok yeah, so it's unordered | 19:16 |
shardy | unless we remove that | 19:17 |
skramaja | yes shardy | 19:17 |
*** fragatina has quit IRC | 19:17 | |
skramaja | shardy: i guess standalone also was the as from UI | 19:17 |
skramaja | because in UI they wanted to provide confirmation | 19:18 |
skramaja | before deploying the genereated paramters | 19:18 |
skramaja | jtomasek: ^ | 19:18 |
shardy | skramaja: Ok, I think I missed the confirmation requirement last week | 19:18 |
*** sdake has quit IRC | 19:19 | |
skramaja | shardy: i actually wanted to talk to you regarding the stack flattening change (from javascript to tripleo-common). i wanted to work on it. | 19:19 |
skramaja | if you have not started already :) | 19:19 |
shardy | skramaja: sure, please go ahead | 19:20 |
shardy | I took a quick look but not started anything | 19:20 |
shardy | I suspect it can be done via a mistral workflow with a yaql expression, but we might want a custom action if that's faster | 19:20 |
skramaja | ofcource i will need your guidance, but doing it will give me indepth understanding of stack execution. | 19:20 |
shardy | if you look in t-h-t we're doing a bunch of flatten/groupBy queries which I think are pretty similar to what's needed here | 19:21 |
shardy | skramaja: sure, please feel free, let me know if I can help | 19:21 |
skramaja | shardy: ok.. sure. i will look into it the samples and will come back to you.. right now neck depth in ovs2.6 migration. | 19:23 |
shardy | skramaja: sure, sounds good, thanks! | 19:23 |
*** chlong has joined #tripleo | 19:25 | |
*** rlandy|brb is now known as rlandy | 19:26 | |
dprince | mandre, Slower, flaper87: wierd error on the undercloud-containers job. WOndering if the Amazon S3 outage is effecting docker-hub perhaps | 19:26 |
flaper87 | dprince: it is, docker hub is down | 19:27 |
flaper87 | dprince: can't download images or push them | 19:27 |
EmilienM | amazon s3 outage is only one zone (experimental afik) | 19:28 |
*** sdake has joined #tripleo | 19:28 | |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates stable/ocata: Adding keystone parameters for Tacker https://review.openstack.org/439125 | 19:28 |
EmilienM | docker-hub relies on a single s3 zone? | 19:28 |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates stable/ocata: Adding keystone parameters for Congress https://review.openstack.org/439127 | 19:28 |
dprince | EmilienM: very suspicious, my music stopped playing, docker hub is down. I think I'm just going to take the rest of the day off | 19:28 |
EmilienM | lol | 19:29 |
EmilienM | wait spotify is down? | 19:29 |
* dprince has since restarted music from Apple servers instead | 19:29 | |
dprince | EmilienM: don't know about spotify, it might be fine | 19:29 |
EmilienM | dprince: tripleo ci is maybe down because S3, we don't know | 19:30 |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates stable/ocata: Adding keystone parameters for Congress https://review.openstack.org/439127 | 19:33 |
slagle | anyone know what happened to the warez server? | 19:33 |
slagle | sorry, ww | 19:33 |
openstackgerrit | Dan Radez proposed openstack/tripleo-heat-templates stable/ocata: Adding keystone parameters for Tacker https://review.openstack.org/439125 | 19:34 |
bkero | win 86 | 19:34 |
EmilienM | slagle: AFIK the process crassed during the tripleo promotion, you can restart it with puppet | 19:35 |
rook | gfidente|afk dtantsur|afk is there a way to nuke the root disk? | 19:38 |
*** shardy has quit IRC | 19:38 | |
slagle | EmilienM: i was making a haha about the s3 outage :) | 19:39 |
*** dougbtv_ is now known as dougbtv | 19:39 | |
slagle | EmilienM: maybe you were too :) | 19:39 |
EmilienM | slagle: yeah I know :P | 19:39 |
EmilienM | slagle: but still, please restart the warez server | 19:39 |
slagle | it's on its way up | 19:40 |
slagle | productivity will resume | 19:40 |
EmilienM | :D | 19:40 |
*** fragatina has joined #tripleo | 19:42 | |
* trown did not know about the DCC feature of openstackbot | 19:43 | |
*** thrash|biab is now known as thrash | 19:44 | |
openstackgerrit | Lars Kellogg-Stedman proposed openstack/tripleo-quickstart master: define non_root_group explicitly https://review.openstack.org/439133 | 19:45 |
*** skramaja has quit IRC | 19:46 | |
openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart-extras master: Use pull-puppet-modules from upstream https://review.openstack.org/423190 | 19:46 |
*** fragatina has quit IRC | 19:47 | |
*** fragatina has joined #tripleo | 19:50 | |
*** fragatina has quit IRC | 19:50 | |
*** fragatina has joined #tripleo | 19:51 | |
bnemec | Hmm, only 15 check-triple jobs running? http://zuul-status.nemebean.com/ | 19:52 |
bnemec | Something seems to be amiss. | 19:52 |
*** cdearborn has joined #tripleo | 19:52 | |
*** tobias_fiberdata has quit IRC | 19:53 | |
EmilienM | bnemec: things look pretty green http://tripleo.org/cistatus.html | 19:54 |
bnemec | EmilienM: Yeah, but new jobs aren't spawning so the queue just keeps growing. | 19:56 |
bnemec | I see pabelanger noted some issues starting Jenkins slaves on rh1. | 19:56 |
*** dspano has quit IRC | 19:57 | |
dmsimard | stevebaker, therve: o/ I'm a complete noob to heat-agents, have a question | 20:01 |
dmsimard | suppose I want to add a toggle (disabled by default?) in https://github.com/openstack/heat-agents/blob/master/heat-config-ansible/install.d/50-heat-config-hook-ansible -- is there a way to do that ? | 20:01 |
stevebaker | dmsimard: hey | 20:02 |
dmsimard | It would be to add the ARA Ansible callback ( provides reports that look like this: http://46.231.133.111/reports/index.html ) | 20:02 |
stevebaker | dmsimard: what does the toggle do? | 20:03 |
dmsimard | Well I'm just not sure ARA should be used by default | 20:03 |
stevebaker | dmsimard: this is the actual hook https://github.com/openstack/heat-agents/blob/master/heat-config-ansible/install.d/hook-ansible.py | 20:03 |
dmsimard | So it'd be to enable ARA | 20:03 |
stevebaker | dmsimard: so you want to make the hook itself ARA-aware? | 20:04 |
dmsimard | ara is a native ansible callback, ara needs to be installed and then ansible just needs to be configured to look for it | 20:04 |
dmsimard | to configure ansible to look for the ara callback, it can be done by the ANSIBLE_CALLBACK_PLUGINS envvar or through an ansible.cfg file | 20:05 |
*** dsariel has joined #tripleo | 20:05 | |
*** mcornea has quit IRC | 20:05 | |
dmsimard | then it'll automatically collect data when the ansible-playbook command runs | 20:05 |
dmsimard | and the report can be generated when it playbook run is complete with a command | 20:06 |
dmsimard | s/it/the/ | 20:06 |
*** noslzzp has joined #tripleo | 20:06 | |
stevebaker | it sounds like you want to add that logic to hook-ansible.py, and enable it in the heat template by setting a config resource 'options' value https://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Heat::SoftwareConfig-prop-options | 20:06 |
*** noslzzp has quit IRC | 20:07 | |
*** gatuus has quit IRC | 20:07 | |
*** sdake has quit IRC | 20:07 | |
stevebaker | dmsimard: the hook already does other things with 'options' https://github.com/openstack/heat-agents/blob/master/heat-config-ansible/install.d/hook-ansible.py#L53 | 20:07 |
*** noslzzp has joined #tripleo | 20:07 | |
dmsimard | right, but the actual software installation happens in that bash file yes ? | 20:07 |
*** sdake has joined #tripleo | 20:08 | |
*** ukalifon has quit IRC | 20:08 | |
bnemec | EmilienM: Looks like it was a nodepool problem. Infra restarted it and now nodes are spawning again. | 20:09 |
EmilienM | oops | 20:09 |
bnemec | Which probably means we're about to crash rh1 with all the new jobs starting, but what can you do? | 20:09 |
dmsimard | stevebaker: the most "logic" I've found to determine what to install was here https://github.com/openstack/heat-agents/blob/d802d177c722718136d97915d28db133c5499bc6/heat-config-docker-compose/install.d/50-heat-config-hook-docker-compose | 20:10 |
bnemec | Hmm, not getting fips though. | 20:11 |
bnemec | That's also going to be a problem. | 20:11 |
stevebaker | dmsimard: ah, 50-heat-config-hook-ansible does the installation for diskimage-builder without RPMs, but I bet basically nobody uses that. Whatever you do to install ARA in 50-heat-config-hook-ansible will also need to be done in RDO packaging (is ARA packaged in RDO?) | 20:11 |
mwhahaha | EmilienM: https://review.openstack.org/#/c/439052/ fyi passes scenario 004 now | 20:11 |
EmilienM | mwhahaha: yeah I saw, I'll look after the TC meeting | 20:12 |
*** sdake has quit IRC | 20:13 | |
dmsimard | stevebaker: ARA isn't packaged yet (there's actually a bz for it now https://bugzilla.redhat.com/show_bug.cgi?id=1426193 ), only on pypi. It will probably be packaged by the end of pike -- but to begin with we wanted to explore the opportunities in the context of CI. | 20:13 |
openstack | bugzilla.redhat.com bug 1426193 in Package Review "Review Request: ara - Ansible Run Analysis" [Unspecified,New] - Assigned to apevec | 20:13 |
dmsimard | So that'd probably mean installing ARA through pip in the ansible heat agent or something like that -- except I would have preferred to keep it toggled off by default | 20:14 |
*** sdake has joined #tripleo | 20:14 | |
stevebaker | dmsimard: ok, then you could pip install it in 50-heat-config-hook-ansible and test that by building images with diskimage-builder locally, but in CI you'll likely be pip installing from CI scripts | 20:15 |
*** athomas has quit IRC | 20:15 | |
*** gchamoul is now known as gchamoul|afk | 20:15 | |
dmsimard | stevebaker: so we don't use dib images in gate ? | 20:15 |
dmsimard | stevebaker: otherwise where would that end up occuring in CI scripts ? | 20:16 |
dmsimard | I mean, ara is already set up to run from devstack-gate and ooo-q -- this is particularly for the heat agent portion which I understand takes care of the upgrades | 20:17 |
*** panda|bbl is now known as panda | 20:18 | |
stevebaker | dmsimard: yeah, I think dib is used throughout infra CI, but I'm not sure for a given job when it is appropriate to install during image build or during runtime, maybe ask qa or infra folk? | 20:22 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Update puppet-midonet https://review.openstack.org/438749 | 20:22 |
openstackgerrit | Flavio Percoco proposed openstack/tripleo-heat-templates master: Containerize panko https://review.openstack.org/439138 | 20:22 |
*** pkovar has joined #tripleo | 20:22 | |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Add virtual_packages support to norpm provider https://review.openstack.org/437701 | 20:22 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Enable object-expirer on Swift proxy profile https://review.openstack.org/437541 | 20:22 |
stevebaker | dmsimard: what happens when ARA is enabled in an ansible run but not installed? If its not a hard failure-causing dependency then maybe the logic can be added to the hook, but installing ara can be documented in the README? | 20:26 |
dmsimard | stevebaker: nothing happens if the path doesn't exist | 20:26 |
dmsimard | no exceptions or failures afaik | 20:26 |
dmsimard | We could do that -- add an option to toggle the ara callback in the hook -- and leave the ara installation as an exercise to the reader | 20:27 |
stevebaker | dmsimard: yes, and it sounds like you already hava ara installed where you need it | 20:27 |
dmsimard | stevebaker: ok I'll give a shot at a patch | 20:28 |
stevebaker | dmsimard: sounds good | 20:29 |
stevebaker | bnemec: "nova rescue --image", "nova unrescue" may well do the job for switching to an iPXE image | 20:30 |
*** salmankhan has joined #tripleo | 20:30 | |
dmsimard | stevebaker: what are the odds of a dummy depends-on patch (from say, T-H-T) would pick up that heat-agent change ? | 20:32 |
dmsimard | to test the implementation | 20:32 |
bnemec | stevebaker: I've started the process of rebuilding my OVB cloud to test with recent nova. I'm not sure what version will be required, but I figure I'll start with latest and work backwards. | 20:32 |
stevebaker | bnemec: it looks like (un)rescue has been there since at least Kilo, so we should be good | 20:33 |
stevebaker | bnemec: I can try on a kilo cloud once you have something that might work | 20:33 |
bnemec | stevebaker: Sounds good. I'm still hoping we can drop kilo support sometime soon. :-) | 20:34 |
*** liverpooler has joined #tripleo | 20:34 | |
*** salmankhan has quit IRC | 20:35 | |
*** jprovazn has quit IRC | 20:35 | |
stevebaker | dmsimard: is this for local testing or automated testing? If its local testing its probably easiest to just drop your copy of the hook into your overcloud node before running - its only one file. Otherwise once the change has landed it will end up in RDO DLRN repos fairly quickly | 20:36 |
dmsimard | stevebaker: automated testing -- but I figured I'd test it before trying to land the patch :) | 20:37 |
dmsimard | I probably have a way to test it, I'll check it out. | 20:39 |
stevebaker | dmsimard: I see, I'm not sure which jobs build an image first, it would be nice if DependsOn: does the right thing but I've always assumed that image building happens out-of-band from the current job but I could be wrong | 20:39 |
dmsimard | yeah that confused me quickly too :) | 20:40 |
bnemec | stevebaker: Are these ovb jobs? Those build images if any project is in the dependency chain that has an impact on image building. | 20:41 |
bnemec | Also, even for projects that don't impact image build we update the cached images with any changes. | 20:41 |
bnemec | Although I guess if it's a dib element you're updating that would fall under the "needs image build" category. | 20:42 |
stevebaker | bnemec: its only laid out as a dib element, I think basically all consumers will be installing via RPM | 20:43 |
*** liverpooler has quit IRC | 20:43 | |
*** liverpooler has joined #tripleo | 20:44 | |
*** dsariel has quit IRC | 20:45 | |
*** gfidente|afk is now known as gfidente | 20:45 | |
bnemec | load average: 64.61, 54.34, 32.13 | 20:46 |
bnemec | Ouch | 20:46 |
*** sdake_ has joined #tripleo | 20:46 | |
jrist | aieee | 20:47 |
*** athomas has joined #tripleo | 20:47 | |
*** ansmith has quit IRC | 20:47 | |
*** sdake has quit IRC | 20:50 | |
openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart-extras master: Use openstack deploy to install the undercloud https://review.openstack.org/419040 | 20:51 |
*** dsariel has joined #tripleo | 20:58 | |
dmsimard | stevebaker: do you know how the options are passed from t-h-t ? | 20:58 |
dmsimard | like say I'm looking at http://git.openstack.org/cgit/openstack/tripleo-heat-templates/tree/puppet/services/nova-api.yaml#n185 | 20:59 |
stevebaker | dmsimard: like this https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/upgrade_config.yaml#L38 | 21:00 |
dmsimard | ahh, thanks, you're a pro | 21:01 |
stevebaker | dmsimard: thats the kicker, you can't specify options in the composable services config | 21:01 |
dmsimard | right, but for example there's a default module_path | 21:01 |
dmsimard | that's exactly what I was looking for | 21:01 |
stevebaker | ok | 21:01 |
openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 21:01 |
*** ansmith has joined #tripleo | 21:07 | |
*** ansmith has quit IRC | 21:08 | |
*** bandini has quit IRC | 21:12 | |
*** bandini has joined #tripleo | 21:12 | |
*** tbonds has quit IRC | 21:20 | |
*** ansmith has joined #tripleo | 21:20 | |
*** ansmith has quit IRC | 21:20 | |
*** bfournie has quit IRC | 21:24 | |
*** artom has quit IRC | 21:27 | |
*** chlong has quit IRC | 21:27 | |
*** artom has joined #tripleo | 21:32 | |
*** artom has quit IRC | 21:32 | |
*** artom has joined #tripleo | 21:33 | |
*** dsneddon has quit IRC | 21:35 | |
openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates master: Containerize gnocchi services https://review.openstack.org/439170 | 21:35 |
openstackgerrit | Guil Barros proposed openstack/puppet-tripleo stable/newton: Add Contrail settings to haproxy.pp https://review.openstack.org/439172 | 21:37 |
*** chlong has joined #tripleo | 21:39 | |
*** ckyriakidou has joined #tripleo | 21:41 | |
openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates master: Containerize panko api service https://review.openstack.org/439138 | 21:44 |
*** dparkes has joined #tripleo | 21:44 | |
openstackgerrit | David Moreau Simard proposed openstack-infra/tripleo-ci master: Do not merge: Test python-heat-agents package build https://review.openstack.org/439178 | 21:44 |
*** sdake_ has quit IRC | 21:45 | |
*** fultonj has quit IRC | 21:45 | |
openstackgerrit | David Moreau Simard proposed openstack/tripleo-heat-templates master: Do not merge: Test python-heat-agents package build https://review.openstack.org/439179 | 21:46 |
*** gfidente is now known as gfidente|afk | 21:46 | |
*** jayg is now known as jayg|g0n3 | 21:47 | |
*** fultonj has joined #tripleo | 21:47 | |
weshay | leifmadsen, someone just pointed me at toad, very cool man! +1 | 21:47 |
leifmadsen | weshay: :) | 21:48 |
leifmadsen | getting there... a little side project | 21:48 |
weshay | looks good and fun | 21:48 |
leifmadsen | we (yolanda, me) wants to move it into openstack-infra, and then figure out a way of having it consume oooq in different scenarios | 21:48 |
openstackgerrit | Guil Barros proposed openstack/puppet-tripleo stable/newton: Add Contrail settings to haproxy.pp https://review.openstack.org/439172 | 21:49 |
leifmadsen | that way, we have different setups that we can push through oooq that we run into with partners etc | 21:49 |
leifmadsen | yolanda just wrote up a blog post today on doing oooq+baremetal that I need to step through in my home lab | 21:49 |
weshay | oh nice | 21:50 |
weshay | ya.. that *all* sounds very good to me | 21:50 |
leifmadsen | TOAD runs on a virthost, that uses itself as a jenkins slave to deploy the undercloud via oooq, then deploys a baremetal overcloud | 21:50 |
leifmadsen | I've already gotten the "all in one" scenario works with virthost hosting both TOAD + oooq virtual install | 21:50 |
openstackgerrit | Merged openstack/tripleo-quickstart master: Update repo and image links for master release config https://review.openstack.org/438582 | 21:50 |
leifmadsen | so you install TOAD, load up jenkins, push the easy button, and your virthost then has a virtual under/overcloud install | 21:51 |
weshay | I know some people that will be very interested in your work :) | 21:51 |
weshay | ya.. cool | 21:51 |
leifmadsen | it's still rough around the edges, but the idea being that we want to make it stupid simple to spin up a test/dev environment with one command on the console | 21:51 |
leifmadsen | the whole ELK stuff right now is basically abandoned in favour of just getting everything else easy easy easy | 21:52 |
leifmadsen | but we'll circle back around to that. Might be able to leverage all the stuff browbeat has done and just consume their playbooks | 21:52 |
openstackgerrit | Dirk Mueller proposed openstack/diskimage-builder master: [suse] Handle pip-and-virtualenv install for opensuse https://review.openstack.org/437071 | 21:52 |
*** jkilpatr has quit IRC | 21:52 | |
leifmadsen | but +∞ to oooq | 21:52 |
weshay | leifmadsen, ya.. bringing in browbeat should be fairly simple | 21:53 |
leifmadsen | yep | 21:53 |
leifmadsen | once I get the baremetal stuff solid and documented, browbeat is up next | 21:53 |
weshay | leifmadsen, you can see all their runs here.. https://thirdparty-logs.rdoproject.org/ | 21:53 |
leifmadsen | yea, been in close contact with browbeat guys. I helped get some of their documentation ported to RTD | 21:53 |
leifmadsen | and built via CI | 21:54 |
weshay | very cool | 21:54 |
leifmadsen | very cool stuff there | 21:54 |
leifmadsen | lol | 21:54 |
leifmadsen | we're also lightly consuming the ARA stuff from dmsimard | 21:54 |
leifmadsen | we're all about consuming :) | 21:54 |
leifmadsen | we try not to reinvent any wheels | 21:54 |
weshay | ya.. ara is awesome as well | 21:54 |
*** athomas has quit IRC | 21:54 | |
weshay | opensource in practice :) | 21:54 |
*** bfournie has joined #tripleo | 21:55 | |
*** yamahata has quit IRC | 21:55 | |
leifmadsen | indeed | 21:55 |
dmsimard | leifmadsen: I gotta show you the new ARA UI once it stops crashing my browser :P | 21:56 |
leifmadsen | lol | 21:56 |
leifmadsen | YES! | 21:56 |
dmsimard | ah, here's a non-crashing version http://logs.openstack.org/18/438118/18/check/gate-ara-integration-latest-centos-7/24085b4/logs/build/reports/2 | 21:56 |
leifmadsen | I recently updated to the new ara generate html command :) | 21:56 |
* dmsimard is benchmarking against the huge openstack-ansible playbooks with thousands of things | 21:56 | |
*** trown is now known as trown|outtypewww | 21:57 | |
*** fultonj has quit IRC | 21:57 | |
*** chlong has quit IRC | 21:58 | |
leifmadsen | dmsimard: oooooo | 21:59 |
leifmadsen | pretty :) | 21:59 |
leifmadsen | does not play too well with my half monitor window :) | 21:59 |
dmsimard | Yeah it's a bit width hungry, I'll try to make it fit within 1280ish (right now it starts folding at 1440ish) | 22:00 |
dmsimard | still a WIP tho, not merged/released yet | 22:01 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Support multinic network isolation https://review.openstack.org/437437 | 22:04 |
openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates master: Containerize gnocchi services https://review.openstack.org/439170 | 22:06 |
*** lblanchard has quit IRC | 22:07 | |
dmsimard | stevebaker: thanks for your help -- I added https://review.openstack.org/#/c/439145/2/tests/test_hook_ansible.py but I'm not very optimistic that it does what I think it does | 22:09 |
dmsimard | I'm also attempting to run an actual CI job with that patch it in right now, to see if the envvar makes it down to the ansible process | 22:09 |
*** pkovar has quit IRC | 22:10 | |
dmsimard | iirc you usually need to pass the env in subprocess | 22:10 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Add mistral service support for composable upgrades https://review.openstack.org/405246 | 22:10 |
openstackgerrit | Ben Kero proposed openstack/tripleo-quickstart-extras master: Add release notes for deprecation of multinic https://review.openstack.org/439066 | 22:11 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Add mistral service support for composable upgrades https://review.openstack.org/405246 | 22:11 |
openstackgerrit | Ben Kero proposed openstack/tripleo-quickstart-extras master: Add release notes for deprecation of multinic https://review.openstack.org/439066 | 22:12 |
*** dsneddon has joined #tripleo | 22:13 | |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Add mistral service support for composable upgrades https://review.openstack.org/405246 | 22:14 |
openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/newton: scenario001/newton: deploy cinder with rbd backend https://review.openstack.org/439188 | 22:17 |
openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates master: Containerize panko api service https://review.openstack.org/439138 | 22:18 |
*** noslzzp has quit IRC | 22:18 | |
stevebaker | dmsimard: thanks, I've reviewed | 22:22 |
*** egafford has quit IRC | 22:23 | |
*** panda is now known as panda|zZ | 22:28 | |
*** jkilpatr has joined #tripleo | 22:32 | |
openstackgerrit | Ben Kero proposed openstack/tripleo-quickstart-extras master: Add release notes for deprecation of multinic https://review.openstack.org/439066 | 22:35 |
*** gbarros has quit IRC | 22:39 | |
stevebaker | dmsimard: actually posted the review now | 22:40 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart-extras master: Add release notes for deprecation of multinic https://review.openstack.org/439066 | 22:41 |
openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart master: Add possibility to remove image cache directory https://review.openstack.org/439198 | 22:44 |
*** tbonds has joined #tripleo | 22:47 | |
openstackgerrit | Dan Prince proposed openstack-infra/tripleo-ci master: Undercloud containers job with tripleo.sh https://review.openstack.org/437411 | 22:47 |
*** eck` is now known as eck`gone | 22:50 | |
*** tbonds has quit IRC | 22:51 | |
*** tbonds has joined #tripleo | 22:52 | |
*** jmelvin has quit IRC | 22:55 | |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Update puppet-midonet https://review.openstack.org/438749 | 22:56 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Enable object-expirer on Swift proxy profile https://review.openstack.org/437541 | 23:00 |
openstackgerrit | Alex Schultz proposed openstack/puppet-tripleo stable/newton: Add virtual_packages support to norpm provider https://review.openstack.org/437701 | 23:00 |
openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates master: Use --disable= in subscription-manager to avoid shell expansion. https://review.openstack.org/437404 | 23:02 |
*** tbonds has quit IRC | 23:03 | |
openstackgerrit | Vincent S. Cojot proposed openstack/tripleo-heat-templates master: Use --disable= in subscription-manager to avoid shell expansion. https://review.openstack.org/437404 | 23:08 |
*** cdearborn has quit IRC | 23:09 | |
*** jcoufal has quit IRC | 23:10 | |
openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: Download promoted images in case of stable branch https://review.openstack.org/439206 | 23:12 |
*** dmarlin has quit IRC | 23:12 | |
EmilienM | if anyone can look project-config, it seems upgrade jobs are run in stable/newton and we don't want that. Otherwise I'll look later tonight or tomorrow | 23:15 |
* EmilienM off | 23:15 | |
*** saneax-_-|AFK is now known as saneax | 23:18 | |
*** idlemind has joined #tripleo | 23:20 | |
*** paramite has quit IRC | 23:21 | |
*** pmannidi has joined #tripleo | 23:22 | |
*** dparkes has quit IRC | 23:23 | |
*** dprince has quit IRC | 23:29 | |
*** pradk has quit IRC | 23:32 | |
*** rbrady has quit IRC | 23:36 | |
rickflare | guys I am new to tripleo and trying to stand up my environment. I have three nodes that are not matching hardward but both overcloud nodes support IPMI and the undercloud does not. My overcloud nodes have different nics that when I install centos they get different names such as eno or eth0 or enso1sp0. How do I make this work when I setup my undercloud and the documentation is not clear | 23:37 |
*** gbarros has joined #tripleo | 23:42 | |
openstackgerrit | Steve Baker proposed openstack/tripleo-heat-templates master: Switch to dict format for docker_puppet_tasks https://review.openstack.org/439240 | 23:44 |
*** yamahata has joined #tripleo | 23:47 | |
openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/newton: Temporary UCSM mapping files should be opened with write mode https://review.openstack.org/439241 | 23:47 |
*** diana_clarke has quit IRC | 23:49 | |
rickflare | anyone please? | 23:54 |
*** dmacpher has quit IRC | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!