| rpioso | fungi: https://review.opendev.org/c/openstack/project-config/+/764100 | 00:07 |
|---|---|---|
| fungi | thanks, we're experiencing technical difficulties with gerritbot so appreciate the pointer | 00:08 |
| fungi | i have the reenqueue for the tag ready as soon as that merges | 00:09 |
| rpioso | fungi: Of course! And thank you so much :-) | 00:09 |
| fungi | my pleasure | 00:09 |
| *** outbrito has quit IRC | 00:12 | |
| *** outbrito has joined #openstack-infra | 00:13 | |
| * rpioso is stepping away for dinner | 00:15 | |
| fungi | i'll let you know once the tag has been processed | 00:24 |
| *** openstackgerrit has joined #openstack-infra | 00:25 | |
| openstackgerrit | Merged openstack/project-config master: Add publish-to-pypi template for sushy-oem-idrac https://review.opendev.org/c/openstack/project-config/+/764100 | 00:25 |
| fungi | rpioso: all good looks like: https://pypi.org/project/sushy-oem-idrac/ | 00:35 |
| *** sboyron has quit IRC | 00:52 | |
| *** zzzeek has quit IRC | 01:09 | |
| *** zzzeek has joined #openstack-infra | 01:11 | |
| *** zzzeek has quit IRC | 01:27 | |
| *** zzzeek has joined #openstack-infra | 01:28 | |
| *** tkajinam has quit IRC | 01:31 | |
| *** tkajinam has joined #openstack-infra | 01:32 | |
| *** kden has joined #openstack-infra | 01:34 | |
| *** outbrito has quit IRC | 01:55 | |
| rpioso | fungi: It does :-) Thanks, again. I hope you have a wonderful Thanksgiving. | 02:00 |
| fungi | any time, and same to you! | 02:02 |
| *** Goneri has quit IRC | 02:03 | |
| *** rcernin has quit IRC | 02:19 | |
| *** rcernin has joined #openstack-infra | 02:21 | |
| *** rcernin has quit IRC | 02:23 | |
| *** rcernin has joined #openstack-infra | 02:24 | |
| *** vishalmanchanda has joined #openstack-infra | 02:44 | |
| *** hamalq has joined #openstack-infra | 02:45 | |
| *** zzzeek has quit IRC | 02:45 | |
| *** zzzeek has joined #openstack-infra | 02:47 | |
| *** hamalq has quit IRC | 02:49 | |
| *** hamalq has joined #openstack-infra | 03:00 | |
| *** hamalq has quit IRC | 03:05 | |
| *** piotrowskim has quit IRC | 03:08 | |
| *** zxiiro has quit IRC | 03:23 | |
| *** psachin has joined #openstack-infra | 03:36 | |
| *** psachin has quit IRC | 03:38 | |
| *** psachin has joined #openstack-infra | 03:39 | |
| *** ykarel|away has joined #openstack-infra | 04:13 | |
| *** zzzeek has quit IRC | 04:18 | |
| *** sreejithp has joined #openstack-infra | 04:19 | |
| *** sreejithp has quit IRC | 04:19 | |
| *** zzzeek has joined #openstack-infra | 04:20 | |
| *** sreejithp has joined #openstack-infra | 04:20 | |
| *** sreejithp has quit IRC | 04:21 | |
| *** ykarel|away has quit IRC | 04:34 | |
| *** raukadah is now known as chandankumar | 04:52 | |
| *** zzzeek has quit IRC | 04:58 | |
| *** zzzeek has joined #openstack-infra | 04:58 | |
| *** ykarel|away has joined #openstack-infra | 05:00 | |
| *** zzzeek has quit IRC | 05:05 | |
| *** zzzeek has joined #openstack-infra | 05:06 | |
| *** ramishra has quit IRC | 05:07 | |
| *** ramishra has joined #openstack-infra | 05:07 | |
| *** ykarel|away is now known as ykarel | 05:07 | |
| *** rakhmerov has joined #openstack-infra | 05:13 | |
| *** zzzeek has quit IRC | 05:15 | |
| *** zzzeek has joined #openstack-infra | 05:19 | |
| *** zzzeek has quit IRC | 05:45 | |
| *** zzzeek has joined #openstack-infra | 05:46 | |
| *** evrardjp has quit IRC | 05:47 | |
| *** evrardjp has joined #openstack-infra | 05:48 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/project-config master: Normalize projects.yaml https://review.opendev.org/c/openstack/project-config/+/763930 | 06:07 |
| *** ricolin has joined #openstack-infra | 06:12 | |
| *** matt_kosut has joined #openstack-infra | 06:23 | |
| *** lmiccini has joined #openstack-infra | 06:50 | |
| *** zzzeek has quit IRC | 07:05 | |
| *** zzzeek has joined #openstack-infra | 07:07 | |
| *** ysandeep|away is now known as ysandeep | 07:15 | |
| *** sboyron has joined #openstack-infra | 07:22 | |
| *** lpetrut has joined #openstack-infra | 07:22 | |
| *** ralonsoh has joined #openstack-infra | 07:25 | |
| *** eolivare has joined #openstack-infra | 07:36 | |
| *** dklyle has quit IRC | 07:36 | |
| *** ykarel_ has joined #openstack-infra | 07:37 | |
| *** rpittau|afk is now known as rpittau | 07:37 | |
| *** ykarel has quit IRC | 07:40 | |
| *** iurygregory is now known as iurygregory|afk | 07:48 | |
| *** slaweq has joined #openstack-infra | 07:53 | |
| *** jcapitao has joined #openstack-infra | 07:55 | |
| *** zzzeek has quit IRC | 08:05 | |
| *** hashar has joined #openstack-infra | 08:06 | |
| *** zzzeek has joined #openstack-infra | 08:07 | |
| *** ykarel_ is now known as ykarel | 08:34 | |
| *** rcernin has quit IRC | 08:50 | |
| *** tosky has joined #openstack-infra | 08:50 | |
| *** jpena|off is now known as jpena | 08:55 | |
| *** andrewbonney has joined #openstack-infra | 08:56 | |
| *** lucasagomes has joined #openstack-infra | 09:03 | |
| *** mgoddard has joined #openstack-infra | 09:03 | |
| openstackgerrit | Merged openstack/project-config master: Normalize projects.yaml https://review.opendev.org/c/openstack/project-config/+/763930 | 09:09 |
| *** iurygregory|afk is now known as iurygregory | 09:10 | |
| *** derekh has joined #openstack-infra | 09:13 | |
| *** sean-k-mooney1 has joined #openstack-infra | 09:16 | |
| *** sean-k-mooney has quit IRC | 09:17 | |
| *** d34dh0r53 has quit IRC | 09:17 | |
| *** rcernin has joined #openstack-infra | 09:18 | |
| *** jakeyip has left #openstack-infra | 09:18 | |
| *** dchen has quit IRC | 09:20 | |
| slaweq | hi infra team | 09:20 |
| slaweq | I just noticed on our stable/stein and stable/train jobs that we are hitting again issue with wrong setuptools package, like e.g. https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_c71/763023/1/check/neutron-tempest-iptables_hybrid/c711831/job-output.txt | 09:20 |
| *** d34dh0r53 has joined #openstack-infra | 09:20 | |
| slaweq | it seems to be the issue only on rax-ord provider | 09:21 |
| slaweq | maybe You can somehow clean cache of the packages there? | 09:21 |
| *** rcernin has quit IRC | 09:27 | |
| slaweq | and we saw similar issue on ovh-gra1: https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_291/periodic-stable/opendev.org/openstack/networking-bgpvpn/stable/stein/openstack-tox-py27/2919de9/job-output.txt | 09:34 |
| *** rcernin has joined #openstack-infra | 09:37 | |
| *** zzzeek has quit IRC | 09:40 | |
| *** zzzeek has joined #openstack-infra | 09:43 | |
| *** ociuhandu has joined #openstack-infra | 09:47 | |
| frickler | slaweq: I don't have time to look deeper today, but if that's the same issue as we had some weeks ago, the error comes from pypi and we can't do anything on the mirror side to avoid it | 09:47 |
| *** gfidente|afk is now known as gfidente | 09:47 | |
| frickler | we might consider capping setuptools permanently as in ralonsoh's https://review.opendev.org/c/openstack/devstack/+/755794 | 09:47 |
| frickler | there won't be a newer setuptools version for py2 anyway. but that should be discussed in #openstack-qa | 09:48 |
| ralonsoh | revive the patch if needed, just to avoid future problems | 09:50 |
| ralonsoh | as you said, there won't be newer versions for py2 | 09:51 |
| slaweq | frickler: thx for checking | 09:51 |
| *** ociuhandu has quit IRC | 09:57 | |
| *** ociuhandu has joined #openstack-infra | 10:01 | |
| *** artom has quit IRC | 10:05 | |
| *** artom has joined #openstack-infra | 10:05 | |
| *** rcernin has quit IRC | 10:07 | |
| *** artom has quit IRC | 10:19 | |
| *** rcernin has joined #openstack-infra | 10:27 | |
| *** rcernin has quit IRC | 10:33 | |
| *** dtantsur|afk is now known as dtantsur | 10:33 | |
| *** gbutnaru has joined #openstack-infra | 10:35 | |
| *** artom has joined #openstack-infra | 10:43 | |
| *** hashar has quit IRC | 10:50 | |
| *** zzzeek has quit IRC | 10:58 | |
| *** zzzeek has joined #openstack-infra | 10:59 | |
| *** rcernin has joined #openstack-infra | 10:59 | |
| *** rcernin has quit IRC | 11:04 | |
| *** mgoddard has quit IRC | 11:12 | |
| *** ysandeep is now known as ysandeep|afk | 11:22 | |
| *** mgoddard has joined #openstack-infra | 11:23 | |
| *** kden has quit IRC | 11:25 | |
| *** jcapitao is now known as jcapitao_lunch | 11:35 | |
| *** xek has joined #openstack-infra | 11:36 | |
| *** rfolco|ruck has joined #openstack-infra | 11:36 | |
| *** xek has quit IRC | 11:37 | |
| *** xek has joined #openstack-infra | 11:38 | |
| *** zzzeek has quit IRC | 11:42 | |
| *** zzzeek has joined #openstack-infra | 11:44 | |
| *** rcernin has joined #openstack-infra | 11:47 | |
| *** rcernin has quit IRC | 12:07 | |
| jpena | apevec: lookint at https://zuul.openstack.org/status, 763919 seems to be in the front of a massive queue | 12:23 |
| jpena | soory, wrong channel | 12:23 |
| *** ysandeep|afk is now known as ysandeep | 12:26 | |
| *** slaweq_ has joined #openstack-infra | 12:30 | |
| *** ociuhandu has quit IRC | 12:31 | |
| *** slaweq has quit IRC | 12:33 | |
| *** zzzeek has quit IRC | 12:34 | |
| *** jpena is now known as jpena|lunch | 12:35 | |
| *** zzzeek has joined #openstack-infra | 12:38 | |
| *** xek has quit IRC | 12:44 | |
| *** xek_ has joined #openstack-infra | 12:44 | |
| *** jcapitao_lunch is now known as jcapitao | 12:44 | |
| *** rlandy has joined #openstack-infra | 12:52 | |
| *** ociuhandu has joined #openstack-infra | 13:02 | |
| *** xek__ has joined #openstack-infra | 13:05 | |
| *** ociuhandu has quit IRC | 13:08 | |
| *** xek_ has quit IRC | 13:08 | |
| *** ysandeep is now known as ysandeep|brb | 13:08 | |
| *** outbrito has joined #openstack-infra | 13:09 | |
| *** mgoddard has quit IRC | 13:15 | |
| *** mgoddard has joined #openstack-infra | 13:17 | |
| *** rcernin has joined #openstack-infra | 13:21 | |
| *** rcernin has quit IRC | 13:25 | |
| *** vishalmanchanda has quit IRC | 13:26 | |
| *** tkajinam has quit IRC | 13:31 | |
| *** tkajinam has joined #openstack-infra | 13:32 | |
| *** jpena|lunch is now known as jpena | 13:35 | |
| *** kden has joined #openstack-infra | 13:38 | |
| *** hashar has joined #openstack-infra | 13:38 | |
| *** ociuhandu has joined #openstack-infra | 13:53 | |
| *** ysandeep|brb is now known as ysandeep | 13:55 | |
| *** sshnaidm is now known as sshnaidm|afk | 13:56 | |
| *** ociuhandu has quit IRC | 13:57 | |
| *** derekh has quit IRC | 14:02 | |
| *** derekh has joined #openstack-infra | 14:02 | |
| noonedeadpunk | hey! | 14:03 |
| noonedeadpunk | we've recieved bunch of patches like https://review.opendev.org/c/openstack/openstack-ansible-os_designate/+/764168 | 14:04 |
| noonedeadpunk | which from our prespective is not really valid ones, but I just loved to double check that with more experts in tox :) | 14:04 |
| *** auristor has quit IRC | 14:04 | |
| noonedeadpunk | we have contraints defined in install_command for testenv and we suggest they should be working the same way as suggested change | 14:05 |
| *** walshh_ has joined #openstack-infra | 14:07 | |
| *** auristor has joined #openstack-infra | 14:08 | |
| *** ociuhandu has joined #openstack-infra | 14:09 | |
| *** ociuhandu has quit IRC | 14:14 | |
| fungi | noonedeadpunk: the commit message looks like it's (trying) to say that master branch constraints aren't valid for building stable branch docs, but i can't imagine under what circumstances that would happen anyway | 14:21 |
| jrosser | i left a comment on one of them suggesting that the constraint in [testenv] should (?) apply https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/764007 | 14:23 |
| jrosser | but tbh without actually trying it for real i cant be sure | 14:23 |
| jrosser | on the other hand codesearch.o.o suggests that the pattern we use in these repos is quite widespread | 14:24 |
| *** ociuhandu has joined #openstack-infra | 14:25 | |
| fungi | yes, what you have there is our typical, many-years-old workaround for pip not being able to properly assess multiple version constraints for the same dependency. we pre-resolve a working dependency set and then tell pip to install exactly those versions rather than trying to work out for itself what the best version is to satisfy all consumers. we can start to revisit it with the next pip release, | 14:26 |
| fungi | which will provide a far more sane dep solver | 14:26 |
| *** zzzeek has quit IRC | 14:28 | |
| *** zzzeek has joined #openstack-infra | 14:29 | |
| *** rpittau is now known as rpittau|bbl | 14:29 | |
| jrosser | ah interesting, so thats whats being achieved with install_command? | 14:32 |
| *** sshnaidm|afk is now known as sshnaidm | 14:34 | |
| *** ociuhandu has quit IRC | 14:36 | |
| *** vishalmanchanda has joined #openstack-infra | 14:53 | |
| *** Goneri has joined #openstack-infra | 14:56 | |
| *** slaweq_ is now known as slaweq | 14:58 | |
| kden | Hi, I'm not sure this is the right place but I'd like to ask you experts about the possibility of adding an image type. | 15:00 |
| kden | Tacker, which is one of the official projects, currently runs functional tests on ubuntu-focal (RAM: 8GB) nodes. It requires lots of plugins from other projects to be included on controller nodes, and finally seems to suffer from memory shortage. If something like ubuntu-focal-expanded or ubuntu-focal-32GB is available, it really helps. I'd appreciate if you point me in a right direction. | 15:00 |
| kden | Please note that we of course made sure that there are no services or daemons inappropriately consuming large portion of memory. | 15:00 |
| openstackgerrit | Alec Hothan proposed openstack/project-config master: Remove openstack-python and openstack-docs job for x/vmtp project. https://review.opendev.org/c/openstack/project-config/+/764054 | 15:03 |
| *** ociuhandu has joined #openstack-infra | 15:07 | |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Add missing entries in base.json https://review.opendev.org/c/openstack/ptgbot/+/764207 | 15:11 |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Log exception when bot airbag is activated https://review.opendev.org/c/openstack/ptgbot/+/764208 | 15:11 |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Create functions for is_voiced and is_oper checks https://review.opendev.org/c/openstack/ptgbot/+/764209 | 15:11 |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Add memory-only database mode https://review.opendev.org/c/openstack/ptgbot/+/764210 | 15:11 |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Add tests for PTGbot https://review.opendev.org/c/openstack/ptgbot/+/764211 | 15:11 |
| *** ociuhandu has quit IRC | 15:11 | |
| *** hashar has quit IRC | 15:14 | |
| *** ociuhandu has joined #openstack-infra | 15:14 | |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Add tests for PTGbot https://review.opendev.org/c/openstack/ptgbot/+/764211 | 15:16 |
| fungi | kden: those large memory nodes are not available from most of our providers, so best case those jobs would wait a long time for node assignments, worst case (like now) the couple of providers which have them could be either offline completely or frequently unable to fulfill the boot requests, leading to frequent node_failure results too | 15:27 |
| *** ociuhandu_ has joined #openstack-infra | 15:27 | |
| fungi | kden: there is probably another solution available to you: subdivide the memory load | 15:27 |
| fungi | could you run multiple jobs with different sets of plugins? or a multi-node job with different plugins on different nodes? | 15:28 |
| *** ociuhandu has quit IRC | 15:30 | |
| kden | fungi: All right, thanks! BTW I noticed that due to Airship's requirement, ubuntu-bionic-32GB was recently introduced. I'm just curious but are they suffering from those issues now? | 15:31 |
| fungi | yes | 15:32 |
| fungi | they've been working for a while on breaking up their workload to fit into smaller footprint test nodes, which is a much more flexible solution anyway | 15:32 |
| kden | fungi: unfortunately it already subdivides memory loads across three ubuntu-focal nodes. | 15:33 |
| kden | fungi: thanks for the info. I'll try to dig in more. | 15:33 |
| fungi | citycloud offered to host a few 16gb and 32gb memory nodes because airship contributors had requested them, but the hardware they've put our tenant on is often (if not usually) incapable of scheduling nodes of that size on the compute nodes in that environment | 15:34 |
| fungi | openedge also offered to host some, but they're offline for a while due to network equipment failures | 15:34 |
| *** outbrito has quit IRC | 15:35 | |
| fungi | kden: openstack-helm has also been trying to use those, which is why if you go here you'll see a mix of helm and airship builds: https://zuul.opendev.org/t/openstack/builds?result=NODE_FAILURE | 15:36 |
| kden | fungi: what a detailed info, thank you so much! | 15:37 |
| *** zzzeek has quit IRC | 15:41 | |
| *** zzzeek has joined #openstack-infra | 15:43 | |
| *** rpittau|bbl is now known as rpittau | 15:44 | |
| *** kden has quit IRC | 15:44 | |
| *** ociuhandu_ has quit IRC | 15:44 | |
| *** ociuhandu has joined #openstack-infra | 15:45 | |
| *** sean-k-mooney1 is now known as sean-k-mooney | 15:47 | |
| *** lpetrut has quit IRC | 15:52 | |
| *** ykarel is now known as ykarel|away | 15:52 | |
| sean-k-mooney | i have a meeting at the top of the hour so jsut going to leave this question here and ill check back later. do ye have any recomenation for how to configre the mirrors for cis and what software to use. i would like to try and reuse the mirror rule but ideally i would like to avoid a full mirror and use something like apt-cache-ng to cache any pacakges that are actully used | 15:57 |
| *** dklyle has joined #openstack-infra | 15:57 | |
| sean-k-mooney | *mirror role | 15:58 |
| sean-k-mooney | i have devpy deployed for pypi but not actully enabled it in the jobs yet but tyring to figure out what is the best solution for the os packages | 15:59 |
| *** ociuhandu has quit IRC | 16:00 | |
| *** lpetrut has joined #openstack-infra | 16:01 | |
| fungi | for the most part, i would say use apache mod_proxy or squid, and only consider special-purpose caching proxy services if the simple solutions lead to problems with your jobs | 16:08 |
| fungi | i agree that building full mirrors can be a lot of overhead if your jobs will ever only hit a small percentage of the files you're mirroring | 16:09 |
| fungi | as for apt-cacher-ng, i ran it for a while in some of my personal environments, what it's mainly got going for it is that it has the smarts to know when to keep packages in its cache and when to drop them because packages indices no longer refer to them | 16:10 |
| *** sreejithp has joined #openstack-infra | 16:10 | |
| fungi | that could be seen as over-optimization as it's really only preventing a very small amount of additional network traffic | 16:11 |
| fungi | compared to a simpler and more straightforward time-based or space-based caching solution | 16:12 |
| *** ysandeep is now known as ysandeep|away | 16:12 | |
| fungi | spacetime-based ;) | 16:12 |
| *** ykarel|away has quit IRC | 16:14 | |
| *** ociuhandu has joined #openstack-infra | 16:14 | |
| *** xarses has quit IRC | 16:19 | |
| *** xarses has joined #openstack-infra | 16:21 | |
| *** xarses has quit IRC | 16:22 | |
| *** xarses has joined #openstack-infra | 16:22 | |
| *** ociuhandu has quit IRC | 16:24 | |
| *** ociuhandu has joined #openstack-infra | 16:26 | |
| *** chandankumar is now known as raukadah | 16:30 | |
| *** ociuhandu has quit IRC | 16:31 | |
| *** zxiiro has joined #openstack-infra | 16:31 | |
| *** zxiiro has quit IRC | 16:33 | |
| *** zxiiro has joined #openstack-infra | 16:35 | |
| *** irclogbot_1 has quit IRC | 16:41 | |
| *** irclogbot_3 has joined #openstack-infra | 16:46 | |
| *** dklyle has quit IRC | 16:49 | |
| *** dklyle has joined #openstack-infra | 16:53 | |
| *** d34dh0r53 has quit IRC | 17:00 | |
| *** lucasagomes has quit IRC | 17:00 | |
| *** d34dh0r53 has joined #openstack-infra | 17:03 | |
| *** ociuhandu has joined #openstack-infra | 17:03 | |
| openstackgerrit | Matt McEuen proposed openstack/project-config master: New Project Request: airship/vino https://review.opendev.org/c/openstack/project-config/+/763889 | 17:05 |
| openstackgerrit | Matt McEuen proposed openstack/project-config master: New Project Request: airship/sip https://review.opendev.org/c/openstack/project-config/+/763888 | 17:06 |
| *** ociuhandu has quit IRC | 17:09 | |
| *** d34dh0r53 has quit IRC | 17:20 | |
| portdirect | is there anyway to get the old gerrit ui back? | 17:27 |
| portdirect | new is very pretty | 17:27 |
| portdirect | but... ;) | 17:27 |
| clarkb | no | 17:27 |
| * portdirect mumbles about old dogs etc | 17:27 | |
| portdirect | clarkb: thx :) | 17:27 |
| *** ociuhandu has joined #openstack-infra | 17:29 | |
| openstackgerrit | Matt McEuen proposed openstack/project-config master: New Project Request: airship/vino https://review.opendev.org/c/openstack/project-config/+/763889 | 17:34 |
| openstackgerrit | Thierry Carrez proposed openstack/project-config master: Enable py38 tests for openstack/ptgbot https://review.opendev.org/c/openstack/project-config/+/764223 | 17:34 |
| *** xek__ has quit IRC | 17:37 | |
| *** gfidente is now known as gfidente|afk | 17:43 | |
| *** lmiccini has quit IRC | 17:53 | |
| *** psachin has quit IRC | 17:54 | |
| *** ociuhandu has quit IRC | 18:01 | |
| *** derekh has quit IRC | 18:01 | |
| *** ociuhandu has joined #openstack-infra | 18:02 | |
| *** jpena is now known as jpena|off | 18:03 | |
| *** mgoddard has quit IRC | 18:04 | |
| *** dave-mccowan has quit IRC | 18:04 | |
| *** zzzeek has quit IRC | 18:06 | |
| *** lpetrut has quit IRC | 18:06 | |
| *** zzzeek has joined #openstack-infra | 18:06 | |
| *** ociuhandu has quit IRC | 18:06 | |
| *** dave-mccowan has joined #openstack-infra | 18:09 | |
| *** eolivare has quit IRC | 18:11 | |
| *** vishalmanchanda has quit IRC | 18:21 | |
| *** sshnaidm has quit IRC | 18:23 | |
| *** jcapitao has quit IRC | 18:23 | |
| *** sshnaidm has joined #openstack-infra | 18:24 | |
| *** hashar has joined #openstack-infra | 18:25 | |
| fungi | portdirect: there was a time where gerrit allowed you to switch between the old interface and a premature implementation of the new interface, but after some deliberation we determined that stopping there for a while would 1. delay us getting upgraded to a current gerrit release but also 2. give people an even worse impression of the new interface | 18:26 |
| fungi | at least you're getting the new interface with most of the early bug now fixed ;) | 18:26 |
| portdirect | i found the 'dark' theme | 18:27 |
| fungi | yeah, i've already decided that's my default ;) | 18:27 |
| portdirect | which at least give my poor old rentinas a break :) | 18:27 |
| *** sshnaidm has quit IRC | 18:28 | |
| *** sshnaidm has joined #openstack-infra | 18:28 | |
| *** hashar has quit IRC | 18:30 | |
| *** rosmaita has left #openstack-infra | 18:33 | |
| clarkb | ya we realized the major cost of the upgrade was in the notedb transition so wanted to get that done then get to the latest release | 18:38 |
| clarkb | there are a lot of good features in the UI but it isnt perfect either | 18:38 |
| clarkb | I really like that it has gone back to a change based perspective instead of a patchset perspective | 18:38 |
| clarkb | and the single page diffs are a long awaited feature | 18:38 |
| *** zzzeek has quit IRC | 18:41 | |
| *** zzzeek has joined #openstack-infra | 18:42 | |
| *** rpittau is now known as rpittau|afk | 18:52 | |
| *** andrewbonney has quit IRC | 18:53 | |
| *** hashar has joined #openstack-infra | 19:03 | |
| *** dtantsur is now known as dtantsur|afk | 19:33 | |
| *** ramishra has quit IRC | 19:44 | |
| *** xek__ has joined #openstack-infra | 20:11 | |
| *** gbutnaru has quit IRC | 20:12 | |
| *** sreejithp has quit IRC | 20:16 | |
| *** armax has quit IRC | 20:20 | |
| *** xek_ has joined #openstack-infra | 20:28 | |
| *** xek__ has quit IRC | 20:30 | |
| sean-k-mooney | fungi: thanks for squid/mod_proxy caching do you just set it as the http_proxy or will the mirrors role be smart enough to just set it in the config for the approciate package managers | 20:39 |
| sean-k-mooney | just reading though https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/configure-mirrors now but i assume it configre the proxy option in the packagemanager | 20:41 |
| *** sboyron has quit IRC | 20:41 | |
| fungi | sean-k-mooney: we explicitly use the proxy's url in places where we tell things to fetch files | 20:41 |
| *** ralonsoh has quit IRC | 20:42 | |
| fungi | so not setting http_proxy really, though i suppose that might work for some stuff | 20:42 |
| sean-k-mooney | https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/configure-mirrors/tasks/mirror.yaml#L17-L34 | 20:42 |
| sean-k-mooney | it looks like its seeting the mirros in pip.conf | 20:42 |
| sean-k-mooney | and proably in the apt/yum config | 20:43 |
| fungi | in our case we don't really know what ip addresses things might come from, everything's open to the internet, and we don't want to turn into open proxies for every tom dick and mallory to exploit, so we just reverse-proxy specific resources instead | 20:43 |
| sean-k-mooney | fungi: ah i see | 20:44 |
| *** zzzeek has quit IRC | 20:44 | |
| sean-k-mooney | well in my case ill be running cache/proxy/mirror in a vm on the same openstack as the test vms so i dont need to make that vm public | 20:45 |
| *** zzzeek has joined #openstack-infra | 20:46 | |
| sean-k-mooney | i can lock it down to the subnet used by the ci | 20:46 |
| fungi | yes, your situation is potentially more flexible | 20:46 |
| fungi | while it's great that we have donated resources from so many providers, it does mean that there are some things not quite so much under our control | 20:48 |
| sean-k-mooney | ya | 20:49 |
| sean-k-mooney | unless you can boot all the vms on the same network | 20:49 |
| sean-k-mooney | you cant really lock down the mirrors as well | 20:49 |
| sean-k-mooney | if you can boot them all on the same network then you can make it proxy only on the internal ip and have then use that | 20:50 |
| *** armax has joined #openstack-infra | 20:50 | |
| fungi | in most of our providers we could, but also it's not uncommon for providers to expect you to use fips to reach things if you're not booting on a shared provider network, and we don't necessarily have enough fip quota for our external executors to reach all the nodes in some providers | 20:51 |
| sean-k-mooney | well if they provide the neutorn api expecting you to use fip is not really valid | 20:52 |
| sean-k-mooney | but rackspace dont provide the neutron api | 20:52 |
| sean-k-mooney | so in that case ya | 20:53 |
| fungi | alternatively nodes could talk to the proxy on a shared rfc-1918 network which isn't the global ipv4 network they're reached through, but then you get a weird triangulation problem | 20:54 |
| fungi | works for specific reverse-proxied resources, not so well as a general passthrough | 20:54 |
| *** mihalis68_ has quit IRC | 20:57 | |
| sean-k-mooney | well yes that more or less what i ment. you create a tenant network using a private ipv4 subnet then they talk to the proxy over that tenant netowrk | 20:57 |
| sean-k-mooney | if the proxy has the data it serves it if not it forawds it | 20:58 |
| sean-k-mooney | it does add a point of failure however | 20:58 |
| sean-k-mooney | e.g. the porxy | 20:58 |
| fungi | however if your executor does not have an interface on that network, then is it reaching your nodes via a different path? is the linux kernel smart enough to be able to do return path routing these days? | 20:59 |
| sean-k-mooney | by executor do you mean the zuul executor | 20:59 |
| *** sboyron has joined #openstack-infra | 21:00 | |
| sean-k-mooney | that is sshing to the vm either via the floating ip or a public ipv4/v6 address | 21:00 |
| sean-k-mooney | so its basically independent of what adress/route the packages form apt/dnf take | 21:01 |
| sean-k-mooney | but also ya linux can reply to request on a different interface then it recived it on | 21:01 |
| sean-k-mooney | its configurbale via sysctl | 21:02 |
| sean-k-mooney | you dont really need to cache your zuul executor since its just downsloadign the deleta form its local git repos to the gerrit version so those git repos baiclly act as a cache since it does delta encodeing for the git data | 21:03 |
| sean-k-mooney | anyway https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/configure-mirrors looks like it will do what i want i just need to deploy the cache/reverse proxy | 21:04 |
| fungi | zuul executor, yeah. basically asking if a linux guest configured with its default route through a transparent proxy can still reply to an externally-routed connection through a different interface through which it lacks knowledge of an actual return route to the initiator | 21:07 |
| fungi | though i guess http_proxy envvars are the workaround to not having to make your proxy also your gateway | 21:08 |
| *** rfolco|ruck has quit IRC | 21:08 | |
| sean-k-mooney | yep or setting it in the config files for the different package managers | 21:08 |
| sean-k-mooney | they all support havign a proxy listed that is not your defualt http_proxy | 21:09 |
| sean-k-mooney | so you can leave all normally http/tcp traffic un proxied but just make those package manages use it selectivly | 21:09 |
| sean-k-mooney | which is what the zuul-job configure-mirros role does | 21:09 |
| *** hashar has quit IRC | 21:32 | |
| *** rcernin has joined #openstack-infra | 21:35 | |
| *** rcernin has quit IRC | 21:37 | |
| *** rcernin has joined #openstack-infra | 21:38 | |
| *** zzzeek has quit IRC | 21:45 | |
| *** zzzeek has joined #openstack-infra | 21:47 | |
| *** eharney has quit IRC | 21:50 | |
| *** xek_ has quit IRC | 22:08 | |
| *** xarses_ has joined #openstack-infra | 22:34 | |
| *** xarses has quit IRC | 22:34 | |
| *** dayou has quit IRC | 22:51 | |
| *** dayou has joined #openstack-infra | 22:51 | |
| *** rcernin has quit IRC | 22:54 | |
| *** rcernin has joined #openstack-infra | 22:54 | |
| *** rlandy has quit IRC | 23:07 | |
| *** sboyron has quit IRC | 23:08 | |
| *** slaweq has quit IRC | 23:13 | |
| *** sboyron has joined #openstack-infra | 23:26 | |
| *** dchen has joined #openstack-infra | 23:38 | |
| *** dviroel has quit IRC | 23:39 | |
| *** tosky has quit IRC | 23:51 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!