Tuesday, 2022-03-15

*** dviroel|ruck is now known as dviroel|out00:04
*** rlandy|bbl is now known as rlandy01:05
*** rlandy is now known as rlandy|out01:05
*** pojadhav|afk is now known as pojadhav04:45
*** ysandeep|out is now known as ysandeep05:44
*** anbanerj is now known as frenzyfriday07:38
*** arxcruz|off is now known as arxcruz07:47
dpawlikianw: hey, could you resize flavor for logscraper01.openstack.org to have at least 4GB RAM ?07:52
dpawlikianw: it is on vexxhost ca-ymq-107:52
*** ysandeep is now known as ysandeep|mtg08:08
*** jpena|off is now known as jpena08:38
*** arxcruz is now known as arxcruz|brb09:20
*** ysandeep|mtg is now known as ysandeep09:22
*** ysandeep is now known as ysandeep|afk09:51
opendevreviewAlbin Vass proposed zuul/zuul-jobs master: Add flag to reset submodules  https://review.opendev.org/c/zuul/zuul-jobs/+/83375509:52
opendevreviewdaniel.pawlik proposed opendev/system-config master: DNM Add Opensearch Dashboard information  https://review.opendev.org/c/opendev/system-config/+/83326410:04
opendevreviewAlbin Vass proposed zuul/zuul-jobs master: Add flag to reset submodules  https://review.opendev.org/c/zuul/zuul-jobs/+/83375510:09
opendevreviewAlbin Vass proposed zuul/zuul-jobs master: Add flag to reset submodules  https://review.opendev.org/c/zuul/zuul-jobs/+/83375510:09
*** rlandy|out is now known as rlandy10:26
*** ysandeep|afk is now known as ysandeep10:54
*** pojadhav- is now known as pojadhav11:05
*** dviroel|out is now known as dviroel|ruck11:09
*** marios is now known as marios|biab11:23
*** pojadhav is now known as pojadhav|break11:35
*** arxcruz|brb is now known as arxcruz11:42
*** rlandy is now known as rlandy|mtg12:03
*** pojadhav|break is now known as pojadhav12:08
*** marios|biab is now known as marios12:12
opendevreviewMichal Nasiadka proposed openstack/diskimage-builder master: containerfile: Add support for setting network driver  https://review.opendev.org/c/openstack/diskimage-builder/+/83378413:18
*** rlandy|mtg is now known as rlandy13:27
*** dviroel|ruck is now known as dviroel|ruck|mtg14:16
dpawlikfungi: hey, could you resize flavor for logscraper01.openstack.org to have at least 4GB RAM ?14:35
dpawlikinstance is on vexxhost ca-ymq-114:35
fungidpawlik: i can try, just a moment14:35
dpawlikthanks you14:35
*** ysandeep is now known as ysandeep|dinner14:37
fungidpawlik: i've issued an `openstack server resize --flavor=v3-starter-2 logscraper01.openstack.org` (previous flavor was v2-highcpu-2)14:44
fungiv3-starter-2 claims to have 4096gb ram14:45
fungibut still the same vcpu count (2)14:45
dpawlikfungi: ok, thank you14:51
fungisure thing14:53
*** dviroel|ruck|mtg is now known as dviroel|ruck15:01
clarkbinfra-root if looks good to you then I think we can review https://review.opendev.org/c/opendev/system-config/+/828184/ positively. And maybe land that right after our meeting? or earlier I suppose if we have time15:34
*** dviroel|ruck is now known as dviroel|ruck|lunch15:35
fungichecks out for me. merging after the meeting is fine, though i'm happy doing it sooner if there's more feedback15:43
fungialso the theme is wildly different for the test server, at least in my browser15:43
clarkbhrm it looks the same for me?15:44
clarkbthe one difference I note is the "Home" link appears to have gone away in favor of the logo15:44
clarkbbut the theme itself seems to match production15:44
fungiit may be detecting my firefox preferences15:44
clarkbThere is a download file button if you open a file nwo too which is nice15:45
clarkbbut overall it looks pretty similar in my browser15:45
fungimy ff theme is set to "dark" so maybe that's somehow now taking effect on the new gitea15:46
clarkbah interesting15:46
fungimight have switched to using theme colors in the css15:46
clarkbI'm going to grab breakfast, but then I'll be ready to do some meetpad testing if 1600 ish still works for you15:48
*** ysandeep|dinner is now known as ysandeep15:49
*** dviroel|ruck|lunch is now known as dviroel|ruck16:03
clarkbfungi: ok I'm ready. https://meetpad.opendev.org/isitbroken16:05
fungijust a sec16:06
clarkbI left all my mics muted in pavucontrol then unmuted after I joined. Curious to see if you'll hear me. Then I guess disconnect and repeat from another browser (this is currently chrome)16:06
fungisorry, had to switch rooms16:08
NeilHanlonmorning folks. any objections to https://review.opendev.org/c/opendev/base-jobs/+/828436 ? I'm successfully using the rocky nodes in OSA CI runs16:13
clarkbNeilHanlon: no objections, thank you for helping to push that along16:18
NeilHanlonYou're welcome, thank you (and the rest of the gang) for your help and patience16:19
clarkbok we have learned some things about meetpad.16:42
clarkbFirst up is that firefox autoplay settings affect jitsi meet. Users need to allow audio and video in their firefox settings either for meetpad.opendev.org or globally by default. Otherwise they won't hear or see video from other participants. https://support.mozilla.org/en-US/kb/block-autoplay16:42
clarkbThis is confusing because autoplay settings are there to address news sites autplaying videos and being annoying. I wouldn't expect a webrtc client to be impacted but they are. In any case once fungi and I changed the autoplay settings in firefox to allow audio and video and did a refresh (the refresh is required) everything worked great in firefox16:43
clarkbThe other thing we found is that if the meeting drops to one participant and then another participant joins the original meeting member may have their mic muted. This isn't happening consistently though. I think the lesson here is double check your mute setting if you are trying to talk and no one can hear you16:44
clarkbChrom* seemed to work more consistently and intuitively16:44
clarkbdiablo_rojo_phone: ^ based on that it seems like meetpad is useable for the ptg just maybe less perfect than we'd like it to be. Any thoughts on the best way to communicate these gotchas? maybe it can go out in ptg comms?16:45
diablo_rojo_phoneYeah we can do that16:46
diablo_rojo_phoneor add it to the PTGBot site FAQ16:47
diablo_rojo_phoneor something16:47
clarkbalso note that it didn't unmute me as far as I could tell. It would only toggle from unmuted to muted. SO that behavior while annoying is also "safe"16:47
diablo_rojo_phoneDid you have to refresh?16:48
clarkbdiablo_rojo_phone: yes after setting the autoplay setting to allow audio and video a refresh is required16:49
*** marios is now known as marios|out16:57
*** ysandeep is now known as ysandeep|out16:58
*** jpena is now known as jpena|off17:14
*** arxcruz is now known as arxcruz|off17:55
clarkbI'm working on finding system-config changes to abandon and https://review.opendev.org/c/opendev/system-config/+/445616 is possibly one that we want to forward port to the new ansible var. I note that the ulimit reported by /proc/xyz/limits for the gerrit process is 8192 which is double what we set there as a value of 409618:56
clarkbinfra-root ^ maybe someone else can double check me on that and then push an update to modern system-config?18:57
mnasiadkaclarkb: if you’re willing to do another DIB review - I also need this to be pushed forward - https://review.opendev.org/c/openstack/diskimage-builder/+/83378419:10
opendevreviewMerged opendev/base-jobs master: Add rockylinux nodeset  https://review.opendev.org/c/opendev/base-jobs/+/82843619:45
clarkbI'm going to find lunch then I have a bunch of reviews to do looks like19:55
ianwUS Senate votes unanimously to make daylight savings time permanent - after reading i'm still unclear, does that mean everyone in the USA will no longer jump forward/back an hour?19:59
*** rlandy is now known as rlandy|mtg19:59
ianwhere it varies by state if they do it or not, i figured it would be the same20:00
opendevreviewMerged opendev/system-config master: grafana: proxy websockets  https://review.opendev.org/c/opendev/system-config/+/83288020:10
fungiianw: right now states can opt individually to switch to dst during the warmer months or stay on standard time year-round20:11
fungione of the lawmaking bodies for the usa's federal government has voted to make the dst timeshift into the new standard time and not switch back, but the same bill would still need to pass the house of representatives as well, and is also subject to the possibility of presidential veto20:13
*** dviroel|ruck is now known as dviroel|ruck|brb20:19
ianwoh, right, so it's a senate bill, not one that's already passed.  plenty of time for it to be argued over endlessly then :)20:22
ianwi think we'd call something like that a "private member's bill".  most of them don't go anywhere and i think are more about making a point20:25
clarkbthe fact that it was unanimous gives it some potential though. And ya wouldn't take effect until next year if the house passes it and gets it signed by the president20:28
clarkbI'd really like to stop changing time twice a year personally20:28
fungii'd prefer to just drop dst and use standard time year round, no idea why everyone pushes to make dst permanent instead. it's not like that will magically make it summer all year20:29
clarkbfungi: it is because you get sunlight after work for typical work hours20:30
clarkbthat is true for here except for about a month and a half in the witner where it is dark for all but like 8 hours20:30
fungiexcept it's the opposite. it will be dark even earlier in the winter if dst is made permanent20:31
fungithe proponents of year-round dst claim it's safer for children walking to school in the mornings since they won't have to walk in the dark20:31
fungihrm, no you're right, i've got it backwards20:33
fungii think20:33
clarkbya summer time means darker mornings20:33
fungitime is an illusion anyway20:33
clarkbso permanent summer time means more kids will go to school in the dark in winter. Mine already do20:33
clarkbthey can always shift school a half an hour to fix that though20:33
ianwhaha what if DST is abolished but everything starts switching it's start time to a winter/summer schedule :)20:34
clarkbianw: that wouldn't surprise me20:34
clarkbI think that is fine as long as they shift it enough taht waking and breakfast hours don't have to shift20:35
clarkbthat is where we struggle the most since our kids wake up about 40 minutes before we have to be out the door20:35
clarkbits actually kinda need to get the sunrise for the few days that it coincides exactly with the walk to school20:37
clarkbI have basically never in my life been awake early enough to enjoy them until i started walking my kids to school20:37
clarkbwow this gpg expiration comment might win for best code base note20:41
clarkb"sorry not my problem " :)20:41
clarkbianw: Should I +A https://review.opendev.org/c/zuul/zuul-jobs/+/829853 or let you approve it when convenient for monitoring?20:42
clarkbmostly concerned that if it breaks our prod runs may start having sads20:42
*** dviroel|ruck|brb is now known as dviroel|ruck20:57
opendevreviewMerged openstack/diskimage-builder master: containerfile: Add support for setting network driver  https://review.opendev.org/c/openstack/diskimage-builder/+/83378420:59
fungiclarkb: which expiration comment?21:28
clarkbfungi: the one about it expiring in 2106 so ianw won't be able to fix it :)21:28
fungioh, hah21:29
fungijokes on him, we'll solve the mortality problem in his lifetime21:29
fungijust so's he'll hafta fix it21:29
clarkbI think https://review.opendev.org/c/opendev/infra-specs/+/179713/ can be abandoned?21:43
fungii think so, yes21:43
fungialso the specs index could use revisiting21:43
clarkb++ one thing at a time though :)21:43
*** rlandy|mtg is now known as rlandy21:44
clarkbleaving https://review.opendev.org/c/opendev/system-config/+/249815 beacuse it calls out a file we can delete from system-config21:49
clarkbthis is like reading the history books and getting to be a detective at the same time21:49
*** dviroel|ruck is now known as dviroel|ruck|afk21:51
opendevreviewClark Boylan proposed opendev/system-config master: Remove byobu profile configuration  https://review.opendev.org/c/opendev/system-config/+/52942621:56
clarkbI'm going to go ahead and push up updates to some of the changes I've seen that look possibly important21:56
opendevreviewClark Boylan proposed opendev/system-config master: Increase Gerrit's diff timeout  https://review.opendev.org/c/opendev/system-config/+/26349022:01
opendevreviewClark Boylan proposed opendev/system-config master: Increase git open files and ulimit  https://review.opendev.org/c/opendev/system-config/+/44561622:03
ianwclarkb: thanks, yeah i can do 829853 when i can watch it, better safe22:04
opendevreviewClark Boylan proposed opendev/system-config master: Clean up more puppetmaster remnants  https://review.opendev.org/c/opendev/system-config/+/83393022:06
clarkbbah why didn't that use the existing change id?22:06
opendevreviewClark Boylan proposed opendev/system-config master: Clean up more puppetmaster remnants  https://review.opendev.org/c/opendev/system-config/+/24981522:07
*** rlandy is now known as rlandy|bbl22:27
opendevreviewClark Boylan proposed opendev/system-config master: Remove open-vm-tools from servers  https://review.opendev.org/c/opendev/system-config/+/59625422:29
clarkbfungi: did https://review.opendev.org/c/opendev/system-config/+/596419 stall out or did it happen or ? I'll let you decide what to do with that one22:32
fungiit was waiting for the kata community to agree to a schedule to rename it22:37
fungiwhich they never did22:37
fungii'll just abandon, it's easy enough to recreate if they decide to revisit the notion22:37
clarkb++ and you'll need to do it with the nasible stuff anyway22:38
opendevreviewClark Boylan proposed opendev/system-config master: Remove openstackwatch docs  https://review.opendev.org/c/opendev/system-config/+/50534222:44
clarkbI'm using the system-config-cleanup tag on these changes I'm updating since I'm not the owner so hard for me to track them otherwise: https://review.opendev.org/q/topic:system-config-cleanup+status:open22:46
clarkbfeel free to use the same topic if you end up updating things that are still relevant22:47
clarkbianw: thoughts on https://review.opendev.org/c/opendev/system-config/+/621847 ? Pretty sure that job is failing currently and when we make it work it fails again quickly :/22:50
clarkbinfra-root also you may want to double check that openstackwatch was removed as I've asserted via abandoning https://review.opendev.org/c/opendev/system-config/+/505343 but I looked and couldn't find evidence of it on the current server22:51
ianwclarkb: oh ... well the -devel job is failing ATM because of a real ansible bug https://github.com/ansible/ansible/pull/7678122:53
clarkbianw: looks like the ansible devs made a PR against your fork so that you can update your PR against ansible?22:54
clarkbI think my brain just melted a bit22:55
fungiopenstackwatch was indeed ripped out long, long ago22:55
clarkbfungi: thanks!22:55
fungithe last thing to go, if memory serves, was its cronjob, which did not get ported to ansible22:55
ianwyeah i need to motive myself to jump back into it22:55
clarkbianw: I just find that github workflwos to be odd whenit comes to stuff like this22:56
clarkbthe ultimate goal is to update ansible but we go round in circles instead22:56
ianwno need to convince me :)22:56
ianwi will abandon, i agree that time has proven that it's a good non-voting job22:56
clarkbianw: I suspect https://review.opendev.org/c/opendev/system-config/+/690465 is another that can be abandoned as we've been doing LE openstack.org certs successfully for a bit?23:03
ianwyep, gone :)23:04
opendevreviewClark Boylan proposed opendev/system-config master: Use deb-docker mirror for upload image too  https://review.opendev.org/c/opendev/system-config/+/68985823:06
clarkbfungi: ^ you've reviewed that one before. Also should be self testing if it builds any docker imges23:07
ianwfungi: i didn't notice a theme change you mentioned in the gitea update?23:08
clarkbianw: iirc he thought it was related to local dark theme firefox settings23:09
fungiianw: it seems like it may have picked up my firefox theme defaults where previous versions didn't23:09
clarkbgitea is trying to follow the browser theme I guess?23:09
ianwoh, interesting23:09
ianwit still shows as a jinja project for me in that test site, and python in the main site.  very minor but i'm not sure what it indexes differently23:09
clarkbfungi: ianw: I think eavesdrop is backed up now if you confirm can we abandon https://review.opendev.org/c/opendev/system-config/+/719484 and https://review.opendev.org/c/opendev/system-config/+/8578923:10
clarkbianw: I was wondering if the indexer got smarter23:10
clarkbor perhaps the opposite :)23:10
clarkbI think I'm going to call the system-config spring cleaning for now. I think those eavesdrop backup changes can go away if we confirm we're backing it up already. And then reviews/fixes for https://review.opendev.org/q/topic:system-config-cleanup+status:open are probably also a good idea23:13
ianwi know we've discussed the broken images on pages like https://opendev.org/zuul before but can't remember the conclusion23:13
ianwclarkb: yep, removed those with links to where we added it.  we always get there ... eventually :)23:14
fungithe orgs need avatar icons uploaded somehow23:14
fungior we may be able to stuff them into its db23:14
ianwthough it does have them on the test site23:15
clarkbianw: I think the conclusion was it wasn't related to the specific changes we were making and debugging would need to happen separately. I suspect that some upgrade did it and we didn't catch it at the tiem23:15
clarkbiirc it tries to fetch the resource and it is a 40423:15
clarkbso maybe something in the db?23:15
clarkbbut ya reuploading/resetting the avatars might correct it by overwriting whatever is broken?23:15
ianwyeah, i'll put it on the todo list.  only cosmetic but it does make it look a little "unfinished" 23:15
clarkbI think at one time mordred may have set some values23:16
clarkband I'm wondering if those didn't migrate properly through an upgrade23:16
clarkbianw: thank you for the review on the gitea upgrade. I think it is late enough in my day that I'd prefer to approve it tomorrow23:17
clarkbfungi: you'll be around early tomorrow to help monitor that?23:17
clarkbI won't object if others want to alnd it and watch it23:17
clarkbbut I'm less than an hour to needing to sort out dinner23:17
fungii'm interviewing someone between 13-14z and then have some errands to run after that, but will be around by 15:30z23:19
clarkbsounds good23:20
clarkbI don't even know what time of day my day starts anymore due to DST but its around then :)23:20
clarkbI think I managed to abandon ~110 changes today23:21
ianwon the abandon train, and again something i feel like we've discussed is https://review.opendev.org/c/opendev/gerrit/+/552288 ... opendev/gerrit is no longer used23:42
*** timburke_ is now known as timburke23:42
ianwi think we talked about it coming up before actual gerrit in search results23:42
ianw<fungi> once the system-config change lands, optional follow-ups are retiring the opendev/gerrit repo and ripping track-upstream support out of jeepyb23:45
*** dviroel|ruck|afk is now known as dviroel|ruck23:55
Clark[m]Ya we are 100% upstream Gerrit now23:55
Clark[m]And we can do local patches in our docker image builds. Don't need to carry the whole repo. I think the historical context there may be interesting and there are a lot of branches which makes retirement a bit weird23:56

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!