Monday, 2020-06-01

*** tosky has quit IRC00:20
*** Meiyan has joined #opendev00:58
fungijust a heads up, i e-mailed the openstack-discuss ml with a summary of the jitsi-meet case insensitivity/etherpad case sensitivity situation for meetpad, and suggested we can rename etherpads to their lower-case versions, as well as to remove/substitute punctuation we're not supporting in meetpad right now01:04
fungii'm happy to handle those, though it's basically a single etherpad api call so not complicated01:07
*** xiaolin has joined #opendev01:20
spotzfungi I sent a rename request email to the servie-discuss list for out D&I session for tomorrow but it's in moderation. I suspect most requests being email in are going to get held up01:56
fungispotz: unless they come from list subscribers, yeah. i'll process the moderation queue in a moment and then rename the pad for you01:59
*** diablo_rojo has joined #opendev02:06
*** ysandeep|afk is now known as ysandeep02:08
fungiand per my reply, doesn't look like there's any content in the pad yet, so won't need renaming02:17
fungispotz: mentions an even different name02:20
diablo_rojo_phonThis are the auto-generated etherpads.. not many teams actually use those sadly.02:20
diablo_rojo_phonI think you can override the link there though.02:20
fungiahh, okay02:21
fungialso the diversity-wg link on tomorrow's schedule at still goes to the zoom room02:21
diablo_rojo_phonIt's up to the moderators to change where they are meeting via the bot.02:24
diablo_rojo_phonI sent out details for how to do that.02:24
*** rm_work has quit IRC02:28
*** Eighth_Doctor has quit IRC02:31
*** Eighth_Doctor has joined #opendev02:53
*** ykarel|away is now known as ykarel04:07
openstackgerritIan Wienand proposed openstack/project-config master: Move wheel build role/job outside project-config
ianwAJaeger: thanks, that should cleanup and depend on jobs that cleans up the requirements branches04:16
*** diablo_rojo has quit IRC04:46
*** xiaolin has quit IRC04:47
*** dpawlik has joined #opendev05:18
openstackgerritMerged zuul/zuul-jobs master: ensure-tox: fix tox_executable fact when tox is preinstalled
openstackgerritMerged zuul/zuul-jobs master: ensure-twine: Check executable presence using shell+bin/bash
AJaegerianw: thanks.05:36
AJaegerianw: could you review, and , please?05:37
*** dpawlik has quit IRC05:57
*** dpawlik has joined #opendev05:57
*** dpawlik has quit IRC05:58
ianwAJaeger: what's the context on 731989?  it looks right but i dont' fully understand06:21
*** DSpider has joined #opendev07:00
*** dpawlik has joined #opendev07:00
*** tosky has joined #opendev07:28
jrossergood morning - can i get a hold on a failing job please?07:33
ianwjrosser: which job?07:41
jrosserianw: its 731251, openstack-ansible-lxc-dir-ubuntu-focal, "investigate lxc errors on focal"07:42
openstackgerritMichał Piotrowski proposed openstack/project-config master: Adding elastic/kibana repository
*** ysandeep is now known as ysandeep|lunch07:50
*** gtema has joined #opendev07:51
ianwjrosser: | 0000000133 | openstack | |      openstack-ansible-lxc-dir-ubuntu-focal      |             .*            |     1     |                     jrosser: investigate lxc errors on focal07:56
ianwrecheck should catch it07:56
jrosserexcellent, thanks07:56
openstackgerritGuillaume Chauvel proposed zuul/zuul-jobs master: ensure-twine: Fix missing default value twine_executable
*** moppy has quit IRC08:01
*** moppy has joined #opendev08:01
*** ravsingh has joined #opendev08:02
*** gtema has quit IRC08:02
*** gtema has joined #opendev08:03
gtemaHi guys, I would like to rename "lowercase" 2 etherpads for PTG (use with meetpad): and
*** ykarel is now known as ykarel|lunch08:23
jrosserianw: that job is failed, my key is here
*** ysandeep|lunch is now known as ysandeep08:45
ianwjrosser: umm can you get me the ip quick?08:45
ianwjrosser: ok try that root@08:49
jrossercool works, thankyou08:51
*** ysandeep is now known as ysandeep|brb09:01
hrwAJaeger, ianw: how we are with that wheel cache stuff? I see that patches are in but lacks any updates so something needs to be done still09:07
*** dpawlik has quit IRC09:08
*** ykarel|lunch is now known as ykarel09:10
*** xiaolin has joined #opendev09:16
AJaegerianw: regarding 731989 - this is a partial revert, a wheel profile is not helpful09:32
AJaegerhrw: you can check the log files09:32
*** gtema has quit IRC09:33
AJaeger etc09:34
openstackgerritMerged zuul/zuul-jobs master: ensure-twine: Fix missing default value twine_executable
*** xiaolin has quit IRC09:46
*** ysandeep|brb is now known as ysandeep09:46
ianwhrw / AJaeger: yeah i'm going to have to take a look, because i thought we moved to copying the wheels via the executor but i must be wrong09:53
ianw ... it fails because no afs packages09:53
AJaegerianw: argh ;(09:54
ianwi think the publish jobs should copy the results from the builder to afs, rather than relying on the building node to do it09:56
ianwi *thought* we had done that, but maybe we only talked about it09:57
ianwso i think it's solvable.  also, getting afs clients working and sticking with the status quo isn't impossible either09:57
*** gtema has joined #opendev09:58
*** xiaolin has joined #opendev09:59
*** Meiyan has quit IRC10:04
openstackgerritIan Wienand proposed openstack/project-config master: Move wheel build role/job outside project-config
openstackgerritIan Wienand proposed openstack/project-config master: Reparent wheel-mirror jobs to new openstack-zuul-jobs versions
ianwAJaeger: ^ that is going to help.  we've already merged fixes to the build-cache jobs into o-z-j ... but the periodic jobs are using the old build-mirror jobs still in project-config.10:11
openstackgerritAndreas Jaeger proposed zuul/zuul-jobs master: Document twine_executable
AJaegerianw: thanks.10:25
openstackgerritMichał Piotrowski proposed openstack/project-config master: Adding elastic/kibana repository
*** gtema has quit IRC10:56
openstackgerritAndreas Jaeger proposed zuul/zuul-jobs master: Document twine_executable
*** ysandeep is now known as ysandeep|afk11:12
*** dpawlik has joined #opendev11:13
*** ysandeep|afk is now known as ysandeep11:36
*** gtema_ has joined #opendev11:41
hrwianw: thanks11:43
*** gtema_ has quit IRC11:57
openstackgerritJens Harbott (frickler) proposed zuul/zuul-jobs master: Fix download-logs template
spotzfungi diablo_rojo_phon - we're ok with either as we'll most likely be a regular meeting12:02
*** gtema_ has joined #opendev12:11
*** gtema_ has quit IRC12:25
*** gtema_ has joined #opendev12:41
*** gtema_ has quit IRC12:47
*** priteau has joined #opendev12:48
*** ykarel is now known as ykarel|afk12:51
fungii'll take care of gtema's pad renaming requests in a sec12:53
clarkbI'm making tea, but I'm in the meetpad room12:54
*** ravsingh has quit IRC12:55
fungiwas just sterilizing my earbuds12:56
fungijoining in a sec12:56
* mordred is in but sound is weird - grabbing a different computer12:56
fungihrm, the ptg schedule is still linking to a zoom url... what were we using for meetpad url? i'll update it12:57
AJaegeraccording to
* mordred has now spoken to AJaeger on meetpad, so considers the day successfulk12:59
fungiokay, it's updated12:59
*** xiaolin has quit IRC13:05
*** gtema_ has joined #opendev13:16
fungimy bandwidth must be really terrible13:17
fungii could barely hear anyone until i set the client to low-bandwidth mode, and even then people are cutting in and out13:17
fungieven low-definition video was causing significant audio issues for me13:18
mnasermeetpad doesn't even work for me on safari13:19
fungioh, yeah, i have no idea if it works for safari at all13:19
fungiand i was never able to get it to work with firefox13:19
fungionly chrome has worked for me13:19
clarkbchrom* are definitely recommended13:19
clarkbfirefox has been hit and miss13:19
fungiwell, chromium worked for me i mean13:19
*** gtema_ has quit IRC13:20
fungioh, might not be a bandwidth problem here. i'm running 6% idle across 4 cores. chromium is using basically all my cpu13:22
mnaseryeah it's eating up my battery hard too on chrome even on macos (aka hardware acceleration and friends are tehre)13:24
fungii had firefox running in the background with a few (unrelated) tabs open. closing that seemed to free up a little more cpu so my sound quality has improved slightly13:29
mnaser6% battery drop in 9 minutes13:34
clarkbmnaser: the mobile clients are apparently much more efficient if battery life is a major concern for you13:34
fungi#status renamed etherpad SDK-VictoriaPTG-Planning to sdk-victoriaptg-planning and API-SIG-VictoriaPTG-Planning to api-sig-victoriaptg-planning per gtema's request at 08:07 utc today in #opendev13:35
openstackstatusfungi: unknown command13:35
fungi#status log renamed etherpad SDK-VictoriaPTG-Planning to sdk-victoriaptg-planning and API-SIG-VictoriaPTG-Planning to api-sig-victoriaptg-planning per gtema's request at 08:07 utc today in #opendev13:35
openstackstatusfungi: finished logging13:35
corvuso/ sorry i'm late13:40
*** sgw has quit IRC13:40
mnasercorvus: we could hear you btw13:45
corvusyay i'm back13:45
corvusi accidentally used ff, now i'm on chromium13:46
fungii've noticed that chromium with just a jitsi-meet tab can consume basically all 4 cores of my i7-4500U (1.80GHz), load average hovers around 8. any time cpu idle drops to 8% the room goes entirely silent for me. i'm probably missing some hardware acceleration13:49
fungier, any time it drops to 0% i meant13:49
*** sgw has joined #opendev14:27
*** dtantsur|afk is now known as dtantsur14:48
AJaeger#thanks clarkb for all the note taking in the etherpad!14:49
openstackstatusAJaeger: Added your thanks to Thanks page (
AJaegermordred: regarding service sunsetting - is now ready?14:55
clarkbthanks everyone for joining14:55
AJaeger(that's part of the  puppet sunsetting)14:55
clarkbWe'll be back at 2300UTC starting with discussions on authentication of services14:56
AJaegerclarkb: I won't join the 2300 one unless I really cannot sleep ;)14:56
clarkbAJaeger: understood14:56
mordredAJaeger: yeah- I think so14:56
mordredAJaeger: also - I found 2 more:
AJaegercool. want to wait until PTG is over to push for them?14:57
AJaeger config-core, please review
*** ykarel|afk is now known as ykarel14:58
*** mlavalle has joined #opendev15:01
*** gtema_ has joined #opendev15:17
*** gtema_ has quit IRC15:21
clarkbAJaeger: on it15:42
*** rm_work has joined #opendev15:45
*** ykarel is now known as ykarel|away15:45
openstackgerritMerged zuul/zuul-jobs master: Fix download-logs template
openstackgerritMerged openstack/project-config master: Reparent wheel-mirror jobs to new openstack-zuul-jobs versions
openstackgerritMerged openstack/project-config master: Adding elastic/kibana repository
openstackgerritMerged openstack/project-config master: Add publish-to-pypi to ansible-config_template
openstackgerritMerged openstack/project-config master: Project creation request: airship/charts
*** mlavalle has quit IRC16:12
*** rchurch has joined #opendev16:17
*** dpawlik has quit IRC16:29
clarkbI've been dumping jicofo logs periodically just to see if there is any complains from jitsi meet and havent seen any yet16:30
clarkbwe seem to have reasonable distribution across jvbs as well16:30
*** dpawlik has joined #opendev16:30
clarkbwhat we haven't seen though is a flood of users, will be interesting to see as the week progresses if that happens or not16:30
*** mlavalle has joined #opendev16:32
*** dpawlik has quit IRC16:41
*** dpawlik has joined #opendev16:42
fungiso far i've been in three different meetpad rooms. maybe it's just the groups i choose to associate myself with, but all of them wanted to use meetpad instead of zoom16:52
fungithough i gather the release team wound up deciding to switch to zoom due to some audio connectivity issues16:52
clarkbbased on the numbers diablo_rojo gave us last week I think meetpad intentions are high16:52
*** dtantsur is now known as dtantsur|afk17:01
*** tkajinam has quit IRC17:11
*** tosky has quit IRC17:19
clarkbsmcginnis: if you end up with a hole in your schedule and want to debug your meetpad issues further I'd be happy to help with that17:26
clarkb(on our side it would be good to udnerstand the issues people are having)17:28
*** donnyd_ has joined #opendev17:28
mordredclarkb: right after our session gerrit released 3.217:32
clarkbmordred: ya I saw the email17:34
fungiit's almost like they were listening in17:34
*** ysandeep is now known as ysandeep|away17:35
*** dpawlik has quit IRC17:37
fungitripleo is planning to try meetpad tomorrow, supposedly they had 50 participants on zoom today, so they're going to see how it holds up for a larger group17:47
clarkbfungi: based on our testing for the ussuri celebration I think thats about the limti of what a single jvb can handle17:48
clarkbfungi: dependso n the amount of webcams i think17:48
clarkbdefinitely a good test17:48
*** DSpider has quit IRC18:02
mlavalleThe Neutron team intends to use meetpad. We are seeing zoom as a backup.... so how do I start the session in zoom?18:02
*** DSpider has joined #opendev18:03
clarkbmlavalle: I think you judt dial into zoom with the password18:03
clarkbbut we dont run zoom so not the best group to ask18:04
*** DSpider has quit IRC18:11
diablo_rojo_phonmlavalle: I sent an email about moderation to slaweq18:11
diablo_rojo_phonBut if you look at the ptgbot docs you can find the command to override the zoom URL to be a meetpad URL.18:11
diablo_rojo_phonAll the info wrt zoom is in the email I sent to slaweq being the moderator for neutron.18:13
diablo_rojo_phonmlavalle: ^18:14
fungialso got some jitsi-meet moderation questions in the diversity wg session, looks like the first person to join gets set as the mod, can add other participants as mods, mods can mute and eject participants18:14
fungibasically all you need18:14
fungiplaying it by ear to see if i feel up to joining the wednesday morning opendev session... that's midnight to 2am local where i am, so i may not make it that long18:18
mlavallediablo_rojo_phon: slaweq might not be able to moderate the meeting beause he is experiencing some health problems. he asked me to be ready to moderate in case he doesn't show up. Right now it is night time for him. If you could share that email with me, it would be great18:18
mlavalleI am just preparing myself for tomorrow, in case I have to step up to the plate ;-)18:19
diablo_rojo_phonmlavalle: I'll forward the email if you DM me what email you want it to be sent to.18:20
*** sshnaidm is now known as sshnaidm|afk18:20
clarkbfungi: I successfully tested jitsi meet moderation (muting of participant) today too18:21
clarkband ya first one to join is moderator18:21
mlavalleclarkb: that's good to know18:21
*** DSpider has joined #opendev18:23
mlavallediablo_rojo_phon: got it. Thanks!18:23
*** cloudnull5 has joined #opendev18:25
*** cloudnull has quit IRC18:26
*** cloudnull has joined #opendev18:28
*** cloudnull5 has quit IRC18:30
*** Eighth_Doctor has quit IRC18:41
smcginnisclarkb: I connected to the oslo call from my Fedora workstation and audio was fine. Earlier I tried from my work Windows laptop. It had connected fine, but I couldn't hear anything. Not sure if they could hear me or not.18:43
smcginnisQuite possible it's a weird work firewall or proxy issue.18:43
clarkbsmcginnis: the browser audio settings have tripped me up previously in similar ways18:44
clarkbsmcginnis: you want to ensure you've enabled things like your mic and webcam and ensure the browser has access to the correct devices18:44
clarkbthough taht was always for my input not audio output18:44
smcginnisI did get prompted and accepted access to mic and webcam when I first connected.18:45
clarkbsmcginnis: did the devices look correct ?18:45
smcginnisThen switched over which device was being used. Played test sound and at least that was coming through.18:45
clarkba lot of computers end up with multiple valid options18:45
smcginnisBut I assume that was just a locally played sound.18:45
clarkbalso were you using chrome/chromium?18:45
smcginnisI tried system default, headset, and speakerphone options for mic and audio playback. Didn't seem to make a difference.18:46
smcginnisOn Chrome.18:46
clarkbok so probably not a device issue like I've had in the past18:46
smcginnisI'm just going to stick with my Fedora machine for any future ones. That's my preference anyway. ;)18:49
fungii also found that on my systems with a bunch of different audio capture and sink options, sometimes the ones i wanted to use were muted at the alsa level by default and i had to un-mute them18:49
fungiadding pulseaudio also helped me, since it created merged virtual devices18:50
clarkboh one thing I ran into with chrome is if you plug in your device after chrome starts it won't see it18:50
clarkbI have to restart chrome if I plug in my hedset18:50
smcginnisNot this particular case, but a good thing to watch out for.18:50
fungii know pa has a longstanding reputation as an unholy nightmare, but recentish releases are a lot more stable18:51
mordredsmcginnis: also - fwiw - I have been tripped up several times now by my html5 video-autoplay-blocker. turns out streaming audio/video not so much when you have an autoplay blocker18:53
*** DSpider has quit IRC19:15
*** DSpider has joined #opendev19:15
*** Eighth_Doctor has joined #opendev19:20
*** mlavalle has quit IRC19:29
clarkbdoes anyone know why jvb02-04 have more graphs than meetpad01 and jvb01 in cacti? I'm assuming its related to the network issues the 01 hosts had, do we do a separate oid discovery step?19:29
*** DSpider has quit IRC19:30
mordredclarkb: I do not know the answer to that question19:31
*** mlavalle has joined #opendev19:32
clarkbthe major thing I think we are missing is the bits per second and packets per second measurements19:32
clarkbwe have memory, swap, and cpu though19:32
clarkboh and load19:32
*** DSpider has joined #opendev19:33
mordredclarkb: maybe the oid discovery stuff is triggered by something changing - but we had network issues so the next time the thing didn't change and so we didn't re-run it?19:38
clarkbya thats my hunch19:39
clarkbI bet we can rerun it if we know what it is19:39
clarkbor clone the list from jvb0219:39
mordredclarkb: you wanna see another weird thing?19:49
mordredclarkb: the tox-linters error here makes *NO SENSE*:
corvusclarkb: i think the script should re-run discovery, but maybe we don't run it as much these days19:56
corvusclarkb: it's pretty easy to fix up through the web though,  i'll do that in a sce19:56
clarkbcorvus: thanks19:57
corvusclarkb: the admin ui says we have 32 graphs for 2-4 and 7 graphs for meetpad and 0120:02
clarkbcorvus: that sounds about right20:02
clarkband I expect them to all have ~32 like 2-420:02
corvusclarkb: did you change the 'hostname' for jvb01?20:03
clarkbcorvus: yes20:03
corvusi think it may have been set back20:03
clarkbit had the same problem sa meetpad so I gave it the same updated treatment20:03
clarkb2-4 did not have this problem and were left alone20:03
corvusyeah, the create_graphs script changes it back20:06
clarkbmaybe we sould remove the AAAA record for them? or just leave them be until we can boot a new cacti? Not really great options when networking isn't networking20:06
corvusokay, i changed the 'hostname' to the ip address, clicked the little circle icon for all 3 data queries for the host, then ran "./" which created all except the 3 free space graphs, but then reset the hostname.  so then i updated the hostname again, re-ran the partition query, then clicked "create graphs for this host" then checked the 3 graphs for the partitions, then20:11
corvusclicked create.20:11
corvussame for jvb0120:11
corvusi'm guessing something about the sequencing in that script causes it to re-run the partition query after resetting the hostname20:11
corvusanyway, those hosts should be good starting now (give it 5-10m to collect data)20:12
clarkbyou think it will steady state now that the graphs are all created?20:12
corvusat least, as long as the data collection has been working20:12
corvusclarkb: we might be able to have our script set the device name and hostname explicitly20:13
corvusso just use "ipv4 if available" across the board20:13
corvusphp -q add_device.php --description="$HOST_NAME" --ip="$HOST_NAME" --template=$HOST_TEMPLATE_ID --version=2 --community="public"20:14
clarkbI think the risk with that is if we change the name in dns and don't rerun that script?20:14
corvusthat's what we're running now ^20:14
corvusclarkb: i think it gets run periodically20:14
corvusso yeah, actually "description" and "ip" are already set separately20:15
corvusso i think if we just change --ip="$HOST_NAME" to --ip="$LOOKUP_IPV4 if available else $HOST_NAME" we'll have what we want20:15
corvusi have to venture out for groceries, biab.20:16
mordredcorvus: be safe20:19
*** owalsh has quit IRC20:38
*** priteau has quit IRC20:38
*** owalsh has joined #opendev20:56
fungiinteresting discovery in a session just now... apparently if keyboard focus is on on the shared document, you can toggle soft mute of your input with the "m" key21:23
fungi(for jitsi-meet)21:24
clarkbfungi: there are a number of keyboard shortcuts but I don't think when I was taking notes earlier today that it unmuted when I typed m21:24
clarkbcould be more subtle for when the input is read yb jitsi ?21:25
fungier, i meant if keyboard focus is *not* on the shared document ;)21:25
fungithat is, taking notes in the pad shouldn't toggle your mute state21:25
fungi"t" apparently brings up statistics, which has been mildly annoying since ctrl+t,ctrl+t is how i swap active windows in my wm21:26
mordredfungi: just don't take any notes that have the letter m21:26
fungiwell, thankfully you don't have to actually avoid the letter m... though maybe we could go back to it being a ligature of nn21:27
clarkbthe good news is it seems to filter my loud keyboard well so even if every 'm' toggled mute we'd probably be ok21:30
mordredfungi: I thinking that nnaybe we write a pep8 rule to replace all use of the rascally letter21:31
fungithough apparently my workstation fan noise was not being well filtered earlier... appears my netbook is faring better for this21:31
*** diablo_rojo has joined #opendev21:41
*** tosky has joined #opendev22:32
ianwinfra-root: all dependencies for to finalise the wheel build jobs moving into openstack-zuul-jobs are merged now, appreciate it if we could have a look22:47
ianwworking on publishing via the executors now, ontop of that22:47
*** tkajinam has joined #opendev22:47
corvusit's ptg time again at
*** DSpider has quit IRC23:39
*** mlavalle has quit IRC23:46

Generated by 2.17.2 by Marius Gedminas - find it at!