| *** cheng1 has quit IRC | 00:53 | |
| *** cheng1 has joined #starlingx | 00:54 | |
| *** yaawang has joined #starlingx | 02:52 | |
| *** irclogbot_3 has quit IRC | 06:05 | |
| *** irclogbot_1 has joined #starlingx | 06:07 | |
| *** bengates has joined #starlingx | 07:16 | |
| -openstackstatus- NOTICE: uWSGI made a new release that breaks devstack, please refrain from rechecking until a devstack fix is merged. | 07:44 | |
| *** cheng1 has quit IRC | 07:48 | |
| *** cheng1 has joined #starlingx | 07:50 | |
| *** born2bake has joined #starlingx | 07:50 | |
| born2bake | https://bugs.launchpad.net/starlingx/+bug/1883432 | 07:51 |
|---|---|---|
| openstack | Launchpad bug 1883432 in StarlingX "A-I-O Duplex - PXe cant see the second controller" [Undecided,New] | 07:51 |
| *** cheng1 has quit IRC | 08:12 | |
| *** cheng1 has joined #starlingx | 08:14 | |
| *** cheng1 has quit IRC | 08:40 | |
| *** cheng1 has joined #starlingx | 08:42 | |
| born2bake | https://bugs.launchpad.net/starlingx/+bug/1883432 | 10:13 |
| openstack | Launchpad bug 1883432 in StarlingX "A-I-O Duplex - PXe cant see the second controller" [Undecided,New] | 10:13 |
| bwensley | Hey born2bake - are you sure your second controller is booting from the management interface that is directly connected to controller-0. | 11:22 |
| bwensley | This is almost certainly an issue with your physical networking setup. | 11:22 |
| born2bake | i am not how can I specify from which inmt to boot | 11:24 |
| born2bake | i am not sure...cause on my servers in bios...I can just add embded nic for network boot | 11:24 |
| born2bake | bwensley | 11:25 |
| bwensley | Can't help you there. To make this work, your server needs to boot from the NIC connected to the management interface on controller-0. | 11:26 |
| *** mthebeau has joined #starlingx | 11:48 | |
| born2bake | bwensley okay, thanks a lot... Installation proceeded :/ was wrong interface https://prnt.sc/t01ua8 ; However, I am wondering if its okay that its stuck at initrd.... | 11:53 |
| bwensley | Is the console on your server being directed to a serial port? | 11:57 |
| bwensley | You would need to look there. | 11:57 |
| bwensley | Please set your LP to invalid and explain what the problem was. | 11:58 |
| bwensley | Typically for things like this it is better to send a mail to the mailing list. We don't use LPs for support requests. | 11:58 |
| born2bake | Okay, will do. thank you | 11:59 |
| bwensley | np | 11:59 |
| born2bake | bwensley I dont think it supports serial y... | 12:00 |
| born2bake | however, if it chose serial...does it mean that its still installing? | 12:00 |
| born2bake | or should I reboot machine and choose graphics? | 12:01 |
| born2bake | I could not tick on graphic console cause it automatically started boot | 12:02 |
| bwensley | It is probably still installing. I would just wait. | 12:04 |
| born2bake | okay, thanks | 12:05 |
| bwensley | To use the graphical console, just add "console=" to the end of your host-update command. | 12:06 |
| bwensley | E.g.: system host-update 2 personality=controller console= | 12:06 |
| born2bake | oh ok.. | 12:07 |
| born2bake | I dont think that its gonna help if I add it now | 12:08 |
| born2bake | since its already at initrd etc.. | 12:08 |
| born2bake | btw do you know, if I am installing 3.0 version and 4.0 is coming in July, would I be able to upgrade my setup? | 12:09 |
| bwensley | Correct - you can only do that on the initial host-update command. | 12:10 |
| bwensley | No you can't. You would need a 3.1 maintenance release to do that and I don't believe there are any plans to create that release. | 12:10 |
| *** ijolliffe has joined #starlingx | 12:28 | |
| born2bake | okay then I will just learn and work on 3rd version...and then wait for 4th | 12:37 |
| born2bake | bwensley https://docs.starlingx.io/deploy_install_guides/r3_release/bare_metal/aio_duplex_install_kubernetes.html#configure-controller-1 here in docs I am bit confused....its saying to specify mngmt and oam ports and add them. However, I cant do that casue openrc is not sourced. Additionally, if I run source /etc/platform/openrc its not there either.. so should I run ansible bootstrap on controller-1 as well? | 12:42 |
| bwensley | No - those commands are to be run on controller-0. | 12:47 |
| bwensley | The whole installation/configuration process is driven from controller-0 (the active controller). | 12:48 |
| born2bake | Okay, thank you. | 13:00 |
| born2bake | [sysadmin@controller-0 ~(keystone_admin)]$ system datanetwork-add ${PHYSNET0} flat | 13:00 |
| born2bake | DataNetwork of name physnet0 already exists. | 13:00 |
| born2bake | [sysadmin@controller-0 ~(keystone_admin)]$ system datanetwork-add ${PHYSNET1} flat | 13:00 |
| born2bake | DataNetwork of name physnet1 already exists. | 13:00 |
| born2bake | Thus, on second controller I should give them physnet2 and 3 names right? | 13:01 |
| born2bake | bwensley | 13:08 |
| bwensley | No - the "system datanetwork-add" commands only need to be run once. I expect that is an error in the documentation. | 13:13 |
| bwensley | You can raise an LP for that if you like. | 13:13 |
| born2bake | and for | 13:14 |
| born2bake | https://prnt.sc/t03prj those are needed? | 13:14 |
| bwensley | eys | 13:15 |
| bwensley | yes | 13:15 |
| born2bake | okay, last one I wanted to ask: echo ">>>> Configuring nova-local" | 13:17 |
| born2bake | NOVA_SIZE=34 - size will be taken from 1st, root disk right? I cannot specify separate disk for nova local? | 13:17 |
| bwensley | Not my area of expertise. I would just follow the instructions. | 13:19 |
| born2bake | okay. thank you very much for your help | 13:25 |
| *** cheng1 has quit IRC | 13:27 | |
| *** cheng1 has joined #starlingx | 13:28 | |
| sgw | Morning all | 13:30 |
| born2bake | Hello... bwensley https://prnt.sc/t04dd7 finally yay thank you :) | 13:42 |
| born2bake | Degraded availability is due to ceph reconfiguration? | 13:42 |
| born2bake | Also, in case if I want to use second controller in cojunction with first one...I guess its active/active state...how do I configure it? | 13:43 |
| born2bake | https://prnt.sc/t04f3a fm-alarm list | 13:44 |
| bwensley | The controllers are automatically configured for active/standby. You can't change that. | 13:52 |
| born2bake | so if I run any container workloads or vm's in openstack, are they gonna be spreaded across both aio-controllers? | 13:52 |
| bwensley | yes | 13:53 |
| born2bake | Okay, and in case if I want to extend capacity by worker node....need it to be connected via mngmt interface as well? | 13:55 |
| *** dpanech has joined #starlingx | 13:59 | |
| bwensley | yes | 13:59 |
| born2bake | which means each server needs to have 3 NIC's right :/ | 14:04 |
| *** cheng1 has quit IRC | 14:13 | |
| *** mthebeau has quit IRC | 14:15 | |
| *** cheng1 has joined #starlingx | 14:16 | |
| bwensley | Worker node requirements would be as per https://docs.starlingx.io/deploy_install_guides/r3_release/bare_metal/controller_storage_hardware.html | 14:19 |
| *** abailey has joined #starlingx | 14:21 | |
| *** al_bailey has joined #starlingx | 14:21 | |
| *** al_rules has joined #starlingx | 14:21 | |
| *** al_bailey has quit IRC | 14:21 | |
| *** abailey has quit IRC | 14:24 | |
| *** zbsarash1 has joined #starlingx | 14:28 | |
| sgw | morning all | 14:44 |
| bwensley | morning | 14:45 |
| *** sgw has quit IRC | 14:47 | |
| *** sgw has joined #starlingx | 14:49 | |
| ildikov | sgw: morning | 14:50 |
| sgw | ildikov: Morning! | 14:51 |
| ildikov | sgw: I'm about to upload the blog post about the PTG to GitHub. Do you still would like to make the order changes? | 14:51 |
| sgw | ildikov: nah it's fine, I understand your thought process now | 14:51 |
| *** bengates_ has joined #starlingx | 14:56 | |
| *** bengates has quit IRC | 14:56 | |
| ildikov | sgw: are you sure? I'm happy to move things around if it makes more sense like that :) | 14:58 |
| ildikov | sgw: I just wanted to give background why I wrote it how I did, but it doesn't mean it can't be improved | 14:59 |
| *** slittle1 has quit IRC | 15:00 | |
| *** sgw1 has joined #starlingx | 15:06 | |
| ildikov | sgw: FYI, https://github.com/StarlingXWeb/starlingx-website/pull/80 | 15:43 |
| ildikov | sgw: my colleague will merge it later today | 15:43 |
| born2bake | Hi guys, I am wondering when you specify nova-local size... "NOVA_SIZE=34" - this size can changed and where does it take from? from root disk? can I specify my other disk for openstack/nova service somehow? | 16:01 |
| born2bake | Also, when you specify data int, it can be just one physnet0 right? you dont need to specify 2? or you need to have separate for k8s and o-stack? | 16:18 |
| *** slittle1 has joined #starlingx | 16:24 | |
| sgw | born2bake: I would help if I could, but this is something you might as the -discuss list also | 16:30 |
| sgw | slittle1: did you see zhipeng's email note? have you had a chance to update the ussuri build? | 16:32 |
| *** slittle1 has quit IRC | 16:34 | |
| *** zhipeng[m] has joined #starlingx | 16:38 | |
| zhipeng[m] | Morning | 16:39 |
| sgw1 | zhipeng[m]: hi there! looks like you got it working | 16:47 |
| zhipeng[m] | Yes, I need to use another VPN :) | 16:50 |
| sgw1 | dpanech: I am seeing this right that Master is also failing, is it getting fixed? I don't see scott today (he was here earlier). | 16:51 |
| dpanech | sgw1: looking into it | 16:53 |
| sgw | dpanech: great thanks, let us know if there is some action needed. | 16:54 |
| sgw | dpanech: your looking at master, so the ussri build is not going to happen until master is fixed, correct? | 16:55 |
| sgw | zhipeng[m]: maybe master needs to happen first today. | 16:55 |
| zhipeng[m] | got it | 16:57 |
| zhipeng[m] | From ussuri error log, I just found openstack-helm-infra build fail, that I have fixed in my latest update. | 16:59 |
| sgw | zhipeng[m]: so when scott or davlet file up the next build, they need to make sure to cherrypick your lastest change | 17:00 |
| zhipeng[m] | yes | 17:00 |
| sgw | That's what I wrote in the email also | 17:00 |
| zhipeng[m] | git fetch https://review.opendev.org/starlingx/openstack-armada-app refs/changes/35/720135/12 && git cherry-pick FETCH_HEAD | 17:01 |
| zhipeng[m] | only this update | 17:01 |
| dpanech | sgw: give me a few mins, i'm checking with someone else | 17:02 |
| *** slittle1 has joined #starlingx | 17:05 | |
| dpanech | sgw: the build is still going let's wait a bit. looks like dockerhub was unresponsive in the morning it took ~1hr for the login to succeed. It should be done soon. | 17:08 |
| sgw1 | which build? master or ussuri? | 17:09 |
| dpanech | sgw: master, it just finished | 17:09 |
| *** bengates_ has quit IRC | 17:09 | |
| born2bake | sgw discuss list? | 17:16 |
| sgw | born2bake: starlingx-discuss@lists.starlingx.io | 17:19 |
| sgw | listsed in topic area also | 17:19 |
| born2bake | Basically, I can prepare all my questions and send them there? | 17:20 |
| sgw | born2bake: yes, some could be answered here in more real time, but if folks don't answer then the email list might get attention | 17:21 |
| born2bake | oke doke thanks | 17:21 |
| sgw | born2bake: also join the wednesday community meeting | 17:22 |
| slittle1 | looks like ussuri build on cengn is failing due to patches that don't apply | 17:30 |
| slittle1 | http://mirror.starlingx.cengn.ca/mirror/starlingx/ussuri/centos/monolithic/20200615T120015Z/logs/std/failed-packages/ | 17:30 |
| slittle1 | let me check that the cherry-picks are up to date | 17:30 |
| slittle1 | armada app is out of date, fixing it ... | 17:33 |
| sgw1 | Right, I think that's what zhipeng[m] was trying to point out. | 17:38 |
| *** bengates has joined #starlingx | 17:43 | |
| slittle1 | https://review.opendev.org/#/c/731461/ does not apply | 17:46 |
| slittle1 | perhaps a missing depends-on ? | 17:46 |
| *** bengates has quit IRC | 17:48 | |
| sgw | zhipeng[m]: ^^^??? | 17:52 |
| slittle1 | seems at odds with https://review.opendev.org/#/c/720135/ re openstack-helm-infra/centos/openstack-helm-infra.spec | 17:57 |
| sgw | Honestly not sure, let me try to reach zhipeng[m] | 17:58 |
| *** slittle1 has quit IRC | 18:01 | |
| *** slittle1 has joined #starlingx | 18:03 | |
| *** slittle1 has quit IRC | 18:11 | |
| born2bake | Is there any plans on kernel upgrade in 4.0th version? | 18:18 |
| sgw | born2bake: yes, we are moving the 4.18 CentOS-8 kernel in 4.0 | 18:22 |
| born2bake | 4.0 release planned on July isnt it? | 18:22 |
| sgw | correct | 18:24 |
| *** cheng1 has quit IRC | 18:25 | |
| *** cheng1 has joined #starlingx | 18:27 | |
| *** slittle1 has joined #starlingx | 18:37 | |
| sgw | slittle1: I am looking at those patches and zhipeng[m]'s email from 6:08am (PT) on the Ussuri Test build failed thread | 18:40 |
| sgw | I am trying to understand what he was doing, but it seems something was not pushed or updated correctly. | 18:41 |
| sgw | slittle1: what are you adding to the openstack-armada-app? | 18:46 |
| sgw | I think he just wants thisone: git fetch https://review.opendev.org/starlingx/openstack-armada-app refs/changes/35/720135/12 && git cherry-pick FETCH_HEAD | 18:46 |
| sgw | That should apply cleanly to master/HEAD | 18:47 |
| *** slittle1 has quit IRC | 19:09 | |
| *** slittle1 has joined #starlingx | 19:12 | |
| sgw | slittle1: did you see my note above about just the one fetch/cherry-pick for ussuri? | 19:13 |
| *** slittle1 has quit IRC | 19:17 | |
| born2bake | guys is it that important to have 2 data networks? In docs data networks are always specified as 2 (physnet0 and physnet1). | 19:29 |
| born2bake | Can I use just one right cause its only related to openstack workloads? | 19:29 |
| *** slittle1 has joined #starlingx | 19:31 | |
| *** slittle1 has quit IRC | 19:53 | |
| *** slittle1 has joined #starlingx | 19:58 | |
| sgw | slittle1: have you been able to restart the Ussuri build ? | 20:03 |
| sgw | dpanech: ^^^ maybe your working on it? | 20:03 |
| *** rchurch has quit IRC | 20:44 | |
| *** rchurch has joined #starlingx | 20:45 | |
| *** cheng1 has quit IRC | 21:15 | |
| *** cheng1 has joined #starlingx | 21:17 | |
| *** cheng1 has quit IRC | 21:34 | |
| *** cheng1 has joined #starlingx | 21:34 | |
| *** slittle1 has quit IRC | 21:53 | |
| *** sgw has quit IRC | 21:59 | |
| *** sgw has joined #starlingx | 22:01 | |
| zhipeng[m] | slittle1: you are right, I need update https://review.opendev.org/#/c/731461/ as well. | 22:07 |
| zhipeng[m] | done | 22:07 |
| zhipeng[m] | slittle1: Could you help retrigger build again? thanks! | 22:13 |
| zhipeng[m] | I fogot to update this last patch which is still for openstack-helm-infra, need update after rebasing as well. | 22:30 |
| *** slittle1 has joined #starlingx | 22:43 | |
| born2bake | Is there any way to troubleshoot why second controller is not installing starlingx correctly? First time when I was trying, i connected both servers directly. This time, I connected them all to switch....then I run second controller in network boot mode... Installation started and stuck there forever https://prnt.sc/t0ecpt ; Furthermore, then I checked host-list and I can see that one more machine appeared in there O_O even though controller-1 | 23:04 |
| born2bake | didnt finish deployment. https://prnt.sc/t0ed3f | 23:04 |
| born2bake | bwensley maybe you would be able to help :) if you get any advice please type here i will check logs tmrw | 23:48 |
| *** born2bake has quit IRC | 23:48 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!