| strgout | howdy.. | 00:15 |
|---|---|---|
| strgout | ceph .. when i deploy with ceph what globals options enable creating volumes for VM? | 00:18 |
| strgout | you know.. a better question would be where are are global options documented | 00:20 |
| logan- | strgout: the best place to look for ceph-ansible options is https://github.com/ceph/ceph-ansible/tree/master/group_vars | 00:21 |
| logan- | with regard to the openstack integration, there is a doc here with example configs: https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-prod-ceph.html | 00:22 |
| logan- | look for "cinder_backends" at the bottom | 00:23 |
| *** SmearedBeard has quit IRC | 00:31 | |
| *** andymccr has quit IRC | 00:36 | |
| *** _nyloc_ has joined #openstack-ansible | 00:36 | |
| *** nyloc has quit IRC | 00:36 | |
| *** chyka has quit IRC | 00:40 | |
| *** andymccr has joined #openstack-ansible | 00:40 | |
| *** markvoelker has joined #openstack-ansible | 00:43 | |
| *** markvoelker has quit IRC | 00:45 | |
| *** markvoelker has joined #openstack-ansible | 00:49 | |
| *** markvoelker has quit IRC | 00:50 | |
| *** andymccr has quit IRC | 00:51 | |
| *** andymccr has joined #openstack-ansible | 00:55 | |
| *** kaiokassiano has joined #openstack-ansible | 00:58 | |
| *** dave-mccowan has joined #openstack-ansible | 01:07 | |
| *** markvoelker has joined #openstack-ansible | 01:08 | |
| strgout | hmm so do i need all the options in cinder_backends? | 01:08 |
| *** markvoelker has quit IRC | 01:09 | |
| *** rmcall has joined #openstack-ansible | 01:10 | |
| *** dave-mccowan has quit IRC | 01:42 | |
| *** kaiokassiano has quit IRC | 01:45 | |
| *** gfa has joined #openstack-ansible | 01:50 | |
| *** jafeha__ has joined #openstack-ansible | 01:52 | |
| *** jafeha has quit IRC | 01:54 | |
| *** pbandark has quit IRC | 02:10 | |
| *** chyka has joined #openstack-ansible | 02:12 | |
| strgout | oh i see how it is.... so if I really want volumes I have to enable cinder. pfft. | 02:12 |
| strgout | so one thing i'm unsure about. If i enable a new module (in this case cinder) can i just flip to yes then deploy or do i need to destroy then deploy again? | 02:15 |
| *** dave-mccowan has joined #openstack-ansible | 02:15 | |
| *** chyka has quit IRC | 02:17 | |
| *** rmcall has quit IRC | 02:21 | |
| *** pbandark has joined #openstack-ansible | 02:28 | |
| *** pmannidi has quit IRC | 02:44 | |
| *** dave-mccowan has quit IRC | 02:47 | |
| *** pmannidi has joined #openstack-ansible | 02:48 | |
| *** lbragstad has quit IRC | 02:54 | |
| *** strgout has quit IRC | 03:10 | |
| *** pbandark has quit IRC | 03:16 | |
| cloudnull | evenings | 03:17 |
| cloudnull | strgout if you want to enable a new service you simply add the required config and run the playbooks. | 03:21 |
| *** udesale has joined #openstack-ansible | 03:59 | |
| *** chhavi__ has joined #openstack-ansible | 04:25 | |
| masber | good afternoon, I am trying to setup my first OSA deployment | 04:33 |
| masber | do I need br-mgmt on the compute nodes? | 04:33 |
| pradiprwt | masber: yes, you need br-mgmt in every node | 04:56 |
| masber | thank you! | 04:57 |
| pradiprwt | this interface is used for container management | 04:57 |
| masber | and also bind the external VIP? | 04:57 |
| pradiprwt | yeah, you are right | 04:58 |
| pradiprwt | internal VIP as well | 04:58 |
| masber | but not all services runs on LXC right? nova and cinder and bare metal | 04:58 |
| pradiprwt | yes, but ansible needs to deploy the services on compute and storage also, so it is using br-mgmt for that | 05:00 |
| pradiprwt | I think you can keep IP to interface | 05:01 |
| masber | sorry this is my first OSA installation, I went throught the documentation but I have the feeling taht I need help, I tried before packstack and kolla-ansible and they all configures the underlines network in different way plus I am not an expert in OVS or linux bridges etc... | 05:01 |
| masber | pradiprwt, one question, external VIP is same as the external_lb_vip_address and you said I need to set it up on the compute nodes, then are we talking about a DVR installation? | 05:08 |
| *** poopcat1 has quit IRC | 05:09 | |
| *** brad[] has quit IRC | 05:10 | |
| pradiprwt | masber : Sorry I did't get you.. I was saying about management network for compute host... | 05:17 |
| pradiprwt | you can assign any network IP to external_lb_vip_address, it should be accessible from infra nodes | 05:19 |
| masber | so I put the br-mgmt on the compute node, and then do I also need to bind the external VIP into the br-mgmt:0? | 05:20 |
| pradiprwt | No need to bind external VIP with compute node, just provide as a variable to external_lb_vip_address in openstack_user_config.yml file, OSA will bind it with HAproxy | 05:25 |
| *** nshetty has joined #openstack-ansible | 05:26 | |
| pradiprwt | masber: the steps will be 1) prepare the deployment host which is your ansible node 2) Prepare the target host which are infra, compute and storage nodes then apply the playbooks | 05:28 |
| pradiprwt | make sure networks are configured properly | 05:28 |
| *** chyka has joined #openstack-ansible | 05:29 | |
| pradiprwt | for configuration you can take reference of given example https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-test.html#test-environment-config | 05:29 |
| *** admin0 has quit IRC | 05:33 | |
| masber | hahaha ok! I will try | 05:33 |
| *** chyka has quit IRC | 05:33 | |
| *** nshetty is now known as nshetty|afk | 05:35 | |
| *** woodard has quit IRC | 05:37 | |
| *** markvoelker has joined #openstack-ansible | 05:39 | |
| *** bhujay has joined #openstack-ansible | 05:42 | |
| *** admin0 has joined #openstack-ansible | 05:47 | |
| *** nshetty|afk is now known as nshetty | 05:59 | |
| *** jbadiapa has quit IRC | 06:02 | |
| *** oanson has joined #openstack-ansible | 06:06 | |
| *** oanson has quit IRC | 06:18 | |
| *** oanson has joined #openstack-ansible | 06:22 | |
| *** jbadiapa has joined #openstack-ansible | 06:26 | |
| *** pcaruana has joined #openstack-ansible | 06:27 | |
| *** jbadiapa has quit IRC | 06:30 | |
| *** pcaruana has quit IRC | 06:33 | |
| *** pcaruana has joined #openstack-ansible | 06:34 | |
| *** udesale__ has joined #openstack-ansible | 06:57 | |
| *** udesale has quit IRC | 06:59 | |
| *** arbrandes1 has joined #openstack-ansible | 07:01 | |
| *** arbrandes has quit IRC | 07:03 | |
| *** sar has joined #openstack-ansible | 07:06 | |
| *** sar is now known as sbr | 07:06 | |
| *** sbr is now known as sbra | 07:06 | |
| *** masber has quit IRC | 07:08 | |
| *** jbadiapa has joined #openstack-ansible | 07:18 | |
| *** udesale has joined #openstack-ansible | 07:19 | |
| *** udesale has quit IRC | 07:20 | |
| *** pmannidi has quit IRC | 07:21 | |
| *** udesale__ has quit IRC | 07:21 | |
| *** SmearedBeard has joined #openstack-ansible | 07:27 | |
| *** armaan has joined #openstack-ansible | 07:30 | |
| *** SmearedBeard has quit IRC | 07:32 | |
| *** sbra has quit IRC | 07:34 | |
| *** threestrands has quit IRC | 07:36 | |
| *** mardim has quit IRC | 07:38 | |
| *** sbra has joined #openstack-ansible | 07:45 | |
| *** mardim has joined #openstack-ansible | 07:51 | |
| *** _nyloc_ is now known as nyloc | 07:52 | |
| *** rgogunskiy has joined #openstack-ansible | 07:57 | |
| *** mbuil has joined #openstack-ansible | 08:04 | |
| *** gkadam has joined #openstack-ansible | 08:04 | |
| *** shardy has joined #openstack-ansible | 08:09 | |
| *** ianychoi has quit IRC | 08:35 | |
| *** mbuil has quit IRC | 08:35 | |
| *** mbuil has joined #openstack-ansible | 08:35 | |
| *** epalper has joined #openstack-ansible | 09:04 | |
| *** markvoelker has quit IRC | 09:08 | |
| *** sxc731 has joined #openstack-ansible | 09:11 | |
| *** pbandark has joined #openstack-ansible | 09:19 | |
| *** SmearedBeard has joined #openstack-ansible | 09:32 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone master: Use a venv with previous ansible version for upgrades https://review.openstack.org/532470 | 09:35 |
| *** cshen has quit IRC | 09:35 | |
| *** nshetty has quit IRC | 09:56 | |
| *** nshetty has joined #openstack-ansible | 09:56 | |
| *** cshen has joined #openstack-ansible | 09:56 | |
| *** nshetty_ has joined #openstack-ansible | 10:03 | |
| *** nshetty has quit IRC | 10:04 | |
| *** nshetty_ is now known as nshetty | 10:05 | |
| *** threestrands has joined #openstack-ansible | 10:21 | |
| *** stuartgr has joined #openstack-ansible | 10:36 | |
| *** rpittau has joined #openstack-ansible | 10:54 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone master: DNM - test to see if linters are working https://review.openstack.org/532489 | 10:56 |
| *** rgogunskiy has quit IRC | 10:57 | |
| pradiprwt | Hi Everyone, I have integrated the ceph with OSA, everything is working fine but I am not able to delete images.. any suggestion for this issue >>Traceback >>http://paste.openstack.org/show/642228/ | 11:02 |
| *** Smeared_Beard has joined #openstack-ansible | 11:02 | |
| *** SmearedBeard has quit IRC | 11:05 | |
| *** sbra has quit IRC | 11:06 | |
| *** markvoelker has joined #openstack-ansible | 11:08 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_cinder master: tasks: cinder_qos: Use exact match for QoS rules https://review.openstack.org/532494 | 11:18 |
| openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible-os_neutron master: add ml2 config for networking bgpvpn https://review.openstack.org/522598 | 11:31 |
| *** armaan has quit IRC | 11:33 | |
| *** armaan has joined #openstack-ansible | 11:33 | |
| openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible-os_neutron master: add ml2 config for networking bgpvpn https://review.openstack.org/522598 | 11:33 |
| *** armaan has quit IRC | 11:34 | |
| *** armaan has joined #openstack-ansible | 11:34 | |
| *** chhavi__ has quit IRC | 11:35 | |
| *** markvoelker has quit IRC | 11:42 | |
| *** ThomasSch has joined #openstack-ansible | 11:50 | |
| ThomasSch | hi | 11:50 |
| ThomasSch | i know it is not OSA specific but does anyone know where to configure the action triggered by horizon when you click migrate host? | 11:50 |
| ThomasSch | i would love to have it trigger host-evacuate-live | 11:51 |
| *** drifterza has joined #openstack-ansible | 11:53 | |
| *** shardy is now known as shardy_lunch | 12:03 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_cinder master: tasks: cinder_qos: Use exact match for QoS rules https://review.openstack.org/532494 | 12:18 |
| *** rgogunskiy has joined #openstack-ansible | 12:20 | |
| *** smatzek has joined #openstack-ansible | 12:20 | |
| *** brad[] has joined #openstack-ansible | 12:29 | |
| *** lihi has joined #openstack-ansible | 12:29 | |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 12:36 |
| *** markvoelker has joined #openstack-ansible | 12:39 | |
| bhujay | pradiprwt: home on the glance pool permission is proper . Some hint for the permission - { name: client.glance, key: "$(ceph-authtool --gen-print-key)", mon_cap: "allow r", osd_cap: "allow class-read object_prefix rbd_children, allow rwx pool={{ openstack_glance_pool.name }}", mode: "0600", acls: [] } | 12:42 |
| bhujay | *hope | 12:43 |
| bhujay | pradiprwt: connect from shell using glance key and try creating and deleting some rbd image with glance pool to see if the issue is with ceph or openstack | 12:44 |
| *** jwitko_ has quit IRC | 12:46 | |
| *** shardy_lunch is now known as shardy | 12:47 | |
| mhayden | good morning, folks | 12:52 |
| evrardjp | morning mhayden | 12:56 |
| *** prasen has joined #openstack-ansible | 12:56 | |
| *** yifei has quit IRC | 12:59 | |
| *** yifei has joined #openstack-ansible | 13:00 | |
| *** DELL has joined #openstack-ansible | 13:03 | |
| *** DELL is now known as Guest69945 | 13:04 | |
| *** Guest69945 has quit IRC | 13:06 | |
| *** Sadipan has joined #openstack-ansible | 13:09 | |
| openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible-os_neutron master: add ml2 config for networking bgpvpn https://review.openstack.org/522598 | 13:09 |
| *** smatzek has quit IRC | 13:09 | |
| * mhayden is deploying master on his centos 7 environment | 13:10 | |
| *** markvoelker has quit IRC | 13:12 | |
| *** ivveh has joined #openstack-ansible | 13:13 | |
| *** smatzek has joined #openstack-ansible | 13:14 | |
| *** drifterza has quit IRC | 13:14 | |
| odyssey4me | mhayden https://media.giphy.com/media/kYkQYXkO3XyRa/giphy.gif | 13:17 |
| mhayden | haha | 13:18 |
| odyssey4me | mhayden you have seen that all the upgrade jobs are failing, right? | 13:23 |
| *** yifei1 has joined #openstack-ansible | 13:24 | |
| *** yifei has quit IRC | 13:26 | |
| *** yifei1 is now known as yifei | 13:26 | |
| *** woodard has joined #openstack-ansible | 13:26 | |
| *** shardy is now known as shardy_mtg | 13:28 | |
| *** sxc731 has quit IRC | 13:30 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Fix 'status' typo https://review.openstack.org/532526 | 13:30 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts stable/pike: Allow NetworkManager to work (if present) https://review.openstack.org/532201 | 13:31 |
| mhayden | evrardjp / odyssey4me: if y'all could help me tidy up my mess, i'd be much obliged https://review.openstack.org/#/c/532526/ | 13:31 |
| *** woodard has quit IRC | 13:32 | |
| *** Sadipan has quit IRC | 13:32 | |
| *** woodard has joined #openstack-ansible | 13:33 | |
| odyssey4me | oh dear :/ done | 13:33 |
| *** bhujay has quit IRC | 13:33 | |
| mhayden | thanks sir | 13:34 |
| *** hw_wutianwei has joined #openstack-ansible | 13:35 | |
| *** sbra has joined #openstack-ansible | 13:35 | |
| *** prasen has quit IRC | 13:36 | |
| mhayden | oof i botched it again | 13:37 |
| mhayden | i knew quitting coffee was a bad idea | 13:37 |
| evrardjp | mhayden: you remember what I said about FOCUS? :p | 13:37 |
| *** SerenaFeng has joined #openstack-ansible | 13:38 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Fix typo for wait-online service https://review.openstack.org/532526 | 13:38 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts stable/pike: Allow NetworkManager to work (if present) https://review.openstack.org/532201 | 13:39 |
| mhayden | i guess i'm glad i'm deploying master ;) | 13:39 |
| *** prasen has joined #openstack-ansible | 13:39 | |
| evrardjp | What would be cool, is that we stop introducing features this week, and just fix the whole series of bug we have. | 13:40 |
| evrardjp | like the upgrade ones. | 13:40 |
| evrardjp | :) | 13:40 |
| evrardjp | Just saying! | 13:40 |
| mhayden | haha | 13:40 |
| *** SerenaFeng has quit IRC | 13:40 | |
| mhayden | well i am doing greenfield testing of master on centos this morning in the background while doing other things | 13:40 |
| mhayden | do we plan to address the ansible 2.4 deprecations around "import_tasks/include_tasks" in this cycle? | 13:42 |
| mhayden | not sure if i missed a discussion there | 13:43 |
| *** prasen has quit IRC | 13:46 | |
| evrardjp | we will but that's a minor bug, right now we have so much bigger on our plate :) | 13:47 |
| evrardjp | like real ansible 2.4 regressions and stuff! | 13:47 |
| evrardjp | (working on it) | 13:47 |
| evrardjp | and our inventory issues :/ | 13:47 |
| evrardjp | and our upgrade issues :/ | 13:47 |
| evrardjp | (working on it too) | 13:47 |
| evrardjp | and the opened bugs! | 13:47 |
| evrardjp | I think there is one assigned to you with sudo | 13:48 |
| evrardjp | long story short: I am sure we can find you places to help if you want :D | 13:48 |
| *** shardy_mtg is now known as shardy | 13:59 | |
| *** drifterza has joined #openstack-ansible | 14:00 | |
| *** armaan has quit IRC | 14:00 | |
| *** armaan has joined #openstack-ansible | 14:00 | |
| *** hw_wutianwei has quit IRC | 14:01 | |
| *** sxc731 has joined #openstack-ansible | 14:04 | |
| *** hw_wutianwei has joined #openstack-ansible | 14:08 | |
| *** markvoelker has joined #openstack-ansible | 14:10 | |
| *** cmart has joined #openstack-ansible | 14:11 | |
| cmart | Good morning OSA. Looks like Ubuntu just released kernel patches for the "Meltdown" vulnerability. Has anyone applied these to their deployments yet, and if so, run into any issues? | 14:12 |
| *** jwitko_ has joined #openstack-ansible | 14:19 | |
| *** smatzek has quit IRC | 14:22 | |
| *** smatzek has joined #openstack-ansible | 14:22 | |
| *** smatzek has quit IRC | 14:24 | |
| *** chhavi__ has joined #openstack-ansible | 14:34 | |
| *** lbragstad has joined #openstack-ansible | 14:38 | |
| *** markvoelker has quit IRC | 14:43 | |
| mhayden | cmart: mattt just pointed me at #20 here -> https://blog.online.net/2018/01/03/important-note-about-the-security-flaw-impacting-arm-intel-hardware/ | 14:44 |
| *** afred312 has quit IRC | 14:46 | |
| *** afred312 has joined #openstack-ansible | 14:46 | |
| cmart | "Just got the Meltdown update to kernel linux-image-4.4.0-108-generic but this doesnt boot at all. It just hangs after grub." | 14:46 |
| cmart | wow. oops. glad I didn't apply that one. sounds like a fix was released, 4.4.0-109? | 14:47 |
| *** markvoelker has joined #openstack-ansible | 14:51 | |
| *** prasen has joined #openstack-ansible | 14:52 | |
| *** rgogunskiy has quit IRC | 14:56 | |
| -openstackstatus- NOTICE: Gerrit is being restarted due to slowness and to apply kernel patches | 14:57 | |
| *** openstackgerrit has quit IRC | 15:01 | |
| *** markvoelker has quit IRC | 15:03 | |
| *** tone_z has joined #openstack-ansible | 15:03 | |
| *** smatzek has joined #openstack-ansible | 15:05 | |
| *** smatzek_ has joined #openstack-ansible | 15:06 | |
| *** tone_zrt has quit IRC | 15:07 | |
| *** smatzek has quit IRC | 15:08 | |
| *** woodard has quit IRC | 15:08 | |
| *** dave-mccowan has joined #openstack-ansible | 15:09 | |
| *** dave-mcc_ has joined #openstack-ansible | 15:12 | |
| *** ianychoi has joined #openstack-ansible | 15:14 | |
| *** dave-mccowan has quit IRC | 15:14 | |
| *** pbandark has quit IRC | 15:16 | |
| *** pbandark has joined #openstack-ansible | 15:20 | |
| *** sxc731 has quit IRC | 15:25 | |
| *** bhujay has joined #openstack-ansible | 15:27 | |
| *** hw_wutianwei has quit IRC | 15:29 | |
| *** openstackgerrit has joined #openstack-ansible | 15:36 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Fix typo for wait-online service https://review.openstack.org/532526 | 15:36 |
| *** markvoelker has joined #openstack-ansible | 15:37 | |
| *** germs has joined #openstack-ansible | 15:48 | |
| *** foutatoro has joined #openstack-ansible | 15:50 | |
| *** sxc731 has joined #openstack-ansible | 15:50 | |
| *** nshetty has quit IRC | 15:50 | |
| *** sbra has quit IRC | 16:04 | |
| *** mfisch` has joined #openstack-ansible | 16:06 | |
| *** mfisch has quit IRC | 16:10 | |
| *** woodard has joined #openstack-ansible | 16:10 | |
| *** woodard has quit IRC | 16:10 | |
| *** woodard has joined #openstack-ansible | 16:10 | |
| *** john51 has quit IRC | 16:17 | |
| *** markvoelker has quit IRC | 16:19 | |
| *** markvoelker has joined #openstack-ansible | 16:22 | |
| *** epalper has quit IRC | 16:23 | |
| *** john51 has joined #openstack-ansible | 16:23 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Avoid retrieving IP from a missing bridge https://review.openstack.org/532244 | 16:25 |
| *** markvoelker has quit IRC | 16:27 | |
| *** armaan has quit IRC | 16:27 | |
| *** armaan has joined #openstack-ansible | 16:28 | |
| *** armaan has quit IRC | 16:28 | |
| *** armaan has joined #openstack-ansible | 16:28 | |
| *** chyka has joined #openstack-ansible | 16:31 | |
| *** prasen has quit IRC | 16:41 | |
| mhayden | evrardjp: i slid you a little more keepalived love https://github.com/evrardjp/ansible-keepalived/pull/56 | 16:41 |
| evrardjp[m] | woot | 16:42 |
| evrardjp[m] | I can't check rn, so it will probably be for tomorrow. | 16:43 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Use full paths for keepalived scripts https://review.openstack.org/532573 | 16:45 |
| *** bhujay has quit IRC | 16:46 | |
| *** RandomTech has joined #openstack-ansible | 16:49 | |
| RandomTech | hey, is it normal for rabbitMQ to have an error during setup-infrastructure that doesnt happen after rerunning it | 16:49 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/ansible-hardening master: tasks: auth: Pass --unrestricted to Linux Grub2 entries https://review.openstack.org/532574 | 16:50 |
| hwoarang | mhayden: ^ what do you think | 16:50 |
| mhayden | lookin | 16:51 |
| mhayden | hmm, does that work okay on other distros? | 16:53 |
| mhayden | i've not tested | 16:53 |
| hwoarang | works on ubuntu | 16:53 |
| hwoarang | you need password to edit an entry but you can boot without a password | 16:53 |
| hwoarang | i believe that's the desired behavior | 16:53 |
| mhayden | yeah, i'd certainly hope so | 16:54 |
| mhayden | sheesh | 16:54 |
| mhayden | :P | 16:54 |
| mhayden | 71961 requires auth before single-user and maintenance mode booting | 16:55 |
| mhayden | so i think that's sufficient | 16:55 |
| hwoarang | let me check again what happens in the advanced boot entries | 16:56 |
| hwoarang | but if there is a problem there that affects all distros and it's unrelated :/ | 16:56 |
| hwoarang | *unrelated to this bug | 16:56 |
| mhayden | hwoarang: btw, i backported that fix for networkmanager -> https://review.openstack.org/#/c/532201/ | 16:58 |
| hwoarang | sounds like all entries can be booted without password so a separate fix is needed | 16:59 |
| *** sbra has joined #openstack-ansible | 17:01 | |
| *** dave-mcc_ has quit IRC | 17:11 | |
| *** pcaruana has quit IRC | 17:11 | |
| prometheanfire | do we have more info than https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-ceph.html for integrating ceph into osa? | 17:11 |
| logan- | https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-prod-ceph.html | 17:11 |
| *** pradiprwt has quit IRC | 17:12 | |
| prometheanfire | ah, thanks | 17:14 |
| *** pbandark has quit IRC | 17:15 | |
| *** phalmos has joined #openstack-ansible | 17:15 | |
| mhayden | that feeling when you fight OSA for a half hour only to release you forgot to trunk a vlan | 17:17 |
| mhayden | :| | 17:17 |
| * mhayden razzafrazzas | 17:17 | |
| *** chyka has quit IRC | 17:18 | |
| *** chyka has joined #openstack-ansible | 17:19 | |
| *** dave-mccowan has joined #openstack-ansible | 17:19 | |
| prometheanfire | logan-: that link makes it look like osa is deploying ceph-ansible | 17:19 |
| *** phalmos has quit IRC | 17:20 | |
| logan- | that is what the example is for, yes | 17:20 |
| logan- | are you lookign to integrate existing ceph cluster into osa? | 17:20 |
| prometheanfire | yep | 17:20 |
| *** Smeared_Beard has quit IRC | 17:20 | |
| *** mfisch` has quit IRC | 17:23 | |
| *** germs1 has joined #openstack-ansible | 17:23 | |
| *** mbuil has quit IRC | 17:24 | |
| *** phalmos has joined #openstack-ansible | 17:27 | |
| logan- | prometheanfire: i don't know if there is much info about that in the docs pages, but there is a lot of example info on how to setup the ceph client vars in https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/user_variables.yml | 17:30 |
| prometheanfire | logan-: true, it's just a mater of figuring out exactly which bits to flip | 17:33 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-repo_server stable/newton: Ensure the /var/log/apt-cacher-ng directory created https://review.openstack.org/532581 | 17:35 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-repo_server stable/ocata: Ensure the /var/log/apt-cacher-ng directory created https://review.openstack.org/532582 | 17:35 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-repo_server stable/pike: Ensure the /var/log/apt-cacher-ng directory created https://review.openstack.org/532583 | 17:36 |
| *** cjloader_ has joined #openstack-ansible | 17:36 | |
| *** zerick has joined #openstack-ansible | 17:39 | |
| cloudnull | afternoons ! | 17:40 |
| *** zerick has quit IRC | 17:41 | |
| *** zerick has joined #openstack-ansible | 17:42 | |
| *** zerick has quit IRC | 17:45 | |
| *** zerick has joined #openstack-ansible | 17:46 | |
| *** gkadam has quit IRC | 17:47 | |
| SamYaple | o/ | 17:47 |
| *** armaan has quit IRC | 17:47 | |
| *** armaan has joined #openstack-ansible | 17:48 | |
| *** drifterza has quit IRC | 17:50 | |
| spotz | Hey SamYaple | 17:50 |
| spotz | hey cloudnull | 17:51 |
| cloudnull | yo yo | 17:54 |
| *** sxc731 has quit IRC | 18:10 | |
| *** RandomTech has quit IRC | 18:18 | |
| *** openstacking_123 has joined #openstack-ansible | 18:19 | |
| DimGR | yo | 18:23 |
| *** openstack has joined #openstack-ansible | 18:32 | |
| *** ChanServ sets mode: +o openstack | 18:32 | |
| *** sxc731 has joined #openstack-ansible | 18:32 | |
| *** SmearedBeard has joined #openstack-ansible | 18:44 | |
| *** stuartgr has quit IRC | 18:47 | |
| *** SmearedBeard has quit IRC | 18:52 | |
| *** TxGirlGeek has joined #openstack-ansible | 18:55 | |
| mhayden | i deployed master and xtrabackup's xinetd listener isn't configured to allow traffic from my haproxy node :/ | 19:01 |
| mhayden | so the health checks fail each time | 19:01 |
| *** poopcat has joined #openstack-ansible | 19:06 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/pike: Allow NetworkManager to work (if present) https://review.openstack.org/532201 | 19:07 |
| *** shardy has quit IRC | 19:08 | |
| sxc731 | Hi team, I'm trying to figure out why my OSA-deployed Keystone performs poorly... I've found that the uWSGI processes complain with "!!! no internal routing support, rebuild with pcre support !!!" | 19:08 |
| sxc731 | (this is 16.0.2 on Xenial) | 19:09 |
| sxc731 | Is this serious? Should I raise a defect? According to https://stackoverflow.com/a/22645915/645016 it's just a matter of compiling uWSGI after installing libpcre3-dev, which seems to be missing from the repo containers | 19:10 |
| *** poopcat1 has joined #openstack-ansible | 19:10 | |
| *** SmearedBeard has joined #openstack-ansible | 19:11 | |
| spotz | sxc731: If you would that'd be nice, we can go over it at the next triage | 19:12 |
| *** poopcat has quit IRC | 19:13 | |
| *** poopcat1 is now known as poopcat | 19:14 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-haproxy_server master: Make haproxy stats refresh interval configurable https://review.openstack.org/532607 | 19:14 |
| sxc731 | spotz: thank you! Will do. Alas, I've ruled this out as the cause of my poor KS performance after installing a "fixed" uWSGI (getting ~150ms token validation with some weird degenerate cases at 5000+ms, on lightly loaded metal). I'm trying to profile the processes but uWSGI doesn't seem to make this easy. Any idea as to what could cause this poor performance? | 19:15 |
| spotz | sxc731: I haven't done much with metal in a long time. i do know evrardjp and mauybe cloudnull were doing somoe work on it but not sure what aspect | 19:16 |
| *** SmearedBeard has quit IRC | 19:16 | |
| sxc731 | spotz: sorry for confusion KS is running in containers but the containers are on decent metal as opposed to some VMs (which should yield decent perf) | 19:17 |
| cloudnull | lbragstad: ^ | 19:17 |
| cloudnull | sxc731: I've not seen anything in particular that would create a slow performance situation with Keystone and uWSGI | 19:18 |
| cloudnull | is it only token validation ? | 19:19 |
| lbragstad | sxc731: o/ | 19:19 |
| lbragstad | it is strange that you're seeing this with uwsgi only (if that is the case), were any other configuration changes made? | 19:20 |
| sxc731 | thank you guys! I haven't done any tweaking and uWSGI was the default deployment from what I can tell? | 19:20 |
| sxc731 | Basically deployed OpenStack perf is crap all round and I figured if token validation doesn't perform, then pretty much nothing will, right? Upward from 150ms for each token validation oin lightly loaded HW can't be right, can it? | 19:21 |
| sxc731 | I'm using this from Dolph Matthews to benchmark KS: https://gist.github.com/dolph/02c6d37f49596b3f4298 | 19:22 |
| lbragstad | unfortunately, it depends on your configuration... do you know if you've enabled caching (osa might by default but I haven't checked in a while) | 19:22 |
| cloudnull | sxc731: I've seens act like that when caching was busted? | 19:22 |
| cloudnull | ha. | 19:22 |
| sxc731 | Yes, caching is enabled by default and I've left it at that | 19:22 |
| * cloudnull deffers to lbragstad | 19:23 | |
| cloudnull | :) | 19:23 |
| sxc731 | lbragstad: is there an easy way to figure out if memcached is actually working as it should? | 19:23 |
| sxc731 | Agree, cache was my first thought | 19:24 |
| lbragstad | sxc731: i believe oslo.cache exposes debug logging (which is super noisy) but it should tell you in the service logs if there are hits or misses in the cache | 19:24 |
| lbragstad | i'd also just double check the caching configuration if you know there is traffic going to memcache | 19:25 |
| cloudnull | sxc731: is debug enabled and are you seeing a lot of IO on your hosts? | 19:25 |
| lbragstad | for one reason or another, we offer a million knobs for caching in keystone | 19:25 |
| sxc731 | lbragstad: cool, I suppose enabling global "debug = True" in KS's main conf file isn't sufficient to get this? I'll see if I can figure how to turn it on | 19:25 |
| lbragstad | let me see if i can find the oslo.cache bits | 19:25 |
| cloudnull | i've seen a notable slowdown in openstack when debug is enabled. | 19:26 |
| lbragstad | https://github.com/openstack/oslo.cache/blob/master/oslo_cache/_opts.py#L69-L75 | 19:26 |
| cloudnull | not to say it wasn't slow before debug was enabled. | 19:26 |
| lbragstad | ^ that probably won't fix your performance problems, but it could help diagnose if your cache backend is seeing traffic | 19:26 |
| cloudnull | but it can exacerbate the issue. | 19:27 |
| lbragstad | keystone allows you to toggle caching *per* keystone sub-system, which gives you the flexibility to cache identity data but not cache catalog data (or whatever your case is) | 19:27 |
| sxc731 | Thanks for the suggestion: the issue first appeared when debugging was turned off. Enabling it to diagnose didn't actually make a noticeable difference; suggesting my perf was cr*p to begin with... | 19:27 |
| lbragstad | I usually recommend caching everything though | 19:28 |
| lbragstad | this is a pike deployment? | 19:28 |
| sxc731 | Yes, I did study the caching config; seems that so long as the main toggle is on then all subsystems not explicitly disabled will be on too. But I'm sure you knew this ;-) | 19:28 |
| sxc731 | It is Pike yes | 19:28 |
| lbragstad | ok - yep, that sounds right | 19:28 |
| lbragstad | so long as your cache backend is seeing activity you should be caching everything | 19:29 |
| *** rmcall has joined #openstack-ansible | 19:30 | |
| sxc731 | OK let me try to see if I can prove caching is working using your debug toggle | 19:31 |
| lbragstad | token validation is also strictly made up of read-only database queries (if caching is disabled) | 19:31 |
| sxc731 | FWIW this is keystone.conf (generated by OSA with just some minor tweak by myself): http://paste.openstack.org/show/642486/ | 19:32 |
| spotz | hey lbragstad:) | 19:32 |
| cloudnull | sxc731: also which token type are you using? fernet (the default)? | 19:33 |
| sxc731 | My tweak was adding caching = true for [catalog] and [role] (before I understood they were on by default). Perhaps the longer cache_time for [catalog] would be a nice enhancement for OSA? | 19:33 |
| sxc731 | cloudnull: yep, I have actually not tweaked any of the config, so Fernet it is | 19:33 |
| cloudnull | cool | 19:33 |
| cloudnull | just checking | 19:33 |
| lbragstad | spotz: o/ | 19:33 |
| lbragstad | is this argument the same across all keystone containers? backend_argument = url:172.20.120.247,172.20.68.222,172.20.112.234:11211 | 19:34 |
| sxc731 | At some point I was wondering if this could be caused by invalid certs or not running the KS internal & admin endpoints with TLS | 19:34 |
| sxc731 | lbragstad: yes it's the same | 19:34 |
| lbragstad | ok - cool | 19:34 |
| sxc731 | lbragstad: OSA-generated; I've compared them | 19:35 |
| lbragstad | sxc731: do you have an idea of what the data looks like in keystone (how many users, projects, role assignments, etc..) | 19:35 |
| *** SmearedBeard has joined #openstack-ansible | 19:36 | |
| sxc731 | lbragstad: it's a pretty new default install with just a handful of users/projects. I did set Tempest loose on it and it created some addtn'l shtuff | 19:36 |
| lbragstad | ack - so nothing out of the ordinary there | 19:37 |
| sxc731 | perf was (subjectively, as in, as witnessed in Horizon) crap straight out of the box and I've been trying to figure this out for a good few weeks now :( | 19:37 |
| sxc731 | lbragstad: can you confirm that ~150ms time to validate tokens is substandard? Dolph's gist'ed b-m results have order of 5-6ms! | 19:39 |
| *** mpranjic has joined #openstack-ansible | 19:39 | |
| sxc731 | And also what could be causing these 1-2/100 spikes to over 5s to validate? This is single-threaded benchmarking on a system that's not commissioned yet, ie: no background load. | 19:40 |
| lbragstad | when we did the benchmarking, we found it to be totally subjective based on the environment | 19:40 |
| *** SmearedBeard has quit IRC | 19:41 | |
| sxc731 | Here's a typical run with 100 requests: | 19:41 |
| sxc731 | Percentage of the requests served within a certain time (ms): | 19:41 |
| sxc731 | 50% 163 | 19:41 |
| sxc731 | 66% 579 | 19:41 |
| sxc731 | 75% 588 | 19:41 |
| * lbragstad digs up performance bot results | 19:41 | |
| sxc731 | 80% 596 | 19:41 |
| sxc731 | 90% 5572 | 19:41 |
| sxc731 | 95% 5592 | 19:41 |
| sxc731 | 98% 5592 | 19:41 |
| sxc731 | 99% 5592 | 19:41 |
| sxc731 | BTW, I'm running the b-m's directly against the KS back-ends as opposed to going through haproxy (which didn't make a difference but I wanted to rule out haproxy shenanigans) | 19:42 |
| lbragstad | sure | 19:43 |
| lbragstad | i did have some on-metal performance infrastructure so that we could run performance benchmarks on patches in review | 19:43 |
| *** rmcall has quit IRC | 19:43 | |
| lbragstad | and i dumped all the results https://github.com/lbragstad/keystone-performance/tree/master/results | 19:43 |
| sxc731 | OK just looking at one of those (https://github.com/lbragstad/keystone-performance/blob/master/results/03ba1324c2/20161122103227/benchmark-results.txt) they don't seem too far off mine; albeit you don't see those degenerate cases at 5s! | 19:45 |
| lbragstad | right | 19:45 |
| lbragstad | sxc731: if you deploy keystone with a higher affinity do you see differences? | 19:45 |
| sxc731 | I noticed the validation API spits back the complete catalog (which can be quite big). Presumably this isn't the case everytime an OS service validates through keystonemiddleware? | 19:46 |
| lbragstad | there is a switch in keystonemiddleware for that i believe | 19:46 |
| lbragstad | but that catalog is also cache-able | 19:46 |
| *** SmearedBeard has joined #openstack-ansible | 19:46 | |
| *** rmcall has joined #openstack-ansible | 19:46 | |
| sxc731 | Sure, and it should be cached if my cache is working well... that's still feels like quite a bit of data that flows back to the client (~7k in my case) | 19:47 |
| sxc731 | What's the affinity? | 19:48 |
| sxc731 | (sorry, not familiar with that one) | 19:48 |
| lbragstad | for reference - dolph and i benchmarked a PoC of fernet in a global deployment (~36 app servers backed by ~18 galera nodes) and we saw anywhere from 110 - 160 ms response times for 120k RPM | 19:48 |
| lbragstad | sxc731: affinity is just the number of containers | 19:48 |
| sxc731 | Ah... well I'm running a single b-m client against a single container so presumably this wouldn't make a huge difference | 19:49 |
| lbragstad | that was about the hardest we were able to push it, but certainly noticed it scaled linearly | 19:49 |
| sxc731 | Presumably the disk perf only matters if not hitting cache? Mine are on SSDs anyway... | 19:50 |
| lbragstad | correct | 19:50 |
| lbragstad | in which case, you might notice when the cache expires | 19:50 |
| *** cshen_ has joined #openstack-ansible | 19:50 | |
| sxc731 | But if I summarize what you're saying, order of 150ms is ok; therefore I must look elsewhere as to what's causing every single Horizon page to take order of 10s to load? | 19:51 |
| *** zerick has quit IRC | 19:51 | |
| *** zerick_ has joined #openstack-ansible | 19:51 | |
| lbragstad | sxc731: yeah | 19:52 |
| *** SmearedBeard has quit IRC | 19:52 | |
| ashak | can anyone help with a bit of coaching? I've been looking at why a bunch of logging doesn't work on CentOS, particularly syslog and haproxy, but I think it affects a few of teh other default logs. I know how to fix it but i'm a bit conflicted on how I should fix it. | 19:52 |
| lbragstad | sxc731: but - we do have oslo.cache working into keystonemiddleware and it can be setup the same way you do caching in keystone | 19:52 |
| lbragstad | which means you should only see round-trip performance hits once from the service | 19:52 |
| lbragstad | and that should help | 19:53 |
| sxc731 | lbragstad: now that's quite interesting. Does this need to be set in each service's config? | 19:53 |
| lbragstad | the options would be exposed through keystonemiddleware via oslo.cache, so yeah.. | 19:53 |
| lbragstad | correct, it would need to be defined at the service | 19:53 |
| lbragstad | https://github.com/openstack/keystonemiddleware/commit/9d8e2836fe7fca186e0380d8a532540ff5cc5215 | 19:54 |
| sxc731 | Do you happen to know if this is enabled by default with OSA deployments? I don't suppose the Horizon performance I'm seeing is "works as expected" for a pretty much out-of-box OSA deployment using Fernet? | 19:55 |
| sxc731 | I'm coming from a Fuel-deployed Kilo deployment (UUID tokens) on weaker hardware that performs significantly better | 19:55 |
| lbragstad | sxc731: yeah - the last time i deployed osa i didn't notice that | 19:56 |
| lbragstad | it could be something specific to horizon, too? | 19:56 |
| sxc731 | lbragstad: I agree. Seems like I've been barking up the wrong tree (KS) all along; it was dolph's gist that lead me to believe that my ~150ms validations were off (though these 5s degenerate cases are also intriguing but don't explain my bad overall perf of itself). Thank you _very_ much for all the help, I really appreciate it! | 19:58 |
| spotz | hehe blaiming dolph:) | 19:59 |
| sxc731 | lbragstad: With your pointers, I'm still going to check that the cache works properly and that the inter-service cache is properly enabled | 19:59 |
| lbragstad | sxc731: anytime... i mean, there are probably things we can do in keystone | 19:59 |
| lbragstad | and yeah - i'd start by confirming cache traffic | 20:00 |
| sxc731 | spotz: really didn't mean to blame anyone in the amazing community! | 20:00 |
| spotz | sxc731: We love Dolph no worries:) | 20:01 |
| lbragstad | sxc731: IME if you have 150 ms token validation responses, you shouldn't be having 10s page load issues in horizon | 20:01 |
| sxc731 | lbragstad: one last q: if I want to hunt down those 5s degenerate cases, how would you go about profiling KS running under uWSGI; seems really hard... I've played with --profiler=pycall but this just outputs counts of call trees; not particularly useful | 20:01 |
| lbragstad | dolph had a good writeup using repoze | 20:02 |
| lbragstad | i've used plop in the past | 20:02 |
| lbragstad | https://github.com/bdarnell/plop | 20:02 |
| sxc731 | OK ta, I'll google those | 20:02 |
| lbragstad | http://blog.dolphm.com/performance-profiling-openstack-services-with-repoze-profile/ | 20:03 |
| sxc731 | lbragstad: brilliant, thank you again! (and sorry for hijacking the OSA channel with all these KS q's) | 20:03 |
| lbragstad | sxc731: no worries - come find me in #openstack-keystone if you have any more issues | 20:04 |
| sxc731 | lbragstad: will do, thank you. I'll also raise the defect on "!!! no internal routing support, rebuild with pcre support !!!"; this can't be improving things; though I've confirmed this isn't the cause of my issue | 20:05 |
| *** cshen_ has quit IRC | 20:05 | |
| lbragstad | sounds good | 20:05 |
| *** hybridpollo has joined #openstack-ansible | 20:10 | |
| mhayden | sxc731: which log had that message? | 20:11 |
| *** cshen_ has joined #openstack-ansible | 20:17 | |
| sxc731 | mhayden: sorry was chatting with lbragstad on the KS channel. The msg appears in both keystone-wsgi-admin.log and keystone-wsgi-public.log (KS container obviously). This is on Xenial. Believe the fix is here: https://stackoverflow.com/a/22645915/645016 (ie: make sure libpcre3-dev is installed when building the wheels on the repo container?) | 20:23 |
| mhayden | ah okay, thanks for that | 20:23 |
| *** rmcall has quit IRC | 20:25 | |
| *** dave-mccowan has quit IRC | 20:27 | |
| *** dave-mccowan has joined #openstack-ansible | 20:38 | |
| sxc731 | uWSGI PCRE issue: https://bugs.launchpad.net/openstack-ansible/+bug/1742538 | 20:43 |
| openstack | Launchpad bug 1742538 in openstack-ansible "OSA-deployed uWSGI used by Keystone warns: "!!! no internal routing support, rebuild with pcre support !!!"" [Undecided,New] | 20:43 |
| SamYaple | sxc731: thats not an issue, just a warning | 20:49 |
| *** chyka has quit IRC | 20:49 | |
| SamYaple | with or without pcre isnt changing anything | 20:50 |
| *** chyka has joined #openstack-ansible | 20:50 | |
| SamYaple | (not for the way we use uwsgi that is) | 20:50 |
| sxc731 | SamYaple: sure, I daid as much in the ticket. So Keystone doesn't rely in "internal routing"? | 20:50 |
| sxc731 | ^in^on | 20:51 |
| SamYaple | sxc731: forgive me, its been a while since i reeaserched it, but because of the way we run things (single service and all) the pcre stuff wouldnt get triggered anyway | 20:51 |
| SamYaple | that was my takeaway at the time | 20:51 |
| SamYaple | that said, i still build with libpcre3-dev | 20:52 |
| SamYaple | not discouraging adding it | 20:52 |
| sxc731 | SamYaple: OK ta. Indeed I suppose if there's a reasonably cheap way to fix the issue maybe it should be done (as a low prio) just to avoid causing false alarm? | 20:52 |
| *** Jeffrey4l has quit IRC | 20:52 | |
| SamYaple | agreed 100% | 20:53 |
| sxc731 | Good stuff. I'm sure there's like a gazillion more pressing issues... | 20:54 |
| SamYaple | yea but none with such easy solutions :P | 20:55 |
| sxc731 | Well then, cool low hanging fruit. I have to admit I looked into how to fix this in repo_build (assuming that's even the right place) and couldn't figure it out; so at least for me, not trivial... Ended up installing uWSGI from PyPi to test if it made a difference... | 20:57 |
| *** chyka has quit IRC | 20:59 | |
| *** chyka_ has joined #openstack-ansible | 20:59 | |
| *** cshen_ has quit IRC | 21:00 | |
| *** Jeffrey4l has joined #openstack-ansible | 21:03 | |
| SamYaple | sxc731: you just need to add the dev libs to the repo build package list | 21:07 |
| sxc731 | SamYaple: ok ta. I must be a little thick because I still can't seem to fins this and I'm sure it's obvious... but presumably this should be in there by default, correct? | 21:11 |
| SamYaple | yea | 21:12 |
| SamYaple | probably | 21:12 |
| *** kylek3h has joined #openstack-ansible | 21:16 | |
| *** smatzek_ has quit IRC | 21:19 | |
| *** smatzek has joined #openstack-ansible | 21:20 | |
| *** smatzek_ has joined #openstack-ansible | 21:22 | |
| *** smatzek has quit IRC | 21:24 | |
| *** smatzek_ has quit IRC | 21:26 | |
| *** sbra has quit IRC | 21:27 | |
| *** dave-mccowan has quit IRC | 21:29 | |
| *** sxc731 has quit IRC | 21:31 | |
| *** kylek3h has quit IRC | 21:38 | |
| *** chhavi__ has quit IRC | 21:41 | |
| mhayden | cloudnull: i like neutron agents on the hosts, but dang, i'm in SELinux hell with that right now ;) | 21:41 |
| jrosser | looking at a deploy log i see that pretty much every non skipped task takes a minimum of ~6 seconds, regardless of how trivial it is | 21:41 |
| *** pcaruana has joined #openstack-ansible | 21:42 | |
| jrosser | is this normal? that minimum time seems to entirely dominate the deploy time of ~5hours | 21:42 |
| *** pcaruana has quit IRC | 21:43 | |
| *** armaan has quit IRC | 21:48 | |
| *** woodard_ has joined #openstack-ansible | 21:50 | |
| *** woodard has quit IRC | 21:50 | |
| *** woodard_ has quit IRC | 21:51 | |
| *** threestrands_ has joined #openstack-ansible | 21:52 | |
| *** threestrands_ has quit IRC | 21:53 | |
| *** threestrands_ has joined #openstack-ansible | 21:53 | |
| *** threestrands has quit IRC | 21:54 | |
| *** chyka_ has quit IRC | 21:57 | |
| *** armaan has joined #openstack-ansible | 22:00 | |
| *** smatzek has joined #openstack-ansible | 22:00 | |
| *** masber has joined #openstack-ansible | 22:02 | |
| *** cjloader_ has quit IRC | 22:03 | |
| *** smatzek has quit IRC | 22:05 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_neutron master: Add SELinux policies for bare metal agents https://review.openstack.org/532646 | 22:05 |
| mhayden | cloudnull: ^^ | 22:05 |
| mhayden | antonym: on those glance backports, did you use the cherry pick button in gerrit? | 22:17 |
| -openstackstatus- NOTICE: The zuul system is being restarted to apply security updates and will be offline for several minutes. It will be restarted and changes re-equeued; changes approved during the downtime will need to be rechecked or re-approved. | 22:22 | |
| *** afred312 has quit IRC | 22:27 | |
| *** threestrands_ has quit IRC | 22:41 | |
| *** threestrands has joined #openstack-ansible | 22:55 | |
| *** afred312 has joined #openstack-ansible | 22:58 | |
| *** afred312 has quit IRC | 23:03 | |
| *** afred312 has joined #openstack-ansible | 23:07 | |
| *** rmcall has joined #openstack-ansible | 23:09 | |
| *** threestrands has quit IRC | 23:11 | |
| *** threestrands has joined #openstack-ansible | 23:20 | |
| *** openstacking_123 has quit IRC | 23:39 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests stable/ocata: Skip and echo out gzip log compression error https://review.openstack.org/532692 | 23:40 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests stable/newton: Skip and echo out gzip log compression error https://review.openstack.org/532693 | 23:41 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests stable/pike: Skip and echo out gzip log compression error https://review.openstack.org/532694 | 23:43 |
| *** cjloader_ has joined #openstack-ansible | 23:46 | |
| cloudnull | jrosser: yup. we've an execution plugin that will evaluate tasks and not even load them if we know for certain nothing will be executed for this very issue. | 23:47 |
| *** phalmos has quit IRC | 23:47 | |
| cloudnull | however we can't fully evaluate all tasks so the plugin is quite conservative | 23:48 |
| *** cjloader_ has quit IRC | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
