*** rcernin has joined #openstack-lbaas | 00:11 | |
*** rcernin_ has quit IRC | 00:12 | |
*** rcernin has quit IRC | 00:13 | |
*** rcernin has joined #openstack-lbaas | 00:14 | |
*** aojea has joined #openstack-lbaas | 00:39 | |
openstackgerrit | German Eichberger proposed openstack/octavia master: Refactor the pluggin of the VIP https://review.openstack.org/604479 | 00:39 |
---|---|---|
*** aojea has quit IRC | 00:44 | |
*** dmellado has joined #openstack-lbaas | 00:51 | |
*** hongbin has joined #openstack-lbaas | 01:27 | |
*** celebdor has quit IRC | 01:32 | |
*** hongbin has quit IRC | 01:36 | |
*** hongbin has joined #openstack-lbaas | 01:37 | |
*** hongbin_ has joined #openstack-lbaas | 01:41 | |
*** hongbin has quit IRC | 01:43 | |
*** hongbin has joined #openstack-lbaas | 02:05 | |
*** hongbin_ has quit IRC | 02:07 | |
*** ramishra has joined #openstack-lbaas | 03:12 | |
*** lxkong has quit IRC | 03:15 | |
*** lxkong has joined #openstack-lbaas | 03:15 | |
*** hongbin has quit IRC | 03:52 | |
*** aojea has joined #openstack-lbaas | 04:16 | |
*** yamamoto has quit IRC | 04:17 | |
*** yamamoto has joined #openstack-lbaas | 04:17 | |
*** aojea has quit IRC | 04:20 | |
abaindur | What does it mean when i keep seeing errors like: | 05:00 |
abaindur | Load Balancer def80fd6-67e9-4827-822f-a63981a286f4 is immutable and cannot be updated. | 05:00 |
abaindur | when trying to delete a loadbalancer, a listener, pool, or member? | 05:00 |
abaindur | the loadbalancer is in ERROR state | 05:01 |
*** yboaron_ has joined #openstack-lbaas | 05:21 | |
abaindur | I foud some old bugs from 2016, which indicate this issue was fixed. we are on queens | 05:27 |
*** ccamposr has joined #openstack-lbaas | 05:35 | |
abaindur | also... is there an API to find the active amphora in active/standby topology? | 06:03 |
abaindur | i'm currently using the DB query from operator maintenace guide: "mysql octavia -e 'select id, compute_id, lb_network_ip, role from amphora where status="ALLOCATED" or status="READY";'" | 06:04 |
openstackgerrit | mashuting proposed openstack/neutron-lbaas-dashboard master: Modify the way controls certificates show or hide https://review.openstack.org/609902 | 06:27 |
openstackgerrit | mashuting proposed openstack/neutron-lbaas-dashboard master: Modify the way controls certificates show or hide https://review.openstack.org/609902 | 06:37 |
openstackgerrit | mashuting proposed openstack/neutron-lbaas-dashboard master: Modify the way controls certificates show or hide https://review.openstack.org/609902 | 06:39 |
*** ramishra_ has joined #openstack-lbaas | 06:40 | |
openstackgerrit | mashuting proposed openstack/neutron-lbaas-dashboard master: Modify the way controls certificates show or hide https://review.openstack.org/609902 | 06:40 |
*** ramishra has quit IRC | 06:43 | |
*** pcaruana has joined #openstack-lbaas | 06:56 | |
*** yboaron_ has quit IRC | 07:04 | |
*** rcernin has quit IRC | 07:06 | |
*** rpittau has quit IRC | 07:07 | |
*** rpittau has joined #openstack-lbaas | 07:07 | |
*** abaindur has quit IRC | 07:23 | |
*** ramishra_ is now known as ramishra | 07:33 | |
openstackgerrit | Ian Wienand proposed openstack/octavia master: [wip] Redirect disk-image-builder logs, make verbose https://review.openstack.org/612622 | 07:51 |
*** velizarx has quit IRC | 08:02 | |
*** velizarx has joined #openstack-lbaas | 08:07 | |
*** yboaron_ has joined #openstack-lbaas | 08:15 | |
openstackgerrit | Luis Tomas Bolivar proposed openstack/octavia master: Enabling SG customization on loadbalancer listerners https://review.openstack.org/602564 | 08:31 |
openstackgerrit | Luis Tomas Bolivar proposed openstack/octavia master: Deprecating option to create LBaaS security group on users' project https://review.openstack.org/612631 | 08:31 |
*** celebdor has joined #openstack-lbaas | 08:33 | |
*** dmellado has quit IRC | 08:34 | |
openstackgerrit | Luis Tomas Bolivar proposed openstack/octavia master: Enabling SG customization on loadbalancer listerners https://review.openstack.org/602564 | 08:37 |
openstackgerrit | Luis Tomas Bolivar proposed openstack/octavia master: Deprecating option to create LBaaS security group on users' project https://review.openstack.org/612631 | 08:37 |
*** dmellado has joined #openstack-lbaas | 09:18 | |
*** bzhao__ has quit IRC | 09:20 | |
*** salmankhan has joined #openstack-lbaas | 09:38 | |
*** ianychoi has quit IRC | 10:22 | |
*** ianychoi has joined #openstack-lbaas | 10:25 | |
*** yamamoto has quit IRC | 11:03 | |
*** yamamoto has joined #openstack-lbaas | 11:04 | |
*** yamamoto has quit IRC | 11:08 | |
*** yamamoto has joined #openstack-lbaas | 11:21 | |
*** pck has joined #openstack-lbaas | 11:28 | |
*** aojea_ has joined #openstack-lbaas | 12:42 | |
*** aojea_ has quit IRC | 12:55 | |
*** aojea_ has joined #openstack-lbaas | 12:56 | |
*** yamamoto has quit IRC | 13:02 | |
*** emccormick has joined #openstack-lbaas | 13:05 | |
*** aojea_ has quit IRC | 13:17 | |
*** aojea_ has joined #openstack-lbaas | 13:18 | |
*** aojea_ has quit IRC | 13:22 | |
openstackgerrit | Vadim Ponomarev proposed openstack/octavia master: Add a soft check for the allowed_address_pairs extension. https://review.openstack.org/568546 | 13:32 |
*** yamamoto has joined #openstack-lbaas | 13:35 | |
*** emccormick has quit IRC | 13:40 | |
*** aojea_ has joined #openstack-lbaas | 13:45 | |
*** aojea_ has quit IRC | 13:48 | |
tobias-urdin | johnsom: https://storyboard.openstack.org/#!/story/2004156 | 13:49 |
*** aojea_ has joined #openstack-lbaas | 13:52 | |
*** aojea_ has quit IRC | 13:55 | |
*** aojea_ has joined #openstack-lbaas | 13:56 | |
*** aojea_ has quit IRC | 14:00 | |
*** ramishra has quit IRC | 14:06 | |
*** kobis1 has joined #openstack-lbaas | 14:17 | |
*** kobis1 has quit IRC | 14:17 | |
ltomasbo | ping johnsom: if you have time, please take a look at https://review.openstack.org/#/c/602564 | 14:50 |
openstackgerrit | Vadim Ponomarev proposed openstack/octavia master: Support create amphora instance from volume based. https://review.openstack.org/570505 | 14:53 |
*** yboaron_ has quit IRC | 15:25 | |
*** emccormick has joined #openstack-lbaas | 15:35 | |
*** celebdor has quit IRC | 15:35 | |
*** ianychoi_ has joined #openstack-lbaas | 15:36 | |
*** ianychoi has quit IRC | 15:40 | |
*** ivve has joined #openstack-lbaas | 15:42 | |
*** ccamposr has quit IRC | 15:44 | |
*** phuoc has joined #openstack-lbaas | 16:38 | |
emccormick | Boy oh boy. You wanna get people actively deploying / breaking your project, deprecate the alternative :D | 17:00 |
emccormick | Some of y'all should subscribe to the operators list until the merger happens. Octavia is getting lots of traffic there lately. | 17:02 |
johnsom | I am on the operator mailing list, and replying to any questions that come up. | 17:02 |
johnsom | In fact I'm replying to one right now. | 17:02 |
johnsom | Something about an HMAC issue | 17:03 |
emccormick | oh super :). Thanks. I just spewed a whole bunch of stuff out there on the management network question | 17:03 |
emccormick | maybe the beginnings of a noob guide, at least for kolla people | 17:04 |
*** phuoc has quit IRC | 17:05 | |
*** tomtom001 has joined #openstack-lbaas | 17:08 | |
johnsom | Cool, yeah, over the weekend I had a huge backlog of stuff to look at, e-mail is one | 17:10 |
tomtom001 | Hello, just got Octavia installed and working on Queens... One issue is that the health monitor shows online yet the Operating Status is always offline, I'm creating the LB's through horizon, is this any kind of known issue? Is the health monitor ping'ing from the octavia container or the LB instance itself? | 17:10 |
tomtom001 | Any help is much appreciated thank you. | 17:10 |
emccormick | tomtom01 Are you deploying with Kolla by any chance? | 17:11 |
johnsom | tomtom001 This likely means your lb-mgmt-net is not working correctly and the health heartbeats are not making it back to your controllers | 17:11 |
tomtom001 | No, using Openstack-ansible; ok so the octavia-health-manager on the controllers need to be able to talk to the "web instances" inside the LB? | 17:12 |
tomtom001 | or are you saying that the LB instance needs to send heartbeats back to the controllers? | 17:13 |
emccormick | tomtom01 instances talk to the manager | 17:13 |
tomtom001 | ok LB instances talk back to the health-manager. | 17:13 |
emccormick | tomtom01 well both really. Config happens with controllers talking to the instances, but heartbeats are sent from amphora to health manager | 17:14 |
emccormick | so you need to be able to go both ways | 17:14 |
tomtom001 | right, ok, I will verify that | 17:14 |
johnsom | tomtom001 Yes, the "amphora" instance send heartbeats back to the controllers that includes the health and statistics information. | 17:14 |
emccormick | also make sure your dashboard is using the Octavia dashboard and not the old LBaaS one. I've found the old one will happily create broken load balancers and never touch octavia | 17:15 |
emccormick | verify they exist where they should with the cli | 17:15 |
emccormick | and that the amphorae are getting created | 17:16 |
*** phuoc has joined #openstack-lbaas | 17:18 | |
*** salmankhan has quit IRC | 17:20 | |
tomtom001 | Yep thanks johnsom, emccormick, found a config issue. | 17:22 |
*** phuoc has quit IRC | 17:27 | |
*** phuoc has joined #openstack-lbaas | 17:43 | |
*** aojea has joined #openstack-lbaas | 17:46 | |
*** abaindur has joined #openstack-lbaas | 18:04 | |
*** abaindur has quit IRC | 18:05 | |
*** abaindur has joined #openstack-lbaas | 18:05 | |
*** irclogbot_3 has joined #openstack-lbaas | 18:35 | |
*** nicolasbock has joined #openstack-lbaas | 18:35 | |
nicolasbock | Hi. I am trying to get Octacia to work in an all-in-one OpenStack Ansible. Unfortunately the amphora image seems to struggle to stay alive and dies with a kernel backtrace. | 18:37 |
nicolasbock | If it helps, I can paste a screenshot of at least the last part of that backtrace. | 18:38 |
*** phuoc has quit IRC | 18:56 | |
openstackgerrit | German Eichberger proposed openstack/octavia master: Refactor the pluggin of the VIP https://review.openstack.org/604479 | 19:08 |
*** salmankhan has joined #openstack-lbaas | 19:09 | |
johnsom | nicolasbock Yeah, I can take a look. I wonder if it is the KVM bug. | 19:12 |
nicolasbock | I can launch a cirros though | 19:12 |
nicolasbock | But that's presumably a little bit lighter weight though ;) | 19:12 |
johnsom | Ok, so it's not the KVM bug. It kills everything | 19:12 |
nicolasbock | Ah ok | 19:13 |
nicolasbock | Is there a good place where I can paste the screenshot? | 19:13 |
nicolasbock | (if that helps) | 19:13 |
johnsom | I have not had an amphora with a kernel panic before. | 19:13 |
*** salmankhan has quit IRC | 19:13 | |
johnsom | Hmm, not sure on that. | 19:14 |
nicolasbock | ok | 19:14 |
nicolasbock | I'll try to create a server directly with the image | 19:14 |
nicolasbock | Although I can't quite imagine that this would make a difference | 19:15 |
johnsom | Maybe here: https://imgur.com/upload | 19:15 |
nicolasbock | https://imgur.com/XtqMBhb | 19:15 |
johnsom | Hmmm, not much to go off of there. Can you look at the qemu log for the instance? Is there any thing interesting in there? | 19:17 |
johnsom | Where did you get the image? Did you build it or download it from somewhere? | 19:18 |
nicolasbock | I ran the all-in-one OpenStack Ansible on master | 19:19 |
nicolasbock | It downloads the amphora during setup | 19:19 |
johnsom | Ok, so it grabbed a nightly build image. | 19:19 |
nicolasbock | I can dig through the code to find out where exactly | 19:19 |
johnsom | I know where ansible is getting that. | 19:19 |
nicolasbock | The qemu log looks clean | 19:20 |
johnsom | I can't pull down that test image to see if last nights image build somehow went bad. My VMs are in use at the moment. | 19:22 |
johnsom | I can show you how to build a fresh image if you would like, it's pretty straight forward. | 19:22 |
*** pcaruana has quit IRC | 19:23 | |
johnsom | I'm not sure if there is an ansible playbook to build a new image. xgerman_? | 19:23 |
emccormick | nicolasbock can you copy the entire bootlog to a pastebin somewhere? | 19:36 |
emccormick | go to the instance in horizon, select the log tab, click the full log button, copy and paste. | 19:37 |
nicolasbock | emccormick (IRC): Can I do that on the CLI too? | 19:37 |
nicolasbock | I don't have horizon installed | 19:37 |
emccormick | you can use http://paste.openstack.org/ | 19:37 |
emccormick | openstack console log show <instance name> | 19:39 |
nicolasbock | Awesome, didn't know that :) | 19:40 |
emccormick | you can > somefile.txt and then upload that to the paste thing I think | 19:41 |
nicolasbock | http://paste.openstack.org/show/732852/ | 19:42 |
xgerman_ | johnsom: let me check | 19:50 |
*** openstackgerrit has quit IRC | 20:06 | |
xgerman_ | I removed that: https://github.com/openstack/openstack-ansible-os_octavia/commit/95eee6bc11e97105cb0356d7475bee699d404bda | 20:22 |
xgerman_ | so we could resurrect that | 20:23 |
emccormick | nicolasbock I'm not 100% sure, but I think this is a problem with supported CPU features vs. the Kernel that image is using | 20:23 |
johnsom | Ah, ok, bummer | 20:23 |
xgerman_ | well, slowed down the build and we had the script anyway | 20:24 |
emccormick | nicolasbock If you have the ability to set the CPU in nova.conf, try using this: | 20:24 |
emccormick | cpu_mode=custom | 20:24 |
emccormick | cpu_model=kvm64 | 20:24 |
emccormick | and see if it doesn't boot | 20:24 |
johnsom | Yeah. I think a nice to have might be a var that builds or downloads. | 20:24 |
emccormick | I've run into this with various Ubuntu builds in the past | 20:24 |
xgerman_ | https://github.com/rcbops/rpc-octavia/blob/master/gating/scripts/rpc-build-image.yml | 20:25 |
xgerman_ | is a playbook we use internally | 20:25 |
*** ivve has quit IRC | 20:42 | |
*** salmankhan has joined #openstack-lbaas | 21:03 | |
abaindur | johnsom: if we delete an amphora VM directly, will it not spin up a new amphora? | 21:10 |
johnsom | Yes it will | 21:11 |
johnsom | The health manager will detect it and start a failover flow that replaces the failed amphora | 21:11 |
*** rtjure has quit IRC | 21:39 | |
*** openstackgerrit has joined #openstack-lbaas | 21:47 | |
openstackgerrit | Michael Johnson proposed openstack/octavia-lib master: Initial provider driver library checkin https://review.openstack.org/612815 | 21:47 |
*** aojea has quit IRC | 22:02 | |
openstackgerrit | Michael Johnson proposed openstack/octavia-lib master: Initial provider driver library checkin https://review.openstack.org/612815 | 22:03 |
nicolasbock | I updated `nova.conf` emccormick (IRC) but in the domain I get `cpu mode='host-model' check='partial'` and `<model fallback='allow'/>` | 22:23 |
openstackgerrit | Michael Johnson proposed openstack/octavia-lib master: Initial provider driver library checkin https://review.openstack.org/612815 | 22:23 |
nicolasbock | My `nova.conf` updates don't seem to have worked. | 22:24 |
*** rcernin has joined #openstack-lbaas | 22:24 | |
nicolasbock | I'll spend some more time with it | 22:24 |
nicolasbock | Thanks for the suggestion! | 22:24 |
openstackgerrit | Michael Johnson proposed openstack/octavia-lib master: Initial provider driver library checkin https://review.openstack.org/612815 | 22:34 |
openstackgerrit | Ian Wienand proposed openstack/octavia master: [wip] Redirect disk-image-builder logs, make verbose https://review.openstack.org/612622 | 22:37 |
*** threestrands has joined #openstack-lbaas | 23:02 | |
emccormick | nicolasbock hmm yeah it didn't trigger the change clearly. That's the default I think | 23:15 |
emccormick | host-model passes through what the physical box has which is what you specifically do *not* want | 23:15 |
openstackgerrit | Ian Wienand proposed openstack/octavia master: [wip] Redirect disk-image-builder logs, make verbose https://review.openstack.org/612622 | 23:20 |
emccormick | kinda looks like either nova-compute didn't get restarted or the options didn't make it there | 23:21 |
openstackgerrit | Ian Wienand proposed openstack/octavia-tempest-plugin master: Collect diskimage-builder logs https://review.openstack.org/612830 | 23:21 |
emccormick | cat your nova-compute.log and grep for those variables (cpu_mode and cpu_model) at around the time you restarted nova-compute | 23:24 |
emccormick | note you need to be running with debug = True to confirm | 23:26 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!