| *** cdearborn has quit IRC | 00:28 | |
| *** ricolin has joined #openstack-kolla | 01:14 | |
| *** cah_link1 has joined #openstack-kolla | 01:36 | |
| *** cah_link has quit IRC | 01:38 | |
| *** cah_link1 is now known as cah_link | 01:38 | |
| openstackgerrit | Xing Zhang proposed openstack/kolla-ansible master: Remove task for adding octavia to admin project https://review.opendev.org/720243 | 02:21 |
|---|---|---|
| *** igordc has joined #openstack-kolla | 02:46 | |
| *** igordc has quit IRC | 02:52 | |
| *** igordc has joined #openstack-kolla | 02:52 | |
| *** igordc has quit IRC | 02:57 | |
| *** evrardjp has quit IRC | 04:37 | |
| *** evrardjp has joined #openstack-kolla | 04:37 | |
| *** kevinz has joined #openstack-kolla | 04:42 | |
| *** ykarel|away is now known as ykarel | 04:51 | |
| *** skramaja has joined #openstack-kolla | 04:53 | |
| *** jovial[m] has quit IRC | 05:10 | |
| *** mnasiadka has quit IRC | 05:10 | |
| openstackgerrit | Merged openstack/kolla-ansible master: CI: Set post timeout to half hour https://review.opendev.org/720496 | 05:14 |
| *** jovial[m] has joined #openstack-kolla | 05:15 | |
| *** mnasiadka has joined #openstack-kolla | 05:15 | |
| *** k_mouza has joined #openstack-kolla | 05:19 | |
| *** k_mouza has quit IRC | 05:23 | |
| sri_ | morning | 05:36 |
| *** oklhost has joined #openstack-kolla | 05:39 | |
| oklhost | Hi @all | 05:39 |
| oklhost | I got some issues while deploying monasca with help of kolla-ansible. All containers are up and running, but monasca-api-error.log is full of errors: ImportError: No module named enum. Also after enabling monasca, the central logging seems broken. I would be very grateful to get a tip here. We're planning to migrate a lot of our VMware workloads to OpenStack, but a lot of these monitoring tools not working after deployment. Same for | 05:44 |
| oklhost | cloudkitty, but for now I disabled cloudkitty. | 05:44 |
| *** bline has quit IRC | 05:56 | |
| *** dking has quit IRC | 05:58 | |
| *** nikparasyr has joined #openstack-kolla | 06:18 | |
| *** dougsz has joined #openstack-kolla | 06:48 | |
| *** xinliang has joined #openstack-kolla | 06:55 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 06:55 |
| *** cah_link1 has joined #openstack-kolla | 07:11 | |
| *** cah_link has quit IRC | 07:12 | |
| *** cah_link1 is now known as cah_link | 07:12 | |
| *** jbadiapa has joined #openstack-kolla | 07:19 | |
| *** rpittau|afk is now known as rpittau | 07:19 | |
| *** bengates has joined #openstack-kolla | 07:23 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 07:26 |
| *** bersace has joined #openstack-kolla | 07:28 | |
| *** born2bake has joined #openstack-kolla | 07:29 | |
| *** bengates has quit IRC | 07:34 | |
| *** bengates has joined #openstack-kolla | 07:34 | |
| cosmicsound | connecting to cinder@ceph (conf=/etc/ceph/ceph.conf, timeout=-1). _do_conn /var/lib/kolla/venv/lib/python3.6/site-packages/cinder/volume/drivers/rbd.py:431 | 07:40 |
| cosmicsound | Initialized capabilities list: {'vendor_name': 'Open Source', 'driver_version': '1.2.0', 'storage_protocol': 'ceph', 'total_capacity_gb': 15880.4, 'free_capacity_gb': 15831.53, 'reserved_percentage': 0, 'multiattach': True, 'thin_provisioning_support': True, 'max_over_subscription_ratio': '20.0', 'location_info': | 07:43 |
| cosmicsound | 'ceph:/etc/ceph/ceph.conf:e315504e-ec57-42ae-84dc-24ad28a677a1:cinder:volumes', 'backend_state': 'up', 'volume_backend_name': 'rbd-1', 'replication_enabled': False, 'provisioned_capacity_gb': 320, 'properties': {'thin_provisioning': {'title': 'Thin Provisioning', 'description': 'Sets thin provisioning.', 'type': 'boolean'}, 'compression': {'title': | 07:43 |
| cosmicsound | 'Compression', 'description': 'Enables compression.', 'type': 'boolean'}, 'qos': {'title': 'QoS', 'description': 'Enables QoS.', 'type': 'boolean'}, 'replication_enabled': {'title': 'Replication', 'description': 'Enables replication.', 'type': 'boolean'}}}. init_capabilities | 07:43 |
| cosmicsound | /var/lib/kolla/venv/lib/python3.6/site-packages/cinder/volume/driver.py:791 | 07:43 |
| *** well100 has joined #openstack-kolla | 07:54 | |
| well100 | hello | 07:54 |
| *** threestrands has quit IRC | 07:56 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 08:00 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe master: Docs: Ansible tuning https://review.opendev.org/720542 | 08:01 |
| well100 | you are support bgpvpn? | 08:02 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 08:04 |
| mnasiadka | morning | 08:06 |
| mnasiadka | well100: you mean bagpipe? | 08:06 |
| well100 | mnasiadka: this mean https://docs.openstack.org/networking-bgpvpn/train/user/index.html | 08:07 |
| mnasiadka | well100: I'm asking which driver. | 08:08 |
| well100 | yes bagpipe | 08:08 |
| mnasiadka | I don't see it installed in Kolla Neutron containers - so no support out of the box | 08:09 |
| *** seco has joined #openstack-kolla | 08:09 | |
| mnasiadka | but if you build your own images and add bagpipe with template-override plus override neutron configs in kolla-ansible - it should work | 08:10 |
| hrw | elo | 08:12 |
| hrw | mnasiadka: https://review.opendev.org/#/c/720566/ CI job needs your vote | 08:12 |
| patchbot | patch 720566 - kolla - CI: add kolla-ansible-debian-source-aarch64 job - 2 patch sets | 08:12 |
| *** e0ne has joined #openstack-kolla | 08:13 | |
| hrw | and https://review.opendev.org/#/c/720480/ so there will be working something to test ;d | 08:14 |
| patchbot | patch 720480 - kolla - nova-libvirt: fix after centos 7 removal damage - 1 patch set | 08:14 |
| mnasiadka | hrw: will do - ansible lint long lines needs your vote - https://review.opendev.org/#/c/702429/ :) | 08:18 |
| patchbot | patch 702429 - kolla-ansible - Ansible lint: lines longer than 160 chars - 8 patch sets | 08:18 |
| hrw | will do | 08:19 |
| hrw | approved | 08:20 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 08:23 |
| *** ykarel is now known as ykarel|lunch | 08:26 | |
| oklhost | no one has an idea about my problem? :( | 08:29 |
| *** e0ne_ has joined #openstack-kolla | 08:30 | |
| *** e0ne has quit IRC | 08:30 | |
| cosmicsound | good day | 08:30 |
| mnasiadka | oklhost: this sounds like a bug - please raise one on launchpad.net - please write in the bug what distro you use, what install type (source/binary) and what openstack release (stein/train/whatever) | 08:31 |
| oklhost | thx! | 08:32 |
| oklhost | For me it looks like the apache is configured using wsgi-2.7 but the venv is 3.6 | 08:33 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730698 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] mod_wsgi (pid=29): Target WSGI script '/monasca-api/monasca_api/api/wsgi.py' cannot be loaded as Python module. | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730746 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] mod_wsgi (pid=29): Exception occurred processing WSGI script '/monasca-api/monasca_api/api/wsgi.py'. | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730765 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] Traceback (most recent call last): | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730787 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] File "/monasca-api/monasca_api/api/wsgi.py", line 18, in <module> | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730846 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] File "/var/lib/kolla/venv/lib/python3.6/site-packages/monasca_api/api/server.py", line 21, in <module> | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730893 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] from oslo_config import cfg | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.730900 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] File "/var/lib/kolla/venv/lib/python3.6/site-packages/oslo_config/cfg.py", line 39, in <module> | 08:33 |
| oklhost | [Fri Apr 17 07:31:26.731319 2020] [wsgi:error] [pid 29:tid 139675178346240] [remote 10.38.1.14:35918] import enum | 08:33 |
| hrw | mnasiadka: https://paste.centos.org/view/c39381eb | 08:33 |
| hrw | oklhost: use pastebin.com | 08:33 |
| cosmicsound | anyone seen this on cinder / ceph before? https://mdb.uhlhost.net/uploads/8806154cd23bf430/image.png | 08:33 |
| mnasiadka | oklhost: or paste.openstack.org - or any paste | 08:33 |
| oklhost | Well, sorry | 08:34 |
| mnasiadka | hrw: looks good :) | 08:34 |
| hrw | mnasiadka: run.yml is worse | 08:35 |
| oklhost | http://paste.openstack.org/raw/792300/ | 08:36 |
| mnasiadka | hrw: one thing at a time - I've added E303 to skip list (git used in place of git module) | 08:36 |
| oklhost | I use ubuntu with "source" in globals.yml, downloaded all images from docker hub and pushed it to our local registry | 08:36 |
| mnasiadka | hrw: and I'm thinking if we shouldn't add E305 for run.yml - because that will be a lot of rewriting... | 08:37 |
| hrw | mnasiadka: :) | 08:37 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla-ansible master: Lint CI ansible rules https://review.opendev.org/720664 | 08:38 |
| hrw | small steps | 08:38 |
| hrw | mnasiadka: most of run.yml E305 can probably be command instead of shell | 08:39 |
| hrw | mnasiadka: just need to make sure that scripts are executable | 08:39 |
| hrw | right? | 08:39 |
| mnasiadka | true | 08:40 |
| mnasiadka | mgoddard: rework or skip? ^^ | 08:40 |
| mgoddard | mnasiadka: we don't need to lint our test playbooks | 08:41 |
| cosmicsound | http://paste.openstack.org/show/792302/ | 08:41 |
| mgoddard | at least not right now | 08:41 |
| mgoddard | we've spent enough time linting this cycle :) | 08:42 |
| cosmicsound | this is repeating endlessly | 08:42 |
| cosmicsound | and it seems it has the same request name | 08:42 |
| mnasiadka | mgoddard: makes sense, let me see how to nicely exclude it | 08:43 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: kibana/debian: use upstream package on all architectures https://review.opendev.org/718612 | 08:47 |
| cosmicsound | wheni create a machine the volumes are created. and all works, yet the logs are not positive | 08:48 |
| cosmicsound | http://paste.openstack.org/show/792303/ | 08:51 |
| cosmicsound | Could this be a bug? | 08:51 |
| cosmicsound | it does not make sense one request passes and some periodic repetitive request isnt | 08:52 |
| mnasiadka | cosmicsound: where do you see an error? | 08:52 |
| cosmicsound | 2020-04-17 08:49:47.972 35 DEBUG cinder.volume.drivers.rbd [req-76749fbb-e4dd-4efa-8b04-1784370e2e35 - - - - -] connecting to cinder@ceph (conf=/etc/ceph/ceph.conf, timeout=-1). _do_conn /var/lib/kolla/venv/lib/python3.6/site-packages/cinder/volume/drivers/rbd.py:431 | 08:54 |
| cosmicsound | here | 08:54 |
| cosmicsound | same request repeating and. getting timeout -1? | 08:54 |
| cosmicsound | last lines | 08:54 |
| mnasiadka | no, timeout is set to -1 | 08:54 |
| cosmicsound | ok | 08:55 |
| cosmicsound | so thats all normal | 08:55 |
| mnasiadka | and it's repeating, because it's how it works - it's collecting driver status - better focus on finding logs with ERROR | 08:56 |
| cosmicsound | thanks for the tip | 08:58 |
| cosmicsound | i have none :D | 08:58 |
| cosmicsound | at the moment | 08:58 |
| mnasiadka | cosmicsound: you are the first person to look in the logs when everything works :) | 08:58 |
| cosmicsound | https://mdb.uhlhost.net/uploads/08a46c4a93effa3e/image.png | 08:59 |
| cosmicsound | :D | 08:59 |
| cosmicsound | have some warnings, i mostly need to either check more or open bugs :D | 08:59 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 09:00 |
| *** k_mouza has joined #openstack-kolla | 09:02 | |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible stable/train: Fix service_mapped_to_host filter for common services https://review.opendev.org/720669 | 09:09 |
| *** bornie2bake has joined #openstack-kolla | 09:12 | |
| *** tonythomas has joined #openstack-kolla | 09:17 | |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible master: Remove task for adding octavia to admin project https://review.opendev.org/720243 | 09:24 |
| *** ykarel|lunch is now known as ykarel | 09:25 | |
| openstackgerrit | Merged openstack/kayobe master: Update inventory templates for Ussuri https://review.opendev.org/720521 | 09:28 |
| openstackgerrit | Merged openstack/kolla-ansible master: Ansible lint: lines longer than 160 chars https://review.opendev.org/702429 | 09:29 |
| openstackgerrit | Xing Zhang proposed openstack/kolla-ansible master: Remove task for adding octavia to admin project https://review.opendev.org/720243 | 09:33 |
| openstackgerrit | Xing Zhang proposed openstack/kolla-ansible master: Remove task for adding octavia to admin project https://review.opendev.org/720243 | 09:33 |
| *** k_mouza has quit IRC | 09:47 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: Remove octavia user from admin project https://review.opendev.org/720243 | 09:50 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: Allow OVS bridges to connect directly to interface https://review.opendev.org/705037 | 09:51 |
| *** k_mouza has joined #openstack-kolla | 09:58 | |
| bornie2bake | hi.. maybe its a weird question... but is there any way to make horizon loading faster? :) | 10:02 |
| hrw | bornie2bake: run it of faster machine? | 10:03 |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible stable/train: CentOS 8: CI: Test CentOS 7 to 8 migration https://review.opendev.org/711601 | 10:03 |
| bornie2bake | machine: 64gb ram, 16 cpu's | 10:03 |
| *** numans has quit IRC | 10:08 | |
| *** skramaja_ has joined #openstack-kolla | 10:08 | |
| *** numans has joined #openstack-kolla | 10:10 | |
| *** skramaja has quit IRC | 10:10 | |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: Avoid writing out requirements.txt in kolla-ansible role https://review.opendev.org/707853 | 10:15 |
| *** e0ne_ has quit IRC | 10:21 | |
| *** e0ne has joined #openstack-kolla | 10:22 | |
| *** gfidente|afk is now known as gfidente | 10:22 | |
| mrunge | Hi there, there is still a change for the collectd container up, it makes things easier. could anyone take a quick loog at https://review.opendev.org/#/c/718671 ? | 10:24 |
| patchbot | patch 718671 - kolla (stable/train) - add collectd-pmu, collectd-rtd and collectd-turbos... - 2 patch sets | 10:24 |
| *** xinliang has quit IRC | 10:29 | |
| *** rpittau is now known as rpittau|bbl | 10:30 | |
| *** bornie2bake has quit IRC | 10:31 | |
| hrw | yoctozepto, mgoddard: https://review.opendev.org/#/c/714393/ https://review.opendev.org/#/c/714394/ - ideas for better solutions? | 10:40 |
| patchbot | patch 714393 - kolla-ansible - nova: support dev_mode for API bootstrap - 1 patch set | 10:40 |
| patchbot | patch 714394 - kolla-ansible - nova-cell: for dev_mode it is 'nova' - 1 patch set | 10:40 |
| *** cah_link has quit IRC | 10:41 | |
| *** k_mouza has quit IRC | 10:43 | |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 10:44 |
| openstackgerrit | Merged openstack/kolla-ansible master: Add support for encrypting cinder api. https://review.opendev.org/709663 | 10:45 |
| hrw | yoctozepto: https://paste.centos.org/view/98258cf1 is what you suggested? | 10:45 |
| openstackgerrit | Merged openstack/kolla-ansible master: OVN: Followup changes https://review.opendev.org/720430 | 10:47 |
| *** numans has quit IRC | 10:59 | |
| *** bengates has quit IRC | 10:59 | |
| *** bengates has joined #openstack-kolla | 10:59 | |
| *** bengates has quit IRC | 11:02 | |
| *** bengates has joined #openstack-kolla | 11:03 | |
| *** dannins has quit IRC | 11:03 | |
| *** k_mouza has joined #openstack-kolla | 11:04 | |
| born2bake | manila-related question: I have external ceph deployed via ceph-ansible, mds-cephfs is active. Deployed manila with enable_manila_backend_cephfs_native; https://docs.openstack.org/kolla-ansible/train/reference/storage/manila-guide.html following this guide, creating share using CEPHFS instead of NFS, share is created with error. In the logs: manila-scheduler.log - InvalidParameterValue: You must create a share type in advance, and specify in | 11:05 |
| born2bake | request body or set default_share_type in manila.conf. However, default share type is specified in the first step of the guide - manila type-create default_share_type True | 11:05 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: Improve SSH known host error messages https://review.opendev.org/720687 | 11:08 |
| openstackgerrit | Merged openstack/kayobe stable/train: CentOS 8: seed VM & bifrost https://review.opendev.org/720278 | 11:09 |
| openstackgerrit | Merged openstack/kolla master: nova-libvirt: fix after centos 7 removal damage https://review.opendev.org/720480 | 11:10 |
| openstackgerrit | Merged openstack/kolla master: CI: add kolla-ansible-debian-source-aarch64 job https://review.opendev.org/720566 | 11:10 |
| hrw | down to 17 opened patches ;D | 11:11 |
| *** e0ne has quit IRC | 11:25 | |
| *** e0ne has joined #openstack-kolla | 11:25 | |
| *** dougsz has quit IRC | 11:35 | |
| *** nikparasyr has quit IRC | 11:35 | |
| born2bake | manila-related question: I have external ceph deployed via ceph-ansible, mds-cephfs is active. Deployed manila with enable_manila_backend_cephfs_native; https://docs.openstack.org/kolla-ansible/train/reference/storage/manila-guide.html following this guide, creating share using CEPHFS instead of NFS, share is created with error. In the logs: manila-scheduler.log - InvalidParameterValue: You must create a share type in advance, and specify in | 11:40 |
| born2bake | request body or set default_share_type in manila.conf. However, default share type is specified in the first step of the guide - manila type-create default_share_type True | 11:40 |
| openstackgerrit | Merged openstack/kolla stable/train: Add vitrage-persistor image https://review.opendev.org/719176 | 11:40 |
| openstackgerrit | Merged openstack/kolla stable/stein: Add vitrage-persistor image https://review.opendev.org/719178 | 11:40 |
| *** nikparasyr has joined #openstack-kolla | 11:41 | |
| *** jonaspaulo has joined #openstack-kolla | 11:48 | |
| *** hjensas has quit IRC | 11:52 | |
| *** bengates has quit IRC | 12:04 | |
| *** bengates has joined #openstack-kolla | 12:04 | |
| *** cah_link has joined #openstack-kolla | 12:07 | |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible stable/train: CentOS 8: CI: Test CentOS 7 to 8 migration https://review.opendev.org/711601 | 12:14 |
| *** rpittau|bbl is now known as rpittau | 12:19 | |
| *** hjensas has joined #openstack-kolla | 12:21 | |
| *** e0ne has quit IRC | 12:23 | |
| *** e0ne has joined #openstack-kolla | 12:23 | |
| yoctozepto | hrw: could be, my memory is limited ;-) | 12:32 |
| *** ykarel is now known as ykarel|afk | 12:35 | |
| *** ykarel|afk is now known as ykarel | 12:39 | |
| *** negronjl has quit IRC | 12:42 | |
| *** negronjl has joined #openstack-kolla | 12:43 | |
| yoctozepto | is it just me or stable branches CI is severly butchered? | 12:49 |
| stingrayza | hey folks o/ | 12:50 |
| stingrayza | quick q: the oom-killer on one of my kolla compute nodes nuked a couple of vms - has anyone seen something like this before? surely it shouldn't have been scheduled there if there wasn't enough space to run it? | 12:52 |
| stingrayza | and is there anything I can do to stop it from happening again... | 12:54 |
| openstackgerrit | Will Szumski proposed openstack/kayobe master: Add support for encrypted block devices https://review.opendev.org/720040 | 13:02 |
| openstackgerrit | Merged openstack/kolla-ansible master: Remove octavia user from admin project https://review.opendev.org/720243 | 13:05 |
| *** eliaswimmer has quit IRC | 13:11 | |
| kplant | stingrayza: are you running pinned or shared? | 13:12 |
| stingrayza | kplant: I _think_ pinned (If I understand the question). There's no block storage (not yet running cinder), instances use local storage on each compute node and are generally tied to that compute node for their lifetime. | 13:14 |
| kplant | sorry, i meant pinned or shared cpus :-) | 13:15 |
| stingrayza | np :) I don't know - how would I check? | 13:15 |
| kplant | well, if you haven't configured it - you're using shared vcpus | 13:15 |
| kplant | there's a setting in nova you can set so you can't over subscribe memory | 13:15 |
| kplant | let me find it | 13:16 |
| stingrayza | thanks! | 13:16 |
| kplant | [default] / ram_allocation_ratio | 13:16 |
| kplant | if you set that to <= 1.0, it won't schedule more than the available memory on the machine | 13:17 |
| stingrayza | yup (just stumbled across this) https://docs.openstack.org/nova/latest/admin/cpu-topologies.html | 13:17 |
| kplant | over subscribing vcpus is cool (sometimes), but imo oversubscribing ram isn't :-) | 13:17 |
| kplant | i would also set that ratio to something < 1.0 to leave memory for the os,nova,etc. | 13:18 |
| stingrayza | yeah - that's pretty much the same view as I have... | 13:18 |
| stingrayza | any way to see what the system is using for that value if I haven't explicitly set it? | 13:18 |
| stingrayza | (my old cluster before k-a was set to 1.0 somewhere along the line) | 13:19 |
| kplant | you can use osprofiler to dump the running nova.conf, iirc | 13:19 |
| born2bake | manila-related question: I have external ceph deployed via ceph-ansible, mds-cephfs is active. Deployed manila with enable_manila_backend_cephfs_native; https://docs.openstack.org/kolla-ansible/train/reference/storage/manila-guide.html following this guide, creating share using CEPHFS instead of NFS, share is created with error. In the logs: manila-scheduler.log - InvalidParameterValue: You must create a share type in advance, and specify in | 13:20 |
| born2bake | request body or set default_share_type in manila.conf. However, default share type is specified in the first step of the guide - manila type-create default_share_type True | 13:20 |
| kplant | looks like default for [default]/initial_ram_allocation_ratio is 1.5 | 13:20 |
| stingrayza | ah, gotcha. yeah - that's not going to work for me - our use-case is ram heavy and that's always our first thing to run out of | 13:23 |
| stingrayza | cool. let me set to 1.0 and see how I go from there... | 13:23 |
| stingrayza | thanks, kplant! | 13:23 |
| kplant | yw, gl | 13:24 |
| *** jimcrowleyibm has joined #openstack-kolla | 13:25 | |
| openstackgerrit | Mark Goddard proposed openstack/kolla-ansible stable/train: CentOS 8: CI: Test CentOS 7 to 8 migration https://review.opendev.org/711601 | 13:30 |
| *** ykarel is now known as ykarel|afk | 13:31 | |
| *** cdearborn has joined #openstack-kolla | 13:39 | |
| *** bornie2bake has joined #openstack-kolla | 13:46 | |
| *** averi has joined #openstack-kolla | 13:54 | |
| *** averi has left #openstack-kolla | 13:56 | |
| *** TrevorV has joined #openstack-kolla | 14:15 | |
| *** bengates has quit IRC | 14:27 | |
| *** bengates has joined #openstack-kolla | 14:27 | |
| *** dking has joined #openstack-kolla | 14:28 | |
| *** e0ne has quit IRC | 14:28 | |
| *** jimcrowleyibm has quit IRC | 14:28 | |
| *** jimcrowleyibm has joined #openstack-kolla | 14:30 | |
| *** seco has quit IRC | 14:30 | |
| dking | mgoddard: Would you happen to be around this morning (evening for you)? | 14:32 |
| *** e0ne has joined #openstack-kolla | 14:34 | |
| mgoddard | dking: hi, yes I'm around. What's up? | 14:34 |
| dking | First, I saw your code review. That does seem helpful for me. I think I was also missing the deploy step, but seeing the code from A Universe From Nothing is very helpful. | 14:35 |
| dking | I have a few questions at the moment before the weekend, though. One of them is that I noticed that in https://docs.openstack.org/kayobe/latest/configuration/network.html#configuring-network-roles, should it instead be external_net_names as a list, rather than external_net_name as a string? The inline comments say that the later is deprecated. It would work as it is, but I was wondering if external_net_names might be | 14:37 |
| *** skramaja has joined #openstack-kolla | 14:39 | |
| mgoddard | dking: external_net_names is the preferred approach | 14:40 |
| mgoddard | I'll update the docs | 14:40 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe stable/train: Fix ntp and chrony in mixed CentOS 7 and 8 environments https://review.opendev.org/720731 | 14:40 |
| dking | Thanks. That's a small thing. I just missed the difference in the name, and it caused me to some cconfusion. | 14:40 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: Remove unused rgw_port ceph variable https://review.opendev.org/720732 | 14:42 |
| mgoddard | dking: thanks for raising it | 14:42 |
| dking | Currently, my issue is that I'm getting that error about "The Python 2 bindings for rpm are needed for this module." So, I'm assuming that somewhere, maybe something is trying to use Python 3? I know this is probably more of an Ansible question, but I'm having trouble locating where that is. | 14:42 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe stable/train: Prevent openrc files from using wrong OS_CACERT value https://review.opendev.org/720733 | 14:43 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: Docs: fix reference to deprecated external_net_name https://review.opendev.org/720734 | 14:43 |
| *** yevi has joined #openstack-kolla | 14:43 | |
| mgoddard | dking: which version/branch are you using? | 14:44 |
| mgoddard | and is it CentOS 7 or 8? | 14:44 |
| dking | http://paste.openstack.org/show/792333/ | 14:44 |
| dking | CentOS 7 on both the Ansible host and the seed host. | 14:44 |
| mgoddard | and which version/branch of kayobe? | 14:45 |
| *** skramaja_ has quit IRC | 14:46 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: [DNM] OVN IPv6 https://review.opendev.org/720735 | 14:46 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: [DNM] OVN IPv6 https://review.opendev.org/720735 | 14:47 |
| dking | That may be a good question. I was just looking at that I've been using Train, but I don't know that I specified anything on this particular attempt. This time, I just installed with "pip install kayobe". | 14:47 |
| openstackgerrit | Pierre Riteau proposed openstack/kayobe master: Use upper constraints when installing or updating packages https://review.opendev.org/720736 | 14:47 |
| mgoddard | dking: you should be able to run kayobe --version | 14:48 |
| dking | kayobe 7.0.0 | 14:49 |
| mgoddard | dking: you shouldn't need the arguments: | 14:52 |
| mgoddard | -i $KAYOBE_CONFIG_PATH/inventory --kolla-config-path $KOLLA_CONFIG_PATH --kolla-venv $base_path/venvs/kolla-ansible | 14:52 |
| mgoddard | especially if you sourced kayobe-config/kayobe-env | 14:53 |
| dking | I'm no longer using A Universe From Nothing. I'm just using it as a starting point. I was having an error with virtual seed host, and so I moved back to just using Kayobe. | 14:55 |
| *** bornie2bake has quit IRC | 14:55 | |
| dking | I've just been reading its configs to get an idea of a sane set of configs to start with, and looking at the scripts to get an idea of the right way to deploy. | 14:55 |
| dking | Whatever happened, it must have been my mistake. I just tried running "kayobe control host bootstrap" again, and I got the same error, which I didn't get before. I don't recall making any changes to the python version, though. I just had to pip install selinux. | 14:58 |
| *** seco has joined #openstack-kolla | 14:59 | |
| mgoddard | dking: do you have 'localhost ansible_connection=local' in the inventory? | 14:59 |
| dking | Yes, it's there in the hosts file in the inventory. | 15:00 |
| mgoddard | dking: ok, I'm going to try running through the early install on CentOS 7 | 15:01 |
| *** bengates has quit IRC | 15:01 | |
| *** bengates_ has joined #openstack-kolla | 15:01 | |
| dking | Okay. I may have to try making a new venv and try again. | 15:01 |
| *** seco has quit IRC | 15:03 | |
| *** skramaja has quit IRC | 15:05 | |
| *** ykarel|afk is now known as ykarel | 15:06 | |
| dking | Is there a way to start fresh with Kayobe? | 15:07 |
| dking | Other than Re-OSing the Ansible host? | 15:08 |
| *** bline has joined #openstack-kolla | 15:08 | |
| mgoddard | dking: yeah apart from a few packages everything should be wipeable by removing $base_dir | 15:09 |
| kplant | hey mgoddard - anything on the roadmap to add support for https://docs.openstack.org/adjutant/latest/index.html in kolla/k-a ? | 15:10 |
| kplant | i might take a stab at building a container for it if not | 15:10 |
| mgoddard | kplant: not that I'm aware of. Bear in mind that it had some questions around it during the PTL election: https://etherpad.opendev.org/p/victoria-leaderless | 15:14 |
| mgoddard | kplant: eventually the main lead Adrian remembered to step up, but it highlighted that the project has not yet had an official release | 15:14 |
| mgoddard | dking: I was able to run the following in a CentOS 7.7 VM: http://paste.openstack.org/show/792337/ | 15:17 |
| mgoddard | dking: that may give you something to compare with | 15:18 |
| *** ricolin has quit IRC | 15:19 | |
| openstackgerrit | Merged openstack/kolla stable/train: add collectd-pmu, collectd-rtd and collectd-turbostat to C8 https://review.opendev.org/718671 | 15:19 |
| dking | Thanks. I'm sure I mucked something up myself, probably when I was trying to handle the selinux issue. | 15:20 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla master: Cleanup py27 support https://review.opendev.org/720751 | 15:20 |
| dking | mgoddard: I see that you run "kayobe playbook run venvs/kayobe/share/kayobe/ansible/ip-allocation.yml". So, that's before kayobe seed host configure? | 15:20 |
| mgoddard | dking: it was just a shortcut to run the playbook that failed for you. It's part of the seed host configure command | 15:21 |
| dking | or is that just how to run that one play from "kayobe seed host configure" that I'm having the current trouble executing? | 15:21 |
| dking | Okay. | 15:21 |
| dking | It's good to know that I can do that to run individual parts. | 15:22 |
| dking | Also, https://github.com/openstack/kayobe-config-dev is a different repo than what I was using. I was using the one without "-dev". Is kayobe-config-dev recommended? | 15:23 |
| kplant | mgoddard: i had no idea, i guess i thought the project was further along than it is. i don't want to make work for other people, especially when it's something uncertain | 15:24 |
| mgoddard | dking: that's our testing/development all-in-one config. We use it in CI and local development | 15:25 |
| mgoddard | dking: kayobe-config is the raw unconfigured config, you should use that | 15:25 |
| dking | Okay. I might take a look at it, though, especially if it's what's used in CI. | 15:26 |
| mgoddard | kplant: I don't think it's dead, but I'd suggest doing some research if you think it might be useful for you | 15:26 |
| mgoddard | dking: yeah, might be useful. It's not dissimilar to the a-u-f-n config, just for a single host | 15:26 |
| dking | And I have another question for in the future. I haven't made it that far yet, but I noticed that bifrost runs in a container on the seed host. Once things are up, I'll want to be able to poll the inventory remotely. Would I just be able to expose the port of the running container so that I could make ironic requests? | 15:27 |
| openstackgerrit | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Add ansible-lint to tox https://review.opendev.org/694779 | 15:28 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: Cleanup py27 support https://review.opendev.org/720752 | 15:30 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: Cleanup py27 support https://review.opendev.org/720752 | 15:35 |
| openstackgerrit | Qitao proposed openstack/kolla-ansible master: Add vitrage-persistor service https://review.opendev.org/715364 | 15:35 |
| mgoddard | dking: the container uses host networking, so you can access the APIs | 15:40 |
| kplant | mgoddard: yeah i'm still going to pursue it personally, just want to respect other peoples time before i bother them about getting this into k/ka :-) | 15:40 |
| dking | That's even better! Thanks for your help. | 15:40 |
| *** sean-k-mooney has quit IRC | 15:41 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-cli master: Cleanup py27 support https://review.opendev.org/720754 | 15:42 |
| dking | A quick question, though. I had tried, very briefly, to setup bifrost previously, using instructions that I think were part of a quickstart using kolla-ansible. However, I seem to recall that the image was broken, and was told that it had something to do with changes to Python 2 -> 3. I'm assuming there are checks done for CI for Kayobe that make sure that whatever bifrost image or source is being used is working corr | 15:44 |
| mgoddard | dking: there was a lot of churn while we added support for python 3 & centos 8, but it's now pretty stable. We test a seed deploy in CI | 15:45 |
| dking | Okay. Thank you very much. I feel a bit better knowing that was addressed. | 15:47 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe stable/train: Fix ntp and chrony in mixed CentOS 7 and 8 environments https://review.opendev.org/720731 | 15:55 |
| *** ykarel is now known as ykarel|away | 15:57 | |
| openstackgerrit | Mark Goddard proposed openstack/kayobe stable/train: Fix ntp and chrony in mixed CentOS 7 and 8 environments https://review.opendev.org/720731 | 15:59 |
| openstackgerrit | Radosław Piliszek proposed openstack/kayobe master: Cleanup py27 support https://review.opendev.org/720764 | 16:04 |
| *** e0ne has quit IRC | 16:04 | |
| *** e0ne_ has joined #openstack-kolla | 16:04 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-cli master: Cleanup py27 support https://review.opendev.org/720754 | 16:05 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-cli master: Cleanup py27 support https://review.opendev.org/720754 | 16:06 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: Cleanup py27 support https://review.opendev.org/720752 | 16:07 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla master: Cleanup py27 support https://review.opendev.org/720751 | 16:07 |
| *** rpittau is now known as rpittau|afk | 16:17 | |
| *** gfidente is now known as gfidente|afk | 16:20 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kayobe master: Cleanup py27 support https://review.opendev.org/720764 | 16:26 |
| *** negronjl has quit IRC | 16:29 | |
| *** hamzy__ has joined #openstack-kolla | 16:31 | |
| *** hamzy_ has quit IRC | 16:31 | |
| *** negronjl has joined #openstack-kolla | 16:33 | |
| *** evrardjp has quit IRC | 16:37 | |
| *** bengates_ has quit IRC | 16:37 | |
| *** evrardjp has joined #openstack-kolla | 16:37 | |
| *** bengates has joined #openstack-kolla | 16:38 | |
| *** kevinz has quit IRC | 16:40 | |
| sri_ | cosmicsound: hi are you able solve the the rgw keystone integration ? | 16:43 |
| *** bengates has quit IRC | 16:44 | |
| sri_ | cosmicsound: I've created new lab today, I was able to create swift endpoints and it's working, let me know i can give the my notes | 16:45 |
| *** gfidente|afk has quit IRC | 16:46 | |
| *** igordc has joined #openstack-kolla | 17:04 | |
| *** igordc has quit IRC | 17:09 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-cli master: Cleanup py27 support https://review.opendev.org/720754 | 17:10 |
| yoctozepto | sri_: you can share this even in a bug report so that all benefit | 17:12 |
| sri_ | yoctozepto: sure :) http://paste.openstack.org/show/792344/ | 17:15 |
| *** k_mouza has quit IRC | 17:16 | |
| *** nikparasyr has quit IRC | 17:18 | |
| yoctozepto | sri_: thanks! | 17:21 |
| sri_ | :) | 17:29 |
| born2bake | sri_ did you have any autharization issues? My setup is pretty much similar to yours and its still not working for me :/ | 17:33 |
| born2bake | When I open object store in horizon, getting kicked. when i am trying to use openstack container list as an admin - 401 unauthorized error :/ | 17:34 |
| sri_ | no, did you restart the ceph mon and rgw after updating config ? | 17:34 |
| sri_ | born2bake: can you see any error messages in rgw logs ? | 17:36 |
| born2bake | 2020-04-17T19:08:23.924+0300 7f76f3f9f700 1 op->ERRORHANDLER: err_no=-1 new_err_no=-1 | 17:37 |
| born2bake | 2020-04-17T19:08:23.924+0300 7f76f3f9f700 1 ====== req done req=0x7f7808510870 op status=0 http_status=401 latency=0s ====== | 17:37 |
| *** e0ne_ has quit IRC | 17:38 | |
| *** e0ne has joined #openstack-kolla | 17:38 | |
| sri_ | born2bake: my setup is string forward Installed the ceph with ceph-ansible, then installed the openstack with k-a then updated the ceph.conf in rgw section, restarted the ceph rgw and mon. then create the swift endpoints | 17:45 |
| born2bake | so did I :) how did you restart ceph rgw and mon on the node? | 17:46 |
| sri_ | born2bake: systemctl restart ceph-mon@cloudnodesite1.service, systemctl restart ceph-radosgw@rgw.cloudnodesite1.rgw0.service | 17:47 |
| *** e0ne_ has joined #openstack-kolla | 17:49 | |
| *** e0ne has quit IRC | 17:49 | |
| sri_ | born2bake: FYI this is all-in-one setup, ceph and openstack running in the same server | 17:49 |
| born2bake | same for me :) | 17:49 |
| yevi | For me, with Train release, I use this path `/swift/v1/AUTH_%\(project_id\)s` for rgw keystone integration | 17:51 |
| *** dciabrin has quit IRC | 18:00 | |
| cosmicsound | Sri_ my integration still fails for now | 18:42 |
| cosmicsound | pm details | 18:42 |
| cosmicsound | anyone here running working reverse dns with designate? | 18:42 |
| cosmicsound | i have some issues gettings the right pools details in place. Think there might be some issues with the built in template or is just me done something wrong. | 18:43 |
| sri_ | cosmicsound: hmm, I've done this with stein and train both working, how did rgw keyring was created ? | 18:47 |
| cosmicsound | At initial command to give permissions | 18:49 |
| *** dciabrin has joined #openstack-kolla | 18:52 | |
| sri_ | cosmicsound: I did not give any permissions, installed ceph using c-a, that's it | 18:56 |
| *** e0ne_ has quit IRC | 18:56 | |
| *** e0ne has joined #openstack-kolla | 18:56 | |
| cosmicsound | Right | 18:57 |
| cosmicsound | well something is wrong surely | 18:57 |
| cosmicsound | Everything works except swift | 18:58 |
| cosmicsound | i think my keystone integration or something is broken | 18:58 |
| cosmicsound | sri_ can you share your ceph.conf part with rgw settings? | 18:59 |
| sri_ | cosmicsound: http://paste.openstack.org/show/792344/ | 19:00 |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: WIP: Revove support for CentOS 7 and Python 2 https://review.opendev.org/720818 | 19:02 |
| *** dougsz has joined #openstack-kolla | 19:15 | |
| *** cah_link1 has joined #openstack-kolla | 19:19 | |
| *** cah_link has quit IRC | 19:20 | |
| *** cah_link1 is now known as cah_link | 19:20 | |
| *** e0ne has quit IRC | 19:26 | |
| *** jimcrowleyibm has quit IRC | 19:48 | |
| *** jimcrowleyibm has joined #openstack-kolla | 19:48 | |
| *** cah_link has quit IRC | 19:51 | |
| openstackgerrit | Merged openstack/kolla stable/train: Fix that the cyborg database init failed. https://review.opendev.org/720370 | 19:57 |
| openstackgerrit | Merged openstack/kolla-ansible master: Remove unused rgw_port ceph variable https://review.opendev.org/720732 | 19:57 |
| openstackgerrit | Merged openstack/kolla-ansible master: Add vitrage-persistor service https://review.opendev.org/715364 | 19:57 |
| *** TrevorV has quit IRC | 20:04 | |
| *** dougsz has quit IRC | 20:33 | |
| *** dougsz has joined #openstack-kolla | 20:34 | |
| *** dougsz has quit IRC | 20:38 | |
| openstackgerrit | Mark Goddard proposed openstack/kayobe master: WIP: Revove support for CentOS 7 and Python 2 https://review.opendev.org/720818 | 21:12 |
| *** jimcrowleyibm has quit IRC | 21:28 | |
| yevi | Curiosity question, I am new to Kolla-ansible, but not new to openstack. After doing a couple installs I strated to dig around the configs and noticed that you still have a value for using port 35357 for auth_url which I thought was depricated in queens after switching to identity v3 for the auth_uri over port 5000. Is it still there for backwards compatabilty? | 21:33 |
| *** tonythomas has quit IRC | 21:56 | |
| *** jbadiapa has quit IRC | 21:57 | |
| born2bake | sri_ thanks a lot! swift worked lol :) | 22:15 |
| sri_ | born2bake: glad to hear that 👍 | 22:29 |
| *** igordc has joined #openstack-kolla | 23:08 | |
| openstackgerrit | James Kirsch proposed openstack/kolla master: Add stunnel container for Glance service https://review.opendev.org/720850 | 23:24 |
| *** igordc has quit IRC | 23:27 | |
| *** born2bake has quit IRC | 23:43 | |
| openstackgerrit | James Kirsch proposed openstack/kolla-ansible master: Add support for encrypting glance/heat api https://review.opendev.org/707131 | 23:45 |
| openstackgerrit | James Kirsch proposed openstack/kolla-ansible master: Add support for encrypting glance/heat api https://review.opendev.org/707131 | 23:49 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
