| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Retire Fedora 31 for 32 https://review.opendev.org/751975 | 01:31 |
|---|---|---|
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: only run docker-setup.yaml when installed https://review.opendev.org/747062 | 01:31 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: update-json-file: add role to combine values into a .json https://review.opendev.org/746834 | 01:31 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: Linaro MTU workaround https://review.opendev.org/747063 | 01:31 |
| *** Topner has joined #opendev | 02:33 | |
| *** Topner has quit IRC | 02:37 | |
| openstackgerrit | wu.shiming proposed openstack/diskimage-builder master: Update imp module to importlib https://review.opendev.org/752011 | 02:46 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-kubernetes: workaround slow account creation https://review.opendev.org/752184 | 03:52 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-kubernetes: workaround slow account creation https://review.opendev.org/752184 | 04:11 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Retire Fedora 31 for 32 https://review.opendev.org/751975 | 04:11 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: only run docker-setup.yaml when installed https://review.opendev.org/747062 | 04:11 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: update-json-file: add role to combine values into a .json https://review.opendev.org/746834 | 04:11 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: Linaro MTU workaround https://review.opendev.org/747063 | 04:11 |
| *** Topner has joined #opendev | 04:34 | |
| *** Topner has quit IRC | 04:38 | |
| *** ykarel|away has joined #opendev | 04:43 | |
| *** diablo_rojo has quit IRC | 04:44 | |
| *** ykarel|away is now known as ykarel | 04:46 | |
| *** fressi has joined #opendev | 04:50 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: Linaro MTU workaround https://review.opendev.org/747063 | 04:59 |
| *** fressi has quit IRC | 05:17 | |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: Linaro MTU workaround https://review.opendev.org/747063 | 05:42 |
| openstackgerrit | wu.shiming proposed openstack/diskimage-builder master: Update imp module to importlib https://review.opendev.org/752011 | 05:48 |
| *** mrunge_ is now known as mrunge | 06:00 | |
| *** lpetrut has joined #opendev | 06:15 | |
| *** qchris has quit IRC | 06:21 | |
| openstackgerrit | wu.shiming proposed openstack/diskimage-builder master: Update imp module to importlib https://review.opendev.org/752011 | 06:27 |
| *** qchris has joined #opendev | 06:33 | |
| openstackgerrit | wu.shiming proposed openstack/diskimage-builder master: Update imp module to importlib https://review.opendev.org/752011 | 06:42 |
| *** slaweq has joined #opendev | 06:48 | |
| *** fressi has joined #opendev | 06:58 | |
| *** ysandeep|away is now known as ysandeep | 07:05 | |
| *** hashar has joined #opendev | 07:07 | |
| *** andrewbonney has joined #opendev | 07:35 | |
| *** tosky has joined #opendev | 07:43 | |
| *** DSpider has joined #opendev | 07:47 | |
| *** guillaumec has quit IRC | 07:51 | |
| *** guillaumec has joined #opendev | 07:52 | |
| *** moppy has quit IRC | 08:01 | |
| *** moppy has joined #opendev | 08:01 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Make iscsi-boot element support centos 8 https://review.opendev.org/747878 | 08:11 |
| *** dtantsur|afk is now known as dtantsur | 08:33 | |
| openstackgerrit | Merged openstack/diskimage-builder master: update various gentoo bits https://review.opendev.org/747502 | 08:34 |
| openstackgerrit | Merged openstack/diskimage-builder master: Copy apt gpg keys directly into trusted.gpg.d https://review.opendev.org/747810 | 08:34 |
| *** ykarel_ has joined #opendev | 08:43 | |
| *** ykarel has quit IRC | 08:46 | |
| *** ykarel_ is now known as ykarel | 08:47 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Move centos python3 installation after RHEL subscription https://review.opendev.org/749773 | 08:50 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: update-json-file: add role to combine values into a .json https://review.opendev.org/746834 | 08:59 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: ensure-docker: Linaro MTU workaround https://review.opendev.org/747063 | 08:59 |
| *** hashar has quit IRC | 09:51 | |
| openstackgerrit | Witold Bedyk proposed openstack/project-config master: Revert "Use only noop jobs for openstack/monasca-transform" https://review.opendev.org/752207 | 09:52 |
| *** fressi has quit IRC | 10:25 | |
| *** Topner has joined #opendev | 10:36 | |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Partial address ansible-lint E208 https://review.opendev.org/748480 | 10:36 |
| *** Topner has quit IRC | 10:41 | |
| *** slaweq has quit IRC | 10:41 | |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Add mode to packer https://review.opendev.org/752213 | 10:50 |
| *** slaweq has joined #opendev | 10:54 | |
| *** elod has quit IRC | 11:18 | |
| *** elod has joined #opendev | 11:18 | |
| *** fressi has joined #opendev | 11:25 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul-jobs master: ensure-haskell-stack: ensure all requirements are installed https://review.opendev.org/752225 | 12:00 |
| *** slaweq has quit IRC | 12:01 | |
| *** slaweq has joined #opendev | 12:30 | |
| openstackgerrit | Lon Hohberger proposed openstack/diskimage-builder master: rhel-common: Provide method to select module streams https://review.opendev.org/747732 | 12:31 |
| openstackgerrit | Lon Hohberger proposed openstack/diskimage-builder master: rhel-common: Provide method to select module streams https://review.opendev.org/747732 | 12:40 |
| openstackgerrit | Lon Hohberger proposed openstack/diskimage-builder master: rhel-common: Provide method to select module streams https://review.opendev.org/747732 | 12:43 |
| *** ricolin has quit IRC | 13:20 | |
| *** sshnaidm is now known as sshnaidm|afk | 13:32 | |
| *** ykarel_ has joined #opendev | 13:35 | |
| *** ykarel has quit IRC | 13:37 | |
| smcginnis | clarkb, ianw, fungi: Any update on the pypi index issues? | 13:59 |
| frickler | smcginnis: I assumed pypi fixed things earlier tonight, do you still see issues after that? | 14:00 |
| smcginnis | frickler: Thanks. Just coming online now, so I haven't seen anything. Wanted to make sure before I started approving any patches. | 14:02 |
| *** ykarel_ is now known as ykarel| | 14:04 | |
| *** ykarel| is now known as ykarel | 14:04 | |
| *** owalsh has quit IRC | 14:12 | |
| fungi | smcginnis: last word was that they took the problem bandersnatch mirror offline by 22:30 | 14:15 |
| ildikov | hi all | 14:16 |
| fungi | but that fastly can continue to serve "stale" content for up to 24 hours | 14:16 |
| *** ricolin has joined #opendev | 14:16 | |
| ildikov | I saw this error on a StarlingX governance repo patch: "AttributeError: 'FlakesChecker' object has no attribute 'CONSTANT'" | 14:16 |
| smcginnis | Great. Nice work all tracking that down. That was a really odd one. | 14:16 |
| ildikov | and wanted to check if this is something that others have seen to? | 14:16 |
| smcginnis | ildikov: I think that's a side effect of moving to running on focal nodes, and therefore running under py3.8. | 14:17 |
| smcginnis | ildikov: Likely the hacking version needs to be raised. | 14:17 |
| fungi | yeah, i think the solution has usually been to switch to newer pyflakes (maybe via hacking upgrade if they're also using hacking) | 14:17 |
| ildikov | I only saw hacking in the test-requirements.txt: hacking!=0.13.0,<0.14,>=0.12.0 # Apache-2.0 | 14:18 |
| ildikov | so I figured I come here and ask for suggestions on what to bump :) | 14:18 |
| fungi | ildikov: can you link the build result page? i'm happy to take a look and see what's dragging it in where | 14:19 |
| ildikov | fungi: https://zuul.opendev.org/t/openstack/build/69abb77fb34e4159893176bd48d3fad9 | 14:19 |
| frickler | ildikov: using hacking >= 3.2.0 is the easiest solution I think, did that for neutron-dynamic-routing https://review.opendev.org/750779 | 14:22 |
| AJaeger | ildikov: that hacking is ancient, so going with 3.2.0 as proposed by frickler is the right way forward | 14:24 |
| smcginnis | ildikov: hacking locks down the version of the various other linters it pulls in, so hopefully you should just need to update that version and that will take care of it. | 14:24 |
| ildikov | cool, thanks!! | 14:25 |
| ildikov | I will put a patch and check | 14:25 |
| smcginnis | ildikov: Just a warning, going from that very old version of hacking to something newer will likely pick up a lot of hacking issues. | 14:25 |
| ildikov | I will follow the approach in frickler's patch | 14:25 |
| smcginnis | So there will probably be some cleanup work to do along with the requirements change. | 14:25 |
| smcginnis | AJaeger had a lot of patches taking care of issues like that, so you can probably find some good examples out there if you need it. | 14:26 |
| fungi | yeah, my preferred workflow, if it helps, is to raise to the latest version of your static analyzers, run them to see what they start complaining about, then put all those new rules into an ignore list. that'll get you onto the new versions with minimal pain. after that you can peel back any new rules you like one at a time fixing just the problems they expose | 14:28 |
| *** priteau has joined #opendev | 14:30 | |
| ildikov | the repo is very plain simple so I would hope there won't be a massive amount of things to fix, while I'm sure there are skeletons in there | 14:35 |
| *** fdegir has quit IRC | 14:36 | |
| *** lpetrut has quit IRC | 14:38 | |
| fungi | ildikov: yeah, i mean we just bumped hacking in openstack/election yesterday and all we needed to change to make it happy was this: https://review.opendev.org/#/c/734854/2/openstack_election/cmds/ci_check_all_candidate_files.py | 14:57 |
| fungi | granted we were only updating from 2.0.x to 3.1.x | 14:57 |
| ildikov | nice! | 14:59 |
| *** ykarel is now known as ykarel|away | 15:12 | |
| *** fressi has quit IRC | 15:12 | |
| *** mlavalle has joined #opendev | 15:16 | |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Add mode to packer https://review.opendev.org/752213 | 15:18 |
| *** ysandeep is now known as ysandeep|away | 15:22 | |
| *** owalsh has joined #opendev | 15:24 | |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Add managed jobs to periodic https://review.opendev.org/748682 | 15:24 |
| clarkb | smcginnis: frickler fungi If we see more issues after the 24 hour mark we will probably want to track those down but until then I think we did what we could | 15:26 |
| openstackgerrit | zbr proposed zuul/zuul-jobs master: Add managed jobs to periodic-weekly https://review.opendev.org/748682 | 15:31 |
| *** ykarel|away has quit IRC | 15:37 | |
| zbr | clarkb: can we create the weekly pipeline now? https://review.opendev.org/#/c/748706/ | 15:39 |
| *** slaweq has quit IRC | 15:48 | |
| *** slaweq has joined #opendev | 15:50 | |
| clarkb | zbr: I think that is a better question in #zuul | 15:58 |
| *** Tengu has quit IRC | 16:05 | |
| *** Tengu has joined #opendev | 16:06 | |
| *** tosky has quit IRC | 16:11 | |
| clarkb | does anyone know if pip has a force rebuild wheel flag on install? there is --no-binary and --force-reinstall but what about when you have a wheel linked to an old so and want to rebuild the wheel? | 16:16 |
| clarkb | I've addressed this so far by rm'ing the wheels from my wheel cache | 16:17 |
| clarkb | I double checked the post_failure in the gate and it appears the job may be breaking host networking (its a job that includes dynamic-routing in the name). pip install fails due to no route to host to our mirror when installing stackviz | 16:21 |
| clarkb | I don't think that is related to our post failure problems yesterday. I notified neutrons irc channel too | 16:21 |
| clarkb | infra-root now that nb03.opendev.org is doing zk tls I think we can alnd https://review.opendev.org/#/c/750833/ to block port 2181 on the zk servers. We don't disable 2181 entirely as it is useful for local debugging | 16:25 |
| openstackgerrit | Merged openstack/project-config master: Use only noop jobs for openstack/monasca-transform https://review.opendev.org/751983 | 16:40 |
| openstackgerrit | Merged openstack/project-config master: End project gating for openstack/monasca-analytics https://review.opendev.org/751987 | 16:40 |
| *** ykarel|away has joined #opendev | 16:43 | |
| openstackgerrit | Merged openstack/project-config master: Retired x/osops-* repos https://review.opendev.org/751130 | 16:49 |
| *** ykarel|away has quit IRC | 16:50 | |
| johnsom | Happy dance, the pypa issue seems to be resolved. Thanks for all of the effort that went into tracking down that bad pypa mirror. | 16:53 |
| fungi | clarkb: not sure about forcing wheel rebuilds. i too just blow away my wheel cache and my venvs regularly and then recreate them all (basically any time i build a new default python3 point release) | 16:53 |
| fungi | which reminds me, looks like v3.5.10 and v3.8.6rc1 were tagged since i last refreshed | 16:54 |
| fungi | and i guess 3.5 is technically eol now too | 16:56 |
| smcginnis | Oh wow, as of this week. | 16:57 |
| clarkb | johnsom: I think my favorite part of the whole thing is we stopped running bandersnatch when mirroring pypi hit 2TB of disk use. They just hit 12TB :) | 16:57 |
| fungi | https://www.python.org/dev/peps/pep-0478/ "There are no specific plans for the next release of Python 3.5. Note that Python 3.5 will no longer be supported after September 2020. Therefore, 3.5.10 may be the final release of Python 3.5." | 16:58 |
| smcginnis | 2020-09-13 according to https://devguide.python.org/#branchstatus | 16:58 |
| fungi | cool. i can start planning to drop 3.5 testing for one of my personal projects now | 16:59 |
| fungi | after i make sure it's still working with 3.5.10 that is | 16:59 |
| *** fdegir has joined #opendev | 17:01 | |
| openstackgerrit | Merged openstack/project-config master: Update branch checkout for octavia-lib DIB element https://review.opendev.org/745877 | 17:03 |
| *** dtantsur is now known as dtantsur|afk | 17:13 | |
| *** ysandeep|away is now known as ysandeep | 17:27 | |
| *** ysandeep is now known as ysandeep|away | 17:32 | |
| *** tosky has joined #opendev | 18:51 | |
| *** priteau has quit IRC | 19:12 | |
| *** Tengu has quit IRC | 19:15 | |
| *** Tengu has joined #opendev | 19:17 | |
| clarkb | fungi: https://review.opendev.org/#/c/751919/ I think that is the last piece needed in dib ebfore we can do a release which will enable us to delete nb04 | 19:18 |
| clarkb | note that the tumbleweed job passes there but not in the parent | 19:18 |
| fungi | i should probably subscribe my gertty to openstack/diskimage-builder changes | 19:20 |
| fungi | sidetracked by a kernel upgrade, the openafs dkms build is currently failing with linux 5.8 | 19:22 |
| openstackgerrit | Merged openstack/diskimage-builder master: Install grep before busybox on suse distros https://review.opendev.org/751880 | 19:22 |
| clarkb | I'm not sure its a rush but seeing ^ was about to merge reminded me :) | 19:23 |
| clarkb | and I'm trying to clear my backlog today so that I can get back to new things :) | 19:23 |
| *** andrewbonney has quit IRC | 19:28 | |
| *** owalsh has quit IRC | 20:11 | |
| *** owalsh has joined #opendev | 20:43 | |
| *** owalsh has quit IRC | 20:56 | |
| clarkb | ianw: I'm surprised that rhel is shipping VM images that are EFI based since I think a lot of clouds don't support that yet? | 21:08 |
| clarkb | I guess that is where utilities like dib come in as they can "fix" the image for you if necessary | 21:08 |
| *** slaweq has quit IRC | 21:14 | |
| *** slaweq has joined #opendev | 21:15 | |
| *** owalsh has joined #opendev | 21:18 | |
| *** slaweq has quit IRC | 21:27 | |
| *** slaweq has joined #opendev | 21:33 | |
| *** slaweq has quit IRC | 21:54 | |
| fungi | for the big proprietary public clouds, they have their own tweaked images provided for you so you don't have to worry about it. for everything else, red hat sells cloud software | 21:56 |
| *** tosky_ has joined #opendev | 21:56 | |
| fungi | no doubt whatever red hat is covering with support contracts supports efi boot | 21:57 |
| *** tosky has quit IRC | 21:59 | |
| *** tosky_ is now known as tosky | 22:00 | |
| *** slaweq has joined #opendev | 22:01 | |
| openstackgerrit | Goutham Pacha Ravi proposed openstack/project-config master: Cleanup x/devstack-plugin-glusterfs jobs https://review.opendev.org/752335 | 22:06 |
| *** slaweq_ has joined #opendev | 22:10 | |
| *** slaweq has quit IRC | 22:13 | |
| *** slaweq_ has quit IRC | 22:19 | |
| ianw | clarkb: i think the images are "dual boot" | 22:46 |
| ianw | in theory, dib can create dual-boot efi/bios images too | 22:46 |
| ianw | actually s/can/does/ after a recent change | 22:46 |
| ianw | ... but ... practice i'm not so sure | 22:47 |
| clarkb | ah | 22:47 |
| ianw | again in theory we should be able to try our nodepool tests with efi, and then, in theory, switch on the dual-boot efi and see if that boots both bios and efi | 22:50 |
| clarkb | ianw: if we get https://review.opendev.org/#/c/751919/ in I think that is all we need for tumbleweed fixes | 22:50 |
| clarkb | not sure if there are other changes (like the rhel one) htat you may also want to get into a dib release | 22:50 |
| ianw | no problem; i sent a bunch back through the gate after the recent breakages, havent' checked if it all committed yet :) | 22:51 |
| ianw | a lot of green ticks! | 22:51 |
| clarkb | the parent to that change merged today | 22:51 |
| clarkb | pypi is much happier :) | 22:51 |
| ianw | i double check the queue ,but i think 751919 would be the last outstanding thing for a release now | 22:52 |
| ianw | https://p.datadoghq.com/sb/7dc8b3250-85dcf667bd?from_ts=1600273495941&to_ts=1600277095941&live=true is an interesting page | 22:53 |
| ianw | pypi stats | 22:53 |
| ianw | 942,000 requests/min | 22:54 |
| clarkb | I can't seem to pull that up | 22:54 |
| fungi | ahh, yeah, debian has multiarch iso images which can boot efi/mbr and multiple processor architectures for example, definitely doable | 22:55 |
| clarkb | not getting a tcp connection | 22:55 |
| clarkb | but I believe it :) | 22:55 |
| fungi | it's too busy to give you a tcp connection | 22:56 |
| TheJulia | out of curiosity, are there more log storage issues? | 22:56 |
| fungi | TheJulia: i haven't seen any, but that means nothing. do you have a suspect failure you can link? | 22:57 |
| TheJulia | https://4317fd6ef51d22a4668e-8daa7927094e2ef08ab8f0f95a49fc40.ssl.cf5.rackcdn.com/748315/23/check/metalsmith-integration-glance-localboot-centos8-uefi/79e4b9b/ | 22:57 |
| clarkb | TheJulia: not that I've seen. I tracked down one post failure earlier today to broken host netowrking in a job that tests dynamic routing | 22:57 |
| TheJulia | click controller | 22:57 |
| TheJulia | empty folder :\ | 22:57 |
| TheJulia | or at least that is what I see in two browsers | 22:57 |
| fungi | that's unlikely to be quota related | 22:57 |
| TheJulia | that is kind of what I figured, but I wanted to ask | 22:57 |
| clarkb | fungi: TheJulia thats not the oe cloud either | 22:58 |
| fungi | though our log uploads try to generate indices, right? | 22:58 |
| clarkb | so if it is a swift problem it is a different problem | 22:58 |
| clarkb | fungi: yes we generate our own indices | 22:58 |
| TheJulia | I just don't want to blindly rehceck it | 22:59 |
| fungi | maybe we created an empty index for that (what does our tooling do if the dir is empty?) | 22:59 |
| ianw | are there files there in the manifest? | 22:59 |
| fungi | great point | 22:59 |
| *** tosky has quit IRC | 22:59 | |
| * TheJulia opens the main job log file | 22:59 | |
| fungi | yes, looks like it | 22:59 |
| fungi | "controller" has children listed at least | 23:00 |
| fungi | for example 1d22a4668e-8daa7927094e2ef08ab8f0f95a49fc40.ssl.cf5.rackcdn.com/748315/23/check/metalsmith-integration-glance-localboot-centos8-uefi/79e4b9b/controller/ironic-bm-logs/ | 23:00 |
| TheJulia | https://4317fd6ef51d22a4668e-8daa7927094e2ef08ab8f0f95a49fc40.ssl.cf5.rackcdn.com/748315/23/check/metalsmith-integration-glance-localboot-centos8-uefi/79e4b9b/controller/ironic-bm-logs/ works | 23:00 |
| TheJulia | lol | 23:00 |
| TheJulia | jinx | 23:00 |
| fungi | your cut-n-paste outshined mine | 23:00 |
| fungi | recent firefox likes to highlight stuff in the url bar without copying it to the paste buffer, so i have to unhighlight and rehighlight it. very aggravating | 23:01 |
| ianw | TheJulia: you could grab everything with the download script if you like | 23:01 |
| ianw | we should definitely look into why that index wasn't generated/uploaded | 23:01 |
| fungi | so for some reason we failed to populate the inde | 23:01 |
| fungi | x | 23:01 |
| TheJulia | well, I managed to grab what I need :\ | 23:01 |
| TheJulia | Just went "well this is weird" | 23:02 |
| ianw | so index.html must be there, but it's blank i guess | 23:02 |
| clarkb | ya it doesn't 404 | 23:03 |
| clarkb | you just get back an empty file | 23:03 |
| clarkb | Makes me wonder if this could be related to how our swift backed docker registry would end up with empty blob layers | 23:03 |
| clarkb | maybe there is some sublte behavior thing there? | 23:03 |
| ianw | there was | 23:06 |
| fungi | yeah, given what i've heard about the "swiftness" of rackspace's swift (cloudfiles) i doubt it suffers an identical bug to real swift, which leads me to suspect the problem could be client side | 23:07 |
| ianw | there was https://review.opendev.org/#/c/751973/ which is a bad file on rax | 23:07 |
| ianw | https://58aca06c0a1750aa55ba-25dc1a1096eaf82a51392114c0a4a973.ssl.cf2.rackcdn.com/732435/7/check/heat-functional/852a5bf/controller/logs/screen-barbican-keystone-listener.txt | 23:07 |
| clarkb | fwiw `grep 79e4b9bb79ab429e98b4511227cfb11d /var/log/zuul/executor-debug.log` on ze06 to get hte ansible log for this job | 23:09 |
| clarkb | doesn't show anything abnormal for swift log upload | 23:09 |
| clarkb | and ya if you wget the file its a completely empty empty file but not a 404 | 23:12 |
| clarkb | looking at the zuul-jobs upload-logs-swift role we seem to write html even if there is no other content | 23:13 |
| clarkb | so an empty dir should still have something there | 23:13 |
| clarkb | all that to say it is really weird to have an empty file | 23:15 |
| clarkb | I'd expect a 404 (because upload failed) or something with html in it if not | 23:15 |
| ianw | http://1d22a4668e-8daa7927094e2ef08ab8f0f95a49fc40.ssl.cf5.rackcdn.com/748315/23/check/metalsmith-integration-glance-localboot-centos8-uefi/79e4b9b/controller/ironic-bm-logs/ | 23:15 |
| ianw | i just go access denied on that? ^ | 23:16 |
| ianw | oh, maybe it's the http i've mis-pasted | 23:16 |
| ianw | yep | 23:17 |
| clarkb | its a different "hostname" | 23:17 |
| clarkb | its related to how cdn works with swift aiui | 23:17 |
| clarkb | oh interest fungi and TheJulia have different hostnames in their links | 23:18 |
| clarkb | where did the 1d22a466... come from? | 23:18 |
| TheJulia | Redirect maybe? | 23:18 |
| TheJulia | I got to it in a web browser based on the initial link I copied from the job | 23:19 |
| clarkb | maybe different cdn nodes and only the content after the - is used to hash and verify | 23:19 |
| ianw | https://7598b949b48f30e69b3e-9945c6e94a1b9feee2753a6244e7fd57.ssl.cf5.rackcdn.com/748315/19/check/metalsmith-integration-glance-localboot-centos8-uefi/7084429/controller/ | 23:40 |
| ianw | is the same job, on rax, with a correct index | 23:40 |
| ianw | so i don't know, i guess we just monitor for more and see if we can find a pattern | 23:40 |
| *** DSpider has quit IRC | 23:42 | |
| clarkb | we could have the upload role fetch back content and check the first few bytes | 23:42 |
| clarkb | probably overkill | 23:42 |
| fungi | also lots of extra bandwidth consumption | 23:49 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!