| *** dhill is now known as Guest7200 | 01:27 | |
| *** elodilles is now known as elodilles_pto | 08:23 | |
| corvus | i have a theory why the old images weren't deleted; i think it's because they predate some new fields that zuul uses for image management. i'm going to hit the "delete" button in the web ui, then build some more new images and see if the new-old ones cycle out automatically. | 14:38 |
|---|---|---|
| corvus | also, we need to add the 'state' field to the web ui :) | 14:39 |
| opendevreview | Merged zuul/zuul-jobs master: Use consistent registry type var name across roles https://review.opendev.org/c/zuul/zuul-jobs/+/881524 | 14:42 |
| opendevreview | Jeremy Stanley proposed opendev/bindep master: DNM: Test PBR beta release https://review.opendev.org/c/opendev/bindep/+/940258 | 15:28 |
| opendevreview | Jeremy Stanley proposed zuul/zuul-jobs master: ensure-pyproject-build: Use the pip module https://review.opendev.org/c/zuul/zuul-jobs/+/940314 | 16:01 |
| clarkb | fungi: are we in a position to start landing more bindep changes or do you think we should wait for pbr to get a release or two out the door first? (just not sure what we feel is the right order of operations here) | 16:44 |
| fungi | i'm still fiddling a bit | 16:50 |
| fungi | clarkb: my observation with the latest iteration of 940258 is that pyproject-build isolated build venvs aren't getting setuptools preinstalled by default on older pythons, so we actually need that dependency change to drop the >=3.12 environment marker for older platforms | 16:59 |
| clarkb | thats odd, docs it not use either venv or virtualenv? | 17:00 |
| clarkb | fungi: or we need to keep setuptools in the build system dependency list | 17:01 |
| clarkb | I think I'm fine with either option. Dropping the env marker is probably most efficient in terms of ease of use | 17:01 |
| *** priteau is now known as Guest7267 | 17:04 | |
| *** priteau2 is now known as priteau | 17:04 | |
| fungi | clarkb: https://github.com/pypa/build/blob/main/src/build/env.py#L202 | 17:09 |
| fungi | it explicitly instructs the installer to omit setuptools and wheel from the venv | 17:09 |
| fungi | also uninstalls it if it ends up in there: https://github.com/pypa/build/blob/main/src/build/env.py#L240 | 17:10 |
| fungi | and then the legacy fallback installs setuptools always for non-pep-517 cases: https://github.com/pypa/build/blob/main/src/build/_builder.py#L35 | 17:13 |
| fungi | so yeah, i think that perfectly explains my observations | 17:15 |
| corvus | clarkb: https://stackoverflow.com/questions/20414669/nginx-add-headers-when-returning-400-codes may be the thing we need | 19:36 |
| corvus | add "always" to the add_header | 19:36 |
| clarkb | thanks | 19:36 |
| fungi | infra-root: JayF: as discussed in today's opendev meeting, i've disabled bounce processing for the openstack-discuss mailing list for now, and will raise my present concerns on the mailman-users mailing list to see if it can be improved to the point where it's usable for us there | 20:01 |
| JayF | cool; I hope they're able to figure it out b/c I don't want your efforts here to be for naught :/ | 20:02 |
| clarkb | I'm just hoping that mailman doesn't say something like "well those mail systems are clearly broken and evicting their membership is a feature not a bug" | 20:10 |
| fungi | the main concern i have is that it sends copies of messages (bounces in verp probes, held messages in moderator notifications), and when those included messages contain spam or content likely to be misidentified as spam it creates problems | 20:15 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Update graphite to send CORS headers even on 400 responses https://review.opendev.org/c/opendev/system-config/+/940328 | 20:16 |
| clarkb | fungi: ya verp probes for valid addresses seem like something that should haev neutral content | 20:16 |
| clarkb | hopefully they don't consider that a feature and instead think of it as a bug | 20:16 |
| fungi | well, at a minimum it could be made configurable | 20:17 |
| corvus | clarkb: what about adding always to allow-headers? i do think x-grafana-device-id is sent for that options request | 20:18 |
| fungi | i get that it has the potential to reduce work for list admins if users can find out for themselves why their messages bounced, but it comes at the risk of making the verp probes useless and possibly disclosing more than it needs to in the case of bulk rejections | 20:18 |
| fungi | in the case of moderator notices, forwarding copies of potential spam has the potential to get the list server flagged as a spam source by the providers used by moderators | 20:20 |
| Clark[m] | corvus: ack I can add always to all four | 20:28 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Update graphite to send CORS headers even on 400 responses https://review.opendev.org/c/opendev/system-config/+/940328 | 20:47 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Deploy mariadb for etherpad from opendev's quay mirror https://review.opendev.org/c/opendev/system-config/+/940336 | 21:00 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Upgrade etherpad to v2.2.7 https://review.opendev.org/c/opendev/system-config/+/940337 | 21:00 |
| clarkb | we should probably hold a node for the 2.2.7 update but I want to make sure it builds and runs cleanly against ci before I bother | 21:01 |
| fungi | infra-root: JayF: as a followup, here's the thread i started on mailman-users about the trouble with verp probe messages: https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/thread/OFTHDKATNSDCXO6VDNDFFVR6BI37R64J/#OFTHDKATNSDCXO6VDNDFFVR6BI37R64J | 21:10 |
| clarkb | looks great thank you for putting that together | 21:16 |
| fungi | also the markdown formatter they have turned on in their hyperkitty always seems to mistake my standard signature separator ("-- ") as an underline and turns the last paragraph of the message into a section heading | 21:18 |
| fungi | i try to stick in a blank line before the signature separator when sending to mailman-users, but forgot this time | 21:18 |
| clarkb | https://review.opendev.org/c/opendev/system-config/+/940328 this change passed testing though I'm not sure how robust our graphtie testing is. It probably does fall into the category of thing where landing it and checking on the held node is the easiest way forward though. If it breaks impact should be minimal | 21:19 |
| clarkb | but also I'm about to go outside because it si warm and sunny and that won't last | 21:19 |
| fungi | you should do that thing instead, yes | 21:20 |
| fungi | take advantage of the good weather while you can | 21:20 |
| corvus | the only images listed on https://zuul.opendev.org/t/opendev/image/debian-bullseye are from today \o/ | 23:24 |
| corvus | i think that means all the new image management functionality is working as expected. it's probably safe to start adding periodic image build jobs now. | 23:25 |
| fungi | awesome! | 23:26 |
| clarkb | woot | 23:35 |
| opendevreview | Clark Boylan proposed opendev/system-config master: DNM force etherpad failure to hold node https://review.opendev.org/c/opendev/system-config/+/840972 | 23:37 |
| clarkb | I've put an autohold in place for ^ so that we can sanity check etherpad 2.2.7 | 23:38 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Switch refstack to opendevmirror hosted mariadb image https://review.opendev.org/c/opendev/system-config/+/940347 | 23:40 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Switch keycloak to opendevmirror hosted mariadb image https://review.opendev.org/c/opendev/system-config/+/940348 | 23:42 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Switch mailman3 to opendevmirror hosted mariadb image https://review.opendev.org/c/opendev/system-config/+/940349 | 23:45 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Switch standalone mariadb to opendevmirror hosted mariadb image https://review.opendev.org/c/opendev/system-config/+/940350 | 23:47 |
| clarkb | corvus: ^ that one will have a small impact on zuul when deployed I think | 23:47 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Switch Gerrit to opendevmirror hosted mariadb image https://review.opendev.org/c/opendev/system-config/+/940351 | 23:49 |
| clarkb | I think that is everything using mariadb | 23:49 |
| clarkb | fungi: enjoy your evening but maybe first thing tomorrow we can land the graphite nginx config updaet and then work through ^ | 23:58 |
| clarkb | fungi: also we can land the exim config update too | 23:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!