| *** sdake_ has joined #tripleo | 00:11 | |
| *** ooolpbot has joined #tripleo | 00:11 | |
| ooolpbot | URGENT TRIPLEO TASKS NEED ATTENTION | 00:11 |
|---|---|---|
| ooolpbot | https://bugs.launchpad.net/tripleo/+bug/1663187 | 00:11 |
| *** ooolpbot has quit IRC | 00:11 | |
| openstack | Launchpad bug 1663187 in tripleo "Nova Placement API on IPv6 unreachable from compute nodes" [Critical,In progress] - Assigned to Juan Antonio Osorio Robles (juan-osorio-robles) | 00:11 |
| *** Goneri has joined #tripleo | 00:11 | |
| *** sdake has quit IRC | 00:12 | |
| *** morazi_ has quit IRC | 00:13 | |
| *** Goneri has quit IRC | 00:26 | |
| *** limao has joined #tripleo | 00:37 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Configure the placement API's interface to use the internal endpoint https://review.openstack.org/432864 | 00:38 |
| *** Goneri has joined #tripleo | 00:38 | |
| *** dougbtv_ has quit IRC | 00:51 | |
| *** gatuus has quit IRC | 01:02 | |
| *** Goneri has quit IRC | 01:15 | |
| *** rhallisey has quit IRC | 01:31 | |
| *** saneax is now known as saneax-_-|AFK | 01:35 | |
| *** rhallisey has joined #tripleo | 01:44 | |
| *** dtrainor has joined #tripleo | 01:53 | |
| *** jrist has quit IRC | 01:58 | |
| *** sdake_ is now known as sdake | 02:09 | |
| *** jrist has joined #tripleo | 02:11 | |
| *** jrist has joined #tripleo | 02:11 | |
| *** sdake has quit IRC | 02:17 | |
| *** sdake has joined #tripleo | 02:19 | |
| *** ramishra has joined #tripleo | 02:30 | |
| *** dmacpher has joined #tripleo | 02:34 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Use Keystone internal endpoint instead of admin for services https://review.openstack.org/432419 | 02:41 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Use Keystone internal endpoint instead of admin for nova-ironic https://review.openstack.org/433038 | 02:41 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Add release note for services endpoint change https://review.openstack.org/433181 | 02:41 |
| *** dbecker has joined #tripleo | 02:44 | |
| openstackgerrit | Dan Prince proposed openstack/python-tripleoclient master: Deploy the undercloud with Heat https://review.openstack.org/351351 | 02:44 |
| *** sudipto has joined #tripleo | 02:48 | |
| *** sudipto_ has joined #tripleo | 02:48 | |
| *** dbecker has quit IRC | 02:49 | |
| *** dtrainor has quit IRC | 02:51 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Switch to net-config-multinode https://review.openstack.org/426372 | 02:52 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Switch to net-config-multinode https://review.openstack.org/426372 | 02:52 |
| *** dougbtv has joined #tripleo | 02:53 | |
| openstackgerrit | Merged openstack/puppet-pacemaker master: Add rspec tests for stonith fences https://review.openstack.org/421338 | 02:54 |
| openstackgerrit | Emilien Macchi proposed openstack/python-tripleoclient stable/ocata: stable/ocata CI run - do not merge https://review.openstack.org/426231 | 02:54 |
| *** n4t4n has joined #tripleo | 02:59 | |
| *** n4t4n has quit IRC | 03:01 | |
| *** sshnaidm|off has quit IRC | 03:07 | |
| *** sshnaidm|off has joined #tripleo | 03:08 | |
| *** yamahata has quit IRC | 03:15 | |
| *** rhallisey has quit IRC | 03:30 | |
| *** yamahata has joined #tripleo | 03:49 | |
| *** psahoo has joined #tripleo | 03:51 | |
| openstackgerrit | Michael Chapman proposed openstack/tripleo-quickstart master: Add support for virtual port types https://review.openstack.org/407347 | 03:54 |
| *** david-lyle_ has joined #tripleo | 03:56 | |
| *** david-lyle has quit IRC | 03:58 | |
| *** links has joined #tripleo | 03:59 | |
| *** nmathew has joined #tripleo | 04:06 | |
| openstackgerrit | Michael Chapman proposed openstack/tripleo-quickstart master: Add support for virtual port types https://review.openstack.org/407347 | 04:10 |
| *** fzdarsky|afk has joined #tripleo | 04:12 | |
| *** dougbtv has quit IRC | 04:17 | |
| *** sdake has quit IRC | 04:22 | |
| openstackgerrit | Michael Chapman proposed openstack/tripleo-quickstart master: Add support for virtual port types https://review.openstack.org/407347 | 04:35 |
| *** udesale has joined #tripleo | 04:47 | |
| *** sudipto_ has quit IRC | 04:50 | |
| *** sudipto has quit IRC | 04:50 | |
| *** prateek has joined #tripleo | 04:50 | |
| *** sasha2 has quit IRC | 04:56 | |
| nmathew | hi all | 05:02 |
| nmathew | gm | 05:02 |
| *** sudipto_ has joined #tripleo | 05:26 | |
| *** sudipto has joined #tripleo | 05:26 | |
| *** saneax-_-|AFK is now known as saneax | 05:26 | |
| *** skramaja_afk is now known as skramaja | 05:32 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui master: Cleanup Nodes selectors https://review.openstack.org/431515 | 05:36 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui master: Update DeploymentPlan page node counts https://review.openstack.org/433449 | 05:36 |
| openstackgerrit | Saravanan KR proposed openstack/tripleo-heat-templates master: Fix empty parameter_default error in environment file https://review.openstack.org/431505 | 05:50 |
| *** twi99y has joined #tripleo | 05:50 | |
| *** twi99y__ has joined #tripleo | 05:52 | |
| *** twi99y_ has quit IRC | 05:53 | |
| *** twi99y has quit IRC | 05:54 | |
| *** saibarspeis has joined #tripleo | 05:54 | |
| *** links has quit IRC | 05:55 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-common master: Updated from global requirements https://review.openstack.org/432118 | 06:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-heat-templates master: Updated from global requirements https://review.openstack.org/432119 | 06:02 |
| *** dsneddon has quit IRC | 06:07 | |
| *** jaosorior has joined #tripleo | 06:09 | |
| *** links has joined #tripleo | 06:11 | |
| *** jtomasek has quit IRC | 06:12 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job https://review.openstack.org/426863 | 06:13 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: Deploy novajoin in the TLS-everywhere job https://review.openstack.org/427481 | 06:19 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job (novajoin) https://review.openstack.org/427482 | 06:19 |
| *** pgadiya has joined #tripleo | 06:22 | |
| *** abregman has joined #tripleo | 06:25 | |
| *** sudipto_ has quit IRC | 06:26 | |
| *** sudipto has quit IRC | 06:26 | |
| *** padkrish has joined #tripleo | 06:32 | |
| *** saibarspeis has quit IRC | 06:33 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job (novajoin) https://review.openstack.org/427482 | 06:38 |
| *** padkrish has quit IRC | 06:42 | |
| *** agurenko has joined #tripleo | 06:43 | |
| openstackgerrit | Chandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow https://review.openstack.org/431916 | 06:45 |
| *** links has quit IRC | 06:48 | |
| *** tzumainn has quit IRC | 06:49 | |
| *** ealcaniz has joined #tripleo | 06:49 | |
| *** rcernin has joined #tripleo | 06:54 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: Deploy novajoin in the TLS-everywhere job https://review.openstack.org/427481 | 06:55 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job (novajoin) https://review.openstack.org/427482 | 06:55 |
| *** sudipto has joined #tripleo | 06:56 | |
| *** sudipto_ has joined #tripleo | 06:56 | |
| nmathew | what is the logic behind eth0 -> nic1 mapping in os-net-config, how they are mapped | 06:57 |
| nmathew | i have a machine with 3 interfaces | 06:57 |
| dsneddon_afk | nmathew, First, any NICs with names like eth* are evaluated, or eno* or em* (embedded NICs) by alphanumeric order. Then, any other interfaces (ensp3s24, p1p2, etc.) by alphanumeric order. | 06:58 |
| *** dmacpher has quit IRC | 06:59 | |
| skramaja | nmathew: the logic is in here. https://github.com/openstack/os-net-config/blob/master/os_net_config/utils.py#L139 | 06:59 |
| nmathew | dsneddon_afk, skramaja thanks | 07:01 |
| skramaja | nmathew: worth to note that if there any dpdk bindings done, then the corresponding interface will not be listed. but it will be added via the _DPDK_MAPPING_FILE.. | 07:01 |
| *** links has joined #tripleo | 07:02 | |
| skramaja | nmathew: not applicable in the first fresh deployment though.. | 07:02 |
| nmathew | any doc or link? | 07:03 |
| skramaja | nmathew: it is done in the same function ordered_active_nics() as i posted above. | 07:04 |
| nmathew | ok | 07:04 |
| skramaja | dsneddon_afk: can you approve https://review.openstack.org/#/c/422449/, CI is through.. | 07:04 |
| *** dsariel has joined #tripleo | 07:05 | |
| dsneddon_afk | skramaja, Looks like someone beat me to it | 07:08 |
| *** iranzo has joined #tripleo | 07:08 | |
| *** artom has quit IRC | 07:11 | |
| skramaja | ohh.. thanks dsneddon_afk for taking look. and thanks jaosorior for approving. | 07:12 |
| *** fragatin_ has joined #tripleo | 07:13 | |
| *** tesseract has joined #tripleo | 07:15 | |
| *** fragatina has quit IRC | 07:17 | |
| *** fragatin_ has quit IRC | 07:18 | |
| *** artom has joined #tripleo | 07:18 | |
| chandankumar | Hello | 07:26 |
| *** bkopilov has quit IRC | 07:26 | |
| chandankumar | How can i run tripleo-ci script locally only tempest part? | 07:26 |
| *** bkopilov_ has quit IRC | 07:27 | |
| *** iranzo has quit IRC | 07:27 | |
| *** iranzo has joined #tripleo | 07:27 | |
| *** iranzo has joined #tripleo | 07:27 | |
| *** rasca has joined #tripleo | 07:28 | |
| *** rlandy has quit IRC | 07:28 | |
| *** dsariel has quit IRC | 07:29 | |
| *** fragatina has joined #tripleo | 07:36 | |
| *** fragatina has quit IRC | 07:37 | |
| *** fragatina has joined #tripleo | 07:37 | |
| *** saibarspeis has joined #tripleo | 07:41 | |
| *** rasca has quit IRC | 07:41 | |
| openstackgerrit | Merged openstack/puppet-tripleo master: Add ::ironic::config to Ironic base profile https://review.openstack.org/429290 | 07:42 |
| *** jprovazn has joined #tripleo | 07:43 | |
| *** fragatina has quit IRC | 07:44 | |
| *** fragatina has joined #tripleo | 07:44 | |
| *** abehl has joined #tripleo | 07:44 | |
| *** nijaba has quit IRC | 07:49 | |
| *** links has quit IRC | 07:50 | |
| *** nijaba has joined #tripleo | 07:52 | |
| *** nijaba has quit IRC | 07:52 | |
| *** nijaba has joined #tripleo | 07:52 | |
| *** links has joined #tripleo | 07:52 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/tripleo-ui master: Imported Translations from Zanata https://review.openstack.org/432581 | 07:53 |
| *** shardy has joined #tripleo | 07:58 | |
| *** tremble has joined #tripleo | 08:03 | |
| *** tremble has joined #tripleo | 08:03 | |
| *** pmannidi has quit IRC | 08:04 | |
| openstackgerrit | Merged openstack/os-net-config master: Network service is failing with DPDK https://review.openstack.org/422449 | 08:07 |
| *** sshnaidm|off is now known as sshnaidm | 08:10 | |
| *** pcaruana has joined #tripleo | 08:11 | |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart master: Remove hardcoded ansible user stack https://review.openstack.org/431092 | 08:11 |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart master: Remove hardcoded ansible user stack https://review.openstack.org/431092 | 08:12 |
| *** amoralej|off is now known as amoralej | 08:15 | |
| *** leanderthal|afk is now known as leanderthal | 08:22 | |
| *** shardy has quit IRC | 08:24 | |
| *** stendulker has joined #tripleo | 08:26 | |
| *** rasca has joined #tripleo | 08:29 | |
| *** cylopez has joined #tripleo | 08:30 | |
| *** rwsu has joined #tripleo | 08:31 | |
| *** prateek has quit IRC | 08:31 | |
| *** prateek has joined #tripleo | 08:32 | |
| *** jlinkes has joined #tripleo | 08:35 | |
| *** flepied has quit IRC | 08:36 | |
| *** b00tcat has joined #tripleo | 08:38 | |
| *** Vijayendra has quit IRC | 08:39 | |
| *** jpena|away is now known as jpena | 08:39 | |
| *** jpich has joined #tripleo | 08:41 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Updated from global requirements https://review.openstack.org/432119 | 08:42 |
| openstackgerrit | Saravanan KR proposed openstack/os-net-config stable/newton: Network service is failing with DPDK https://review.openstack.org/433496 | 08:43 |
| *** paramite has joined #tripleo | 08:44 | |
| *** suuuper has joined #tripleo | 08:45 | |
| *** zhihui has joined #tripleo | 08:46 | |
| *** fragatina has quit IRC | 08:47 | |
| *** yamahata has quit IRC | 08:48 | |
| *** karimb has joined #tripleo | 08:50 | |
| *** athomas has joined #tripleo | 08:51 | |
| *** jaganathan has joined #tripleo | 08:52 | |
| *** jpena is now known as jpena|away | 08:52 | |
| *** Vijayendra has joined #tripleo | 08:53 | |
| *** zoli|gone is now known as zoliXXL | 08:58 | |
| *** gfidente has joined #tripleo | 08:59 | |
| *** jpena|away is now known as jpena | 08:59 | |
| *** zzzeek has quit IRC | 09:00 | |
| *** mcornea has joined #tripleo | 09:01 | |
| *** zzzeek has joined #tripleo | 09:01 | |
| mcornea | marios: hey man, good morning | 09:02 |
| *** skramaja has quit IRC | 09:02 | |
| marios | o/ mcornea | 09:02 |
| mcornea | marios: I got stuck yesterday on updating the compute nodes, it looks that the yum update works fine but it fails after that with Error: Evaluation Error: Error while evaluating a Function Call, Could not find data item step in any Hiera data file and no default supplied at /etc/puppet/modules/tripleo/manifests/profile/base/nova/compute/libvirt.pp:27:11 on node overcloud-novacompute-0.localdomain | 09:04 |
| mcornea | marios: any hint what I can do to move forward? | 09:04 |
| marios | mcornea: it could be an issue trying to apply the puppet (did you have https://review.openstack.org/#/c/424661 in the env)? | 09:04 |
| marios | mcornea: ah... yeah so we may need to tweak that (afaik we haven't tested that puppet apply from ^^^ yet) | 09:05 |
| marios | mcornea: and from the sounds of the error we need to remove the 'step' or pass it in when we call the puppet apply (it is normally supplied by the heat templtaes but here we are applying it manually) | 09:05 |
| mcornea | marios: yeah, it looks that I have that manifest but I think it's missing some hieradata, let me show you | 09:07 |
| *** flepied has joined #tripleo | 09:07 | |
| marios | mcornea: (the puppet-tripleo side is expecting the step to be set so we may need to set it in hiera)... would help to know more about what the error was exactly (but i suspect more of us will get to that today hopefully), i'm running v39 of the nova patch | 09:07 |
| mcornea | marios: http://paste.openstack.org/show/598766/ | 09:08 |
| marios | mcornea: thanks (yeah so the error looks like it comes from the ceph client class in puppet-tripleo... sec) | 09:09 |
| mcornea | marios: hiera step returns nil | 09:10 |
| mcornea | marios: aren't we deleting the hieradata at some point? | 09:11 |
| marios | mcornea: yeah its failing here https://github.com/openstack/puppet-tripleo/blob/master/manifests/profile/base/ceph/client.pp#L27 so we need to think about this some more... we can't just puppet apply | 09:11 |
| marios | mcornea: no the problem is the step is set by the tht normally ... we apply the puppet as a number of steps and set the step each time | 09:12 |
| mcornea | marios: gotcha, so how did it work in the past? | 09:12 |
| marios | mcornea: and here we are just puppet apply without setting the step. so it may not be so simple as 'just set the setp' .... we may need a loop to simulate thesteps | 09:12 |
| marios | mcornea: i think this is something we should talk about some more when more people are online in a bit | 09:12 |
| mcornea | marios: oki, thanks for checking it - I also have a bug for it https://bugs.launchpad.net/tripleo/+bug/1664304 | 09:13 |
| openstack | Launchpad bug 1664304 in tripleo "Upgrading compute nodes fails with: Error: Evaluation Error: Error while evaluating a Function Call, Could not find data item step in any Hiera data file and no default supplied at /etc/puppet/modules/tripleo/manifests/profile/base/ceph/client.pp:27:11 on node overcloud-novacompute-0.localdomain" [Critical,Triaged] | 09:13 |
| *** jtomasek has joined #tripleo | 09:13 | |
| marios | mcornea: cool thanks | 09:13 |
| *** dsariel has joined #tripleo | 09:13 | |
| *** aufi has joined #tripleo | 09:17 | |
| *** udesale has quit IRC | 09:26 | |
| *** bkopilov has joined #tripleo | 09:26 | |
| chem | marios: oh boy, I missed a lot of dev -> puppet apply /root/${ROLE}_puppet_config.pp | 09:26 |
| chem | marios: backtracking on the dev ... | 09:26 |
| *** bkopilov_ has joined #tripleo | 09:27 | |
| matbu | chem: hehe | 09:27 |
| marios | chem: yeah we may need a loop in there am poking a bit too | 09:27 |
| matbu | chem: you plan to backport it for osp 8 right ? :D | 09:27 |
| *** dsariel has quit IRC | 09:28 | |
| chem | matbu: ah, if you could test my patch in an ha env with 2 computes, that would be sweet :) | 09:28 |
| *** dsariel has joined #tripleo | 09:28 | |
| chem | marios: a loop and hiera data, but are the hiera data easily available on the platform with the new hiera hook ? | 09:29 |
| marios | chem: good question... i think it should all be available at this point | 09:30 |
| matbu | chem: it should, because we install in the upgrade init | 09:30 |
| marios | chem: because we ran all the deploy steps on all nodes, except the post deploy | 09:30 |
| marios | chem: in the earlier composable steps | 09:30 |
| marios | chem: except the post deploy * on those roles that have the upgrade config disabled | 09:30 |
| marios | chem: (like compute and swift right now) | 09:31 |
| *** lucas-afk is now known as lucasagomes | 09:33 | |
| openstackgerrit | Merged openstack/tripleo-ui master: Imported Translations from Zanata https://review.openstack.org/432581 | 09:37 |
| *** salmankhan has joined #tripleo | 09:38 | |
| chem | marios: do you mind if I work on this one, just to get some fresh air ? | 09:44 |
| chem | marios: my master upgrade env is in composable upgrade step ... | 09:45 |
| marios | chem: not at all please go ahead | 09:45 |
| marios | chem: enjoy! | 09:45 |
| chem | marios: yeah ! | 09:45 |
| marios | chem: :( | 09:45 |
| matbu | chem: you really want me to test your review for osp 8 ? | 09:48 |
| chem | matbu: ah, now you're getting scared .... :) | 09:48 |
| matbu | chem: hehe no :) but i can if you need | 09:49 |
| chem | matbu: nope, that's fine, I have my own env working (with vm migration in it ..!) and somehow qe should verify it as well | 09:50 |
| matbu | chem: yep probably better QE could test it than me :) | 09:50 |
| openstackgerrit | Arx Cruz proposed openstack/tripleo-quickstart-extras master: Skipping tests requiring cinder-backup https://review.openstack.org/433099 | 09:54 |
| *** tesseract-RH has joined #tripleo | 09:56 | |
| *** akrivoka has joined #tripleo | 09:56 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates master: Deploy Manila with CephMDS in scenario004 https://review.openstack.org/423386 | 10:00 |
| *** dsariel has quit IRC | 10:02 | |
| *** dsariel has joined #tripleo | 10:02 | |
| *** bogdando has joined #tripleo | 10:02 | |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 10:03 |
| *** apetrich has quit IRC | 10:04 | |
| *** tesseract has quit IRC | 10:04 | |
| *** b00tcat has quit IRC | 10:05 | |
| *** apetrich has joined #tripleo | 10:06 | |
| *** tosky has joined #tripleo | 10:07 | |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-heat-templates master: Add nova service support for composable upgrades https://review.openstack.org/405241 | 10:08 |
| *** percevalbot has joined #tripleo | 10:09 | |
| *** limao has quit IRC | 10:10 | |
| *** ealcaniz has quit IRC | 10:15 | |
| *** ealcaniz has joined #tripleo | 10:15 | |
| openstackgerrit | Merged openstack/tripleo-common master: Updated from global requirements https://review.openstack.org/432118 | 10:21 |
| *** zoliXXL is now known as zoli|lunch | 10:21 | |
| *** bogdando has quit IRC | 10:27 | |
| *** b00tcat has joined #tripleo | 10:32 | |
| *** b00tcat has joined #tripleo | 10:32 | |
| *** sdake has joined #tripleo | 10:34 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci master: Installs openssh packages that try to workaround sshd start problem https://review.openstack.org/430688 | 10:36 |
| *** salmankhan has quit IRC | 10:37 | |
| *** sdake_ has joined #tripleo | 10:38 | |
| *** sdake has quit IRC | 10:39 | |
| *** sdake__ has joined #tripleo | 10:39 | |
| *** sdake_ has quit IRC | 10:42 | |
| *** adarazs is now known as adarazs_lunch | 10:42 | |
| *** milan has joined #tripleo | 10:44 | |
| *** shardy has joined #tripleo | 10:48 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Add sensu and fluentd client to major upgrade environment https://review.openstack.org/433588 | 10:51 |
| *** sdake has joined #tripleo | 10:51 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Fix empty parameter_default error in environment file https://review.openstack.org/431505 | 10:51 |
| *** sdake__ has quit IRC | 10:53 | |
| ccamacho | hey marios FYI testing the opstools services yesterday, my upgrade failed on step0 because collectd is not installed in newton, im reinstalling all to re-test it, Ill let you know also to paramite. | 10:54 |
| paramite | ack, thanks | 10:54 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates master: Add Ceph RGW to capabilities map https://review.openstack.org/433590 | 10:57 |
| marios | matbu: ok thanks...that is strange... i mean is it in your roles_data.yaml? if it isn't then the upgrade steps aren't included | 10:57 |
| marios | err ccamacho ^^ | 10:57 |
| marios | ccamacho: i.e. we only run the upgrade tasks for those services that are deployed (roles_data.yaml) e.g. https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/major_upgrade_steps.j2.yaml#L194 | 10:59 |
| chem | marios: is the "mistral-db-manage populate" and following commands required upstream as well ? | 11:00 |
| marios | matbu: hmmm am not sure are skipping the puppet apply on the 'disabled' roles.. i just saw a puppet apply fail on my swift node and i have https://review.openstack.org/#/c/431398/ (but it was applied weirdly i just noticed it looked like this http://paste.openstack.org/show/598784/ | 11:01 |
| marios | chem: yeah i think we will either have that in the docs or we will have to hide it in the undercloud upgrade | 11:02 |
| marios | matbu: i am wondering if i should reset my env before going to lunch or try again after fixing that ^^^ (i.e. did you get a clean pass with that stuff?) | 11:02 |
| chem | marios: thanks | 11:02 |
| marios | matbu: hmm so it really wasn't applied properly, going to try a rerun first :) | 11:04 |
| *** cylopez has quit IRC | 11:04 | |
| *** salmankhan has joined #tripleo | 11:07 | |
| ccamacho | marios not 100% sure, im re-deploying it I'll let you know asap | 11:09 |
| openstackgerrit | Alan Pevec proposed openstack-infra/tripleo-ci master: DNM testing RDO dependency updates https://review.openstack.org/422244 | 11:10 |
| *** stendulker has quit IRC | 11:11 | |
| *** dtantsur|afk is now known as dtantsur | 11:12 | |
| *** ccamacho is now known as ccamacho|brb | 11:12 | |
| *** jrist has quit IRC | 11:27 | |
| *** mat128 has joined #tripleo | 11:27 | |
| *** muskan_ has joined #tripleo | 11:32 | |
| *** zoli|lunch is now known as zoliXXL | 11:32 | |
| *** zoliXXL is now known as zoli|FBCI | 11:34 | |
| *** ccamacho|brb is now known as ccamacho | 11:35 | |
| shardy | dciabrin: Hi! Question re https://review.openstack.org/#/q/Ic69de63acbfb992314ea30a3a9b17c0b5341c035,n,z when you're around | 11:35 |
| *** jkilpatr has quit IRC | 11:36 | |
| *** nmathew has quit IRC | 11:36 | |
| michapma_alt | panda|zZ: for https://review.openstack.org/#/c/407347/ do you doubt the mess of filters works, or is there a bug regarding blocked include_role? It works as far as I can tell, but it is terribly ugly :( | 11:37 |
| *** rhallisey has joined #tripleo | 11:38 | |
| *** panda|zZ is now known as panda | 11:38 | |
| panda | michapma_alt: I think the mess of filter is inevitable, I just don't trust include_role enough | 11:39 |
| panda | michapma_alt: we had problems in the past putting include_role near any other construct in ansible | 11:39 |
| *** jrist has joined #tripleo | 11:40 | |
| *** jrist has joined #tripleo | 11:40 | |
| michapma_alt | panda: ah. I was actually going to ask this evening if folks are interested in having jmespath as a dependency to clean up some of the filter soup | 11:40 |
| michapma_alt | panda: I saw on the doc page for older versions the behavior is not the same | 11:40 |
| michapma_alt | panda: are we supporting ansible < 2.2? | 11:40 |
| panda | michapma_alt: I don't think we have an official version list we support, but if this change is breaking in the past version, it needs to be accompanied by a release note | 11:42 |
| michapma_alt | panda: makes sense. The other options would be to just put the dependencies inside the block - there's only two anyway, or to have ovs as a dependency even when it isn't used. | 11:44 |
| openstackgerrit | Chandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow https://review.openstack.org/431916 | 11:45 |
| michapma_alt | panda: the latter will need changes to the CI env I guess because it's not available for install on the virthost - I already tried earlier. | 11:45 |
| michapma_alt | panda: where should I put a release note? | 11:50 |
| panda | michapma_alt: you know how to user reno ? | 11:51 |
| panda | use* | 11:51 |
| michapma_alt | panda: I've never even heard of reno :) | 11:52 |
| *** sudipto has quit IRC | 11:52 | |
| *** sudipto_ has quit IRC | 11:52 | |
| panda | michapma_alt: hope EmilienM doesn't hear you. | 11:52 |
| michapma_alt | panda: haha, I found it in pip, I'll figure it out. Thanks for the help. | 11:53 |
| panda | I hate it when he lets steam out of his nose. | 11:53 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Set mysql client bind address via ServiceNetMap https://review.openstack.org/433607 | 11:54 |
| shardy | michapma_alt: see http://docs.openstack.org/developer/reno/usage.html | 11:56 |
| shardy | it's pretty straight forward, just generates a template which you then fill out | 11:56 |
| michapma_alt | shardy: cheers | 11:57 |
| *** sshnaidm has quit IRC | 11:59 | |
| *** dbecker has joined #tripleo | 12:02 | |
| EmilienM | Good morning folks! | 12:03 |
| shardy | Morning EmilienM! | 12:05 |
| *** sshnaidm has joined #tripleo | 12:06 | |
| *** jkilpatr has joined #tripleo | 12:06 | |
| *** dbecker has quit IRC | 12:12 | |
| *** suuuper has quit IRC | 12:13 | |
| *** muskan_ has quit IRC | 12:18 | |
| *** b00tcat has quit IRC | 12:18 | |
| *** zoli|FBCI is now known as zoli | 12:18 | |
| openstackgerrit | Chandan Kumar proposed openstack/tripleo-quickstart-extras master: Use Ocata Tempest rpm with new tempest workflow https://review.openstack.org/431916 | 12:21 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 12:24 |
| EmilienM | panda: I did :-P | 12:25 |
| panda | d'oh | 12:27 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/instack-undercloud master: Fix ntp configuration. https://review.openstack.org/433623 | 12:27 |
| *** pkovar has joined #tripleo | 12:28 | |
| *** dbecker has joined #tripleo | 12:29 | |
| chem | EmilienM: not sure if this one is important or not https://bugs.launchpad.net/tripleo/+bug/1664537 | 12:30 |
| openstack | Launchpad bug 1664537 in tripleo "ntp is not properly configured on the undercloud." [Undecided,In progress] - Assigned to Sofer Athlan-Guyot (sofer-athlan-guyot) | 12:30 |
| *** nmathew has joined #tripleo | 12:30 | |
| *** nmathew- has joined #tripleo | 12:31 | |
| shardy | http://logs.openstack.org/19/432119/6/check/gate-tripleo-ci-centos-7-multinode-upgrades-nv/3c6af4e/console.html#_2017-02-14_06_35_08_089816 | 12:31 |
| chem | matbu: marios so my composite upgrade just hang in the air, first I though that it may be related to ntp configuration, but really, nothing is happening at some point. Is this a know issue and I'm missing a patch or something local to my env ? | 12:31 |
| shardy | hrm | 12:31 |
| shardy | we seem to have broken the upgrade job, and I'm not sure why | 12:32 |
| EmilienM | chem: it is I think | 12:32 |
| shardy | $TRIPLEO_ROOT/tripleo-ci should exist, but it looks like it doesn't | 12:32 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Configuring a default ntp server. https://review.openstack.org/420573 | 12:32 |
| *** nmathew- has quit IRC | 12:33 | |
| *** lucasagomes is now known as lucas-hungry | 12:34 | |
| EmilienM | shardy: the only recent change that could have broke would be https://github.com/openstack-infra/tripleo-ci/commit/cd23c1d034c83b704f0d12c115c18b498414c118 | 12:36 |
| *** cylopez has joined #tripleo | 12:36 | |
| marios | chem: heh could be ... i mean i have a lot of patches applied here... do you remove the old hiera data in your upgrade init? | 12:36 |
| *** dsariel has quit IRC | 12:37 | |
| shardy | EmilienM: yeah, I can't currently see how though | 12:37 |
| shardy | EmilienM: that patch did pass the experimental upgrades job, right? | 12:37 |
| marios | chem: what step is it hanging on? | 12:37 |
| *** thrash|g0ne is now known as thrash | 12:38 | |
| chem | marios: not twice at the same place | 12:38 |
| shardy | Ok, actually it didn't, I may try a revert to see if it helps | 12:38 |
| EmilienM | shardy: wait | 12:39 |
| *** nmathew- has joined #tripleo | 12:39 | |
| EmilienM | I saw a successful job passing after my patch, let me confirm that | 12:39 |
| chem | marios: I've bump into this ntp problem where my overcloud what moving date/time while my undercloud did not. do you have time dif on your side ? | 12:39 |
| chem | marios: but not sure it's related | 12:39 |
| shardy | http://logs.openstack.org/19/432419/1/check/gate-tripleo-ci-centos-7-multinode-upgrades-nv/2d0a9cd/ | 12:40 |
| *** adarazs_lunch is now known as adarazs | 12:40 | |
| shardy | that appears to be the first failure with this problem | 12:40 |
| *** derekh has joined #tripleo | 12:40 | |
| *** sdake_ has joined #tripleo | 12:41 | |
| marios | chem: yeah don't think it is related either (mine does seem to have a diff now between over/undercloud hadn't noticed if it was the case before actually) because mine isn't hanging (though it is failing :) now on puppet)... but i think i'm going to go and reset my env didn't do it for 3 days now and i have too many things applied | 12:41 |
| *** dougbtv has joined #tripleo | 12:41 | |
| EmilienM | shardy: go ahead | 12:41 |
| marios | chem: chem in the meantime while is reset/deploying i will poke at the puppet apply.. we should try and get something out today i think | 12:42 |
| *** nmathew has quit IRC | 12:42 | |
| chem | marios: I've something ready for this but I couldn't run it | 12:42 |
| marios | chem: i mean we can both look there and syncup (ah ok cool, i thought you were waiting) | 12:42 |
| openstackgerrit | Carlos Camacho proposed openstack-infra/tripleo-ci master: Removing multinode_major_upgrade.yaml https://review.openstack.org/433201 | 12:43 |
| chem | marios: I'll push the patch to review in a few minute | 12:43 |
| marios | chem: why don't you just post it with WIP ? | 12:43 |
| marios | chem: yeah OK thanks | 12:43 |
| *** bkopilov has quit IRC | 12:43 | |
| EmilienM | shardy: looking at logstash, it's more than upgrade job | 12:43 |
| *** bkopilov_ has quit IRC | 12:43 | |
| chem | marios: got distracted by the ntp problem -> https://review.openstack.org/#/c/433623/1 | 12:43 |
| *** sdake has quit IRC | 12:43 | |
| openstackgerrit | Steven Hardy proposed openstack-infra/tripleo-ci master: Revert "toci_gate_test: allow to run upgrade jobs for scenarios" https://review.openstack.org/433627 | 12:43 |
| EmilienM | shardy: http://logstash.openstack.org/#/dashboard/file/logstash.json?query=build_name:%20*tripleo-ci*%20AND%20build_status:%20FAILURE%20AND%20message:%20%5C%22504%20Gateway%20Time-out%5C%22 | 12:44 |
| EmilienM | shardy: we have it since days | 12:44 |
| shardy | EmilienM: sorry, I'm confused, the error is "/opt/stack/new/tripleo-ci/deploy.env: No such file or directory" | 12:45 |
| shardy | not a gateway timeout | 12:45 |
| *** b00tcat has joined #tripleo | 12:45 | |
| shardy | http://logs.openstack.org/19/432419/2/check/gate-tripleo-ci-centos-7-multinode-upgrades-nv/bff6853/console.html#_2017-02-14_09_32_01_394710 | 12:45 |
| shardy | it looks like the directory is missing, but I don't see how it can be when we're running the script from /opt/stack/new/tripleo-ci | 12:46 |
| EmilienM | shardy: sorry wrong copy paste | 12:46 |
| EmilienM | http://logstash.openstack.org/#dashboard/file/logstash.json?query=build_name%3A%20*tripleo-ci*%20AND%20build_status%3A%20FAILURE%20AND%20message%3A%20%5C%22%2Fopt%2Fstack%2Fnew%2Ftripleo-ci%2Fdeploy.env%3A%20No%20such%20file%20or%20directory%5C%22 | 12:46 |
| EmilienM | better now ^ | 12:46 |
| EmilienM | we had it the first time on 2017-02-05T05:25:10.000-05:00 | 12:46 |
| shardy | Hmm, strange | 12:47 |
| shardy | Ok thanks EmilienM I'll WIP the revert | 12:47 |
| *** lblanchard has joined #tripleo | 12:47 | |
| EmilienM | shardy: i'm investigating, let's keep the revert (I don't mind) | 12:47 |
| *** karimb has quit IRC | 12:48 | |
| EmilienM | the first occurence was during a periodic job, which means during a promotion | 12:48 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 12:48 |
| marios | chem: cool thanks reviewed | 12:49 |
| *** dougbtv has quit IRC | 12:49 | |
| shardy | EmilienM: I think it only started yesterday, the logstash query is picking up some different "cp" errors | 12:50 |
| shardy | not only the common_functions.sh: line 317 one | 12:50 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/instack-undercloud master: Fix ntp configuration. https://review.openstack.org/433623 | 12:50 |
| EmilienM | shardy: what query do you use? | 12:50 |
| shardy | build_name: *tripleo-ci* AND build_status: FAILURE AND message: "line 317: /opt/stack/new/tripleo-ci/deploy.env: No such file or directory" | 12:51 |
| shardy | 2017-02-14T10:40:33.736+00:00 is the first failure like that | 12:51 |
| *** dougbtv has joined #tripleo | 12:51 | |
| EmilienM | shardy: http://logs.openstack.org/72/426372/13/check/gate-tripleo-ci-centos-7-multinode-upgrades-nv/2f8d048/console.html#_2017-02-11_02_18_02_641828 | 12:52 |
| EmilienM | same error, 3 days ago | 12:52 |
| EmilienM | shardy: but I used Depends-On the patch you're reverting | 12:53 |
| EmilienM | shardy: so I think it makes sense | 12:53 |
| shardy | Aha | 12:53 |
| EmilienM | shardy: I'm looking for the root cause now | 12:53 |
| shardy | EmilienM: Ok, well lets see what the check experimental says on the revert | 12:53 |
| EmilienM | shardy: if I don't find it before CI finish, let's merge the revert ok? | 12:53 |
| shardy | I still can't see why your patch would cause this | 12:53 |
| *** lblanchard has quit IRC | 12:54 | |
| shardy | EmilienM: yep lets see if the revert fixes it and we can investigate in the meantime | 12:54 |
| * shardy going to grab lunch now | 12:54 | |
| * EmilienM grabs pancakes | 12:54 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Configuring a default ntp server. https://review.openstack.org/420573 | 12:54 |
| *** flepied has quit IRC | 13:00 | |
| *** jayg|g0n3 is now known as jayg | 13:01 | |
| *** athomas has quit IRC | 13:01 | |
| openstackgerrit | Alan Pevec proposed openstack-infra/tripleo-ci master: DNM testing RDO dependency updates https://review.openstack.org/422244 | 13:02 |
| *** pkovar has quit IRC | 13:02 | |
| *** psahoo has quit IRC | 13:05 | |
| *** suuuper has joined #tripleo | 13:05 | |
| *** athomas has joined #tripleo | 13:06 | |
| *** bfournie has quit IRC | 13:07 | |
| *** trown|outtypewww is now known as trown | 13:08 | |
| EmilienM | shardy: I think I found why | 13:08 |
| *** liverpooler has joined #tripleo | 13:08 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Move rabbit port to string instead of number. https://review.openstack.org/410395 | 13:09 |
| *** dsariel has joined #tripleo | 13:10 | |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 13:10 |
| *** morazi has joined #tripleo | 13:11 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates master: Configuring a default ntp server. https://review.openstack.org/420573 | 13:13 |
| *** karimb has joined #tripleo | 13:16 | |
| EmilienM | shardy: i'm working on a patch | 13:16 |
| openstackgerrit | Dougal Matthews proposed openstack/python-tripleoclient master: [WIP] Testing Kombu https://review.openstack.org/433639 | 13:16 |
| *** dbecker has quit IRC | 13:17 | |
| *** flepied has joined #tripleo | 13:18 | |
| *** morazi has quit IRC | 13:18 | |
| *** dsariel has quit IRC | 13:18 | |
| matbu | marios: does the common works ? | 13:20 |
| thrash | dtantsur: can you explain a bit more your comment on https://review.openstack.org/#/c/430343/3? | 13:20 |
| matbu | marios: i didn't reach the point yet (many issue with other things today) | 13:20 |
| dtantsur | thrash, why do we need this validation if we already have hypervisor stats check? | 13:21 |
| *** lucas-hungry is now known as lucasagomes | 13:21 | |
| dtantsur | node count alone means quite nothing | 13:21 |
| *** pkovar has joined #tripleo | 13:21 | |
| thrash | dtantsur: I honestly don't know the motivation for the original check. It's actually a just a part of a "validation" but that validation had side effects. I didn't like that. | 13:22 |
| *** fultonj has joined #tripleo | 13:22 | |
| marios | matbu: so i think so because it completed those steps afaics (yeah it failed again for me on my swift node, i wasn't properly skipping puppet because i didn't have https://review.openstack.org/#/c/431398 applied properly... then last time it failed on a weird db error so yeah i would have ssen it by that stage if it wasnt workign i think | 13:22 |
| thrash | dtantsur: https://github.com/openstack/python-tripleoclient/blob/master/tripleoclient/utils.py#L474 | 13:22 |
| marios | matbu: so it looks good | 13:22 |
| dtantsur | thrash, I'd not object, but exposing an action in mistral is an API addition, and we should take API additions seriously | 13:22 |
| thrash | dtantsur: that's where it originally came from. | 13:22 |
| matbu | mcornea: marios cool | 13:23 |
| marios | matbu: am trying to move on for now and resist reset on my env before eod becuase otherwise it wasts like 1.5 hours... | 13:23 |
| matbu | mcornea: marios btw if you have a chance of being able to test the transport_url.stdout in the nova patch, it would be awesome | 13:23 |
| marios | matbu: yeah will grab latest of that nove patch on next run | 13:24 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: [WIP] Apply puppet in non-controller script in step. https://review.openstack.org/433641 | 13:24 |
| marios | matbu: can you update for those nits please so i don't have to manually fixup after i apply the patch? | 13:24 |
| matbu | marios: hehe, im redeploying atm cause i'v maade a mistake, and broke every thing :/ | 13:24 |
| thrash | dtantsur: so, do you think we can just roll that up in the check statistics? Or just get rid of it altogether? | 13:24 |
| *** pgadiya has quit IRC | 13:24 | |
| matbu | marios: which ? on nova patch ? | 13:24 |
| dtantsur | thrash, yep, this is my suggestion | 13:25 |
| thrash | dtantsur: ...? which one? :) | 13:25 |
| marios | matbu: yeah some stuff we keep forgetting like update on step 2 is that necessary? | 13:25 |
| matbu | marios: i saw your comment, i think the step6 can be removed, but not sure for the step2 | 13:25 |
| marios | matbu: also on the nova-compute one remove the step6 | 13:25 |
| marios | matbu: ah ok | 13:25 |
| marios | matbu: cool | 13:25 |
| matbu | marios: yep i need to rechec, but i was for the all-in-one | 13:25 |
| marios | matbu: yeah the step6 isn't run anyway right now | 13:25 |
| chem | mcornea: marios matbu this https://review.openstack.org/#/c/433641/ is my attempt at fixing compute (and non-controller by the way) problem | 13:25 |
| matbu | marios: we need to upgrade those package in step2 | 13:25 |
| marios | matbu: ah ok thanks revoting then | 13:26 |
| chem | mcornea: marios matbu completely untested | 13:26 |
| marios | chem: thanks will look in a sec | 13:26 |
| matbu | chem: hehe ack cool i'll look | 13:26 |
| *** sudipto_ has joined #tripleo | 13:26 | |
| *** sudipto has joined #tripleo | 13:26 | |
| ansiwen | marios: I'm confused (again :-) ), where in M->N _overcloud_ upgrade is the "nova-manage db sync" step? | 13:27 |
| *** ccamacho is now known as ccamacho|lunch | 13:27 | |
| matbu | marios: im testing the change for the restart at step0 also, it's probably unnecessary if we provide the transport-url with nova-manange | 13:27 |
| ansiwen | marios: because I assume for the overcloud M->N upgrade the online_data_migration is still missing, right? | 13:28 |
| ansiwen | marios: for reference: https://etherpad.openstack.org/p/tripleo-nova-online-data-migration | 13:28 |
| *** links has quit IRC | 13:28 | |
| ansiwen | marios: almost everything has been merged already | 13:29 |
| marios | ansiwen: yeah so we ran some migrations that were necessary for upgrade in the controller_pacemaker manifests (will get you link) whilst others were dont as usual puppet side during the converge step | 13:29 |
| ansiwen | marios: but M->N overcloud upgrade is still missing I think | 13:29 |
| ansiwen | marios: I have to know where the db-sync happens, because the online_data_migration has to happen after that | 13:30 |
| marios | ansiwen: https://github.com/openstack/tripleo-heat-templates/blob/stable/newton/extraconfig/tasks/major_upgrade_controller_pacemaker_3.sh#L65 | 13:30 |
| marios | matbu: cool | 13:30 |
| dtantsur | thrash, ah, sorry :) lemme check something before answering | 13:30 |
| ansiwen | marios: oh, it's there already! | 13:30 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: [WIP] Apply puppet in non-controller script in step. https://review.openstack.org/433641 | 13:30 |
| ansiwen | marios: that's good news | 13:30 |
| marios | ansiwen: yeah looks like | 13:30 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: toci_gate_test.sh: fix logic for upgrade jobs https://review.openstack.org/433646 | 13:32 |
| EmilienM | shardy: ^ | 13:32 |
| dtantsur | thrash, ok, I stand corrected: counting actual ironic nodes does matter. I'm still not sure it should be two different checks vs one check doing two things, as the goal is the same: ensure we have enough resources to deploy. | 13:32 |
| *** morazi has joined #tripleo | 13:32 | |
| *** ccamacho|lunch is now known as ccamacho | 13:32 | |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 13:32 |
| chem | mcornea: marios matbu a bit tested, the 2 review should work. In my env (stuck in composable upgrade) the /etc/puppet/hieradata is not properly populated, if you're past this could you check that all yaml files are there (or paste me the list) ? | 13:33 |
| EmilienM | ccamacho: there is no release notes for tripleo-ci (and we don't want to I guess, since we don't release it) | 13:33 |
| *** shardy has quit IRC | 13:33 | |
| EmilienM | ccamacho: you can remove it from https://review.openstack.org/#/c/433201/2 | 13:33 |
| EmilienM | ccamacho: also, I already proposed this patch 2 weeks ago: https://review.openstack.org/#/c/426375 | 13:34 |
| thrash | dtantsur: I can get on board with that. I'll just collapse them into a single check. | 13:34 |
| EmilienM | ccamacho: still under review | 13:34 |
| *** rlandy has joined #tripleo | 13:34 | |
| marios | will do chem | 13:34 |
| *** shardy has joined #tripleo | 13:34 | |
| ccamacho | EmilienM ack thanks! | 13:36 |
| dtantsur | thrash, yep. leaving a few minor comments now | 13:36 |
| thrash | dtantsur: thanks. | 13:36 |
| openstackgerrit | Chris Jones proposed openstack/python-tripleoclient master: Fix fencing action parameter name. https://review.openstack.org/433649 | 13:37 |
| chem | marios: thanks | 13:37 |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Use Keystone internal endpoint instead of admin for services https://review.openstack.org/432419 | 13:39 |
| openstackgerrit | Chris Jones proposed openstack/python-tripleoclient master: Fix fencing action parameter name. https://review.openstack.org/433649 | 13:39 |
| dtantsur | thrash, sorry for the noise :) I'm learning something new about nova-ironic interaction every day | 13:40 |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Use Keystone internal endpoint instead of admin for nova-ironic https://review.openstack.org/433038 | 13:40 |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Add release note for services endpoint change https://review.openstack.org/433181 | 13:40 |
| thrash | dtantsur: no worries. I have even less knowledge about it. I'm just basically porting this stuff. So it makes sense to improve while we can. | 13:40 |
| dtantsur | yeah. I'm only worried about public interfaces - the internals can be rewritten later | 13:41 |
| openstackgerrit | Flavio Percoco proposed openstack/python-tripleoclient master: Deploy the undercloud with Heat https://review.openstack.org/351351 | 13:43 |
| mcornea | chem: thanks, I'll try it on my env | 13:44 |
| chem | mcornea: could you paste me the output of /etc/puppet/hieradata on the compute node before non-controller upgarde (after composable upgrade though) ? | 13:46 |
| EmilienM | tripleo meeting in 15 min, feel free to add topics: https://etherpad.openstack.org/p/tripleo-meeting-items | 13:46 |
| mcornea | chem: sure, in a sec | 13:46 |
| *** bfournie has joined #tripleo | 13:46 | |
| mcornea | chem: http://paste.openstack.org/show/598819/ | 13:46 |
| *** morazi has quit IRC | 13:46 | |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-heat-templates master: Containers: Add required EndpointMap parameter https://review.openstack.org/433655 | 13:48 |
| *** jpena is now known as jpena|lunch | 13:48 | |
| chem | mcornea: thanks | 13:48 |
| chem | mcornea: excellent, the patch should work :) | 13:49 |
| *** trown is now known as trown|brb | 13:49 | |
| *** amoralej is now known as amoralej|lunch | 13:50 | |
| EmilienM | gfidente: ++ thx for your work on https://review.openstack.org/#/c/423386 | 13:51 |
| EmilienM | gfidente: I hope we can have it on time for ocata :D | 13:51 |
| EmilienM | gfidente: but not sure your patch in heat will be backported | 13:51 |
| openstackgerrit | Merged openstack/tripleo-ui master: Bump TripleO-UI to 3.0.0 for Ocata https://review.openstack.org/433248 | 13:51 |
| openstackgerrit | Merged openstack/tripleo-ui master: Add i18n to node driver fields https://review.openstack.org/432979 | 13:51 |
| *** sudipto has quit IRC | 13:52 | |
| *** sudipto_ has quit IRC | 13:52 | |
| gfidente | EmilienM yeah we need someone to merge that | 13:52 |
| gfidente | first | 13:52 |
| ansiwen | marios: so, the composable upgrades here will replace these major_upgrade... scripts? https://review.openstack.org/#/c/405241 | 13:52 |
| gfidente | shardy https://review.openstack.org/#/c/424061/ how does it look? | 13:52 |
| marios | ansiwen: right ... in parituclar you would be interested in nova-api for the migrations like https://review.openstack.org/#/c/405241/40/puppet/services/nova-api.yaml@277 | 13:53 |
| EmilienM | therve, zaneb: can you guys look https://review.openstack.org/#/c/424061/ please? | 13:53 |
| ansiwen | marios: yeah, sure, that part I know already | 13:53 |
| EmilienM | gfidente: we'll need a bug report in launchpad to backport it in stable/ocata though | 13:53 |
| shardy | gfidente: lgtm | 13:54 |
| *** pkovar has quit IRC | 13:54 | |
| ansiwen | marios: ok, looks as soon as 424061 is merged, we are all set with online_data_migrations | 13:55 |
| *** jcoufal has joined #tripleo | 13:56 | |
| *** ansmith has joined #tripleo | 13:56 | |
| marios | matbu: just grabbed latest/v40 for nova fyi... i re-enabled the pre-upgrade steps (I had them in SkipUpgradeConfigTags and it failed for me with http://paste.openstack.org/raw/598822/ i think you mentioned v40 fixes the transport url so hope itwill be able to start nova-api this time lets see | 13:56 |
| *** pkovar has joined #tripleo | 13:56 | |
| marios | nice one ansiwen | 13:57 |
| marios | matbu: yeah looks like https://review.openstack.org/#/c/405241/39..40/puppet/services/nova-api.yaml | 13:57 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 13:58 |
| matbu | marios: yep on the previous one, the transport_url was badly build | 13:58 |
| matbu | and i forgot the .stdout | 13:58 |
| shardy | bandini: thanks for the comment on https://review.openstack.org/#/c/433607, I'm thinking of another way to do it that will work with custom roles | 13:59 |
| marios | matbu: cool is starting now (it takes ~7/8 minutes after i hit return to get a heat stack update output | 13:59 |
| marios | matbu: actually more | 13:59 |
| EmilienM | meeting time | 13:59 |
| *** nyechiel has joined #tripleo | 13:59 | |
| matbu | marios: i was so tired yesterday, i didn't understand why the url was wrong, i just discovered this morning it was because of the .stdout on the port number | 13:59 |
| bandini | shardy: ah that'd be awesome, thanks! | 13:59 |
| *** sasha2 has joined #tripleo | 13:59 | |
| marios | thanks EmilienM | 13:59 |
| openstackgerrit | Julie Pichon proposed openstack/tripleo-ui master: Add support for German language https://review.openstack.org/433200 | 14:00 |
| *** fultonj has quit IRC | 14:01 | |
| *** fultonj has joined #tripleo | 14:03 | |
| *** cdearborn has joined #tripleo | 14:04 | |
| *** jprovazn has quit IRC | 14:04 | |
| *** trown|brb is now known as trown | 14:07 | |
| *** tzumainn has joined #tripleo | 14:08 | |
| *** suuuper has quit IRC | 14:08 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Apply puppet in non-controller script in step. https://review.openstack.org/433641 | 14:08 |
| *** suuuper has joined #tripleo | 14:08 | |
| openstackgerrit | Merged openstack/tripleo-quickstart-extras master: Update build nodes task to use the cloud config https://review.openstack.org/428193 | 14:09 |
| *** salmankhan has quit IRC | 14:13 | |
| *** dtrainor has joined #tripleo | 14:13 | |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart-extras master: Fix overcloud-scale failure msg and regression https://review.openstack.org/425352 | 14:15 |
| ccamacho | hey paramite I just finished to test all the services again, its safe to un-comment the collectd lines, I had an issue with my repost and was using the wrong paths when upgrading, overall LGTM for the upgrades job. The only thing is that my env gets crazy and super slow after upgrading the packages.. (Not sure if related to sensu-client) | 14:15 |
| marios | chem: sorry i _still_ didn't check out your review btw.. keep getting distracted but will do properly in a bit | 14:15 |
| marios | chem: first pass looks pretty good fwiw | 14:17 |
| *** salmankhan has joined #tripleo | 14:18 | |
| *** lblanchard has joined #tripleo | 14:19 | |
| chem | marios: it would be my first trully blind review :) | 14:19 |
| marios | chem: really really nice i like it (is cool you already have the $ROLE available and can pass in all the required params to the puppet apply | 14:19 |
| marios | chem: lets see how it goes in practice :) | 14:19 |
| openstackgerrit | Harry Rybacki proposed openstack/tripleo-quickstart-extras master: Remove hardcoded instances of `stack` from roles https://review.openstack.org/421463 | 14:20 |
| zaneb | EmilienM: approved | 14:20 |
| chem | marios: going to try to fix my env, so that I can still have 3 day of upstream productivity \o/ | 14:20 |
| EmilienM | zaneb: thx. Do you think we can backport it to stable/ocata & release it? cc gfidente | 14:20 |
| zaneb | EmilienM: I don't see why not | 14:21 |
| marios | chem: i think escape nit at https://review.openstack.org/#/c/433641/3/extraconfig/tasks/tripleo_upgrade_node.sh | 14:21 |
| chem | marios: you're correct | 14:22 |
| chem | marios: fixing it | 14:22 |
| marios | chem: cool just updated again i think the next one too | 14:23 |
| EmilienM | zaneb: thanks! | 14:23 |
| *** toure|biab is now known as toure | 14:23 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Apply puppet in non-controller script in step. https://review.openstack.org/433641 | 14:23 |
| *** bkopilov has joined #tripleo | 14:24 | |
| *** bkopilov_ has joined #tripleo | 14:24 | |
| mcornea | chem: /root/tripleo_upgrade_node.sh: line 117: ROLE: unbound variable | 14:25 |
| openstackgerrit | Julie Pichon proposed openstack/puppet-tripleo master: Enable languages in UI config https://review.openstack.org/432328 | 14:25 |
| chem | mcornea: I've just fixed it in the last review, marios spotted it just a minute ago. | 14:25 |
| mcornea | chem: ah, cool, will update.thanks | 14:26 |
| owalsh | matbu, marios: I'm sick today, on IRC just for the weekly meeting. I'll review https://review.openstack.org/405241 now but might not be around much after | 14:26 |
| marios | thanks owalsh... sorry about that upgrades sometimes do that to people. get well soon! | 14:26 |
| owalsh | :-D the transport_url jinga does that to me | 14:27 |
| openstackgerrit | Michael Glaser proposed openstack/puppet-tripleo master: Make quotes consistent to match the sample config https://review.openstack.org/432406 | 14:27 |
| owalsh | but it's self-inflicted... | 14:28 |
| matbu | owalsh: lol agree, i tried to find something more "userfriendly" | 14:28 |
| matbu | but no luck yours was the good :) | 14:28 |
| *** egafford has quit IRC | 14:28 | |
| owalsh | matbu: IPv6 is still an issue yea? | 14:28 |
| matbu | owalsh: hm i guess | 14:28 |
| *** saneax is now known as saneax-_-|AFK | 14:28 | |
| *** links has joined #tripleo | 14:29 | |
| matbu | owalsh: but there is a bunch of tool to catch that with ansible | 14:29 |
| matbu | owalsh: just need to find the right way | 14:29 |
| owalsh | matbu: ok, so we raise a bug for IPv6 and merge what we have now? | 14:30 |
| *** egafford has joined #tripleo | 14:33 | |
| matbu | owalsh: yep im fine w/ that | 14:35 |
| *** jprovazn has joined #tripleo | 14:36 | |
| *** nmathew- has quit IRC | 14:37 | |
| openstackgerrit | Merged openstack/tripleo-quickstart master: Fix and improve flake8 test. https://review.openstack.org/428689 | 14:37 |
| jrist | I was going to bring you popcorn at the ptg EmilienM | 14:40 |
| jrist | but now I'm not going to | 14:40 |
| chem | matbu: so I'm stuck on hiera group config (compute, controller), what am I missing ? | 14:40 |
| jrist | :) | 14:40 |
| jrist | mostly because I'm not going to the PTG | 14:40 |
| EmilienM | jtomasek, jrist, ccamacho: I'm preparing an email to Heidi - if you see that I missed something feel free to reply (on openstack-dev) | 14:41 |
| EmilienM | jrist: /me sad | 14:41 |
| jrist | talk to the guys who make the budget! | 14:41 |
| *** NobodyCam has quit IRC | 14:42 | |
| *** patrickeast has quit IRC | 14:42 | |
| *** Ng has quit IRC | 14:42 | |
| *** tbarron has quit IRC | 14:43 | |
| *** patrickeast has joined #tripleo | 14:44 | |
| chem | matbu: I don't have python-heat-agent-hiera installed and cannot find it in the repo I've configured | 14:45 |
| *** jmelvin has joined #tripleo | 14:45 | |
| *** flepied has quit IRC | 14:45 | |
| chem | matbu: could you paste/point me to your upgrade repo configuration ? | 14:45 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 14:45 |
| EmilienM | slagle: I haven't taken time yet but dprince and I shared the same idea that we could reduce the number of steps to 4 because of the recent changes that we did | 14:45 |
| EmilienM | slagle: regarding my recent changes in puppet-tripleo regarding steps, I had no choice because of the nova issues we had. Maybe I missed another approach, in that case I'm very open to change it | 14:46 |
| *** bkopilov_ has quit IRC | 14:47 | |
| *** bkopilov has quit IRC | 14:47 | |
| *** NobodyCam has joined #tripleo | 14:47 | |
| *** tbarron has joined #tripleo | 14:48 | |
| *** amoralej|lunch is now known as amoralej | 14:49 | |
| *** jpena|lunch is now known as jpena | 14:52 | |
| chem | matbu: oki, nevermind, I think I found my mistake | 14:53 |
| matbu | chem: sorry im on mtg | 14:55 |
| chem | matbu:np | 14:55 |
| jrist | EmilienM: perfect reply | 14:55 |
| *** derekh has quit IRC | 14:56 | |
| *** Goneri has joined #tripleo | 14:56 | |
| shardy | chem: http://paste.fedoraproject.org/558039/08416714 is my UpgradeInitCommand | 14:57 |
| chem | shardy: thanks | 14:57 |
| EmilienM | jrist: thx | 14:57 |
| shardy | chem: some of it can be removed when https://review.openstack.org/#/c/424715/ lands | 14:57 |
| *** flepied has joined #tripleo | 15:00 | |
| *** dtantsur is now known as dtantsur|mtg | 15:00 | |
| slagle | EmilienM: i dont know what we could do in terms of quick fixes | 15:00 |
| slagle | EmilienM: personally, i'd like to see us not have to reassert the world on every step | 15:01 |
| slagle | but that probably violates some puppet mantra :) | 15:01 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates master: Adds newton_ocata_upgrade_init_common.sh to the upgrade init https://review.openstack.org/424715 | 15:01 |
| slagle | e.g., we do "if step >= x" | 15:01 |
| EmilienM | slagle: yes, we would break some rules | 15:01 |
| slagle | well, if it's an initial deployment we already know the previous steps have succeeded | 15:02 |
| EmilienM | slagle: I don't say it wouldn't work, I just say it would not meet usual Puppet deployments | 15:02 |
| slagle | right | 15:02 |
| slagle | i mean, it probably takes keystone just as long to query for an endpoint as it does to create one | 15:02 |
| EmilienM | the power of Puppet is to make sure a system is in a specific state | 15:02 |
| EmilienM | the problem I see is resource dependency | 15:02 |
| EmilienM | if we start doing : if step == 3; then manage keystone endpoints; fi | 15:03 |
| EmilienM | I'm afraid some resources will fail at step 4 | 15:03 |
| EmilienM | because dependencies won't be found | 15:03 |
| *** hewbrocca_afk is now known as hewbrocca | 15:04 | |
| *** bogdando has joined #tripleo | 15:04 | |
| openstackgerrit | yolanda.robla proposed openstack/tripleo-quickstart-extras master: WIP: Add disk root device hints to OOOQ https://review.openstack.org/433691 | 15:06 |
| openstackgerrit | Arx Cruz proposed openstack-infra/tripleo-ci master: Removing enable_cinder_backup from periodic job https://review.openstack.org/433692 | 15:07 |
| *** bogdando has quit IRC | 15:07 | |
| *** ipsecguy has quit IRC | 15:07 | |
| *** bogdando has joined #tripleo | 15:07 | |
| openstackgerrit | Bernard Cafarelli proposed openstack/diskimage-builder master: pip-and-virtualenv: also handle rhel distros https://review.openstack.org/433693 | 15:08 |
| *** abregman is now known as abregman|brb | 15:08 | |
| openstackgerrit | Arx Cruz proposed openstack/tripleo-quickstart-extras master: Skipping tests requiring cinder-backup https://review.openstack.org/433099 | 15:09 |
| *** owalsh is now known as owalsh_sick | 15:09 | |
| *** dmarlin has joined #tripleo | 15:11 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: Deploy novajoin in the TLS-everywhere job https://review.openstack.org/427481 | 15:12 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job (novajoin) https://review.openstack.org/427482 | 15:12 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 15:12 |
| *** ealcaniz has quit IRC | 15:14 | |
| *** yamahata has joined #tripleo | 15:15 | |
| EmilienM | shardy: let's merge https://review.openstack.org/#/c/433627/ | 15:15 |
| EmilienM | shardy: unless you like my fix https://review.openstack.org/#/c/433646/1 | 15:15 |
| EmilienM | shardy: but matbu -1 | 15:15 |
| *** bkopilov has joined #tripleo | 15:15 | |
| *** bkopilov_ has joined #tripleo | 15:15 | |
| EmilienM | shardy: fyi my fix pass CI (look in zuul) | 15:16 |
| shardy | EmilienM: Ok, I'm fine with whichever you prefer | 15:16 |
| shardy | we can land the revert then revert the revert later if you like | 15:17 |
| *** prateek has quit IRC | 15:17 | |
| EmilienM | matbu: any opinion? | 15:17 |
| EmilienM | shardy: I think matbu is right, I can improve my patch and I don't mind doing it. Let's land your revert now, so we can make progress | 15:19 |
| shardy | EmilienM: Ok, wfm, thanks! | 15:20 |
| EmilienM | can someone approve this revert please? https://review.openstack.org/#/c/433627/ | 15:20 |
| EmilienM | trown: ^ | 15:23 |
| *** saibarspeis has quit IRC | 15:23 | |
| *** twi99y__ has quit IRC | 15:24 | |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Set mysql client bind address via ServiceNetMap https://review.openstack.org/433607 | 15:24 |
| trown | I reverted solely on green CI... did not actually look at the revert | 15:24 |
| *** twi99y has joined #tripleo | 15:24 | |
| matbu | EmilienM: sorry i was on mtg | 15:24 |
| EmilienM | trown: thanks! | 15:25 |
| *** bogdando has quit IRC | 15:25 | |
| *** twi99y_ has joined #tripleo | 15:26 | |
| *** twi99y_ has quit IRC | 15:27 | |
| *** twi99y has quit IRC | 15:27 | |
| *** jcromer has joined #tripleo | 15:27 | |
| *** twi99y has joined #tripleo | 15:28 | |
| jcromer | hello, i would like to ask a question about an issue i am having with validation when attempting to deploy an overcloud | 15:28 |
| ansiwen | shardy, marios, matbu: regarding the post-upgrade online_data_migration: this tool will actually run in a loop, trying to migrate as much as possible, and end if everything has been migrated, or it cannot further migrate the missing parts. in the second case we have to give feedback to the operator that manual steps are necessary to finish the upgrade. do we have a possibility to do that? | 15:28 |
| chem | does someone know how ntp was configured on the undercloud before, there is the --ntp-server for overcloud but for undercloud ? | 15:29 |
| marios | ansiwen: if you mean 'signal' not that i can think of (maybe in logs) but i'd think it would be more of a docs thing perhaps? (speking of which we should start some of those soon....) | 15:29 |
| trown | sshnaidm: it is a bit annoying that for the stable releases we have current-tripleo-$release but for master we have current-tripleo (not current-tripleo-master) | 15:29 |
| trown | sshnaidm: means we require some logic to check if release is master and do special case | 15:30 |
| sshnaidm | trown, we can do symlink | 15:30 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Revert "Revert "toci_gate_test: allow to run upgrade jobs for scenarios"" https://review.openstack.org/433715 | 15:30 |
| ansiwen | marios: well, if the tool is successful, there would be no manual step necessary. so it would really depend on the result of the tool. | 15:30 |
| ansiwen | marios: or you mean to document: check the logs, and if they say, unmigrated stuff still existing, then doing the manual migration? | 15:31 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Set mysql client bind address via ServiceNetMap https://review.openstack.org/433607 | 15:32 |
| sshnaidm | trown, http://66.187.229.139/builds/current-tripleo-master/ | 15:32 |
| trown | sshnaidm: hmm... maybe I just make it with some redundant information as first pass... I think I can make it without logic that way | 15:32 |
| openstackgerrit | Merged openstack/tripleo-ui master: Add support for German language https://review.openstack.org/433200 | 15:32 |
| marios | ansiwen: yeah though i was thinkging of what you said too... i mean we could have a step like the last step to check particular thing/file which tells us the migrations completed... is there such a thing? | 15:32 |
| trown | sshnaidm: ya, but the link is on RDO DLRN side too | 15:32 |
| marios | ansiwen: but i don't know if that would be enough time for the imgrations to complete (probably not) | 15:32 |
| sshnaidm | trown, ok, let's handle this after the mtg | 15:32 |
| shardy | marios, ansiwen: would one possibility be having some sort of check that only happens during the converge update? | 15:33 |
| shardy | so you start migrating things after the control-plane upgrade, then check it's done after the computes are all upgraded and you converge? | 15:34 |
| marios | shardy: so i thought we werent doing that i mean converge | 15:34 |
| openstackgerrit | Merged openstack/tripleo-quickstart master: Do not check for nested params if virt is not enabled https://review.openstack.org/429212 | 15:34 |
| marios | shardy: i.e. we 'fully' upgrade the controllers (package update and then puppet) and we do the same for computes (package update and puppet) | 15:34 |
| marios | shardy: but yeah if we do, then that would be the place to check/do it | 15:34 |
| shardy | marios: Oh, previous discussion with owalsh_sick gave me the impresion we still had to do some nova things after the computes were upgraded | 15:35 |
| marios | shardy: hopefully that gives enough time | 15:35 |
| marios | shardy: i missed that conversation sorry | 15:35 |
| shardy | I've been out of date on the latest with the nova things though | 15:35 |
| owalsh_sick | marios, shardy: just need to remove the upgrade pin afterwards and bounce the services IIRC | 15:35 |
| shardy | marios: Ok, it was a couple of weeks ago, let me find logs to remember why | 15:35 |
| marios | shardy: yeah i wasn't aware of a new need for a converge step. I mean, it should hurt to re-apply puppet everywhere but its an extra step and if it isn't going to do anything/serve some purpose then we should skip it | 15:35 |
| *** dtrainor_ has joined #tripleo | 15:36 | |
| marios | owalsh_sick: so the unpin should be ok now we are doing auto or do we still need to remove it? | 15:36 |
| shardy | marios: yeah IIRC "auto" doesn't actually do what we expected | 15:36 |
| marios | owalsh_sick: services bounce, yeah possibly we miss that part by doing only yum update and puppet apply | 15:36 |
| marios | shardy: oh damn so we need to use the upgrade pin after all ? | 15:36 |
| *** dtrainor has quit IRC | 15:37 | |
| *** funzo has joined #tripleo | 15:37 | |
| marios | funzo: o/ :D | 15:38 |
| funzo | marios: hey bud! | 15:38 |
| EmilienM | funzo: :-O omg | 15:38 |
| EmilienM | is it a funzo here? | 15:39 |
| *** jprovazn has quit IRC | 15:40 | |
| ansiwen | shardy: yes, the online_data_migrations can only finish, when all computes are upgraded, for example | 15:41 |
| *** jlinkes has quit IRC | 15:41 | |
| owalsh_sick | marios: #3 in http://docs.openstack.org/developer/nova/upgrade.html#rolling-upgrade-process. Actually we just need a SIGHUP with version pinned to auto | 15:42 |
| rook | i heard the rumor about funzo! | 15:42 |
| *** dtrainor_ has quit IRC | 15:42 | |
| rook | wb funzo | 15:42 |
| funzo | hey rook! thanks very much | 15:43 |
| *** jpena is now known as jpena|brb | 15:43 | |
| marios | ansiwen: cool, so that tells us that "converge" would definitely be the place to check (i.e. after compute upgrade) but not that we need the converge yet... thanks owalsh_sick looking | 15:43 |
| rook | shardy: heya i wanted to get your thoughts on a bug regarding docs which i think we should patch upstream too.. Right now we have zero recommendations around deployment, ie # of nodes to deploy at a time or introspect. | 15:44 |
| shardy | marios: Ok so looking at logs I think we said the initial upgrade had to either pin to "newton" or set "auto", then we need to set the version to "" in converge | 15:44 |
| shardy | Sorry I dropped looking into that due to PTO but we'll need to give it a try | 15:45 |
| shardy | I guess we'll also want to adjust the CI job to do two updates when nova lands | 15:45 |
| marios | shardy: right so sounds like regardless of whether auto is working, we still need to unpin after all the things are upgraded | 15:46 |
| shardy | rook: Hey, yeah that would be very useful to add to the docs if you've got numbers to share | 15:46 |
| ansiwen | marios, shardy: unfortunately I'm not sure if the compute nodes are not upgraded yet, if the tool would give up retrying... then maybe it would be even better to do it only in the converge step, when we know that all computes are upgraded? | 15:46 |
| *** skramaja has joined #tripleo | 15:46 | |
| rook | shardy yeah wrt deployment, I think it makes sense to say deploy a rack at a time, so recommend 32 nodes at a time | 15:46 |
| marios | owalsh_sick: so the shigup we could deliver with teh ansible steps i expect? and we are already pinning in nova-api i think for the nova review | 15:46 |
| rook | obviously we can do more (though not much more). | 15:47 |
| marios | owalsh_sick: i mean we have a full rstart for the nova-api already | 15:47 |
| shardy | ansiwen: is this the thing we have to run an indeterminate number of times? | 15:47 |
| shardy | marios: I thought the sighup came after the upgrade to unpin (or rather re-detect the version), but I may be mistaken | 15:47 |
| *** links has quit IRC | 15:47 | |
| shardy | marios: in which case we can't use the ansible steps as the converge should disable the ansible upgrade workflow | 15:48 |
| shardy | and just re-apply puppet | 15:48 |
| marios | shardy: i see well both i guess, to pin and unpin... yeah we can't for the unpin i mean use ansible | 15:48 |
| shardy | marios: yeah, perhaps it'll be easier to go back to explicitly pinning, then I suppose puppet will restart things when needed | 15:48 |
| marios | i mean i expect we need the sighup both for pinning | 15:48 |
| marios | (and unpinning) | 15:49 |
| owalsh_sick | shardy: going from no-pin -> auto -> no-pin is the same as explicit pinning I assume | 15:49 |
| *** dtrainor has joined #tripleo | 15:49 | |
| owalsh_sick | as far as puppet is concerned ^^^ | 15:50 |
| panda | oh, today is valentine day. No wonder qpid is working so well today. | 15:50 |
| shardy | owalsh_sick: I guess so, I was thinking puppet would bounce the services whenever we change the pin value | 15:50 |
| shardy | so we could do "newton" -> "" or "newton" -> "auto", or whatever | 15:50 |
| *** abregman|brb is now known as abregman | 15:50 | |
| owalsh_sick | as it stands I think we go from "" -> "auto" (upgrade) -> "" (converge) | 15:51 |
| *** ansmith has quit IRC | 15:52 | |
| *** gatuus has joined #tripleo | 15:53 | |
| shardy | owalsh_sick: Ok, and that should work for any version provided we do the sequence of updates e.g add the converge? | 15:53 |
| *** abregman is now known as abregman|mtg | 15:54 | |
| owalsh_sick | shardy: yes. Maybe in future we could replace converge with an ansible postupgrade step but it would need to run on the computes too | 15:54 |
| *** ramishra has quit IRC | 15:54 | |
| *** jaosorior has quit IRC | 15:54 | |
| shardy | owalsh_sick: Ok, cool, sounds like it's workable then, thanks for clarifying | 15:55 |
| shardy | marios: ^^ so we'll do the upgrade -> converge workflow as before, at least for this release | 15:55 |
| marios | shardy: ack, so regardless of what we pin to for the duration of hte upgrade, we need the converge just to unpin | 15:56 |
| marios | shardy: we could consider disabling the postdeploysteps (puppet) if we don't need them though | 15:56 |
| marios | shardy: i mean for time, assuming we already ran that everywhere right | 15:56 |
| *** jtomasek has quit IRC | 15:59 | |
| openstackgerrit | Attila Darazs proposed openstack/tripleo-quickstart-extras master: GATE CHECK https://review.openstack.org/430277 | 16:00 |
| *** jaganathan has quit IRC | 16:03 | |
| *** zoli is now known as zoli|brb | 16:03 | |
| openstackgerrit | Martin Mágr proposed openstack/tripleo-heat-templates master: Add upgrade tasks for opstools services https://review.openstack.org/431513 | 16:03 |
| openstackgerrit | Flavio Percoco proposed openstack/python-tripleoclient master: Deploy the undercloud with Heat https://review.openstack.org/351351 | 16:05 |
| shardy | marios: I don't think we can disable the puppet steps for the converge, only the upgrade ansible steps | 16:05 |
| *** pkovar has quit IRC | 16:05 | |
| shardy | marios: otherwise any subsequent update would have to remember to re-enable the puppet steps? | 16:06 |
| *** abregman|mtg is now known as abregman | 16:06 | |
| shardy | I think we want to leave things post-converge as they were before the upgrade process, e.g puppet enabled and no ansible steps? | 16:06 |
| *** tremble has quit IRC | 16:06 | |
| *** gbarros has joined #tripleo | 16:07 | |
| *** gbarros has quit IRC | 16:07 | |
| marios | shardy: OK, i was thinking a converge env file where OS::TripleO::PostDeploySteps:None | 16:07 |
| *** ansmith has joined #tripleo | 16:07 | |
| *** dtantsur|mtg is now known as dtantsur | 16:07 | |
| shardy | marios: yeah, I think the issue with that is we want puppet to bounce the nova services when we remove the pin | 16:07 |
| marios | shardy: but sure lets run puppet as well on convrge it can't hurt its just a atime factor i was thinking of | 16:08 |
| shardy | marios: and also that we don't want folks to need to remember to back out the None mapping later | 16:08 |
| shardy | marios: yeah, agreed, I guess we could consider it for CI if we have to | 16:08 |
| *** smalleni has quit IRC | 16:09 | |
| ansiwen | shardy: well, in theory you have to run it indefinite times, and possibly manually fix instances which are in transitional states, before it can finish it's job. but dansmith told me, that the tool would run in a loop, until it cannot further migrate stuff, or finished all migrations. so the idea was to let it run once at least, and then report to the operator, if manual intervention is necessary. | 16:10 |
| *** rook has quit IRC | 16:10 | |
| mcornea | chem: could we turn off puppet debug? it's quite verbose and I don't think it's necessary by default | 16:10 |
| *** lucasagomes has quit IRC | 16:10 | |
| ansiwen | shardy: because in many cases it would probably succeed. | 16:10 |
| *** rcernin has quit IRC | 16:11 | |
| ansiwen | shardy: what does "bounce the nova services" mean? | 16:11 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Add py35 to tox envlist https://review.openstack.org/431666 | 16:11 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Stop using deprecated mockpatch module https://review.openstack.org/431667 | 16:11 |
| *** rook has joined #tripleo | 16:12 | |
| shardy | ansiwen: restart them after the configuration (e.g the RPC pin) is changed | 16:12 |
| chem | mcornea: oki, will make the non-debug the default and try to find a way to have on demand | 16:12 |
| *** rook is now known as Guest47001 | 16:12 | |
| *** sai_ has joined #tripleo | 16:12 | |
| ansiwen | shardy: ack | 16:12 |
| *** lucasagomes has joined #tripleo | 16:12 | |
| *** sai_ is now known as smalleni | 16:13 | |
| *** twi99y_ has joined #tripleo | 16:13 | |
| *** dsariel has joined #tripleo | 16:14 | |
| *** twi99y has quit IRC | 16:16 | |
| *** fzdarsky|afk is now known as fzdarsky | 16:19 | |
| *** pkovar has joined #tripleo | 16:22 | |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci master: Nothing to see here https://review.openstack.org/433755 | 16:22 |
| chem | mcornea: beside the debug output does it seems to work ? | 16:23 |
| mcornea | chem: we've got some network interruption happening | 16:24 |
| chem | mcornea: arghh, oki :) | 16:24 |
| mcornea | chem: during upgrade the computes are not able to reach the rabbitmq servers | 16:25 |
| mcornea | chem: but I think your patch is working as expected, I'm not seeing missing hiera errors | 16:25 |
| *** zoli|brb is now known as zoli | 16:25 | |
| chem | mcornea: does it converge back at the end, ie they can reach rabbitmq eventually and everything fine after the compute upgrade ? | 16:26 |
| *** david-lyle has joined #tripleo | 16:26 | |
| mcornea | chem: if I restore connectivity yes, it completes but I'm trying to see why it breaks | 16:27 |
| *** iranzo has quit IRC | 16:27 | |
| *** jpena|brb is now known as jpena | 16:28 | |
| openstackgerrit | Numan Siddique proposed openstack/tripleo-heat-templates master: OVN bridge mappings for tripleo https://review.openstack.org/402293 | 16:28 |
| openstackgerrit | Merged openstack/tripleo-quickstart-extras master: Creating docs for repo setup role https://review.openstack.org/419523 | 16:28 |
| chem | mcornea: ack | 16:28 |
| *** david-lyle_ has quit IRC | 16:29 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci master: Revert "toci_gate_test: allow to run upgrade jobs for scenarios" https://review.openstack.org/433627 | 16:29 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Switch to net-config-multinode https://review.openstack.org/426372 | 16:29 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Switch to net-config-multinode https://review.openstack.org/426372 | 16:30 |
| gfidente | EmilienM https://review.openstack.org/#/c/425149 ? | 16:30 |
| EmilienM | sshnaidm: for https://review.openstack.org/#/c/433209/1/scripts/deploy.sh do you mind pushing over my patch? | 16:31 |
| mcornea | chem: I think it's openvswitch: br-infra<->tcp:127.0.0.1:6633: connection failed (Connection refused) | 16:31 |
| EmilienM | gfidente: looking | 16:31 |
| openstackgerrit | yolanda.robla proposed openstack/tripleo-quickstart-extras master: WIP: Add disk root device hints to OOOQ https://review.openstack.org/433691 | 16:31 |
| EmilienM | chem, mcornea, matbu, shardy: the revert landed, you can "recheck" patches for upgrades | 16:31 |
| matbu | ack | 16:32 |
| gfidente | EmilienM thanks | 16:33 |
| openstackgerrit | John Trowbridge proposed openstack-infra/tripleo-ci master: update promotion scripts for ocata https://review.openstack.org/427385 | 16:34 |
| openstackgerrit | John Trowbridge proposed openstack-infra/tripleo-ci master: update promotion scripts for ocata https://review.openstack.org/427385 | 16:36 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates master: Add batch upgrade configs to output & clarify disabled configs https://review.openstack.org/428349 | 16:36 |
| *** jprovazn has joined #tripleo | 16:37 | |
| openstackgerrit | Michele Baldessari proposed openstack/tripleo-heat-templates master: Release notes ha composable https://review.openstack.org/433765 | 16:40 |
| gfidente | lucasagomes you around? | 16:41 |
| lucasagomes | gfidente, yo, yes | 16:41 |
| *** Ng has joined #tripleo | 16:41 | |
| *** ChanServ sets mode: +v Ng | 16:41 | |
| ansiwen | shardy, marios: do we support to delay the compute node upgrades? | 16:44 |
| marios | ansiwen: well they are definitely distinct steps | 16:45 |
| marios | ansiwen: the compute upgrade happens at the operator discretion | 16:45 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Remove test-environments/multinode_major_upgrade.yaml https://review.openstack.org/426375 | 16:45 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: upgrades: switch 'newton' to $UPGRADE_RELEASE https://review.openstack.org/433209 | 16:45 |
| ansiwen | marios: [16:51:27] <dansmith> sven: since the operator could be delaying compute node upgrades, doesn't it make sense to not hold that until they're all upgraded and instead just tell them they need to re-run later? | 16:45 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: Deploy novajoin in the TLS-everywhere job https://review.openstack.org/427481 | 16:46 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci master: DO NOT MERGE - testing TLS everywhere job (novajoin) https://review.openstack.org/427482 | 16:46 |
| leanderthal | shardy++ | 16:48 |
| gfidente | lucasagomes that bug about ironic doing the disk partitions | 16:49 |
| *** rcernin has joined #tripleo | 16:49 | |
| openstackgerrit | Michael Glaser proposed openstack/puppet-tripleo master: Make quotes consistent to match the sample config https://review.openstack.org/432406 | 16:49 |
| lucasagomes | gfidente, from nokia ? | 16:49 |
| sshnaidm | EmilienM, you mean to change https://review.openstack.org/#/c/433209/1/scripts/deploy.sh ? | 16:50 |
| gfidente | yeah I think they have big root disks and want to use them with ceph too | 16:50 |
| gfidente | so I think they would like the root partition to end up being the size of the flavor | 16:50 |
| gfidente | (and growfs within that partition) | 16:50 |
| gfidente | but keep the rest of the disk space available to ceph | 16:50 |
| gfidente | now was wondering, isn't this exactly the behavior ironic has today? | 16:51 |
| gfidente | or is it creating the root partition as big as the entire disk? | 16:51 |
| lucasagomes | gfidente, very similar, actually Ironic just makes sure the root partition is the last one | 16:51 |
| EmilienM | sshnaidm: yes please | 16:51 |
| sshnaidm | EmilienM, sure | 16:51 |
| lucasagomes | and cloud-init grows it (see growpart) | 16:51 |
| gfidente | so the partition is of the size of the nova flavor, but it's the last one? | 16:52 |
| lucasagomes | gfidente, I thought they wanted to do things like, have a separated /var/ partition | 16:52 |
| *** Lokesh_Jain_ has joined #tripleo | 16:53 | |
| lucasagomes | gfidente, correct, the root is the last one and is the size of the "disk_gb" property in ironic (which is matched against a flavor) | 16:53 |
| lucasagomes | gfidente, I root being the last one because you also could have a swap and an ephemeral partition | 16:53 |
| *** trown is now known as trown|lunch | 16:54 | |
| gfidente | and is it using gpt or dos label? | 16:54 |
| EmilienM | sshnaidm: thx | 16:54 |
| *** agurenko has quit IRC | 16:56 | |
| *** jpich has quit IRC | 16:58 | |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 17:00 |
| EmilienM | owalsh_sick, matbu : you want https://github.com/openstack/puppet-openstacklib/commit/7ef2f7b0d726f8657d3c370979c52c5bd6440ea1 in newton? | 17:00 |
| *** jkilpatr_ has joined #tripleo | 17:00 | |
| *** sdake has joined #tripleo | 17:01 | |
| EmilienM | I don't think that's going to be helpful for upgrades, for folks who already deployed newton :-/ | 17:01 |
| *** sdake_ has quit IRC | 17:02 | |
| matbu | EmilienM: yep | 17:02 |
| matbu | i don't think too | 17:02 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: upgrades: switch 'newton' to $UPGRADE_RELEASE https://review.openstack.org/433209 | 17:02 |
| sshnaidm | EmilienM, ^^ | 17:02 |
| EmilienM | sshnaidm: thanks for the helpo | 17:02 |
| matbu | EmilienM: but it's not a matter the little puppet piece is working fine, so i will update the review, in test it in a full run | 17:02 |
| *** jkilpatr has quit IRC | 17:02 | |
| openstackgerrit | yolanda.robla proposed openstack/tripleo-quickstart-extras master: Add disk root device hints https://review.openstack.org/433691 | 17:02 |
| matbu | erf just saw the latest comment /me stupid have been tested it with ocata packages | 17:03 |
| ansiwen | flaper87: I'm just playing with flaper87/tripleo-undercloud-init-container | 17:04 |
| ansiwen | flaper87: but it fails with Unable to find image 'tripleo-undercloud-init-container:latest' locally | 17:04 |
| *** sdake has quit IRC | 17:05 | |
| openstackgerrit | yolanda.robla proposed openstack/tripleo-quickstart-extras master: Add disk root device hints https://review.openstack.org/433691 | 17:05 |
| owalsh_sick | EmilienM: yea, that occurred to me in the subsequent comments | 17:05 |
| *** panda is now known as panda|bbl | 17:07 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart master: Update ci-script to enable TripleO gate dependencies https://review.openstack.org/431764 | 17:07 |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci master: Add gate-tripleo-ci-centos-7-ovb-updates into promotion pipeline https://review.openstack.org/433796 | 17:07 |
| *** ggillies has quit IRC | 17:07 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Clone tripleo-quickstart dependant change to tripleo-quickstart-gate-repo https://review.openstack.org/431760 | 17:07 |
| *** sdake has joined #tripleo | 17:08 | |
| EmilienM | bnemec: FYI https://review.openstack.org/#/c/433796/ | 17:08 |
| *** milan has quit IRC | 17:09 | |
| bnemec | EmilienM: Nice, thanks | 17:09 |
| openstackgerrit | Merged openstack/instack-undercloud master: Fix ntp configuration. https://review.openstack.org/433623 | 17:09 |
| EmilienM | owalsh_sick: what is the plan? propose a backport? | 17:09 |
| *** dtantsur is now known as dtantsur|afk | 17:09 | |
| *** ggillies has joined #tripleo | 17:09 | |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart-extras master: Add composable upgrade workflow for tripleo-quickstart https://review.openstack.org/416480 | 17:11 |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart master: Add composable deployment config files and mixed release https://review.openstack.org/410831 | 17:12 |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart master: Add composable deployment config files and mixed release https://review.openstack.org/410831 | 17:13 |
| *** ansmith has quit IRC | 17:13 | |
| Lokesh_Jain_ | Guys.. how is ml2_conf and ml2_conf_sriov enabled on the neutron server at the same time on CentOS.. I can add the config manually.. but is there a way to do that automatically through some configuration parameter? | 17:14 |
| owalsh_sick | matbu, marios: could we move the tranport_url conf after step3? | 17:15 |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart-extras master: Implement overcloud sanity checks role https://review.openstack.org/429716 | 17:15 |
| *** leanderthal is now known as leanderthal|afk | 17:18 | |
| *** flepied has quit IRC | 17:19 | |
| *** gbarros has joined #tripleo | 17:20 | |
| *** pkovar has quit IRC | 17:20 | |
| *** skramaja has quit IRC | 17:21 | |
| *** yamahata has quit IRC | 17:22 | |
| *** dougbtv has quit IRC | 17:23 | |
| *** zoli is now known as zoli|gone | 17:23 | |
| *** gbarros has quit IRC | 17:24 | |
| *** karimb has quit IRC | 17:25 | |
| *** ansmith has joined #tripleo | 17:25 | |
| *** dougbtv has joined #tripleo | 17:25 | |
| *** sudipto has joined #tripleo | 17:28 | |
| *** sudipto_ has joined #tripleo | 17:28 | |
| *** pcaruana has quit IRC | 17:31 | |
| openstackgerrit | mathieu bultel proposed openstack/tripleo-quickstart-extras master: Implement overcloud sanity checks role https://review.openstack.org/429716 | 17:31 |
| *** hewbrocca is now known as hewbrocca_afk | 17:32 | |
| openstackgerrit | Jon Schlueter proposed openstack/tripleo-validations stable/newton: Don't rely on overcloudrc https://review.openstack.org/433812 | 17:32 |
| ansiwen | flaper87: so, the problem is, that in you run.sh you only use tripleo-undercloud-init-container as image name, but if you make a fresh install with your proposed line, the image only has the name flaper87/tripleo-undercloud-init-container | 17:34 |
| ansiwen | flaper87: after I gave the image an alias with docker tag it worked | 17:34 |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Describe use of Manila/CephFS in capabilites_map https://review.openstack.org/425149 | 17:34 |
| *** abehl has quit IRC | 17:35 | |
| *** pkovar has joined #tripleo | 17:35 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Disallow CIDR changes on undercloud update https://review.openstack.org/432388 | 17:37 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Change _member_role_exists to work with current upgrade flow https://review.openstack.org/432435 | 17:37 |
| *** b00tcat has quit IRC | 17:37 | |
| *** dsariel has quit IRC | 17:37 | |
| *** rasca has quit IRC | 17:37 | |
| *** fragatina has joined #tripleo | 17:38 | |
| *** b00tcat has joined #tripleo | 17:39 | |
| *** b00tcat has joined #tripleo | 17:39 | |
| *** fragatina has quit IRC | 17:43 | |
| *** abregman has quit IRC | 17:43 | |
| openstackgerrit | Matthew Flusche proposed openstack/tripleo-heat-templates master: FQDN validation https://review.openstack.org/433820 | 17:44 |
| *** rajinir_ has joined #tripleo | 17:46 | |
| *** gatuus has quit IRC | 17:47 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/instack-undercloud master: Handle service upgrades correctly when enable_telemetry=false https://review.openstack.org/433824 | 17:48 |
| *** gatuus has joined #tripleo | 17:48 | |
| *** myoung has quit IRC | 17:51 | |
| *** jschlueter|afk has quit IRC | 17:51 | |
| *** mhenkel_ has joined #tripleo | 17:55 | |
| *** gfidente is now known as gfidente|afk | 17:55 | |
| *** jschluet has joined #tripleo | 17:56 | |
| *** lucasagomes is now known as lucas-afk | 17:57 | |
| *** dbecker has joined #tripleo | 17:57 | |
| *** Guest47001 is now known as rook | 17:58 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Disallow CIDR changes on undercloud update https://review.openstack.org/432388 | 17:59 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Change _member_role_exists to work with current upgrade flow https://review.openstack.org/432435 | 17:59 |
| *** dsneddon has joined #tripleo | 18:00 | |
| *** jkilpatr has joined #tripleo | 18:01 | |
| *** jkilpatr_ has quit IRC | 18:01 | |
| *** morazi has joined #tripleo | 18:03 | |
| *** dbecker has quit IRC | 18:05 | |
| *** yamahata has joined #tripleo | 18:06 | |
| *** fragatina has joined #tripleo | 18:09 | |
| trozet | shardy: ping? | 18:15 |
| shardy | trozet: Hi! | 18:15 |
| trozet | shardy: hey man...for containerization, any pointers for me on how to start migrating ODL to that? | 18:15 |
| trozet | shardy: i guess we need 2 parts, build the kolla image, and patches to OOO for the service. If you have patchsets or a guide I can reference that would be great | 18:16 |
| shardy | trozet: I'd say chat with flaper87, dprince or Slower as I'm not currently working on container things much | 18:16 |
| *** tosky has quit IRC | 18:16 | |
| shardy | trozet: yeah I think those are basically the steps, flaper87 & co should be able to point to you specific examples | 18:17 |
| trozet | shardy: thanks. let's see if they can help me | 18:17 |
| *** dbecker has joined #tripleo | 18:18 | |
| *** athomas has quit IRC | 18:18 | |
| EmilienM | trozet: you can take example from dprince's patch on adding keystone container: https://review.openstack.org/#/c/416605/ | 18:20 |
| EmilienM | trozet: IIUC, you need to patch Kolla to support ODL to build a container with the right packages | 18:20 |
| trozet | EmilienM: so you think it will be acceptable to build the ODL container in the Kolla project? | 18:21 |
| EmilienM | trozet: other than that, there is nothing much to do, as it will re-use puppet to do what it does on classic deployments | 18:21 |
| EmilienM | trozet: why not? | 18:21 |
| trozet | EmilienM: since it isn't technically openstack, I didn't know if they would just incorporate that into the build system | 18:22 |
| *** ccamacho is now known as ccamacho|Zzz | 18:22 | |
| EmilienM | trozet: they have rabbimq containers | 18:22 |
| EmilienM | and it's not OpenStack | 18:22 |
| EmilienM | it's part of the ecosystem. TBH I don't know but I'm pretty sure it would be accepted. | 18:22 |
| trozet | EmilienM: ok. So it looks like current puppet-tripleo configuration doesn't need to change | 18:22 |
| EmilienM | trozet: not afik | 18:23 |
| trozet | EmilienM: ok cool, thanks | 18:24 |
| *** lblanchard has quit IRC | 18:24 | |
| *** salmankhan has quit IRC | 18:25 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/instack-undercloud master: Handle service upgrades correctly when enable_telemetry=false https://review.openstack.org/433824 | 18:25 |
| *** tesseract-RH has quit IRC | 18:26 | |
| *** cylopez has quit IRC | 18:27 | |
| *** lblanchard has joined #tripleo | 18:28 | |
| *** thrash is now known as thrash|biab | 18:28 | |
| *** trown|lunch is now known as trown | 18:32 | |
| *** sudipto_ has quit IRC | 18:33 | |
| *** sudipto has quit IRC | 18:33 | |
| trown | sshnaidm: I think https://review.openstack.org/427385 should now work with your stable release directories | 18:34 |
| sshnaidm | trown, will look | 18:34 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: Add option for scenarios jobs promotion https://review.openstack.org/425819 | 18:35 |
| *** sudipto has joined #tripleo | 18:35 | |
| *** sudipto_ has joined #tripleo | 18:35 | |
| *** sudipto_ has quit IRC | 18:35 | |
| *** sudipto has quit IRC | 18:35 | |
| sshnaidm | trown, can you name these $1, $2, $3? it'll be more readable what happens in the script.. | 18:39 |
| openstackgerrit | Ana Krivokapic proposed openstack/tripleo-common master: Enhance plan creation and update with plan-environment https://review.openstack.org/414169 | 18:39 |
| *** saibarspeis has joined #tripleo | 18:39 | |
| openstackgerrit | Ana Krivokapic proposed openstack/tripleo-common master: Add plan export action and workflow https://review.openstack.org/422789 | 18:40 |
| trown | sshnaidm: sure, they are named in the usage comment, but I will add separate comments for them | 18:40 |
| openstackgerrit | Ana Krivokapic proposed openstack/python-tripleoclient master: Add plan export command https://review.openstack.org/425858 | 18:41 |
| *** jpena is now known as jpena|away | 18:42 | |
| openstackgerrit | John Trowbridge proposed openstack-infra/tripleo-ci master: update promotion scripts for ocata https://review.openstack.org/427385 | 18:42 |
| trown | sshnaidm: like that ^ | 18:42 |
| sshnaidm | trown, thanks | 18:44 |
| *** b00tcat has quit IRC | 18:44 | |
| openstackgerrit | Ana Krivokapic proposed openstack/tripleo-heat-templates master: Add plan-environment.yaml https://review.openstack.org/417385 | 18:44 |
| *** diana_clarke has joined #tripleo | 18:45 | |
| *** saibarspeis has quit IRC | 18:45 | |
| *** b00tcat has joined #tripleo | 18:46 | |
| *** b00tcat has joined #tripleo | 18:46 | |
| *** suuuper has quit IRC | 18:47 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart-extras master: Add build-images role https://review.openstack.org/414336 | 18:51 |
| *** akrivoka has quit IRC | 18:53 | |
| *** myoung has joined #tripleo | 18:53 | |
| *** pkovar has quit IRC | 18:53 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci master: Collect only logs we need https://review.openstack.org/425459 | 18:56 |
| openstackgerrit | Merged openstack/tripleo-quickstart-extras master: Support exclude list for logs collecting https://review.openstack.org/425395 | 18:57 |
| *** fultonj has quit IRC | 18:58 | |
| *** thrash|biab is now known as thrash | 18:58 | |
| *** fultonj has joined #tripleo | 18:59 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart master: Update ci-script to enable TripleO gate dependencies https://review.openstack.org/431764 | 18:59 |
| openstackgerrit | Merged openstack/tripleo-quickstart-extras master: Skipping tests requiring cinder-backup https://review.openstack.org/433099 | 18:59 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart-extras master: Add image build directory to default collect list https://review.openstack.org/433859 | 19:00 |
| trown | panda|bbl: I think https://review.openstack.org/433859 is the only thing we need to collect the image build logs | 19:00 |
| *** shardy has quit IRC | 19:00 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Clone tripleo-quickstart dependant change to tripleo-quickstart-gate-repo https://review.openstack.org/431760 | 19:01 |
| *** iranzo has joined #tripleo | 19:03 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Add upgrade tasks for opstools services https://review.openstack.org/431513 | 19:05 |
| *** dbecker has quit IRC | 19:05 | |
| *** saneax-_-|AFK is now known as saneax | 19:06 | |
| *** b00tcat has quit IRC | 19:07 | |
| *** mcornea has quit IRC | 19:08 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Disallow CIDR changes on undercloud update https://review.openstack.org/432388 | 19:10 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Change _member_role_exists to work with current upgrade flow https://review.openstack.org/432435 | 19:10 |
| *** sshnaidm is now known as sshnaidm|afk | 19:11 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Disallow CIDR changes on undercloud update https://review.openstack.org/432388 | 19:15 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Change _member_role_exists to work with current upgrade flow https://review.openstack.org/432435 | 19:15 |
| *** rlandy is now known as rlandy|brb | 19:19 | |
| beagles | bnemec: should we wait until we have more info on ha jobs timeout before rechecking or just recheck away? | 19:19 |
| bnemec | beagles: Should be fine to recheck. They're not all failing, but we're close enough to the timeout that some of them are. | 19:21 |
| beagles | bnemec: ack, thanks | 19:21 |
| *** gbarros has joined #tripleo | 19:22 | |
| *** amoralej is now known as amoralej|off | 19:23 | |
| *** gbarros has quit IRC | 19:26 | |
| *** nyechiel has quit IRC | 19:39 | |
| flaper87 | ansiwen: a-ha, good catch. I should fix that! | 19:40 |
| flaper87 | trozet: hey, so, what can I help you with? | 19:40 |
| trozet | flaper87: hi | 19:40 |
| trozet | flaper87: I just need some guidance modifying the opendaylight service to be containerized | 19:41 |
| trozet | flaper87: EmilienM mentioned getting opendaylight built in the Kolla project as a container, so if you have any pointers there that may be a good start | 19:41 |
| *** rlandy|brb is now known as rlandy | 19:41 | |
| flaper87 | trozet: I can point you to some patches and some docs. Both things are still WIP but I hope they'll help to get you started. I'm happy to help you out if those turn out to not be enough | 19:42 |
| flaper87 | trozet: but I'd love for you to provide some feedback on both | 19:42 |
| flaper87 | trozet: 2s | 19:42 |
| trozet | flaper87: please do | 19:42 |
| ansiwen | flaper87: now it fails with ERROR: [Errno 21] Is a directory: '/usr/lib/python2.7/site-packages/tripleoclient/noauth_policy.json' | 19:42 |
| *** fragatina has quit IRC | 19:42 | |
| flaper87 | trozet: so, if there's no kolla image for odl, then yeah, you likely need to first create one | 19:43 |
| ansiwen | flaper87: and in fact that IS a directory, created in the host fs, although tripleoclient is not installed there | 19:43 |
| ansiwen | flaper87: why is it messing with the host filesystem? | 19:43 |
| trozet | flaper87: yeah we just build ODL into an rpm right now | 19:43 |
| *** fragatina has joined #tripleo | 19:43 | |
| flaper87 | ansiwen: yeah, that might be a new thing. Not sure if something changed in the last couple of days. Note that all this is very much under development | 19:43 |
| flaper87 | ansiwen: did you check if that file exists in the container? | 19:44 |
| ansiwen | after the error I have an empty directory named /usr/lib/python2.7/site-packages/tripleoclient/noauth_policy.json | 19:44 |
| flaper87 | trozet: ok, so, I recently created an image for redis, lemme get the bp | 19:44 |
| trozet | flaper87: thanks | 19:44 |
| flaper87 | trozet: https://blueprints.launchpad.net/kolla/+spec/redis-container | 19:44 |
| ansiwen | flaper87: I get that this is alpha stuff, just wanna understand the concept so I can help | 19:44 |
| flaper87 | trozet: I just created that bp, went to the kolla channel, got it approved and implemented the whole thing | 19:44 |
| *** trown is now known as trown|brb | 19:45 | |
| *** mhenkel_ has quit IRC | 19:45 | |
| trozet | flaper87: nice, I will use this as a reference | 19:45 |
| trozet | flaper87: so then after that...will need to come up with a patch for THT right? | 19:45 |
| *** mhenkel_ has joined #tripleo | 19:45 | |
| flaper87 | ansiwen: coolio, so, the `docker run` command, runs another container with extra mounts. This new container installs everything that's needed in the container and starts the install process | 19:45 |
| flaper87 | ansiwen: that file there should probably exist in the second container | 19:46 |
| flaper87 | ansiwen: the undercloud-deploy one | 19:46 |
| flaper87 | trozet: yes, lemme get some links for that | 19:46 |
| trozet | flaper87: EmilienM pointed me to a containerized keystone patch that I can reference for that. https://review.openstack.org/#/c/416605/33/docker/services/keystone.yaml Any other things I might need to take care of? | 19:46 |
| ansiwen | flaper87: https://bpaste.net/show/cafacc391ddd | 19:47 |
| flaper87 | trozet: https://review.openstack.org/#/c/422999/ | 19:47 |
| flaper87 | trozet: that's neutron's example. You may want to refer to that one since it's somewhat related | 19:47 |
| trozet | flaper87: great | 19:47 |
| flaper87 | trozet: https://review.openstack.org/#/c/430201/ these are some docs | 19:47 |
| flaper87 | trozet: sorry for stressing this so much but, while I don't think there will be major changes to the architecture, I do want to make it clear that this is in alpha stage | 19:48 |
| trozet | flaper87: so then how do the containers get deployed. Are they built into RDO and stored in the overcloud? | 19:48 |
| flaper87 | ansiwen: a-ha, I think I know why that's happening. 2s | 19:49 |
| ansiwen | flaper87: so, the mounts of this second container, are the bound to the first container or to the host system? I never run docker in a docker, I have no idea how this is meant to work | 19:49 |
| mwhahaha | bnemec: you aware of teh ovb ha failures? http://logs.openstack.org/85/431185/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/05312d4/console.html#_2017-02-14_18_39_33_651121 | 19:50 |
| *** mhenkel_ has quit IRC | 19:50 | |
| mwhahaha | bnemec: not sure if related, http://logs.openstack.org/85/431185/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/05312d4/logs/undercloud/var/log/ironic/ironic-conductor.txt.gz#_2017-02-14_18_07_17_979 | 19:50 |
| openstackgerrit | Oliver Walsh proposed openstack/tripleo-heat-templates master: Add nova service support for composable upgrades https://review.openstack.org/405241 | 19:52 |
| flaper87 | ansiwen: they are bound to paths in the host. This is new indeed. The example you saw on the email was running heat-all from the undercloud-deploy container, which is probably what we should do in this case. | 19:52 |
| flaper87 | ansiwen: the default is running it as a container | 19:52 |
| flaper87 | ansiwen: https://review.openstack.org/#/c/351351/32/tripleoclient/v1/undercloud_deploy.py | 19:53 |
| bnemec | mwhahaha: Hmm, that's strange. It looks like ironic was having trouble making IPMI calls, but I don't see any errors in the console log for that bmc. | 19:53 |
| flaper87 | ansiwen: L369 | 19:53 |
| flaper87 | we basically need to pass `--heat-native` to the openstack deploy command | 19:54 |
| mwhahaha | bnemec: ok i'll recheck, maybe it's transient | 19:54 |
| flaper87 | ansiwen: if you want to hack on this, I'd say you can put it in the scripts that are in here: https://github.com/flaper87/tripleo-undercloud-init-container | 19:54 |
| *** jcoufal_ has joined #tripleo | 19:54 | |
| flaper87 | ansiwen: if you only want to test this containers stuff, then I'd recommend you to get a fresh vm, clone https://github.com/dprince/undercloud_containers and run `./doit.sh` and then `./run.sh` | 19:55 |
| ansiwen | flaper87: I wanna do both, test the container deployment, and then help with it | 19:55 |
| bnemec | mwhahaha: Yeah, we see those occasionally, which is why we started capturing the bmc logs. Not too helpful in this case though. :-( | 19:56 |
| flaper87 | ansiwen: then try the undercloud_containers for now, familiarize with it and then we can hack on the init-container thing | 19:56 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Add composable upgrade workflow for tripleo-quickstart https://review.openstack.org/416480 | 19:56 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Implement overcloud sanity checks role https://review.openstack.org/429716 | 19:56 |
| ansiwen | flaper87: I don't understand the concept though, I'm running this on a bare metal host, and I expected that the host filesystem will stay untouched, because everything runs inside the container | 19:56 |
| *** jcoufal has quit IRC | 19:57 | |
| flaper87 | ansiwen: yeah, not everything. That's the problem. As I said in the email, there are still some paths that need to be used because the path mounts are on the host | 19:57 |
| flaper87 | A container running another container uses the docker daemon running on the host | 19:57 |
| flaper87 | so, as far as the docker daemon is concerned, the paths that the mounts refer to are in the host | 19:58 |
| ansiwen | flaper87: I see... | 19:58 |
| flaper87 | ansiwen: I think we can get to the point where the host is left untouched | 19:58 |
| flaper87 | but that's not top prio right now, although this is looking good already | 19:58 |
| flaper87 | :D | 19:58 |
| * flaper87 jumps on a meeting | 19:59 | |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 20:00 |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-docs master: Don't limit creation of undercloud.conf to baremetal https://review.openstack.org/433904 | 20:00 |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-docs master: Reference the undercloud configuration wizard in install docs https://review.openstack.org/433905 | 20:00 |
| ansiwen | flaper87: why not just using the conatiner name of the first container as the target, then it will be mounted from container to container... no host mounts required | 20:03 |
| *** trown|brb is now known as trown | 20:03 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Add sanity checks to the baremetal overcloud and upgrades playbooks https://review.openstack.org/433914 | 20:12 |
| *** saneax is now known as saneax-_-|AFK | 20:14 | |
| *** openstackgerrit has quit IRC | 20:18 | |
| *** dbecker has joined #tripleo | 20:20 | |
| *** sdake has quit IRC | 20:22 | |
| *** sdake has joined #tripleo | 20:22 | |
| *** jprovazn has quit IRC | 20:23 | |
| *** aufi has quit IRC | 20:24 | |
| *** twi99y has joined #tripleo | 20:31 | |
| *** flepied has joined #tripleo | 20:31 | |
| *** twi99y_ has quit IRC | 20:34 | |
| *** abregman has joined #tripleo | 20:39 | |
| EmilienM | I'm investigating the ovb-ha timeouts | 20:39 |
| EmilienM | http://logstash.openstack.org/#dashboard/file/logstash.json?query=build_name%3A%20*tripleo-ci*%20AND%20build_status%3A%20FAILURE%20AND%20message%3A%20%5C%22CREATE_FAILED%20%20Create%20timed%20out%5C%22 | 20:39 |
| EmilienM | sounds like it started on Feb 4th | 20:39 |
| EmilienM | with https://review.openstack.org/#/c/428783/ | 20:40 |
| EmilienM | I'll continue the investigation after tc meeting though, if anyone is interested to look ^ | 20:40 |
| *** iranzo has quit IRC | 20:47 | |
| *** athomas has joined #tripleo | 20:49 | |
| *** gbarros has joined #tripleo | 20:53 | |
| mwhahaha | https://review.openstack.org/#/c/431045/ +A anyone? :D | 20:54 |
| *** trown is now known as trown|brb | 20:55 | |
| *** gbarros has quit IRC | 20:58 | |
| EmilienM | mwhahaha: done | 21:01 |
| EmilienM | it sounds like deploying nova in eventlet causes performance issue | 21:03 |
| EmilienM | I'm going to investigate | 21:03 |
| mwhahaha | HEH | 21:03 |
| *** rhallisey has quit IRC | 21:04 | |
| *** fultonj has quit IRC | 21:07 | |
| *** jayg is now known as jayg|g0n3 | 21:08 | |
| *** fultonj has joined #tripleo | 21:11 | |
| *** jcromer has quit IRC | 21:12 | |
| *** jcromer has joined #tripleo | 21:13 | |
| *** jcromer has quit IRC | 21:15 | |
| *** jcromer has joined #tripleo | 21:16 | |
| *** eck` is now known as jeckersb_gone | 21:17 | |
| EmilienM | so in fact, logstash says the first timeout we have (in 1 month) was this one: | 21:17 |
| EmilienM | http://logs.openstack.org/78/428778/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/844707b/console.html#_2017-02-05_00_04_19_821326 | 21:17 |
| EmilienM | which seems unrelated to steps, since it happens before | 21:17 |
| *** jcromer has quit IRC | 21:17 | |
| *** openstackgerrit has joined #tripleo | 21:18 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: Add sanity checks to the baremetal overcloud and upgrades playbooks https://review.openstack.org/433914 | 21:18 |
| *** jkilpatr has quit IRC | 21:20 | |
| openstackgerrit | Martin Mágr proposed openstack/tripleo-heat-templates master: Add logging agents deployment to CI https://review.openstack.org/410210 | 21:21 |
| openstackgerrit | Martin Mágr proposed openstack/tripleo-heat-templates master: Add monitoring agents deployment to CI https://review.openstack.org/433931 | 21:21 |
| *** abregman has quit IRC | 21:22 | |
| *** trown|brb is now known as trown | 21:23 | |
| EmilienM | so I see 3 different problems : | 21:25 |
| EmilienM | 1. What bnemec reported: https://bugs.launchpad.net/tripleo/+bug/1664418 | 21:25 |
| openstack | Launchpad bug 1664418 in tripleo "HA deployments taking close to the CI timeout" [High,Triaged] | 21:25 |
| EmilienM | this one looks related to performances that recently got affected by steps changes (probably) | 21:25 |
| EmilienM | http://logstash.openstack.org/#dashboard/file/logstash.json?query=build_name%3A%20*tripleo-ci*%20AND%20build_status%3A%20FAILURE%20AND%20message%3A%20%5C%22%5Bovercloud.AllNodesDeploySteps.ControllerDeployment_Step5%5D%3A%20CREATE_FAILED%20%20CREATE%20aborted%5C%22 | 21:26 |
| openstackgerrit | Ben Nemec proposed openstack/tripleo-docs master: Update overcloud node specs https://review.openstack.org/433933 | 21:26 |
| *** paramite has quit IRC | 21:26 | |
| EmilienM | it first happenned on https://review.openstack.org/#/c/430357/ | 21:26 |
| EmilienM | which is a patch for upgrades, so not used by ovb-ha jobs | 21:27 |
| *** gatuus has quit IRC | 21:27 | |
| EmilienM | what bnemec reported is "only" 15 hits in 6 days | 21:28 |
| EmilienM | 2. what mwhahaha reported with http://logs.openstack.org/85/431185/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/05312d4/logs/undercloud/var/log/ironic/ironic-conductor.txt.gz#_2017-02-14_18_07_17_979 | 21:29 |
| EmilienM | mwhahaha: any bug report already on ^ ? | 21:30 |
| mwhahaha | i only spotted the one | 21:30 |
| mwhahaha | i rechecked to see if it persisted | 21:30 |
| EmilienM | only 5 hits in 4 days | 21:31 |
| EmilienM | http://logstash.openstack.org/#dashboard/file/logstash.json?query=build_name%3A%20*tripleo-ci*%20AND%20build_status%3A%20FAILURE%20AND%20message%3A%20%5C%22Message%3A%20Failed%20to%20provision%20instance%5C%22 | 21:31 |
| EmilienM | and 3. what I just found: http://logs.openstack.org/78/428778/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/844707b/console.html#_2017-02-05_00_04_19_821326 | 21:31 |
| EmilienM | another timeout but before the steps | 21:31 |
| *** athomas has quit IRC | 21:32 | |
| EmilienM | mwhahaha: could you please file a bug? i'm going to create elastic rechecks | 21:32 |
| mwhahaha | k | 21:32 |
| EmilienM | 3. worries me, 113 hits in 10 days | 21:34 |
| EmilienM | affecting all ovb jobs, not only HA | 21:34 |
| slagle | pradk: aodh-evaluator is tracebacking in my environment because python-redis is not installed | 21:35 |
| slagle | pradk: i'm not using overcloud-full images, and I deploy with EnablePackageInstall: True | 21:35 |
| slagle | pradk: so should the packaging require python-redis? | 21:35 |
| EmilienM | sounds like a bug in aodh spec | 21:35 |
| bnemec | EmilienM: That last one looks similar to the hang we had in the multinode jobs a while back. | 21:36 |
| EmilienM | slagle: https://github.com/rdo-packages/aodh-distgit/search?utf8=%E2%9C%93&q=redis | 21:36 |
| pradk | slagle, hmm its not a direct dependency its only needed when coordination is configured | 21:36 |
| EmilienM | pradk: imho it should be a dep | 21:36 |
| EmilienM | pradk: we can't check the config to see if user has coordination enabled | 21:36 |
| slagle | pradk: yes, but it's barfing tracebacks every couple of seconds | 21:36 |
| bnemec | EmilienM: https://bugs.launchpad.net/tripleo/+bug/1651267 | 21:37 |
| openstack | Launchpad bug 1651267 in tripleo "Timeouts in multinode jobs" [High,Triaged] | 21:37 |
| EmilienM | slagle: you have coordination enabled? | 21:37 |
| pradk | EmilienM, question is why is redis not installed | 21:37 |
| pradk | we have it in the puppetry to install it afair | 21:37 |
| EmilienM | bnemec: mhh | 21:37 |
| EmilienM | bnemec: how to debug this one? | 21:37 |
| slagle | EmilienM: it's pretty much a stock deployment. do we enable it by default? | 21:37 |
| EmilienM | pradk, slagle: is it in t-p-e ? | 21:37 |
| slagle | it wouldn't matter if it was | 21:38 |
| EmilienM | slagle: can I see aodh.conf please? | 21:38 |
| slagle | 16:35 < slagle> pradk: i'm not using overcloud-full images, and I deploy with EnablePackageInstall: True | 21:38 |
| slagle | EmilienM: ^^^ | 21:38 |
| EmilienM | ah right | 21:38 |
| EmilienM | well problem is 1. puppet-aodh won't install it for you AFIK and 2. it's not in t-p-e so you need something that installs it | 21:38 |
| EmilienM | and 3. it's not in aodh deps | 21:38 |
| pradk | EmilienM, it is in tpe | 21:38 |
| slagle | this is a bug then | 21:38 |
| slagle | it should be required by packaging, or installed by puppet | 21:39 |
| EmilienM | pradk: ok it's maybe in tpe but slagle doesn't use it | 21:39 |
| EmilienM | slagle: yes, +1 | 21:39 |
| mwhahaha | EmilienM: https://bugs.launchpad.net/tripleo/+bug/1664727 | 21:39 |
| openstack | Launchpad bug 1664727 in tripleo "ovb fails with CREATE_FAILED ResourceInError: resources.Controller: Went to status ERROR due to "Message: Failed to provision instance ... Timeout reached while waiting for callback"" [High,New] | 21:39 |
| EmilienM | imho we should just fix it in aodh-distgit | 21:39 |
| bnemec | EmilienM: It might be different though. I don't see issues with os-collect-config in http://logs.openstack.org/78/428778/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/844707b/logs/compute-0-tripleo-ci-a-test/var/log/messages | 21:39 |
| EmilienM | mwhahaha: thank you. /me create logstash queries now for elastic recheck | 21:39 |
| pradk | we can put it in the spec.. but i dont think it should be a direct dep | 21:39 |
| EmilienM | pradk: why? | 21:40 |
| pradk | i can use zookeeper instead of redis, should we add zookeeper too? | 21:40 |
| pradk | its a driver that tooz uses | 21:40 |
| EmilienM | mhh | 21:40 |
| slagle | does missing zookeeper cause tracebacks every 2 seconds? | 21:40 |
| slagle | or, fix the tracebacking :) | 21:40 |
| slagle | i don't care :) | 21:40 |
| EmilienM | slagle: please show the aodh.conf | 21:40 |
| pradk | if the config is using it then yes it will | 21:40 |
| pradk | i assume your url is redis:: | 21:40 |
| EmilienM | slagle: it's showing the redis backtracking because I'm sure your coordination url is set on redis URL | 21:41 |
| EmilienM | pradk: we need to fix it in puppet-aodh then? by introspecting the coordination url | 21:41 |
| slagle | EmilienM: is that what we configure by default? | 21:41 |
| EmilienM | pradk: like we do for database url | 21:41 |
| mwhahaha | pradk: EmilienM: i think you want something similar to what we do for memcache, https://github.com/openstack/puppet-oslo/blob/e0b9dd93e21803eae8e9b52f458c563fe028b58f/manifests/cache.pp#L152 | 21:42 |
| pradk | EmilienM, that would be the right way yea | 21:42 |
| EmilienM | slagle: https://github.com/openstack/puppet-tripleo/blob/1e26963bef085c7ec16b342594da6286235fe7da/manifests/profile/base/aodh/evaluator.pp#L34 | 21:42 |
| EmilienM | slagle: if you have redis deployed, then yes this flag is enabled | 21:43 |
| EmilienM | pradk: exactly what mwhahaha said! | 21:43 |
| *** rcernin has quit IRC | 21:44 | |
| pradk | yea i agree .. we should fix this in puppet | 21:44 |
| pradk | i can look into it in a bit | 21:45 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci master: [WIP] Quickstart transition: replace toci_gate_test.sh https://review.openstack.org/431029 | 21:45 |
| EmilienM | pradk: thanks | 21:47 |
| *** ansmith has quit IRC | 21:48 | |
| slagle | EmilienM: do you still want to see my aodh.conf? | 21:48 |
| EmilienM | slagle: yes | 21:48 |
| EmilienM | just making sure | 21:48 |
| slagle | what pastebin is everyone using these days, that I can curl to? | 21:49 |
| EmilienM | slagle: at least the coordination url | 21:49 |
| EmilienM | slagle: paste the coordination url here | 21:49 |
| slagle | do you mean backend_url? | 21:49 |
| jidar | gist that I have an API key plugged into sublime text, so a Select Text -> cmd-k, cmd-p, and then the url is in my clipboard | 21:50 |
| jidar | that's how I pastebin these days anyhow | 21:50 |
| EmilienM | slagle: yes sorry, it changed recently | 21:50 |
| slagle | backend_url = redis://:kUTzH6fN9G6XFYyngqkArTbHJ@192.168.25.11:6379/ | 21:50 |
| slagle | snap, now you have my redis password | 21:50 |
| slagle | i fell for a phishing attempt | 21:50 |
| mwhahaha | but not the username :D | 21:51 |
| pradk | i have a patch coming up .. working on rspec tests | 21:51 |
| mwhahaha | which shouldn't we check before using? | 21:51 |
| bandini | lol | 21:51 |
| mwhahaha | that doesn't seem like a valid setting | 21:51 |
| EmilienM | slagle: kUTzH6fN9G6XFYyngqkArTbHJ is very secure though, good job. | 21:51 |
| slagle | thanks, it was my first dog's name | 21:52 |
| jidar | looool | 21:52 |
| slagle | i wanted to be the only one who could call him | 21:52 |
| mwhahaha | you too? i prefered kUTzH6fN9G4XFYyngqkArTbHJ | 21:52 |
| jidar | https://www.xkcd.com/327/ | 21:52 |
| bandini | ahahah | 21:52 |
| jidar | relevant xkcd | 21:53 |
| *** mhenkel_ has joined #tripleo | 21:55 | |
| EmilienM | bnemec: do you have the url handy of the old multinode job bug timeout ? | 21:58 |
| *** tbonds has quit IRC | 21:59 | |
| bnemec | EmilienM: https://bugs.launchpad.net/tripleo/+bug/1651267 | 21:59 |
| openstack | Launchpad bug 1651267 in tripleo "Timeouts in multinode jobs" [High,Triaged] | 21:59 |
| bnemec | EmilienM: This looks different though. In this case it looks like os-collect-config is calling home just fine, but for some reason Heat isn't proceeding. | 22:00 |
| EmilienM | bnemec: thx | 22:01 |
| mwhahaha | that sounds like the what happens if the client hook is not installed | 22:01 |
| *** jkilpatr has joined #tripleo | 22:03 | |
| EmilienM | bnemec: by reading your analysis, it sounds like step4 is the worst | 22:03 |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Reduce memcached memory configuration https://review.openstack.org/431045 | 22:04 |
| EmilienM | from 683s to 1070s | 22:04 |
| EmilienM | and step4 now manages keystone resources | 22:04 |
| EmilienM | I wonder if we can noop them in step5 | 22:04 |
| *** sdake has quit IRC | 22:04 | |
| EmilienM | OR manage them ONLY at step 4 | 22:04 |
| bnemec | EmilienM: step 3 is actually worse. 509 to 1109. | 22:05 |
| *** sshnaidm|afk is now known as sshnaidm | 22:06 | |
| EmilienM | wait, we moved keystone resources to step3 in fact | 22:06 |
| bnemec | Reverting the puppet-tripleo step changes moves that to step 4. | 22:06 |
| *** lblanchard has quit IRC | 22:06 | |
| bnemec | Well, your point still stands then. :-) | 22:06 |
| bnemec | It's funny that step 5 is now faster than 3 and 4 though. | 22:07 |
| EmilienM | bnemec: I wonder how your reverts passed CI | 22:07 |
| EmilienM | bnemec: it should have failed on cells/placement thing | 22:07 |
| EmilienM | I'm lost | 22:07 |
| EmilienM | because I didn't do these patches for fun i swear | 22:07 |
| bnemec | EmilienM: Maybe the double temprevert cancelled out one of them and left enough of the changes for it to pass? | 22:08 |
| EmilienM | it's just cells required nova-compute that required nova-placement that required nova keystone resources | 22:08 |
| EmilienM | so step 5, then 4, then 3 | 22:08 |
| bnemec | Those functions don't get a lot of use these days so I won't promise that they behave as expected. :-) | 22:08 |
| EmilienM | tbh, I would avoid reverts. My patches are setting pretty strong orchestration on the nova things | 22:08 |
| EmilienM | I'm investigating if that's the keystone resources that take time | 22:09 |
| bnemec | That's why they're temp. :-) | 22:09 |
| EmilienM | and if yes, investigate if we can noop them | 22:09 |
| EmilienM | bnemec: yes I like your testing fwiw | 22:09 |
| bnemec | That's just a debugging patch, not something I think we should do. | 22:09 |
| EmilienM | yeah np | 22:09 |
| EmilienM | bnemec: I wish nova would be easier to deploy | 22:09 |
| bnemec | I actually didn't expect it to pass CI. I just wanted to see the heat deploy times. | 22:09 |
| EmilienM | but I've heard it's working fast in devstack | 22:09 |
| bnemec | Been there, done that, got the t-shirt. | 22:10 |
| EmilienM | I'll have it all next week | 22:10 |
| EmilienM | "It worked in DevStack, not in TripleO though" | 22:10 |
| EmilienM | oh we need that one ^ | 22:10 |
| bnemec | I forget, do we get useful timestamps on the puppet output for the overcloud nodes or does it all get dumped out at once? | 22:11 |
| mwhahaha | at once | 22:12 |
| mwhahaha | cause it's the stdout | 22:12 |
| EmilienM | bnemec, mwhahaha: I'll cheat and look in keystone logs | 22:12 |
| bnemec | Bleh. So we can't profile what steps are taking the longest that way. | 22:12 |
| bandini | if you have puppet debug enabled you have the proper timestamps in /var/log/puppet/heat-debug.log | 22:12 |
| *** gfidente|afk has quit IRC | 22:12 | |
| EmilienM | I think we have that no? | 22:12 |
| mwhahaha | we may not capture it | 22:13 |
| * mwhahaha added that feature | 22:13 | |
| bnemec | Not by default. | 22:13 |
| bandini | I thought we started adding it to ovb very recently? | 22:13 |
| bnemec | I think bandini actually brought it up on the ci debug mode patch. | 22:13 |
| mwhahaha | i added the heat hook to write it out :D | 22:13 |
| bandini | bnemec: I think we merged another variant | 22:13 |
| bandini | mwhahaha: nice ;) | 22:13 |
| bandini | lemme see | 22:14 |
| panda|bbl | bandini: bad news. | 22:14 |
| bnemec | bandini: Yeah, we backported the Debug env, but not the puppet one. | 22:14 |
| bandini | meh | 22:14 |
| bandini | panda|bbl: which one :) | 22:14 |
| bnemec | I don't even remember what the puppet parameter was called... | 22:14 |
| bandini | I have a few | 22:14 |
| mwhahaha | there's also a puppet verbose option if you don't want debug | 22:15 |
| mwhahaha | enable_debug or enable_verbose | 22:15 |
| mwhahaha | https://github.com/openstack/heat-templates/tree/master/hot/software-config/elements/heat-config-puppet | 22:15 |
| bnemec | Oh look: environments/config-debug.yaml: ConfigDebug: true | 22:16 |
| bnemec | I can work with that. | 22:16 |
| panda|bbl | /ban dini | 22:16 |
| panda|bbl | bandini: sshd is still failing | 22:17 |
| bandini | that's the one. I thought we enabled it recently though | 22:17 |
| bandini | panda|bbl: that's not bad news, that's the feedback the openssh folks wanted | 22:17 |
| bandini | now they can sit down and try to fix the damn thing | 22:17 |
| openstackgerrit | Ben Nemec proposed openstack-infra/tripleo-ci master: TEST: Enable ConfigDebug https://review.openstack.org/433953 | 22:17 |
| bnemec | ^moar data | 22:18 |
| EmilienM | quick question about upgrades, do we run all puppet steps after ansible actions? | 22:20 |
| EmilienM | marios, matbu ^ | 22:20 |
| mwhahaha | yes | 22:20 |
| mwhahaha | i asked that question | 22:20 |
| openstackgerrit | Merged openstack/instack-undercloud master: Bump required undercloud memory to 8 GB https://review.openstack.org/432395 | 22:20 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo master: tuning: manage keystone resources only at step3 https://review.openstack.org/433954 | 22:21 |
| EmilienM | mwhahaha: that ^ | 22:21 |
| EmilienM | unless we have a missing dependency in the catalog at step 4 or 5 I think it can work | 22:21 |
| pradk | EmilienM, can you check https://review.openstack.org/#/c/433824/2 we'll need this for osp11 upgrades | 22:22 |
| EmilienM | pradk: looking | 22:23 |
| EmilienM | pradk: failing OVB :/ | 22:23 |
| *** gatuus has joined #tripleo | 22:23 | |
| mwhahaha | http://logs.openstack.org/24/433824/2/check-tripleo/gate-tripleo-ci-centos-7-ovb-ha/05e0a29/console.html#_2017-02-14_19_55_18_611840 bad relationship when it's disabled | 22:23 |
| bandini | gah we restart a truckload of services on subsequent overcloud deploys | 22:24 |
| EmilienM | bandini: details? | 22:24 |
| bandini | EmilienM: https://bugs.launchpad.net/tripleo/+bug/1664650/comments/4 | 22:24 |
| openstack | Launchpad bug 1664650 in tripleo "Minor updates cause control-plane downtime" [High,Triaged] | 22:24 |
| mwhahaha | that usually means the configs are not idempotent between steps | 22:24 |
| bnemec | Hmm, I have an email from a random site inviting me to assess my manager. | 22:24 |
| bnemec | https://www.youtube.com/watch?v=4F4qzPbcFiA | 22:24 |
| pradk | EmilienM, ovb failures are not related to patch afaict .. i recheck'ed it | 22:25 |
| slagle | EmilienM: i'm checking on stackoverflow, but is the ensure_packages syntax right here? https://review.openstack.org/#/c/433951/1/manifests/evaluator.pp | 22:25 |
| EmilienM | lol | 22:25 |
| slagle | stackoverflow is telling me it takes an array or hash | 22:25 |
| pradk | slagle, thats what , does | 22:26 |
| EmilienM | slagle: syntax is good | 22:26 |
| EmilienM | pradk: recheck yeah | 22:26 |
| pradk | yea we'll see | 22:26 |
| slagle | ok | 22:26 |
| mwhahaha | slagle: use ensure_resource('package', 'name', {...}) | 22:26 |
| bandini | mwhahaha: yeah that seems likely. At least for the httpd restart I see it is due to /etc/httpd/conf/ports.conf having changed | 22:26 |
| mwhahaha | slagle: ensure_packages has issues which are tl;dr for me to want to explain | 22:27 |
| bandini | I'll need to track down who the heck is doing it though | 22:27 |
| slagle | mwhahaha: it's not my patch, i was just trying to review it | 22:27 |
| slagle | so i was googling for knowledge | 22:27 |
| slagle | it's how I review puppet patches | 22:27 |
| mwhahaha | slagle: sure, but -1 for ensure_packages :D | 22:27 |
| EmilienM | slagle: you want to be core on puppet modules? we need more folks :P | 22:28 |
| mwhahaha | bandini: if you have the puppet debug log it should tell you what scheduled the refresh | 22:28 |
| *** sdake has joined #tripleo | 22:28 | |
| bandini | mwhahaha: so at https://bugs.launchpad.net/tripleo/+bug/1664650/comments/3 I know it is ports.conf not sure who is doing it though, yet | 22:29 |
| openstack | Launchpad bug 1664650 in tripleo "Minor updates cause control-plane downtime" [High,Triaged] | 22:29 |
| *** dbecker has quit IRC | 22:30 | |
| EmilienM | ports.conf is touched by services run in WSGI | 22:30 |
| bnemec | EmilienM: We should just give stackoverflow core on the puppet modules. ;-) | 22:30 |
| EmilienM | like gnocchi, etc | 22:30 |
| EmilienM | bnemec: lol | 22:30 |
| *** sdake has quit IRC | 22:30 | |
| bandini | mwhahaha: although this http://paste.openstack.org/show/598898/ gives me some hints I guess (aka gnocchi, aodh, ceilometer) | 22:30 |
| bandini | EmilienM: ^ | 22:30 |
| *** sdake has joined #tripleo | 22:31 | |
| pradk | bandini, yep those are telemetry ports | 22:31 |
| pradk | causing issues? | 22:31 |
| EmilienM | it's always telemetry anyway | 22:32 |
| EmilienM | :P | 22:32 |
| pradk | lol | 22:32 |
| bandini | pradk: ehehe we're investigating https://bugs.launchpad.net/tripleo/+bug/1664650/ | 22:32 |
| openstack | Launchpad bug 1664650 in tripleo "Minor updates cause control-plane downtime" [High,Triaged] | 22:32 |
| mwhahaha | so it's probably cause it gets deconfigured in an earlier step | 22:33 |
| mwhahaha | and reconfigured later | 22:33 |
| mwhahaha | which is a problem | 22:33 |
| mwhahaha | we need to isolate those types of configs into a place where they are always done in the same phases | 22:33 |
| *** fpan_ has joined #tripleo | 22:33 | |
| *** funzo_ has joined #tripleo | 22:33 | |
| *** dsneddon_ has joined #tripleo | 22:34 | |
| mwhahaha | so we need to do all our wsgi configs in the same steps at once | 22:34 |
| mwhahaha | what's probably happening is that we h ave keystone in 3 and everything else in 4 | 22:34 |
| mwhahaha | or something of the sort | 22:34 |
| *** trown is now known as trown|outtypewww | 22:34 | |
| bandini | ah that would be a sensible explanation, yes | 22:35 |
| EmilienM | mwhahaha: yes because puppetlabs-apache configure ports when it needs | 22:35 |
| EmilienM | mwhahaha: my guess is step3 setups keystone and keystoen ports | 22:35 |
| EmilienM | and later other ports | 22:35 |
| EmilienM | which means we need to configure all wsgi services at the same step | 22:36 |
| EmilienM | to avoid useless restart | 22:36 |
| EmilienM | problem is, some services like nova-placement wants keystone resources to be ready though it might retry | 22:36 |
| *** sdake_ has joined #tripleo | 22:36 | |
| *** ianw has quit IRC | 22:36 | |
| EmilienM | in fact, nova scheduler & nova compute actually need it, but they run later | 22:36 |
| *** beekneemech has joined #tripleo | 22:36 | |
| *** migi_ has joined #tripleo | 22:37 | |
| EmilienM | does it make sense? | 22:37 |
| *** Tyrantelf_ has joined #tripleo | 22:37 | |
| bandini | we set up keystone on step3 only on the bootstrap node, though? | 22:38 |
| mwhahaha | We'll all API configs should be in 3 and that should be ok | 22:38 |
| mwhahaha | We need to split the API vs scheduler/compute | 22:38 |
| *** rmcadams_ has joined #tripleo | 22:39 | |
| EmilienM | bandini: when is galera cluster ready? end of step2 right? | 22:39 |
| bandini | EmilienM: correct | 22:40 |
| *** fultonj has quit IRC | 22:41 | |
| bandini | couldn't puppetlabs-apache ports handling be made a bit more flexible? Like not using the same ports.conf file for all the ports or something? | 22:41 |
| *** jefrite_ has joined #tripleo | 22:41 | |
| bandini | (although I guess this issue could hit us in other places as well) | 22:41 |
| *** sdake__ has joined #tripleo | 22:42 | |
| mwhahaha | This is the fact multiple things are configuring the same conf file | 22:42 |
| *** gchamoul- has joined #tripleo | 22:42 | |
| mwhahaha | And we aren't doing it consistently for the whole process | 22:42 |
| *** zoli has joined #tripleo | 22:42 | |
| *** sdake has quit IRC | 22:42 | |
| *** jkilpatr has quit IRC | 22:42 | |
| *** dsneddon has quit IRC | 22:42 | |
| *** funzo has quit IRC | 22:42 | |
| *** tbarron has quit IRC | 22:42 | |
| *** egafford has quit IRC | 22:42 | |
| *** tzumainn has quit IRC | 22:42 | |
| *** sshnaidm has quit IRC | 22:42 | |
| *** leanderthal|afk has quit IRC | 22:42 | |
| *** rmcadams has quit IRC | 22:42 | |
| *** bnemec has quit IRC | 22:42 | |
| *** mnaser has quit IRC | 22:42 | |
| *** jefrite has quit IRC | 22:42 | |
| *** fpan has quit IRC | 22:42 | |
| *** Tyrantelf has quit IRC | 22:42 | |
| *** zoli|gone has quit IRC | 22:42 | |
| *** gchamoul has quit IRC | 22:42 | |
| *** vkmc has quit IRC | 22:42 | |
| *** dmanchad has quit IRC | 22:42 | |
| *** greghaynes has quit IRC | 22:42 | |
| *** saneax-_-|AFK has quit IRC | 22:42 | |
| *** migi has quit IRC | 22:42 | |
| *** fungi has quit IRC | 22:42 | |
| mwhahaha | If ports.conf took a directory include it would be easier to switch | 22:43 |
| *** dmanchad has joined #tripleo | 22:44 | |
| *** sdake_ has quit IRC | 22:44 | |
| EmilienM | bandini, slagle: can you guys review https://review.openstack.org/#/c/433715/ please? it's passing gate-tripleo-ci-centos-7-multinode-upgrades-nv and it will help me to make progress on upgrade Ci jobs with scenarios | 22:45 |
| *** rmcadams_ is now known as rmcadams | 22:45 | |
| *** migi_ has quit IRC | 22:46 | |
| *** sasha2 has quit IRC | 22:46 | |
| *** karthiks has quit IRC | 22:46 | |
| *** arxcruz has quit IRC | 22:46 | |
| *** zigo has quit IRC | 22:46 | |
| *** afazekas has quit IRC | 22:46 | |
| *** chlong has quit IRC | 22:46 | |
| *** matbu has quit IRC | 22:46 | |
| *** andreaf has quit IRC | 22:46 | |
| *** lyarwood has quit IRC | 22:46 | |
| *** Slower has quit IRC | 22:46 | |
| *** soc_off_ has quit IRC | 22:46 | |
| *** fabbione has quit IRC | 22:46 | |
| *** marios has quit IRC | 22:46 | |
| *** rdopiera has quit IRC | 22:46 | |
| *** afazekas has joined #tripleo | 22:46 | |
| *** lyarwood has joined #tripleo | 22:46 | |
| *** vkmc has joined #tripleo | 22:47 | |
| bandini | done | 22:47 |
| *** dsariel has joined #tripleo | 22:47 | |
| *** mnaser has joined #tripleo | 22:47 | |
| bandini | I guess httpd could have a /etc/httpd/conf/ports.conf.d/<service>.conf and ports.conf would just Include ports.conf.d/*.conf | 22:48 |
| *** greghaynes has joined #tripleo | 22:48 | |
| bandini | doesn't strike me as a quick change though | 22:49 |
| *** fungi has joined #tripleo | 22:49 | |
| *** leanderthal|afk has joined #tripleo | 22:49 | |
| *** sshnaidm has joined #tripleo | 22:50 | |
| *** jkilpatr has joined #tripleo | 22:50 | |
| *** tbonds has joined #tripleo | 22:50 | |
| *** tzumainn has joined #tripleo | 22:51 | |
| *** egafford has joined #tripleo | 22:51 | |
| *** karthiks has joined #tripleo | 22:52 | |
| *** zigo has joined #tripleo | 22:52 | |
| *** migi_ has joined #tripleo | 22:52 | |
| *** sasha2 has joined #tripleo | 22:52 | |
| *** arxcruz has joined #tripleo | 22:52 | |
| *** chlong has joined #tripleo | 22:52 | |
| *** matbu has joined #tripleo | 22:52 | |
| *** andreaf has joined #tripleo | 22:52 | |
| *** Slower has joined #tripleo | 22:52 | |
| *** soc_off_ has joined #tripleo | 22:52 | |
| *** fabbione has joined #tripleo | 22:52 | |
| *** marios has joined #tripleo | 22:52 | |
| *** rdopiera has joined #tripleo | 22:52 | |
| *** jmelvin has quit IRC | 22:52 | |
| *** karthiks has quit IRC | 22:52 | |
| *** karthiks has joined #tripleo | 22:52 | |
| *** zigo is now known as Guest27057 | 22:54 | |
| EmilienM | owalsh_sick: you still around? https://review.openstack.org/#/c/405241/ sounds passing CI | 22:56 |
| owalsh_sick | EmilienM: yea | 22:58 |
| *** panda|bbl is now known as panda|zZ | 22:58 | |
| EmilienM | owalsh_sick: see my comment | 22:58 |
| * owalsh_sick looking | 22:59 | |
| *** morazi has quit IRC | 23:04 | |
| *** tbonds has quit IRC | 23:05 | |
| *** bfournie has quit IRC | 23:06 | |
| *** cdearborn has quit IRC | 23:08 | |
| owalsh_sick | EmilienM: it's DEFAULT in puppet-nova https://github.com/openstack/puppet-nova/blob/master/manifests/api.pp#L411 | 23:08 |
| owalsh_sick | EmilienM: but shouldn't this be in nova-api.yaml? | 23:10 |
| EmilienM | owalsh_sick: why are we updating an api param in conductor service? | 23:11 |
| EmilienM | owalsh_sick: it doesn't make sense to me. | 23:11 |
| owalsh_sick | EmilienM: yea, I think it would fail if the api services was on a different hosts | 23:12 |
| EmilienM | owalsh_sick: also, why do we configure workers in ansible? | 23:14 |
| EmilienM | owalsh_sick: while it's done by puppet | 23:14 |
| EmilienM | I'm still lost why we do config management with ansible in the upgrade process | 23:14 |
| owalsh_sick | EmilienM: ah, IIRC the upgrade tasks were starting the services initially. Now it's left to puppet so we can probably remove this | 23:16 |
| EmilienM | I'm out a bit | 23:16 |
| owalsh_sick | ok, I'll push a quick change to remove this | 23:17 |
| *** fragatin_ has joined #tripleo | 23:22 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Disallow IP changes on undercloud update https://review.openstack.org/432388 | 23:23 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Change _member_role_exists to work with current upgrade flow https://review.openstack.org/432435 | 23:23 |
| openstackgerrit | Oliver Walsh proposed openstack/tripleo-heat-templates master: Add nova service support for composable upgrades https://review.openstack.org/405241 | 23:23 |
| *** fragatina has quit IRC | 23:25 | |
| *** tbonds has joined #tripleo | 23:27 | |
| *** jcoufal_ has quit IRC | 23:28 | |
| *** jcoufal has joined #tripleo | 23:32 | |
| *** pmannidi has joined #tripleo | 23:40 | |
| *** ianw has joined #tripleo | 23:40 | |
| *** sshnaidm has quit IRC | 23:41 | |
| *** jcoufal has quit IRC | 23:43 | |
| *** tbonds has quit IRC | 23:43 | |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Disallow IP changes on undercloud update https://review.openstack.org/432388 | 23:44 |
| openstackgerrit | Ben Nemec proposed openstack/instack-undercloud master: Change _member_role_exists to work with current upgrade flow https://review.openstack.org/432435 | 23:44 |
| *** dmarlin has quit IRC | 23:45 | |
| *** bfournie has joined #tripleo | 23:47 | |
| *** sshnaidm has joined #tripleo | 23:47 | |
| *** sshnaidm is now known as sshnaidm|off | 23:48 | |
| *** tbarron has joined #tripleo | 23:49 | |
| *** sdake__ is now known as sdake | 23:52 | |
| *** tbonds has joined #tripleo | 23:55 | |
| *** jjung has joined #tripleo | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!