| *** raildo has quit IRC | 00:08 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 00:19 |
|---|---|---|
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: DNM: Skip old-style interface/bridge config for new rhel/centos https://review.opendev.org/729415 | 00:22 |
| *** jmasud has joined #tripleo | 00:34 | |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: Allow overriding InterfaceDefaultRoute with ips_from_pool template https://review.opendev.org/732698 | 00:34 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-quickstart-extras master: DNM: Skip old-style interface/bridge config for new rhel/centos https://review.opendev.org/729415 | 00:42 |
| openstackgerrit | melissaml proposed openstack/diskimage-builder master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/732490 | 00:46 |
| *** jmasud has quit IRC | 01:26 | |
| *** ysandeep|away is now known as ysandeep | 01:38 | |
| *** jmasud has joined #tripleo | 01:41 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-ci master: DNM Test py3 centos7 image build https://review.opendev.org/733035 | 01:59 |
| openstackgerrit | Rabi Mishra proposed openstack/tripleo-heat-templates master: Use empty string for overcloud InternalTLSCAFile param https://review.opendev.org/733036 | 02:00 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ansible master: Max percentage failure support https://review.opendev.org/731846 | 02:01 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ansible master: WIP - Molecule testing for TripleO Strategy Plugins https://review.opendev.org/731848 | 02:01 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ansible master: Molecule testing for TripleO Strategy Plugins https://review.opendev.org/731848 | 02:02 |
| EmilienM | zbr: nevermind the RFE for molecule, we managed to do without it ^ | 02:02 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart master: need more than one test defined for 062 https://review.opendev.org/733038 | 02:03 |
| *** apetrich has quit IRC | 02:11 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Add options for container builds https://review.opendev.org/733039 | 02:16 |
| *** rcernin has quit IRC | 02:24 | |
| *** jdandrea has quit IRC | 02:30 | |
| *** rcernin has joined #tripleo | 02:31 | |
| *** sshnaidm has quit IRC | 02:34 | |
| *** rlandy has quit IRC | 02:43 | |
| *** rfolco|rover has quit IRC | 02:57 | |
| *** jmasud has quit IRC | 03:13 | |
| *** rcernin has quit IRC | 03:23 | |
| *** jmasud has joined #tripleo | 03:31 | |
| *** rcernin has joined #tripleo | 03:32 | |
| *** sshnaidm has joined #tripleo | 03:33 | |
| *** psachin has joined #tripleo | 03:34 | |
| *** jmasud has quit IRC | 03:43 | |
| *** jmasud has joined #tripleo | 03:46 | |
| *** rcernin has quit IRC | 03:46 | |
| openstackgerrit | Merged openstack/puppet-tripleo master: Finish HAProxy config for metrics_qdr https://review.opendev.org/713627 | 03:58 |
| *** dmellado has quit IRC | 04:16 | |
| *** rcernin has joined #tripleo | 04:24 | |
| *** rcernin has quit IRC | 04:28 | |
| *** rcernin has joined #tripleo | 04:29 | |
| *** ykarel|away is now known as ykarel | 04:31 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #tripleo | 04:33 | |
| *** ccamacho has quit IRC | 04:41 | |
| *** sshnaidm_ has joined #tripleo | 04:43 | |
| *** sshnaidm has quit IRC | 04:45 | |
| *** saneax has joined #tripleo | 04:49 | |
| openstackgerrit | yatin proposed openstack/tripleo-ci master: Adds new tripleo-ci-buildimage-jobs template for branchless repos https://review.opendev.org/730763 | 04:52 |
| *** abdysn has joined #tripleo | 04:57 | |
| *** jmasud has quit IRC | 05:06 | |
| *** jmasud has joined #tripleo | 05:09 | |
| *** waleedm has joined #tripleo | 05:12 | |
| *** skramaja has joined #tripleo | 05:12 | |
| *** udesale has joined #tripleo | 05:18 | |
| ramishra | Hmm.. No timestamps for ansible tasks and it just times out..very difficult to understand what's going on https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_479/733036/1/check/tripleo-ci-centos-8-scenario000-multinode-oooq-container-updates/47983dc/logs/undercloud/home/zuul/undercloud_install.log | 05:20 |
| ramishra | it seems like the new callback plugins | 05:20 |
| ykarel | looks like time is there but format changed | 05:29 |
| ykarel | tripleo_firewall : Stop and disable firewalld | 0:01:51.309 | 0.51s | 05:29 |
| *** dmellado has joined #tripleo | 05:31 | |
| ykarel | but looks like timing is not correct | 05:32 |
| *** lmiccini has joined #tripleo | 05:34 | |
| *** jmasud has quit IRC | 05:35 | |
| ykarel | hmm that's timing is different, not timestamp | 05:36 |
| *** rcernin has quit IRC | 05:36 | |
| ramishra | ykarel: yes, it does not align with https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_479/733036/1/check/tripleo-ci-centos-8-scenario000-multinode-oooq-container-updates/47983dc/logs/undercloud/var/log/tripleo-container-image-prepare.log | 05:37 |
| *** rcernin has joined #tripleo | 05:37 | |
| ramishra | bc764e10-27fb-edc9-21d3-000000000106 | TASK | Run tripleo_container_image_prepare logged to: /var/log/tripleo-container-image-prepare.log | 05:38 |
| ramishra | bc764e10-27fb-edc9-21d3-000000000105 | TIMING | tripleo_container_image_prepare : Write role data file | 0:03:11.133 | 0.70s | 05:38 |
| ykarel | yes | 05:40 |
| ramishra | mwhahaha: ^^ fyi | 05:40 |
| ykarel | looks like tripleo-container image prepare stuck for some reason | 05:40 |
| *** sanjayu_ has joined #tripleo | 05:40 | |
| *** saneax has quit IRC | 05:41 | |
| openstackgerrit | Rabi Mishra proposed openstack/tripleo-heat-templates master: Use empty string for overcloud InternalTLSCAFile param https://review.opendev.org/733036 | 05:48 |
| *** marios has joined #tripleo | 05:49 | |
| *** rcernin has quit IRC | 05:51 | |
| *** rcernin has joined #tripleo | 05:56 | |
| *** jfrancoa has joined #tripleo | 06:00 | |
| *** cfontain has joined #tripleo | 06:01 | |
| *** cfontain has quit IRC | 06:07 | |
| *** rcernin has quit IRC | 06:10 | |
| *** bandini has joined #tripleo | 06:14 | |
| *** rcernin has joined #tripleo | 06:14 | |
| *** rcernin has quit IRC | 06:28 | |
| *** cfontain has joined #tripleo | 06:30 | |
| *** dtantsur|afk is now known as dtantsur | 06:43 | |
| openstackgerrit | Luca Miccini proposed openstack/tripleo-heat-templates master: Ensure net.ipv6.conf.lo.disable_ipv6=0 if KernelDisableIPv6=1 https://review.opendev.org/732098 | 06:51 |
| openstackgerrit | Luca Miccini proposed openstack/tripleo-ansible master: Do not set lo.disable_ipv6=0 by default https://review.opendev.org/732168 | 06:51 |
| *** slaweq has joined #tripleo | 06:51 | |
| *** tkajinam has quit IRC | 06:56 | |
| *** tkajinam has joined #tripleo | 06:56 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-ci master: Adds new tripleo-ci-buildimage-jobs template for branchless repos https://review.opendev.org/730763 | 06:59 |
| openstackgerrit | Moshe Levi proposed openstack/tripleo-common master: Revert "Add kolla template override to remove python3-networking-mlnx" https://review.opendev.org/733052 | 07:04 |
| *** apetrich has joined #tripleo | 07:10 | |
| *** mmethot_ has joined #tripleo | 07:19 | |
| *** jcapitao has joined #tripleo | 07:20 | |
| *** xek__ has joined #tripleo | 07:21 | |
| *** mmethot has quit IRC | 07:21 | |
| *** tosky has joined #tripleo | 07:24 | |
| *** lbragstad has quit IRC | 07:27 | |
| *** bogdando has joined #tripleo | 07:27 | |
| *** amoralej|off is now known as amoralej | 07:33 | |
| openstackgerrit | Sandeep Yadav proposed openstack/tripleo-ci master: DNM Test Patch https://review.opendev.org/733057 | 07:33 |
| *** ravsingh has joined #tripleo | 07:34 | |
| openstackgerrit | Sandeep Yadav proposed openstack/tripleo-ci master: DNM Test Patch https://review.opendev.org/733057 | 07:37 |
| *** bandini has quit IRC | 07:43 | |
| *** ravsingh has quit IRC | 07:44 | |
| *** jpena|off is now known as jpena | 07:53 | |
| openstackgerrit | Sandeep Yadav proposed openstack/tripleo-ci master: DNM Test Patch https://review.opendev.org/733057 | 07:54 |
| *** bandini has joined #tripleo | 07:56 | |
| openstackgerrit | Sandeep Yadav proposed openstack/tripleo-ci master: DNM Test Patch https://review.opendev.org/733057 | 07:56 |
| *** yolanda has joined #tripleo | 08:02 | |
| *** jpich has joined #tripleo | 08:03 | |
| *** lucasagomes has joined #tripleo | 08:05 | |
| openstackgerrit | Cédric Jeanneret (Tengu) proposed openstack/tripleo-ansible master: Ensure shortname matches ctlplane IP https://review.opendev.org/732491 | 08:06 |
| *** alexmcleod has joined #tripleo | 08:12 | |
| *** fmount|afk is now known as fmount | 08:14 | |
| *** jaosorior has joined #tripleo | 08:17 | |
| *** ravsingh has joined #tripleo | 08:17 | |
| *** xek__ has quit IRC | 08:18 | |
| *** xek__ has joined #tripleo | 08:18 | |
| openstackgerrit | Luca Miccini proposed openstack/tripleo-heat-templates master: Ensure net.ipv6.conf.lo.disable_ipv6=0 https://review.opendev.org/732098 | 08:21 |
| *** pkopec has joined #tripleo | 08:23 | |
| *** marios has quit IRC | 08:31 | |
| *** redrobot has quit IRC | 08:35 | |
| ykarel | hjensas, hi | 08:38 |
| ykarel | hjensas, /me was looking at https://etherpad.opendev.org/p/tripleo-enable-net-iso-by-default | 08:38 |
| ykarel | is the proposal there improving something that is not covered by https://opendev.org/openstack/python-tripleoclient/src/branch/master/tripleoclient/v1/overcloud_deploy.py#L457-L458 ? | 08:39 |
| *** ysandeep is now known as ysandeep|lunch | 08:39 | |
| ykarel | didn't find any reference of this handling in etherpad so wondering if this was related | 08:40 |
| *** xek_ has joined #tripleo | 08:40 | |
| *** xek__ has quit IRC | 08:43 | |
| *** dciabrin has joined #tripleo | 08:45 | |
| *** paramite has joined #tripleo | 08:46 | |
| openstackgerrit | Martin Mágr proposed openstack/instack-undercloud stable/queens: Configure SNMPD when telemetry is used https://review.opendev.org/731529 | 08:48 |
| *** ccamacho has joined #tripleo | 08:50 | |
| *** tkajinam has quit IRC | 08:53 | |
| *** tkajinam has joined #tripleo | 08:54 | |
| openstackgerrit | Martin Mágr proposed openstack/tripleo-heat-templates stable/stein: Disable LoggingConfiguration by default https://review.opendev.org/733074 | 09:04 |
| *** gfidente|afk is now known as gfidente | 09:07 | |
| *** tkajinam has quit IRC | 09:09 | |
| *** maciejjozefczyk has quit IRC | 09:10 | |
| *** cfontain has quit IRC | 09:15 | |
| *** cfontain has joined #tripleo | 09:16 | |
| *** maciejjozefczyk has joined #tripleo | 09:20 | |
| *** cfontain has quit IRC | 09:20 | |
| *** marios has joined #tripleo | 09:27 | |
| *** aedc has joined #tripleo | 09:28 | |
| hjensas | ykarel: I had missed that we had that validation in there. The validation covers the problem. | 09:30 |
| hjensas | ykarel: I do however think enabling net-iso by default will improve user experience, we can remove the need to include both network-isolation.yaml and network-environment.yaml. Consolidating network configuration to network_data.yaml and roles_data.yaml instead. | 09:31 |
| openstackgerrit | Sergii Golovatiuk proposed openstack/tripleo-heat-templates master: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733077 | 09:32 |
| *** ysandeep|lunch is now known as ysandeep | 09:33 | |
| hjensas | ykarel: In the discussion yesterday I was made aware of a couple of things, the existing validation, and ramishra talked about how overcloud_resource_registry.py updates the resource registry. | 09:33 |
| hjensas | ykarel: I think we may be able to switch to use a default network_data.yaml with no networks defined without breaking updates/upgrades. But I will have to do some research and PoC, to see if that is doable. | 09:36 |
| ykarel | hjensas, Thanks for confirmation, hmm on re looking i found in Note: wrt existing validations | 09:41 |
| ykarel | s/Note:/Notes:/ | 09:41 |
| *** cfontain has joined #tripleo | 09:43 | |
| *** cfontain has quit IRC | 09:44 | |
| *** cfontain has joined #tripleo | 09:46 | |
| *** jmasud has joined #tripleo | 09:46 | |
| openstackgerrit | Lukas Bezdicka proposed openstack/puppet-tripleo stable/ussuri: Remove selinux relabel mount option for neutron https://review.opendev.org/733079 | 09:46 |
| openstackgerrit | Lukas Bezdicka proposed openstack/puppet-tripleo stable/train: Remove selinux relabel mount option for neutron https://review.opendev.org/733081 | 09:47 |
| openstackgerrit | Juan Badia Payno proposed openstack/tripleo-ansible stable/train: BaR: Set default container cli to podman https://review.opendev.org/732118 | 09:48 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/tripleo-quickstart-extras master: Refactor buildtestpackages for downstream purpose https://review.opendev.org/727200 | 09:49 |
| *** cfontain has quit IRC | 09:50 | |
| openstackgerrit | Lukas Bezdicka proposed openstack/tripleo-heat-templates stable/train: FFWD: post Leapp package fixups https://review.opendev.org/729358 | 09:52 |
| *** udesale_ has joined #tripleo | 09:53 | |
| *** udesale has quit IRC | 09:56 | |
| *** jtomasek has joined #tripleo | 09:58 | |
| *** udesale_ has quit IRC | 10:00 | |
| openstackgerrit | Bogdan Dobrelya (bogdando) proposed openstack/tripleo-heat-templates stable/ussuri: Add reserved ports for some services https://review.opendev.org/733087 | 10:10 |
| openstackgerrit | Bogdan Dobrelya (bogdando) proposed openstack/tripleo-heat-templates stable/train: Add reserved ports for some services https://review.opendev.org/733088 | 10:10 |
| hjensas | bandini: mind taking a look at https://review.opendev.org/732491 ? I put a comment there, I'm a bit worried about the effect on RabbitMQ? | 10:17 |
| bandini | hjensas: oh thanks for the heads up. is that change moving the shortname from one network to the other? | 10:20 |
| hjensas | bandini: yes, we have in ServiceNetMap the "HostnameResolvNetwork" which is what control's where the shortname is. | 10:21 |
| bandini | ah ok yeah that is bad, let me add a comment | 10:22 |
| hjensas | bandini: that change basically moves it to the ctlplane. | 10:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/tripleo-heat-templates master: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733077 | 10:23 |
| *** jpich has quit IRC | 10:26 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-ci master: WIP emit-releases for rhos_releases https://review.opendev.org/732170 | 10:26 |
| *** jpich has joined #tripleo | 10:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/tripleo-heat-templates stable/ussuri: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733090 | 10:27 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/tripleo-heat-templates stable/train: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733091 | 10:28 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/tripleo-quickstart-extras master: Refactor buildtestpackages for downstream purpose https://review.opendev.org/727200 | 10:30 |
| *** jcapitao is now known as jcapitao_lunch | 10:35 | |
| *** xek_ has quit IRC | 10:46 | |
| *** xek_ has joined #tripleo | 10:46 | |
| abdysn | Hi all, so a general question, what is the proper way for updating an overcloud image? i mean i tried to use openstack container prepare with a yaml file containing the modify command. and it succeeded, then i tried to update the overcloud using openstack overcloud external-update run --tags container_image_prepare, and the file contains the new tag. but the overcloud fails to update because of a missing blob. so i assume i am missing something, can | 11:18 |
| abdysn | anyone help me with that? | 11:18 |
| abdysn | there is a file: | 11:23 |
| abdysn | which is pointing to a blob image: | 11:23 |
| abdysn | URI: sha256:5644e225cbed91eed4566c1b7fc7eb94c475fc82cd2ce0fa778730efbdfe3ef2/index.json | 11:23 |
| abdysn | which is NOT in the blob directory | 11:23 |
| abdysn | the file in the neutron manifests | 11:24 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/tripleo-quickstart-extras master: Refactor buildtestpackages for downstream purpose https://review.opendev.org/727200 | 11:30 |
| *** jpena is now known as jpena|lunch | 11:31 | |
| *** cfontain has joined #tripleo | 11:33 | |
| *** cfontain_ has joined #tripleo | 11:34 | |
| openstackgerrit | Pooja Jadhav proposed openstack/tripleo-ci master: Checkout diskimage-builder to 2.38.0 tag for CentOS-7 https://review.opendev.org/733114 | 11:34 |
| *** cfontain has quit IRC | 11:38 | |
| *** saneax has joined #tripleo | 11:39 | |
| *** cfontain_ has quit IRC | 11:39 | |
| *** sanjayu_ has quit IRC | 11:40 | |
| *** xek_ has quit IRC | 11:42 | |
| *** rfolco|rover has joined #tripleo | 11:42 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-ci master: WIP emit-releases for rhos_releases https://review.opendev.org/732170 | 11:46 |
| *** jcapitao_lunch is now known as jcapitao | 11:49 | |
| *** fultonj has quit IRC | 11:50 | |
| *** fultonj has joined #tripleo | 11:51 | |
| openstackgerrit | Pooja Jadhav proposed openstack/tripleo-ci master: Checkout diskimage-builder to 2.38.0 tag for CentOS-7 https://review.opendev.org/733114 | 11:52 |
| *** rh-jelabarre has joined #tripleo | 11:53 | |
| *** raildo has joined #tripleo | 11:56 | |
| *** rpittau|afk is now known as rpittau | 11:57 | |
| *** amoralej is now known as amoralej|lunch | 11:58 | |
| cloudnull | mornings | 12:03 |
| *** morazi has quit IRC | 12:04 | |
| *** ysandeep is now known as ysandeep|afk | 12:06 | |
| *** pojadhav|ruck is now known as pojadhav|batery- | 12:07 | |
| *** pojadhav|batery- is now known as pooja|batery-die | 12:08 | |
| *** morazi has joined #tripleo | 12:18 | |
| *** xek has joined #tripleo | 12:27 | |
| *** rlandy has joined #tripleo | 12:32 | |
| *** artom has quit IRC | 12:37 | |
| *** artom has joined #tripleo | 12:37 | |
| *** jpena|lunch is now known as jpena | 12:39 | |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: Allow overriding InterfaceDefaultRoute with ips_from_pool template https://review.opendev.org/732698 | 12:40 |
| *** Goneri has joined #tripleo | 12:45 | |
| weshay|ruck | odyssey4me, ur up first today :) | 12:47 |
| *** skramaja has quit IRC | 12:48 | |
| odyssey4me | weshay|ruck: yup! | 12:49 |
| *** lbragstad has joined #tripleo | 12:53 | |
| *** amoralej|lunch is now known as amoralej | 12:54 | |
| *** ysandeep|afk is now known as ysandeep | 12:55 | |
| *** tkajinam has joined #tripleo | 12:57 | |
| *** ekultails has joined #tripleo | 12:59 | |
| weshay|ruck | PTG | 13:00 |
| weshay|ruck | PTG | 13:00 |
| weshay|ruck | PTG | 13:00 |
| mwhahaha | ykarel, ramishra: qhat's up? I'm open to suggestions on ways to improve it but going forward when things aren't lienar the best way to find stuff is via task uuid. if we need to prepend the current time we can | 13:02 |
| *** pooja|batery-die is now known as pojadhav|ruck | 13:03 | |
| mwhahaha | ramishra: timing has always been after the next task starts btw so to find out what's running it's the last task in the logs | 13:03 |
| ekultails | weshay: As some feedback on the TripleO PTG, I believe it would be great to give everyone a guaranteed 5 minute break for every 30 minutes of discussions/meetings for bio break, stretching, mental health, etc. :-) | 13:04 |
| ramishra | mwhahaha: I always use the timestamp to correlate between logs. Though it's not linear it would be good to have the current timestamp I think | 13:05 |
| mwhahaha | ramishra: k i'll look to adding the current time to the begining. i think ansible.log has timestamps tho | 13:06 |
| mwhahaha | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_479/733036/1/check/tripleo-ci-centos-8-scenario000-multinode-oooq-container-updates/47983dc/logs/undercloud/home/zuul/ansible.log | 13:07 |
| mwhahaha | in the overcloud that gets captured but doesn't seem to for standalone | 13:07 |
| mwhahaha | we might want to fix that | 13:07 |
| mwhahaha | and you could use that instead | 13:07 |
| ramishra | mwhahaha: ok, I like verbose logs though I understand we've limitations around log size. The best we can do for better readability of logs and correlate them is a bonus here | 13:09 |
| mwhahaha | i think timestamp might be useful so i'll look int appending that because I could see matching times against other logs | 13:11 |
| mwhahaha | i wasn't doing it for stdout because it's already captured in the ansible.log | 13:11 |
| mwhahaha | so to stop duping timestamps | 13:11 |
| mwhahaha | but it looks like ansible.log is inconsitent for the standalone | 13:11 |
| mwhahaha | it might have ended in up in /root | 13:11 |
| raildo | mwhahaha, https://review.opendev.org/#/c/729326/ can we backport for stable/queens? We're hitting this ec2 failure on it as well | 13:12 |
| mwhahaha | raildo: yea it's primarily a test env issue but it should be ok to backport | 13:12 |
| raildo | mwhahaha, awesome | 13:13 |
| *** psachin has quit IRC | 13:14 | |
| *** ramishra has quit IRC | 13:21 | |
| *** Guest27280 has joined #tripleo | 13:22 | |
| *** Guest27280 is now known as redrobot | 13:25 | |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: Allow overriding InterfaceDefaultRoute with ips_from_pool template https://review.opendev.org/732698 | 13:26 |
| *** TrevorV has joined #tripleo | 13:31 | |
| EmilienM | odyssey4me, chem : schedule updated for you, tomorrow 13UTC | 13:38 |
| Tengu | hjensas: heya! guess the shortname thing has in fact uncovered an issue with that internalapi thing... ? I'm trying to understand how it's supposed to work right now and I'm just lost. | 13:38 |
| raildo | mwhahaha, it doesn't seems to be a clean backport for stable/queens, can you help me with it after the PTG meetings? | 13:39 |
| mwhahaha | yea | 13:39 |
| mwhahaha | i'll get them ina few | 13:39 |
| raildo | mwhahaha, thank you sir :) | 13:40 |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: Allow overriding InterfaceDefaultRoute with ips_from_pool template https://review.opendev.org/732698 | 13:41 |
| hjensas | Tengu: the way it used to work, was that we wrote the thing we write now to overcloud nodes only. And things just worked. | 13:41 |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: Allow overriding InterfaceDefaultRoute with ips_from_pool template https://review.opendev.org/732698 | 13:42 |
| hjensas | Tengu: Now we write it to the undercloud as well, and we can't use the shortname there because it's on the internalapi. | 13:42 |
| hjensas | Tengu: we can't move the shortname to another network in the overcloud, it would move the traffic to the ctlplane network. And things will break, as services does'nt listen on the ctlplane. | 13:43 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/stein: Improve facter cache reliability https://review.opendev.org/733145 | 13:44 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/rocky: Improve facter cache reliability https://review.opendev.org/733146 | 13:44 |
| *** pkopec has quit IRC | 13:44 | |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/queens: Improve facter cache reliability https://review.opendev.org/733147 | 13:44 |
| mwhahaha | raildo: -^ | 13:44 |
| raildo | mwhahaha, well, that was easy haha, thank you! | 13:45 |
| hjensas | Tengu: we never could log into a ceph node and ping the controller node using the shortname alias either. It is'nt/was'nt perfect. | 13:46 |
| lbragstad | sweet thanks mwhahaha | 13:47 |
| *** pkopec has joined #tripleo | 13:49 | |
| *** xek_ has joined #tripleo | 13:50 | |
| *** xek has quit IRC | 13:53 | |
| Tengu | hjensas: hmmm the UndercloudHostnameResolveNetwork might be set to "ctlplane" - and apparently it IS seeing the condition in network/service_net_map.j2.yaml - but I don't understand how it's used in the /etc/hosts generation.... it probably isn't, in fact, right? | 13:53 |
| Tengu | hjensas: so maybe we should get 2 generations - one for the undercloud, and the second one, with that run_once thing, for the overcloud - unless there's a way to take advantage of the HostnameResolveNetwork in the generation itself.... | 13:56 |
| openstackgerrit | Jose Luis Franco proposed openstack/tripleo-heat-templates stable/train: Check transfer data flag to skip pacemaker normal upgrade. https://review.opendev.org/732842 | 13:56 |
| hjensas | Tengu: Can we not make TLS use the FQDN? Or the shortname.ctlplane alias? | 13:57 |
| Tengu | hjensas: the main issue is CU being used to connect from undercloud using "shortname" :(. | 13:58 |
| hjensas | Tengu: I mean, do we want the same alias to resolve to different addresses depending on which host we are on? | 13:58 |
| Tengu | hjensas: apparently we do -.-. | 13:59 |
| Tengu | and I'm not really happy with that, but as said.... some ppl are crying apparently. | 13:59 |
| *** xek_ is now known as xek | 13:59 | |
| *** jmasud has quit IRC | 13:59 | |
| Tengu | note that this "feature" might explode once freeIPA manages the DNS :] | 14:00 |
| Tengu | (can't wait seeing this happening) | 14:00 |
| hjensas | Tengu: users used to put the alias in manually, they can put it in above the ansible managed block? | 14:00 |
| Tengu | nope, because the ansible block is set to be put at the very beginning of the file | 14:01 |
| Tengu | have my PTG session right now | 14:01 |
| mwhahaha | i knew that'd cause problems | 14:01 |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 14:02 |
| * Tengu sees bad practices everywhere | 14:02 | |
| Tengu | mwhahaha: that /etc/hosts management IS a problem in itself ;) | 14:02 |
| hjensas | yeah, repeating myself our name resolving stuffs is a mess. | 14:02 |
| mwhahaha | the problem is mixed management of things | 14:02 |
| Tengu | letting ppl inject things manually in there is........ bleh. | 14:02 |
| mwhahaha | why did we move it to the beginning again? | 14:03 |
| Tengu | ekultails: -^^ | 14:03 |
| Tengu | :) | 14:03 |
| Tengu | I don't recall the whole thing. probably an issue with ordering or something. | 14:03 |
| mwhahaha | probably to prevent a customer from overriding something we actually need | 14:03 |
| mwhahaha | e.g. this | 14:03 |
| openstackgerrit | Jose Luis Franco proposed openstack/tripleo-heat-templates master: Check transfer data flag to skip pacemaker normal upgrade. https://review.opendev.org/732722 | 14:04 |
| hjensas | I think we could make things smart enough to put the shortname alias on the hostname_resolve_network for each role, instead of having the alias on the primary ip entry. | 14:05 |
| hjensas | The logic will be a mess, a ansible filter written in python may be a good way. | 14:06 |
| *** ramishra has joined #tripleo | 14:09 | |
| openstackgerrit | Jose Luis Franco proposed openstack/tripleo-heat-templates stable/ussuri: Check transfer data flag to skip pacemaker normal upgrade. https://review.opendev.org/732844 | 14:09 |
| gfidente | tbarron so I am not sure if we want to adopt ganesha | 14:11 |
| gfidente | I think it's our only option | 14:11 |
| gfidente | but we'd be duplicating code which is in mgr | 14:11 |
| gfidente | fultonj fmount ^^ | 14:12 |
| gfidente | so I feel more like we're forced to do so, but it might bring some positives with it | 14:12 |
| gfidente | like we have more control over which image to use, how, bla bla bla | 14:12 |
| openstackgerrit | Pooja Jadhav proposed openstack/tripleo-ci master: Checkout diskimage-builder to 2.38.0 tag for CentOS-7 https://review.opendev.org/733114 | 14:13 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/stein: Improve facter cache reliability https://review.opendev.org/733145 | 14:13 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/rocky: Improve facter cache reliability https://review.opendev.org/733146 | 14:14 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates stable/queens: Improve facter cache reliability https://review.opendev.org/733147 | 14:15 |
| fmount | gfidente: tbarron having it as a separated entity in TripleO helps and you can have more control on the parameters | 14:16 |
| fmount | so yeah, +1 on this idea ^ | 14:16 |
| gfidente | but then again we have to maintain it | 14:16 |
| gfidente | gouthamr left some more comments in the pad | 14:16 |
| fmount | yeah | 14:17 |
| *** abdysn has quit IRC | 14:17 | |
| gfidente | weshay|ruck fmount fultonj tbarron gouthamr I put ganesha there for 16:30 https://etherpad.opendev.org/p/tripleo-ptg-victoria | 14:19 |
| fultonj | thanks gfidente | 14:19 |
| fultonj | sorry weshay|ruck i guess we have more :) | 14:19 |
| fultonj | actually i'm glad, it's a good topic | 14:19 |
| fmount | gfidente: ack thanks | 14:19 |
| gfidente | I guess an alternative is to keep ceph-ansible around | 14:20 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Deprecate EnablePaunch and remove Paunch support https://review.opendev.org/731565 | 14:20 |
| fultonj | gfidente: can you please write in line 81 ? | 14:20 |
| fultonj | https://etherpad.opendev.org/p/tripleo-ceph | 14:20 |
| fmount | gfidente: let's go through different options and see if we can end up w/ a plan | 14:20 |
| fultonj | gfidente: i asked what i think is an obvious question which we should answer | 14:20 |
| gfidente | fultonj so that is a good point as well | 14:20 |
| gfidente | another alternative | 14:20 |
| gfidente | but basically in the scenario where ceph is external | 14:21 |
| fultonj | wait | 14:21 |
| fultonj | write it in the etherpad | 14:21 |
| gfidente | the overcloud nodes should become part of the external mgr | 14:21 |
| gfidente | sure yeah | 14:21 |
| fultonj | i'm watching it i'll read (promise) | 14:21 |
| fultonj | :) | 14:21 |
| gfidente | this is basically a bit like clients config | 14:22 |
| gfidente | fultonj++ good exploring scenarios | 14:24 |
| fultonj | tbarron: by the time we get cephadm tripleo integration stable we might be on pacific | 14:27 |
| tbarron | fultonj: even there it is tricky, though i'd love the target. plan of record is to use k8s for ganesha HA. | 14:28 |
| tbarron | fultonj: #1) they may think k8s has more HA than it does | 14:28 |
| gfidente | they prefer to do that | 14:29 |
| gfidente | but wasn't ganesha 3.6 ha safe | 14:29 |
| gfidente | storing state in the rbd itself? | 14:29 |
| tbarron | it guarantees eventual restart but not HA, e.g. fencing, etc. | 14:29 |
| gfidente | right | 14:29 |
| fultonj | k8s is out of scope with tripleo-cephadm integration | 14:29 |
| gfidente | while storing state in rbd was actual a/a | 14:29 |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: Allow overriding InterfaceDefaultRoute with ips_from_pool template https://review.opendev.org/732698 | 14:29 |
| tbarron | #2) we need to support integrated TripleO deployment, not just external ceph cluster | 14:29 |
| fultonj | if ganesha needs k8s to get off pacemaker, it won't get to the edge until k8s does | 14:30 |
| tbarron | fultonj: so i see pacific time frame as good for experiments with external ceph cluster which happens to have k8s | 14:30 |
| tbarron | fultonj: gfidente ^^^ if that works well, we could look at doing something similar with integrated ceph cluster deployment | 14:31 |
| tbarron | but who knows what will be happening with k8s/openstack integration by then | 14:31 |
| tbarron | fultonj: gfidente bottom line is that I think we cannot count on the need for pacemaker control to go away by then | 14:32 |
| fultonj | gfidente: i'm trying to understand the scope your proposal into the larger timeframe | 14:32 |
| fultonj | tbarron: you ok then with ganesha not getting to the edge until k8s does? | 14:32 |
| tbarron | fultonj: gfidente though we can be *open* to opportunities to simplify our lives and just use the mgr | 14:32 |
| gfidente | tbarron yeah | 14:33 |
| gfidente | so maybe this is a topic we should discuss | 14:33 |
| gfidente | not in tripleo? | 14:33 |
| fultonj | "just use the mgr" is what i'm totally into | 14:33 |
| gfidente | not sure really | 14:33 |
| tbarron | fultonj: no but I don't have a good alternative | 14:33 |
| gfidente | exactly | 14:33 |
| gfidente | <gfidente> so I feel more like we're forced to do so, but it might bring some positives with it | 14:33 |
| openstackgerrit | Merged openstack/tripleo-docs master: [troubleshooting] Remove QEMU workaround. https://review.opendev.org/701862 | 14:34 |
| tbarron | fultonj: gfidente one thing we can explore is whether cephfs native is acceptable for lots of edge use cases | 14:34 |
| gfidente | yeah | 14:34 |
| fultonj | tbarron: that's a good idea | 14:34 |
| fultonj | tbarron: no pcs required right? | 14:34 |
| tbarron | fultonj: gfidente e.g. telco workloads may belong to trusted VMs | 14:34 |
| gfidente | or the virtiofs thing | 14:34 |
| tbarron | fultonj: correct, cephfs native relies only on mds which manages its own ha | 14:35 |
| gfidente | but then having 2 different implementations is not nice | 14:35 |
| fultonj | i like it | 14:35 |
| Tengu | hjensas: or maybe abandon my change and make it a doc bug because ppl should use hostname.ctlplane instead of shortname from the undercloud? | 14:35 |
| gfidente | I'd keep ceph-ansible to do it in the controlplane | 14:35 |
| gfidente | s/having/maintaining/ | 14:35 |
| * fultonj going to listen to luke now | 14:35 | |
| *** waleedm has quit IRC | 14:35 | |
| tbarron | gfidente: fultonj yes we will research virtio, where cephfs access is mediated by hypervisor (like rbd) rather than allowing VMs network access to Ceph daemons | 14:36 |
| hjensas | Tengu: yes, I personally think thats okay. They should use hostname.ctlplane or even hostname.ctlplane.cloudname i.e the fqdn. | 14:37 |
| * tbarron listens to luke too | 14:37 | |
| Tengu | hjensas: this might be the right thing to do in order to avoid weird situation. And iirc, freeIPA might be in charge of the DNS soon, with the new tls-everywhere deploy (ping lbragstad, xek, ade_lee) - so there won't be this shortname possibility anyway (unless we keep the /etc/hosts thing in place, and I'm pretty sure it will stick until we correct the clusters in order to use an fqdn instead of | 14:45 |
| Tengu | shortname. bleh) | 14:45 |
| Tengu | (or properly set the "search" in resolv.conf..) | 14:45 |
| fultonj | rename the feature: "composable tripleo role" | 14:49 |
| openstackgerrit | Martin Kopec proposed openstack/ansible-role-collect-logs master: Add artcl_build_url to infrared plugin.spec https://review.opendev.org/733158 | 14:50 |
| EmilienM | fultonj++ | 14:51 |
| fultonj | ekultails: good stuff | 14:57 |
| tkajinam | ekultails, do you mind to check the permission of your slides ? | 14:58 |
| ekultails | fultonj: Thanks! The auto scale stuff could probably be something we work on together. It would be good to at least bounce ideas off each other. | 14:58 |
| fultonj | ekultails: i'd be happy to | 14:59 |
| ekultails | tkajinam: I thought I opened it up for read/comment privileges. Will double check on that. | 14:59 |
| tkajinam | ekultails, I tried to open the link in etherpad but seems like it is not public, I'm afraid | 14:59 |
| tkajinam | ekultails, thx | 14:59 |
| ade_lee | ekultails, ping me later this week maybe re: autoscaling and tls-e/ipa | 14:59 |
| ekultails | Updated slides link! https://drive.google.com/file/d/1c2D67QZ5UGBYRONogZJKMief5wJxQAIt/view?usp=sharing | 15:00 |
| ekultails | ^ tkajinam | 15:00 |
| ade_lee | ekultails, I'll need to understand exactly what is/is not being run to see if ipa is affected | 15:00 |
| ekultails | ade_lee: Will do, thanks for offering to help! | 15:00 |
| fultonj | ekultails: i'd like to hook this in with how we scale ceph services | 15:00 |
| tkajinam | ekultails, thanks a lot. now I can see that :-) | 15:00 |
| fultonj | ekultails: and i know it's something which comes up a lot downstream | 15:00 |
| fultonj | and in the scale lab | 15:00 |
| ekultails | fultonj: I was thinking the same thing when I saw your presentation earlier this week. :-) | 15:01 |
| fultonj | https://github.com/openstack/tripleo-ansible/blob/master/tripleo_ansible/roles/tripleo_ceph_run_ansible/tasks/main.yml#L50 | 15:03 |
| fultonj | ekultails: ^ hopefully we can do better than just ^ ;) | 15:03 |
| fultonj | but that bandaid does help | 15:03 |
| openstackgerrit | Bogdan Dobrelya (bogdando) proposed openstack/tripleo-ansible master: Aggregate data for podman-managed containers https://review.opendev.org/729260 | 15:04 |
| openstackgerrit | Martin Kopec proposed openstack/ansible-role-collect-logs master: Run container collection in parallel https://review.opendev.org/733165 | 15:04 |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 15:11 |
| openstackgerrit | yatin proposed openstack/tripleo-ci master: Checkout diskimage-builder to 2.38.0 tag for CentOS-7 https://review.opendev.org/733114 | 15:14 |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-quickstart-extras master: WIP: Remove NetworkSecGroupTest from skip list (train) https://review.opendev.org/733170 | 15:17 |
| *** dmellado has quit IRC | 15:18 | |
| *** dmellado has joined #tripleo | 15:20 | |
| openstackgerrit | Merged openstack/tripleo-upgrade master: Run ceph upgrade after converge https://review.opendev.org/731466 | 15:20 |
| fultonj | gfidente fmount ^ review 731466 | 15:24 |
| *** rh-jelabarre has quit IRC | 15:24 | |
| *** rh-jelabarre has joined #tripleo | 15:24 | |
| Madkiss | hm. "openstack overcloud node delete" fails because the machines are unreachable | 15:40 |
| Madkiss | can I tell thbe thing to ignore that? | 15:40 |
| mwhahaha | Madkiss: what's teh error? | 15:41 |
| *** bandini has quit IRC | 15:41 | |
| *** lmiccini has quit IRC | 15:42 | |
| Madkiss | fatal: [b52-41-2-46-3]: UNREACHABLE! | 15:42 |
| mwhahaha | it's supposed to ignore that so i need teh whole log snippet | 15:43 |
| mwhahaha | that means nothign to me | 15:43 |
| Madkiss | http://paste.debian.net/hidden/6f5ec2b0/ | 15:43 |
| mwhahaha | [WARNING]: Found variable using reserved name: ignore_unreachable | 15:44 |
| mwhahaha | probably related | 15:44 |
| mwhahaha | EmilienM: -^ | 15:44 |
| openstackgerrit | Bogdan Dobrelya (bogdando) proposed openstack/tripleo-quickstart-extras master: Print extracted stats for containers managment https://review.opendev.org/731578 | 15:44 |
| mwhahaha | is the command actually erroring? or are you just seeing that output and assuming problems | 15:44 |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Update scn003 to exercise ExtraFirewallrules capabilities https://review.opendev.org/731611 | 15:44 |
| openstackgerrit | Merged openstack/tripleo-heat-templates stable/train: Force container deletion if namespace does not exist in service_kill https://review.opendev.org/731121 | 15:44 |
| Madkiss | Scale-down configuration failed. -- and I still see the servers in the server list of the undercloud. | 15:45 |
| mwhahaha | k plz file a bug | 15:45 |
| mwhahaha | is this pre-deployed servers that were previously provisioned? seems odd that it'd be an ssh key error | 15:46 |
| Madkiss | no. i cancelled deployment earlier. i have been fiddling with this stuff the whole day again already. | 15:46 |
| mwhahaha | if those are the servers it's trying to delete, it shouldn't fail if they are unreachable | 15:47 |
| Madkiss | How do i get rid of the nodes? I need to go on with deployment. There must be a way to cleanly delete these nods from the setup. | 15:47 |
| mwhahaha | though i wonder because they are actually up that it is a different error | 15:47 |
| Madkiss | I mean. let's just assume these servers blew up into flames. | 15:49 |
| mwhahaha | if they blew into flames it wouldnt' be a key error | 15:49 |
| mwhahaha | it's supposed to ignore failures | 15:49 |
| Madkiss | situation is not different if the host is unreachable. | 15:50 |
| Madkiss | it will still fail | 15:50 |
| Madkiss | had that in the past. | 15:50 |
| mwhahaha | you'd think that, but ansible likes to do weird things | 15:50 |
| mwhahaha | it's likely becaus e[WARNING]: Found variable using reserved name: ignore_unreachable and ansible changed how it handles that | 15:51 |
| mwhahaha | the bug is that probably needs to be renamed and adjusted | 15:51 |
| mwhahaha | rename that var in common/deploy-steps.j2 and common/deploy-steps-playbook-common.yaml in THT to something else (and consistent) | 15:52 |
| mwhahaha | and it'll probably work | 15:52 |
| Tengu | ade_lee: do you think the ansible-tripleo-ipa-server might be a collection?? please! | 15:52 |
| Tengu | ade_lee: that would allow me to re-use it in tripleo-lab | 15:52 |
| Tengu | \o/ | 15:53 |
| Madkiss | so ignore_unreachable_YOLO: true for all of them should work you say? | 15:54 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates master: Fix reserved name variable https://review.opendev.org/733187 | 15:54 |
| mwhahaha | Madkiss: -^ | 15:54 |
| mwhahaha | yes | 15:54 |
| Madkiss | Let's see | 15:54 |
| *** derekh has joined #tripleo | 15:54 | |
| *** karthiks has joined #tripleo | 15:56 | |
| openstackgerrit | Bogdan Dobrelya (bogdando) proposed openstack/tripleo-quickstart-extras master: Print extracted stats for containers managment https://review.opendev.org/731578 | 15:58 |
| *** Goneri has quit IRC | 16:00 | |
| openstackgerrit | Takashi Kajinami proposed openstack/tripleo-heat-templates master: DNM: Test gate results for another implementation of 732698 https://review.opendev.org/733191 | 16:01 |
| Madkiss | mwhahaha: changed it in /usr/share, still "[WARNING]: Found variable using reserved name: ignore_unreachable" | 16:05 |
| mwhahaha | you might need to change it in /var/lib/mistral/overcloud | 16:05 |
| mwhahaha | somewhere | 16:05 |
| mwhahaha | i don't think we re-generate the playbooks for that command | 16:05 |
| *** ykarel is now known as ykarel|away | 16:07 | |
| *** dtantsur is now known as dtantsur|afk | 16:13 | |
| *** lucasagomes has quit IRC | 16:14 | |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 16:14 |
| Madkiss | nope. | 16:14 |
| Madkiss | [WARNING]: Found variable using reserved name: ignore_unreachable | 16:14 |
| Madkiss | my changes in /var/lib/mistral have also been overriden. | 16:15 |
| karthiks | mwhahaha, EmilienM can you take a look at https://review.opendev.org/#/c/729891/ | 16:20 |
| EmilienM | karthiks: sure | 16:20 |
| *** jpich has quit IRC | 16:21 | |
| weshay|ruck | https://review.opendev.org/#/c/731529/ | 16:24 |
| *** rpittau is now known as rpittau|afk | 16:24 | |
| Madkiss | https://bugs.launchpad.net/tripleo/+bug/1881452 | 16:25 |
| openstack | Launchpad bug 1881452 in tripleo "Scale down unreachable host fails" [Medium,In progress] - Assigned to Brendan Shephard (bshephar) | 16:25 |
| Madkiss | oh. | 16:25 |
| mwhahaha | i'll rebase on top of that | 16:26 |
| mwhahaha | i think the var needs to be renamed | 16:26 |
| Madkiss | https://review.opendev.org/#/c/732000/ | 16:27 |
| weshay|ruck | fmount, gfidente fultonj ready for the follow up on ganesha? | 16:27 |
| mwhahaha | Madkiss: https://review.opendev.org/#/c/723382/ | 16:27 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-heat-templates master: Fix reserved name variable https://review.opendev.org/733187 | 16:28 |
| fultonj | weshay|ruck: i'm here, though i think gfidente needs to lead this discussion | 16:29 |
| weshay|ruck | aye | 16:29 |
| gfidente | weshay|ruck ready tes | 16:29 |
| fmount | fultonj: weshay|ruck yeah | 16:29 |
| gfidente | thanks for pinging | 16:29 |
| Madkiss | okay, now I need to figure out how to make it regenerate the code from /usr/share. I have no idea where the stuff in /var/lib/mistral is coming from ... | 16:32 |
| *** ysandeep is now known as ysandeep|away | 16:33 | |
| mwhahaha | Madkiss: we don't use ignore_unreachable in many places so you should just be able to grep for it in /var/lib/mistral/<cloudname> | 16:34 |
| Madkiss | I did, and I changed it there, but it gets regenerated during hte call apparently. | 16:34 |
| Madkiss | as aid. I changed that, and then I ran the delete command, and after that the files were back to their previous satete | 16:34 |
| mwhahaha | oh it's in the container | 16:34 |
| mwhahaha | stupid mistral | 16:34 |
| mwhahaha | make the change in the mistral executor container | 16:34 |
| mwhahaha | podman exec -it -u root mistral_executor /bin/bash | 16:35 |
| mwhahaha | and make the change in that /usr/share/ | 16:35 |
| Madkiss | no "patch" in the container | 16:36 |
| Madkiss | jesus christ. | 16:36 |
| mwhahaha | aren't containers great? | 16:36 |
| Tengu | you can podma copy | 16:37 |
| Tengu | that way you copy the final file in-place in the container. | 16:37 |
| Tengu | or even podman mount <container> | 16:37 |
| Madkiss | the patch seems to be present in the container frm what I can see | 16:38 |
| Tengu | and you get the container FS exposed on the host, then you can use patch|whatever, and then "podman umount <container>" | 16:38 |
| mwhahaha | shouldn't be unless you patched it | 16:38 |
| fultonj | http://blog.johnlikesopenstack.com/2019/07/notes-on-testing-tripleo-common-mistral.html | 16:38 |
| fultonj | FWIW ^ is how i patch the mistral undercloud container | 16:38 |
| fultonj | take it or leave it :) | 16:39 |
| Madkiss | it is. | 16:39 |
| Madkiss | i mean. unless this is somehow mounted of course. | 16:40 |
| Madkiss | let's try restarting the container. | 16:40 |
| Madkiss | nope. | 16:46 |
| openstackgerrit | Merged openstack/tripleo-ansible stable/ussuri: Fix conditions tuned profile https://review.opendev.org/732503 | 16:56 |
| Madkiss | nope. | 16:56 |
| Madkiss | this does not work. the code in the container is good now, but the error remains the same. | 16:56 |
| Madkiss | the files in /var/lib/mistral on the host are still erorneous. | 16:57 |
| Madkiss | where are these generated from? | 16:57 |
| *** karthiks has quit IRC | 16:58 | |
| *** jpena is now known as jpena|off | 17:04 | |
| *** marios has quit IRC | 17:04 | |
| mwhahaha | tbarron, gfidente, fultonj: i'd be happy to join another meeting about this if you want to continue another time. we've generally been trying to stop compensating for lack of features in other tools in tripleo if possible | 17:05 |
| gfidente | mwhahaha I think it's good input | 17:05 |
| gfidente | we can review this with ceph people more | 17:05 |
| gfidente | the lack of ha is a ceph issue | 17:05 |
| gfidente | tbarron ^^ | 17:05 |
| *** amoralej is now known as amoralej|off | 17:06 | |
| openstackgerrit | Bogdan Dobrelya (bogdando) proposed openstack/tripleo-ansible master: Aggregate data for podman-managed containers https://review.opendev.org/729260 | 17:06 |
| fultonj | mwhahaha: thanks | 17:07 |
| fultonj | ceph rbd will work with the new tool | 17:07 |
| fultonj | so will rgw | 17:08 |
| fultonj | it's for manila with ganesha that we have the problem | 17:08 |
| tbarron | i'd be saying the same as mwhahaha in his shoes, just am stuck with this ganesha thing | 17:09 |
| tbarron | which is a very usefult thing for users | 17:09 |
| mwhahaha | you should have said no in the beginning :D | 17:09 |
| Madkiss | mwhahaha: Do you happen to know when and from where stuff in /var/lib/mistral is generated? | 17:09 |
| Madkiss | it *is* generated, but I do not know from where | 17:09 |
| mwhahaha | Madkiss: not off the top of my head | 17:09 |
| mwhahaha | Madkiss: not for the node delete command i'd have to look | 17:09 |
| tbarron | mwhahaha: well then we wouldn't have the useful feature | 17:09 |
| tbarron | which it was my job to integrate into openstack | 17:10 |
| mwhahaha | tbarron: a half implemented useful feature is still half implemented and will enver end up fully implemented | 17:10 |
| mwhahaha | i think i'm struggling with where/how this needs to run | 17:10 |
| tbarron | mwhahaha: you live in a better world than mine :D | 17:11 |
| mwhahaha | if it's an external ceph cluster, then deploying k8s for maangement should be a ceph thing? | 17:11 |
| mwhahaha | pffft my world is terrible, we've been saying yes for years | 17:11 |
| mwhahaha | the fact we integrated it with our pacemaker seems weird | 17:12 |
| Madkiss | anything that involves pacemaker is weird. | 17:12 |
| tbarron | mwhahaha: if it's external then we have no k8s dependency | 17:12 |
| tbarron | mwhahaha: i'm not convinced that its HA will *work* yet but that's another matter | 17:13 |
| gfidente | mwhahaha it's totally awesome | 17:13 |
| gfidente | until then the manila backend driver was 90% netapp | 17:13 |
| tbarron | mwhahaha: k8s guarantees eventual restart, not HA | 17:13 |
| gfidente | mwhahaha after that the manila backend driver was 75% ceph | 17:13 |
| gfidente | mwhahaha so that bit was actually totally useful :D | 17:13 |
| gfidente | weird | 17:13 |
| mwhahaha | my future with ceph: https://my1.fr/files/emilien-right-now-2.gif | 17:13 |
| gfidente | but | 17:13 |
| gfidente | 75% less weird | 17:13 |
| tbarron | mwhahaha: but that's somebody else's responsibility if it's extenal, might as well be a NetApp or EMC storage device | 17:13 |
| fultonj | back to the proposal: so for anyone wanting to use ganesha they need ceph-ansible on their undercloud even when the rest of the ceph services are managed by cephadm/orc | 17:14 |
| fultonj | ^ is the least worst if we're keeping ganesha | 17:14 |
| fultonj | right? | 17:14 |
| gfidente | fultonj that is in victoria/octopus | 17:14 |
| gfidente | we' | 17:14 |
| fultonj | because we don't want to do redo what ceph-ansible did in tripleo | 17:14 |
| gfidente | we'll see at next ptg what happens | 17:14 |
| fultonj | i agree w/ mwhahaha we shouldn't | 17:14 |
| gfidente | basing on more datra | 17:14 |
| tbarron | fultonj: but did it do much? | 17:14 |
| fultonj | tbarron: sorry... what's "it" ? | 17:15 |
| *** bogdando has quit IRC | 17:15 | |
| tbarron | fultonj: ceph-ansible | 17:15 |
| tbarron | for ganesha | 17:15 |
| tbarron | it sets up config for the ganesha container | 17:15 |
| tbarron | it sets up some ceph pools and a service user key | 17:16 |
| tbarron | anything else? | 17:16 |
| tbarron | maybe some ceph.conf | 17:16 |
| gfidente | then there is upgrade | 17:16 |
| Madkiss | EmilienM: do you happen to know? | 17:16 |
| tbarron | it does *not* manage the life cycle of ceph-nfs/ganesha daemon | 17:17 |
| gfidente | tbarron I think it's fair to review this with ceph people first | 17:17 |
| fultonj | tbarron: i'm assuming it did soemthing you needed it to do that nothing else does | 17:17 |
| fultonj | something we still need done and soemthing we're trying to find a way to do without ceph-ansible | 17:18 |
| EmilienM | Madkiss: to know what? | 17:18 |
| *** jcapitao has quit IRC | 17:18 | |
| tbarron | fultonj: i *think* just the things i said, and we used ceph-ansible b/c that was the canonical way to do those thiings and it was already doing those thhings (and more) for the other ceph daemons | 17:18 |
| tbarron | gfidente: +1 on reviewing with ceph people | 17:18 |
| Madkiss | EmilienM: Where does mistral generate its ansible files from for the scale command? I changed stuff in /usr/share, but the files in /var/lib/mistral are, albeit being regenerated, generated from old content | 17:19 |
| mwhahaha | it just runs config_download_deploy workflow | 17:19 |
| mwhahaha | which i think should already have existing playbooks | 17:19 |
| tbarron | fultonj: agree, and i guess i just want to know whether the "something" that ceph-ansible did for ganesha is worth keeping the machinery for ansible to do | 17:19 |
| fultonj | tbarron: me too! | 17:19 |
| fmount | gfidente: tbarron fultonj this is part of the feature gap review we worked on and we should continue to work w/ ceph people | 17:19 |
| fmount | ceph-ansible vs cephadm | 17:20 |
| mwhahaha | Madkiss: i think the issue is that we're regenerating the playbooks but you can't update them w/o patching the stack | 17:20 |
| Madkiss | awesome. | 17:20 |
| fultonj | Madkiss: version? | 17:20 |
| fmount | tbarron: gfidente at least the role used by ceph-ansible can be "reworked" in this new project | 17:20 |
| *** ravsingh has quit IRC | 17:20 | |
| Madkiss | fultonj: 16 | 17:20 |
| *** ysandeep|away is now known as ysandeep | 17:20 | |
| *** hjensas is now known as hjensas|afk | 17:21 | |
| fultonj | https://docs.openstack.org/project-deploy-guide/tripleo-docs/latest/deployment/ansible_config_download.html#manual-config-download | 17:21 |
| fultonj | Madkiss: ^ might this explain | 17:21 |
| fultonj | Heat is authoritative for your config, it's translated to ansible | 17:22 |
| tbarron | fmount: sure, for some value of "rework" :S | 17:22 |
| Madkiss | fultonj: alright, how do I change it? | 17:22 |
| * fultonj has serious problems with the word "it" today | 17:23 | |
| fultonj | change what? | 17:23 |
| fultonj | downlaod the ansible and change it directly | 17:23 |
| fultonj | up update your heat | 17:23 |
| mwhahaha | it's a scale down action | 17:23 |
| mwhahaha | he can't do that | 17:23 |
| Madkiss | I want to delete five hosts from my goddamn cloud. | 17:23 |
| mwhahaha | if the hosts are up, fix the ssh key | 17:23 |
| mwhahaha | so they can connect as a workaround | 17:23 |
| Madkiss | I don't think I have any chance to login to these. | 17:24 |
| mwhahaha | then you'd need to re-run a deployment with the patched THT to fix this going forward | 17:24 |
| openstackgerrit | Merged openstack/tripleo-validations master: Stop to use the __future__ module. https://review.opendev.org/732971 | 17:25 |
| fultonj | gfidente: so we have a ganesha problem to solve | 17:25 |
| * mwhahaha afk | 17:26 | |
| *** jmasud has joined #tripleo | 17:26 | |
| fultonj | gfidente: do you want to update the spec with a ganesha section? | 17:26 |
| gfidente | not yet but I think we should yes | 17:26 |
| gfidente | let's find a plan first | 17:26 |
| fultonj | ok | 17:28 |
| fultonj | we'll pick it up later then | 17:29 |
| openstackgerrit | Martin Kopec proposed openstack/ansible-role-collect-logs master: Run container collection in parallel https://review.opendev.org/733165 | 17:30 |
| *** gfidente is now known as gfidente|afk | 17:31 | |
| *** fmount is now known as fmount|afk | 17:32 | |
| *** aedc has quit IRC | 17:36 | |
| *** waleedm has joined #tripleo | 17:39 | |
| mwhahaha | Madkiss: I might have a solution for you. Gimme a few | 17:40 |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 17:40 |
| EmilienM | mwhahaha, weshay|ruck : are we going to wait post-ptg to land https://review.opendev.org/#/c/723665/ ? | 17:43 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/ussuri: Update scn003 to exercise ExtraFirewallrules capabilities https://review.opendev.org/733290 | 17:44 |
| mwhahaha | Madkiss: try rerunning the deploy with --stack-only (to update the stack but skip runnings the deploy steps) then you should be able to node delete | 17:45 |
| Madkiss | thank you | 17:45 |
| Madkiss | am I just blind, or is that option not documented? | 17:46 |
| openstackgerrit | Merged openstack/tripleo-upgrade stable/train: Remove ffu converge and perform stack update only. https://review.opendev.org/731758 | 17:47 |
| mwhahaha | Madkiss: it's in the --help ? | 17:47 |
| mwhahaha | i might be wrong on the name | 17:47 |
| mwhahaha | it might be --no-config-download | 17:47 |
| * mwhahaha can enver remember | 17:47 | |
| mwhahaha | you want the one that updates the stack that doesn't do the deploy | 17:48 |
| mwhahaha | it changed between some versions i think | 17:48 |
| Madkiss | i'll give it a go | 17:48 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common stable/ussuri: Switch to tripleo stdout callback https://review.opendev.org/731749 | 17:48 |
| EmilienM | mwhahaha: I tried --stack-only the other day and it worked fine | 17:50 |
| EmilienM | fultonj thought we had an issue but we didn't | 17:50 |
| * EmilienM pulls the lp again | 17:51 | |
| EmilienM | https://bugs.launchpad.net/bugs/1880577 | 17:52 |
| openstack | Launchpad bug 1880577 in tripleo "--stack-only --no-config-download are broken" [Undecided,Invalid] | 17:52 |
| mwhahaha | yea i just didn't remember what the cmdline was | 17:52 |
| EmilienM | --stack-only is supposed to be fine | 17:52 |
| EmilienM | fultonj said --no-config-download is also ifne, but i thought this one would be deprecated at some point | 17:53 |
| mwhahaha | --stack-only still does teh config download bits it hink | 17:53 |
| mwhahaha | it's all too complex for me | 17:53 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Deprecate EnablePaunch and remove Paunch support https://review.opendev.org/731565 | 17:53 |
| EmilienM | mwhahaha: no it doesn't | 17:54 |
| EmilienM | I think | 17:54 |
| EmilienM | let me check again | 17:54 |
| mwhahaha | see :D | 17:54 |
| mwhahaha | *i think* should be something we all put on a t-shirt | 17:54 |
| EmilienM | it just creates the Heat stack | 17:54 |
| Madkiss | Not sure if happy or sad about the fact the its developers call all this "too complex". | 17:55 |
| Madkiss | :P | 17:55 |
| mwhahaha | it's less complex than it used to be | 17:55 |
| EmilienM | ha! --no-config-download and --stack-only are the same options | 17:55 |
| mwhahaha | ah | 17:55 |
| EmilienM | and it has action='store_false' on config_download | 17:56 |
| EmilienM | so it creates the plan, create the stack, creates overcloudrc and that's it | 17:57 |
| mwhahaha | yea that's what he needs | 17:57 |
| mwhahaha | cause the stack has the bad tasks in it still | 17:57 |
| EmilienM | right | 17:57 |
| EmilienM | sorry i just read the backlog and I confirm that's what you need to do Madkiss | 17:57 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-image-elements stable/queens: Removing debug output of 51-hosts https://review.opendev.org/731338 | 17:58 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/ussuri: Remove Ceph{Admin,Mon,Mds}Key parameters https://review.opendev.org/731051 | 17:58 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/ussuri: Add additional files to ipa standalone test https://review.opendev.org/731048 | 17:58 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: Introduce tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/729824 | 17:59 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: Introduce tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/729824 | 17:59 |
| weshay|ruck | EmilienM, nope wf | 18:00 |
| weshay|ruck | re: the container build spec | 18:00 |
| *** ysandeep is now known as ysandeep|away | 18:01 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-quickstart-extras master: standalone: add support to build containers before deployment https://review.opendev.org/729465 | 18:03 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-quickstart-extras master: standalone: add support to build containers before deployment https://review.opendev.org/729465 | 18:04 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-quickstart-extras master: standalone: add support to build containers before deployment https://review.opendev.org/729465 | 18:05 |
| openstackgerrit | Merged openstack/tripleo-specs master: Add spec for simple container generation https://review.opendev.org/723665 | 18:07 |
| EmilienM | weshay|ruck: thanks | 18:08 |
| * fultonj reads backlog | 18:17 | |
| fultonj | --no-config-download and --stack-only are the same options | 18:19 |
| fultonj | yep | 18:19 |
| *** derekh has quit IRC | 18:19 | |
| fultonj | EmilienM: they both work, i just had a user error | 18:19 |
| Madkiss | openstack overcloud node delete: error: unrecognized arguments: --stack-only | 18:20 |
| Madkiss | aha | 18:21 |
| mwhahaha | Madkiss: --stack-only is a deploy option | 18:21 |
| mwhahaha | you need to run a deploy | 18:21 |
| mwhahaha | then you can delete | 18:21 |
| Madkiss | ah, uh. okay. | 18:21 |
| *** chem has quit IRC | 18:22 | |
| Madkiss | "openstack stack resource show overcloud CephStorageServers -f json -c attributes" shows nodes that are not there anymore. how do I delete these? they do not appear in "openstack server list". | 18:22 |
| * mwhahaha is not sure what this has to do with anything | 18:22 | |
| Madkiss | That makes two of us. | 18:22 |
| *** yolanda has quit IRC | 18:23 | |
| *** chem has joined #tripleo | 18:24 | |
| *** jmasud has quit IRC | 18:25 | |
| openstackgerrit | Andreas Jaeger proposed openstack/os-net-config master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733299 | 18:27 |
| *** jmasud has joined #tripleo | 18:28 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates master: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733077 | 18:29 |
| openstackgerrit | Andreas Jaeger proposed openstack/paunch master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733300 | 18:29 |
| openstackgerrit | Merged openstack/tripleo-heat-templates stable/train: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733091 | 18:29 |
| openstackgerrit | Merged openstack/tripleo-heat-templates stable/ussuri: Revert "Only enable leapp tasks when distribution is correct" https://review.opendev.org/733090 | 18:29 |
| openstackgerrit | Andreas Jaeger proposed openstack/puppet-pacemaker master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733302 | 18:31 |
| openstackgerrit | Andreas Jaeger proposed openstack/puppet-tripleo master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733303 | 18:33 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-ansible master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733304 | 18:35 |
| Madkiss | How do I get rid of these instances that are in the CephControllers but do not exist? | 18:36 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-common master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733305 | 18:36 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-image-elements master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733306 | 18:37 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-operator-ansible master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733307 | 18:38 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-puppet-elements master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733308 | 18:39 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-quickstart master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733309 | 18:39 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-specs master: Switch to newer openstackdocstheme version https://review.opendev.org/733310 | 18:41 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-upgrade master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733311 | 18:42 |
| openstackgerrit | Andreas Jaeger proposed openstack/tripleo-validations master: Switch to newer openstackdocstheme and reno versions https://review.opendev.org/733312 | 18:43 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-quickstart-extras master: standalone: add support to build containers before deployment https://review.opendev.org/729465 | 18:44 |
| openstackgerrit | Dave Wilde (d34dh0r53) proposed openstack/tripleo-heat-templates master: Add composible service for tls enrollment https://review.opendev.org/729653 | 18:49 |
| *** waleedm has quit IRC | 18:50 | |
| mwhahaha | weshay|ruck: do we have a fix for teh failed horizon job in stable/train? https://review.opendev.org/#/c/731357/ if not can we remove it for now | 18:58 |
| openstackgerrit | Sergii Golovatiuk proposed openstack/tripleo-heat-templates stable/ussuri: Check transfer data flag to skip pacemaker normal upgrade. https://review.opendev.org/732844 | 18:59 |
| beagles | anyone have thoughts on this error https://zuul.opendev.org/t/openstack/build/605406995c8443dc8bba5e619a02a548/log/logs/undercloud/home/zuul/standalone_deploy.log#3789 | 19:06 |
| weshay|ruck | mwhahaha, /me looks | 19:06 |
| mwhahaha | beagles: it's been fixed | 19:07 |
| * mwhahaha digs up review | 19:07 | |
| mwhahaha | beagles: https://review.opendev.org/#/c/732642/ | 19:07 |
| mwhahaha | you need that | 19:07 |
| beagles | mwhahaha: thanks! | 19:07 |
| weshay|ruck | mwhahaha, I see | 19:08 |
| mwhahaha | beagles: rebase | 19:08 |
| beagles | ack | 19:08 |
| openstackgerrit | Michael Johnson proposed openstack/tripleo-ansible master: Add support for Octavia amphora log offloading https://review.opendev.org/724190 | 19:09 |
| openstackgerrit | Brent Eagles proposed openstack/tripleo-ansible master: Add support for Octavia amphora log offloading https://review.opendev.org/724190 | 19:09 |
| beagles | matter/anti-matter collision - there goes another reality | 19:09 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: Implement a new Container Images layout https://review.opendev.org/722486 | 19:11 |
| johnsom | Maybe this reality will have gates that never break! | 19:13 |
| weshay|ruck | mwhahaha, so if we only execute the horizon tempest tests.. and horizon is not discovered.. it will fail | 19:14 |
| weshay|ruck | we can say.. meh | 19:14 |
| weshay|ruck | and merge https://review.opendev.org/733381 | 19:14 |
| mwhahaha | meh | 19:15 |
| mwhahaha | so maybe a tempestconf bug | 19:15 |
| weshay|ruck | or the service is not running | 19:15 |
| weshay|ruck | 1 of the two | 19:15 |
| Madkiss | Does anybody in here know how to delete individual entries from the above attributes json? is that in mysql or in the deployment plan or what? | 19:15 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: Implement a new Container Images layout https://review.opendev.org/722486 | 19:17 |
| *** sshnaidm_ is now known as sshnaidm | 19:19 | |
| weshay|ruck | mwhahaha, ERROR:scss.ast:Function not found: twbs-font-path:1 | 19:19 |
| EmilienM | cloudnull: among other comments, 2 things mwhahaha said during the review of our tripleo-common patch: https://review.opendev.org/#/c/722486/74/container-images/tcib/base/collectd/collectd.yaml and https://review.opendev.org/#/c/722486/73/container-images/tcib/base/dependencies/neutron-base/neutron-base.yaml | 19:19 |
| weshay|ruck | working version: https://80cd7359a330b8308cec-4db038fcb6d2aaec6902b77d2fb6be2d.ssl.cf2.rackcdn.com/731633/1/check/puppet-horizon-tripleo-standalone/abbe95c/logs/undercloud/var/log/extra/podman/containers/horizon/stdout.log | 19:19 |
| * cloudnull looking | 19:19 | |
| weshay|ruck | failed: https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_163/731357/1/check/puppet-horizon-tripleo-standalone/1635cbb/logs/undercloud/var/log/extra/podman/containers/horizon/stdout.log | 19:19 |
| mwhahaha | weshay|ruck: it didn't fail | 19:20 |
| mwhahaha | 3eb43cd9b02a 192.168.24.1:8787/tripleotrain/centos-binary-horizon:ba27b93bcaf31634d02a40682c62c741c304a687_33622c19-updated-20200528203514 kolla_start 16 minutes ago Up 16 minutes ago horizon 49MB (virtual 1.36GB) | 19:20 |
| EmilienM | cloudnull: we could have a linter that checks for long lines | 19:21 |
| weshay|ruck | 2020-05-29 02:16:20.003 230667 INFO tempest.test_discover.plugins [-] Loading tests from Tempest plugin: horizon | 19:22 |
| weshay|ruck | is not found in https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_163/731357/1/check/puppet-horizon-tripleo-standalone/1635cbb/logs/undercloud/home/zuul/tempest/tempest.log | 19:22 |
| cloudnull | EmilienM mwhahaha for https://review.opendev.org/#/c/722486/73/container-images/tcib/base/dependencies/neutron-base/neutron-base.yaml we link the source to the system directory to ensure this doesn't break in the gate: https://review.opendev.org/#/c/724193/39/roles/build-containers/templates/build.sh.j2@57 | 19:22 |
| weshay|ruck | I'll run it by tempest folks.. to double check | 19:22 |
| cloudnull | its not a great solution , but its working for test ? | 19:22 |
| mwhahaha | cloudnull: yea i know why we do it, just it's not ideal | 19:22 |
| mwhahaha | i'm uncertain if there's an easy way to make it relative tho | 19:23 |
| mwhahaha | hardcoding path always lead to a bad time :D | 19:23 |
| cloudnull | yup | 19:23 |
| * cloudnull doesn't know of a way to get the path of a var file from within the same var file | 19:24 | |
| cloudnull | i can get the playbook path, and role path, but im not sure about the var_file path | 19:25 |
| cloudnull | EmilienM all of those files were generated with some python, I guess we could line wrap them manually? | 19:25 |
| cloudnull | we'd just need to make sure not to regen the files using the import tool | 19:26 |
| *** jmasud has quit IRC | 19:30 | |
| mwhahaha | i mean it can be a longer linter line requirement | 19:32 |
| mwhahaha | like 180 chars or something | 19:32 |
| mwhahaha | so you'd only have to wrap the really long stuff, but it needs a linter first to figure out what lines need wrapping (by hands at first) | 19:32 |
| mwhahaha | we should split out the tripleo-common/container-images/* to it's own package too btw (if it isn't already) | 19:33 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: Introduce tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/729824 | 19:37 |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 19:37 |
| EmilienM | mwhahaha: I'm working on it now | 19:37 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-quickstart-extras master: standalone: add support to build containers before deployment https://review.opendev.org/729465 | 19:39 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: Introduce tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/729824 | 19:39 |
| *** pkopec has quit IRC | 19:41 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: container-images: add tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/733388 | 19:41 |
| *** jmasud has joined #tripleo | 19:41 | |
| openstackgerrit | Emilien Macchi proposed openstack/python-tripleoclient master: container-images: add tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/733389 | 19:43 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: DNM - test new standalone job when touching container images https://review.opendev.org/733390 | 19:44 |
| openstackgerrit | Merged openstack/openstack-virtual-baremetal master: Stop to use the __future__ module. https://review.opendev.org/732886 | 19:45 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-quickstart-extras master: standalone: add support to build containers before deployment https://review.opendev.org/729465 | 19:49 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: Introduce tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/729824 | 19:50 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: container-images: add tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/733388 | 19:50 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: DNM - test new standalone job when touching container images https://review.opendev.org/733390 | 19:50 |
| openstackgerrit | Emilien Macchi proposed openstack/python-tripleoclient master: container-images: add tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/733389 | 19:51 |
| openstackgerrit | Emilien Macchi proposed openstack/python-tripleoclient master: container-images: add tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/733389 | 19:51 |
| EmilienM | mwhahaha: openstack-tripleo-common-containers | 19:56 |
| EmilienM | we have it already I think | 19:56 |
| mwhahaha | k | 19:56 |
| EmilienM | let me try it | 19:56 |
| EmilienM | mwhahaha: http://paste.openstack.org/show/794332/ | 19:57 |
| EmilienM | since queens | 19:57 |
| EmilienM | however it changes the directory | 19:57 |
| EmilienM | /usr/share/openstack-tripleo-common-containers/container-images | 19:57 |
| mwhahaha | fun | 19:58 |
| EmilienM | which can be problematic is one tries to build the neutron-base image | 19:58 |
| EmilienM | cloudnull: I have an idea for the path issue thing that mwhahaha reported | 19:59 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Remove centos-specific settings in testenv/ovb https://review.opendev.org/732421 | 19:59 |
| *** jmasud has quit IRC | 19:59 | |
| EmilienM | cloudnull: we could use the workdir variable, set it in vars from tripleoclient and default to the right location like if the package would install it | 19:59 |
| EmilienM | so at least we can override it | 19:59 |
| EmilienM | also from a user standpoint it's not transparent :/ | 20:00 |
| EmilienM | one has to know that this rpm installs the configs in that directory by default | 20:00 |
| EmilienM | i wonder if we could just change the default location | 20:00 |
| mwhahaha | that's where relative might be helpful | 20:00 |
| EmilienM | in the distgit | 20:00 |
| EmilienM | mwhahaha: relative to what? | 20:01 |
| mwhahaha | the .yaml | 20:01 |
| EmilienM | oh right | 20:01 |
| EmilienM | or that yeah | 20:01 |
| EmilienM | cloudnull: when not busy with meetings, let us know what you think about ^ | 20:02 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates stable/train: MaxFailPercentage: default to 0 https://review.opendev.org/732415 | 20:03 |
| Madkiss | mwhahaha: looks like that did the trick, thanks a bunch. If you and EmilienM ever seek to actually consume all the beers i owe to you guys, it'll be chapter 11 for me ... | 20:03 |
| mwhahaha | np | 20:04 |
| * EmilienM blames mwhahaha | 20:04 | |
| mwhahaha | you wrote this stupid thing :D | 20:04 |
| mwhahaha | IT'S YOUR FEATURE | 20:04 |
| EmilienM | i probably had too much beers | 20:04 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ansible master: Max percentage failure support https://review.opendev.org/731846 | 20:07 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ansible master: Molecule testing for TripleO Strategy Plugins https://review.opendev.org/731848 | 20:07 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: non-voting job for tripleo-build-containers-ubi-8 https://review.opendev.org/724193 | 20:07 |
| EmilienM | cloudnull: ^ rechecking with latest changes in tripleo-common, to make sure images still build fine | 20:08 |
| *** jmasud has joined #tripleo | 20:09 | |
| *** maciejjozefczyk has quit IRC | 20:09 | |
| *** maciejjozefczyk has joined #tripleo | 20:09 | |
| *** maciejjozefczyk has quit IRC | 20:12 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-ci master: update file triggers for tls https://review.opendev.org/733392 | 20:12 |
| weshay|ruck | rlandy, ade_lee https://review.opendev.org/733392 | 20:12 |
| rlandy | weshay|ruck: you'll need to merge the reviews that consume the templates | 20:14 |
| weshay|ruck | ? | 20:14 |
| rlandy | https://review.opendev.org/#/q/owner:pojadhav%2540redhat.com+status:open | 20:15 |
| rlandy | https://review.opendev.org/#/c/730321/2/zuul.d/layout.yaml | 20:15 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-ansible master: Prepend timestamp callback output https://review.opendev.org/733394 | 20:17 |
| mwhahaha | ramishra: -^ timestamps for you | 20:17 |
| *** maciejjozefczyk has joined #tripleo | 20:17 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-ci master: Adds new tripleo-ci-buildimage-jobs template for branchless repos https://review.opendev.org/730763 | 20:32 |
| *** aedc has joined #tripleo | 20:34 | |
| openstackgerrit | Merged openstack/tripleo-ansible stable/ussuri: Exclude callbacks env vars for ceph-ansible https://review.opendev.org/732924 | 20:35 |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-ansible stable/train: Exclude callbacks env vars for ceph-ansible https://review.opendev.org/733399 | 20:37 |
| mwhahaha | exclude all the things | 20:38 |
| *** alexmcleod has quit IRC | 20:39 | |
| openstackgerrit | Alex Schultz proposed openstack/tripleo-common stable/train: Switch to tripleo stdout callback https://review.opendev.org/733400 | 20:40 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates master: Deprecate some services https://review.opendev.org/733405 | 20:55 |
| EmilienM | removing all the things | 20:55 |
| *** slaweq has quit IRC | 20:57 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart-extras master: Trigger a job to run a standalone deploy with IPA on multinode https://review.opendev.org/725250 | 21:03 |
| openstackgerrit | Rafael Folco proposed openstack/tripleo-ci master: DNM: Test py3 on c7 https://review.opendev.org/733006 | 21:06 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: Implement a new Container Images layout https://review.opendev.org/722486 | 21:06 |
| weshay|ruck | rlandy, comments on https://review.opendev.org/#/c/731206/ unless I'm misunderstanding | 21:06 |
| EmilienM | mwhahaha, cloudnull ^ i fixed a typo, caught by the CI job: https://512d659ac2ea151243c6-91134e82258f3c2139d09ede1eb5f315.ssl.cf1.rackcdn.com/733390/2/check/tripleo-ci-centos-8-standalone-build-containers/aaeb214/logs/undercloud/home/zuul/container-builds/c06c98a3-4f68-4b0e-a6d1-3af96d7ec446/base/base-build.log | 21:06 |
| mwhahaha | pffft CI doing it's job? unheard of | 21:07 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: container-images: add tripleo-ci-centos-8-standalone-build-containers https://review.opendev.org/733388 | 21:07 |
| EmilienM | weshay|ruck is trying | 21:07 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-common master: DNM - test new standalone job when touching container images https://review.opendev.org/733390 | 21:07 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-ci master: non-voting job for tripleo-build-containers-ubi-8 https://review.opendev.org/724193 | 21:07 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-ci master: trigger multinode-ipa against train https://review.opendev.org/733407 | 21:08 |
| weshay|ruck | EmilienM, ? | 21:09 |
| weshay|ruck | ah | 21:09 |
| weshay|ruck | mwhahaha, tomorrow let me know why we're 100% sure a service is running if the container is running.. not sure I grok that to such certainty | 21:10 |
| mwhahaha | weshay|ruck: because the container fails if the service isn't running | 21:10 |
| * mwhahaha shrugs | 21:11 | |
| rlandy | weshay|ruck: should that be branched train | 21:11 |
| mwhahaha | so it's at least running http, now if horizon is actually error itself that's a nother story but would not be related the service not running | 21:11 |
| mwhahaha | so the service is up, available is not the same thing | 21:12 |
| mwhahaha | weshay|ruck: also, https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_163/731357/1/check/puppet-horizon-tripleo-standalone/1635cbb/logs/undercloud/var/log/containers/httpd/horizon/horizon_access.log | 21:13 |
| mwhahaha | it's responding with a 302/200 | 21:13 |
| mwhahaha | it's 'up' and 'working' | 21:13 |
| *** ccamacho has quit IRC | 21:15 | |
| openstackgerrit | Merged openstack/tripleo-ci master: Stop to use the __future__ module. https://review.opendev.org/732965 | 21:17 |
| *** xek has quit IRC | 21:23 | |
| *** raildo has quit IRC | 21:39 | |
| openstackgerrit | Martin Kopec proposed openstack/ansible-role-collect-logs master: Run container collection in parallel https://review.opendev.org/733165 | 21:44 |
| weshay|ruck | +1 aye | 21:45 |
| *** jfrancoa has quit IRC | 21:48 | |
| *** jmasud has quit IRC | 21:53 | |
| *** holser has quit IRC | 22:05 | |
| *** holser has joined #tripleo | 22:05 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates stable/ussuri: Change the :Z mount flag to :z https://review.opendev.org/732483 | 22:07 |
| openstackgerrit | Merged openstack/puppet-tripleo stable/ussuri: Remove selinux relabel mount option for neutron https://review.opendev.org/733079 | 22:07 |
| openstackgerrit | Merged openstack/puppet-tripleo stable/train: Remove selinux relabel mount option for neutron https://review.opendev.org/733081 | 22:07 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Add options for container builds https://review.opendev.org/733039 | 22:08 |
| *** ekultails has quit IRC | 22:22 | |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Add options for container builds https://review.opendev.org/733039 | 22:23 |
| openstackgerrit | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Add options for container builds https://review.opendev.org/733039 | 22:42 |
| openstackgerrit | Merged openstack/tripleo-ansible master: Stop to use the __future__ module. https://review.opendev.org/732963 | 22:42 |
| *** jmasud has joined #tripleo | 22:43 | |
| *** jmasud has quit IRC | 22:54 | |
| *** jmasud has joined #tripleo | 23:00 | |
| *** rcernin has joined #tripleo | 23:14 | |
| *** gfidente|afk has quit IRC | 23:16 | |
| *** TrevorV has quit IRC | 23:49 | |
| *** tosky has quit IRC | 23:50 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
