| opendevreview | Julia Kreger proposed opendev/glean master: Detect ironic, and skip partition mount. https://review.opendev.org/c/opendev/glean/+/917278 | 00:36 |
|---|---|---|
| opendevreview | Merged openstack/diskimage-builder master: Add Ubuntu 24.04 (noble) build to testing https://review.opendev.org/c/openstack/diskimage-builder/+/915915 | 07:51 |
| opendevreview | Jan Marchel proposed openstack/project-config master: Add new repository for NebulOuS testing data https://review.opendev.org/c/openstack/project-config/+/916876 | 13:00 |
| frickler | mnaser: did you ever get to look at the unplanned shutdown of review.opendev.org? | 13:57 |
| fungi | from roughly a month ago, 2024-03-26 according to my notes | 14:02 |
| fungi | dmesg indicates the reboot occurred at 03:40:41 utc, but no idea how accurate the clock was that early in the boot process | 14:03 |
| frickler | infra-root: I'm in the process of setting up the new afs volume mirror.ceph-deb-reef, but I'm a bit confused by the acl part https://docs.opendev.org/opendev/system-config/latest/afs.html#mirrors | 14:18 |
| frickler | from what I see for existing volumes, we don't use a specific account, but service.reprepro for all of these? | 14:19 |
| fungi | good question. i mainly use that document as a source of hints to remind me what the various commands and subcommands are, but for things like quotas and acls i tend to compare to other existing volumes that have a similar purpose | 14:22 |
| frickler | that would have been my idea, too. so I guess I'll go ahead with that and see how it works out | 14:23 |
| fungi | since we create volumes for a variety of different purposes, and for mirrors specifically we have a variety of scripts relying on different tools and grouped in different ways, i'm not surprised that the doc might have e.g. been written based on an rsync mirror and reprepro mirrors are done a different way | 14:25 |
| *** dhill is now known as Guest2598 | 14:38 | |
| clarkb | fungi: if you have time https://review.opendev.org/c/opendev/glean/+/916952 is part of the xenial cleanup work and is also a logical followon to our recent glean udpates (basically get the ci jobs to match something closer what our priorities are) | 14:54 |
| clarkb | did zigo and tkajinam's emails arrive out of order for anyone else? I suspect this is because tkajinam was copied directly so wasn't affected by delay in lists deliveries but then zigo's mail server delivered the email to lists.openstack.org much later which was then compounded by the lists.openstack.org delivery delays? | 14:58 |
| clarkb | or maybe it is a problem with my mailserver and lists.openstack.org is only having trouble deliverying to me? | 14:59 |
| zigo | clarkb: My fault, my mail went moderated because using the wrong email address, sorry. | 14:59 |
| clarkb | oh that would explain it. Just making sure we don't have any problem to investigate on our side. Thank you for the clarification | 15:00 |
| zigo | Is there a way to get my personal email address whitelisted, even though it's not subscribed? | 15:00 |
| zigo | Because that happens to me often ... :/ | 15:00 |
| clarkb | you may be able to associate it with your mailman3 account, but I'm not sure | 15:00 |
| clarkb | it does look like you can have multiple addresses for each mailman3 user account, but that addresses are explicitlypart of lists and not just the account | 15:01 |
| clarkb | so you'd have to add the email to the account then add the email to the list? | 15:01 |
| fungi | zigo: one option is to subscribe it but set the subscription's delivery method to disabled | 15:07 |
| zigo | fungi: Ok, thanks. | 15:10 |
| fungi | i'm headed out to lunch, back soon | 15:10 |
| zigo | BTW, "subscriber only post" is a bad idea. It's annoying well behaving persons when posting, only because of bad spammers, which should never be done. | 15:11 |
| clarkb | until you're the person who has to combat the spam | 15:12 |
| zigo | I respectfully don't agree! :) | 15:13 |
| JayF | Yeah you have no idea how many crap messages fungi and I (mostly fungi) have to reject outta the queue | 15:13 |
| JayF | The list would be unusable if we allowed unrestricted non-subscriber mosting | 15:13 |
| JayF | **posting | 15:13 |
| * zigo wonders .oO( how the debian listmasters are doing ) | 15:14 | |
| clarkb | the upstream mailing lists for mailman even require first post moderation. Which has been discussed here to help mitigate against spam coming directly via the web ui | 15:20 |
| clarkb | but that seems to have remained infrequent enough | 15:20 |
| clarkb | frickler: all of the existing noble related changes have merged right? | 15:22 |
| clarkb | looks like both dib changes have landed and I think the project-config change for element updates also landed | 15:22 |
| frickler | clarkb: I think so, yes | 15:34 |
| clarkb | thanks for confirming. Want to make sure I'm not letting thigns slip after the weekend | 15:34 |
| frickler | tests on the ceph reef patch are passing now. I would like to approve it now and assume that the mirror is small enough that manual first-seeding isn't necessary https://review.opendev.org/c/opendev/system-config/+/916552 | 15:47 |
| clarkb | looks like the other ceph volumes are under 5GB I agree it should be fine to let that run automatically without manual seeding | 15:50 |
| opendevreview | Merged opendev/glean master: Update zuul config to drop xenial jobs https://review.opendev.org/c/opendev/glean/+/916952 | 16:19 |
| fungi | zigo: perhaps surprisingly, we mostly catch the spam posts because they're not from subscribers, and catch the ones spoofed from subscribers because they're over the default length limit. i agree a more robust solution would involve something like spamassassin and clamav as early filters in mailman before handing off to mailman, but that does require someone to take the time to set up and | 16:51 |
| fungi | maintain it | 16:51 |
| JayF | (and right now, until I volunteered to help, fungi was the only person even doing the moderation in the panel ... so evidence points to little interest in helping with maintenance of the ML) | 16:52 |
| fungi | actually the upstream mailman lists require posts to come from subscribers, but *in addition* they screen initial posts from those new subscribers to make sure they're not spam, and then switch them from moderated to allowed. they immediately reject posts from non-subscribers | 16:52 |
| fungi | zigo: the debian mailing lists frontend everything with separate spam filters, which has the alternative downside of annoying legitimate senders who send something with a disallowed pattern inside their message content | 16:54 |
| AlbinVass[m] | Is it possible to use opendevs zuul for personal GitHub projects? I wanna use zuul but I don't want to host it myself :) | 16:56 |
| clarkb | AlbinVass[m]: no we are not a general purpose ci system for github | 16:58 |
| opendevreview | Merged opendev/system-config master: Ubuntu 22.04: Add ceph quincy and reef to mirror https://review.opendev.org/c/opendev/system-config/+/916552 | 16:58 |
| clarkb | we do support third party testing of things hosted in opendev against things hosted in github but that requires an opendev project host the jobs and pipeline definitions and work through the problems in github if/when they arise | 16:58 |
| fungi | AlbinVass[m]: the original summary of our position on that can be found in https://lists.openstack.org/pipermail/openstack-infra/2019-January/006269.html | 17:00 |
| AlbinVass[m] | clarkb: yeah I thought so. Worth asking though, but I should probably set it up myself. The uptime probably won't be as good | 17:00 |
| fungi | the "kata experiment" mentioned in that post was wound down about a year later when the project involved seemed disinterested in helping with opendev's side of things | 17:01 |
| fungi | our (albeit limited) experience has been that users who don't have a vested interest in interacting with the code review and code hosting platforms we maintain are also disinterested in assisting with the ci/cd system we've integrated with them | 17:02 |
| fungi | and tend to be purely consumers, not collaborators | 17:03 |
| fungi | mirror config deploys are failing because (unused) mirror01.sjc1.vexxhost is unavailable | 17:08 |
| fungi | openstack server show indicates the instance's status is "active" | 17:10 |
| fungi | openstack console log show doesn't include anything useful (ends with a login prompt) | 17:12 |
| fungi | i'll try rebooting it from the api | 17:13 |
| fungi | doing `openstack server reboot` doesn't seem to have made it reachable, even though `openstack console log show` indicates that it did reboot | 17:20 |
| fungi | mnaser: guilhermesp: ^ anything odd going on in vexxhost sjc1? | 17:20 |
| mnaser | fungi: do you have vm uuid handy? | 17:21 |
| fungi | mnaser: a23a4761-58f3-4237-a2da-30717f96a6fa | 17:22 |
| mnaser | ah, I think that HV got stuck somehow a few days ago. | 17:23 |
| fungi | mnaser: thanks for checking it! | 17:23 |
| mnaser | fungi: how about now ? | 17:24 |
| fungi | mnaser: still seeing "ssh: connect to host mirror01.sjc1.vexxhost port 22: No route to host" | 17:25 |
| mnaser | You might need to give it a reboot it probably timed out on dhcp attempts | 17:25 |
| fungi | ping to its ipv6 address is returning "Destination unreachable: Address unreachable" | 17:25 |
| fungi | okay, thanks, will reboot it again now | 17:26 |
| fungi | yay, another reboot and it's responding to ping again | 17:27 |
| fungi | yep, i can ssh into it again | 17:28 |
| fungi | thanks mnaser! | 17:28 |
| mnaser | No problem. Sorry about that | 17:28 |
| fungi | no worries, just weren't sure what was going on | 17:28 |
| clarkb | Looking at more xenial cleanup stuff openstack-python-jobs runs on bionic for python2.7 so that one isn't impacted thankfully | 17:32 |
| clarkb | https://codesearch.opendev.org/?q=openstack-python35-jobs&i=nope&literal=nope&files=&excludeFiles=&repos= shows the reach of the python35 jobs on master though. A couple of those jump out to me: the already mentioned devstack-gate and bashate | 17:34 |
| clarkb | bashate is no longer developed on opendev right? I Guess no one took the time to retire it after they moved it to github? | 17:34 |
| clarkb | oh and openstack/releases is using that template | 17:34 |
| clarkb | though I don't see the jobs reporting to releases so maybe it isn't actually used? | 17:35 |
| clarkb | oh its in a comment so ya its just example code in releases. Nevermind that one | 17:36 |
| clarkb | As a heads up I've had a meh start to the week. GOing to try and reboot my day over lunch and go for a bike ride and generally try to reset things | 18:32 |
| fungi | sounds good, not a ton of stuff going on anyway. good luck! | 18:35 |
| opendevreview | Julia Kreger proposed openstack/diskimage-builder master: ConfigDrive: Disable cloud-init auto dhcp by default https://review.opendev.org/c/openstack/diskimage-builder/+/899886 | 18:39 |
| opendevreview | Clark Boylan proposed opendev/afsmon master: Update afsmon python testing to modern versions https://review.opendev.org/c/opendev/afsmon/+/917470 | 18:40 |
| clarkb | that repo would be an easy one to move to the opendev tenant | 18:41 |
| fungi | oh, yeah probably so | 18:42 |
| clarkb | I think for many of the other projects with python35 templates in their zuul confings I may just propose we drop them from zuul's tenant config | 18:42 |
| clarkb | spot checking them many haven't had any updatse since the opendev reorg and assocaited .gitreview file updates | 18:43 |
| frickler | seems the sjc1 issue also caused our periodic jobs fail early at infra-prod-base for the last three days https://zuul.opendev.org/t/openstack/builds?job_name=infra-prod-base&project=opendev/system-config , only confirmed in the log for the last run though | 18:43 |
| fungi | makes sense | 18:46 |
| fungi | those should hopefully get back to working again | 18:46 |
| frickler | do we have ppl from starlingx around to ping about https://zuul.opendev.org/t/openstack/config-errors?project=starlingx%2Fzuul-jobs&skip=0 ? | 19:21 |
| fungi | slittle: ^ | 19:44 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Update etherpad to v2.0.3 https://review.opendev.org/c/opendev/system-config/+/914119 | 21:18 |
| opendevreview | Clark Boylan proposed opendev/system-config master: DNM force etherpad failure to hold node https://review.opendev.org/c/opendev/system-config/+/840972 | 21:18 |
| clarkb | there is a 2.0.3 release now so I've udpated that stack to make it mergeable. I have also rotated the autohold for the DNM change | 21:20 |
| clarkb | now to sort out a meeting agenda for tomorrow | 21:20 |
| clarkb | ok I'ev got my edits into the meeting agenda | 21:28 |
| clarkb | I'll send that out later today when others have had a chance to edit | 21:28 |
| opendevreview | Clark Boylan proposed opendev/afsmon master: Update afsmon python testing to modern versions https://review.opendev.org/c/opendev/afsmon/+/917470 | 21:43 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!