2026-09-01 04:57:47 <ikke> great, somehow aports got pulled in the go package index: https://pkg.go.dev/gitlab.alpinelinux.org/alpine/aports
2026-09-01 05:00:45 <lotheac> :S
2026-09-01 05:01:16 <lotheac> libzstd_system.go exists in both testing/victoria-logs and community/victoria-metrics... that's enough to cause that?
2026-09-01 05:02:18 <ikke> Well, someone would need to ask go to pull that as a module 
2026-09-01 07:38:16 <jvvv> gitaly seems to be down
2026-09-01 10:44:54 <omni> I think both build-edge-aarch64 and -riscv64 are stuck
2026-09-01 17:12:31 <omni> ikke: would you min kicking them?
2026-09-01 17:12:41 <omni> (also not sure who else can)
2026-09-01 19:13:29 <omni> build-edge-aarch64 is moving again
2026-09-01 19:13:56 <ikke> yeah, kicked it
2026-09-01 19:14:34 <omni> <3
2026-09-01 19:46:55 <omni> https://gitlab.com/gitlab-org/gitlab/-/work_items/592536
2026-09-01 19:49:48 <omni> so there was a technical reason for dumbing down the ui
2026-09-01 19:50:41 <ikke> Is there a kind of irony that that page takes long to load for me?
2026-09-02 00:42:51 <iggy> https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/testing/victoria-logs/libzstd_system.go and https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/community/victoria-metrics/libzstd_system.go ... good reason not to have raw go files in aports
2026-09-02 15:41:56 <jvvv> ikke: The new mirror I've got staged for adding has offerred to be a new tier0 mirror. For ref, see WorkItem alpine/infra/mirrors#1225
2026-09-02 15:44:23 <ikke> jvvv: would be t1 (t0 is dl-master).
2026-09-02 15:44:51 <ikke> We haven't added 3rd parties as t1 mirrors, and that would only work if they have short sync times
2026-09-02 15:47:50 <jvvv> I haven't asked any thing at this point. They offerred and I wasn't sure how to respond. I can ask the contact about it or you can step in if you would like
2026-09-02 15:51:24 <jvvv> It's 20Gbps and I've verified that the mirror looks in good shape. Also, they actually are offerring to be rsync.alpinelinux.org. I expect that is out of scope, but IDK
2026-09-02 15:52:59 <ikke> rsync.a.o is a round-robbin geo-dns service, so we can add more backing servers
2026-09-02 15:53:29 <ikke> The way the current t1 servers work is that they subscribe to msg.a.o and update immediately, with an additional sync every 15m
2026-09-02 16:03:54 <jvvv> Is it ok to quote your last two lines in the workitem?
2026-09-02 17:53:07 <jvvv> ikke: When you get a chance, I think you are going to need to step into the the discussion with mirror.zyner.org
2026-09-02 17:53:20 <jvvv> I think I'm out of my depth
2026-09-03 11:57:41 <ikke> ptrc: ftr, I have been applying some changes manually on algitbot, but it somehow refuses to ignore commits on alpine/rust
2026-09-03 13:52:53 <ptrc> ikke: hm, maybe i messed something up 
2026-09-03 14:02:34 <ikke> ptrc: I forgot to rebuild the container (thought the script was mounted, but it's including in the image)
2026-09-03 15:13:24 <omni> build-3-24-x86 seems stuck
2026-09-03 15:14:04 <ikke> kicked
2026-09-03 16:06:48 <omni> <3
2026-09-04 12:20:13 <omni> are the three build-edge-a* in some kind of limbo?
2026-09-04 14:46:30 <ikke> omni: Doesn't look like it, they appear to be up-to-date and building
2026-09-04 15:03:42 <omni> yeah, perhaps it just looked like it, they began moving not long after I had said that
2026-09-04 15:04:13 <omni> it looked like they had a bit to catch up, but that could be me misinterpreting what I saw
2026-09-04 16:03:59 <invoked> moving from #alpine-linux (and i'm almost out of time)... for shits & giggles (mostly shits) i ran git-sizer on aports: https://bpa.st/3NZ4Y
2026-09-04 16:05:32 <ikke> Trees total size is what's the issue I expect
2026-09-04 16:05:53 <ikke> It's compressed on disk, so not taking a lot of space, but just soo many objects
2026-09-04 16:09:23 <invoked> yeah. dunno what can be done about that
2026-09-04 16:09:48 <invoked> that said, if it's the problem, it'll just get worse
2026-09-04 16:10:47 <ikke> We could decide to truncate history at some point, but that would require some coordination
2026-09-04 16:12:31 <invoked> that could be painful
2026-09-04 16:13:00 <invoked> all the hashes would still change, right
2026-09-04 16:13:12 <ikke> It would, yes
2026-09-04 16:13:40 <invoked> does gitlab have some magic way of handling that?
2026-09-04 16:14:02 <invoked> where it fixes all the links and everything?
2026-09-04 16:17:56 <ikke> It does have some rewriting support, but not sure to what extent it will keep links working
2026-09-04 18:59:43 <achill> mrow mroww https://gitlab.alpinelinux.org/alpine/infra/gitlab-tf/-/merge_requests/67 so i dont get presumeably removed when it gets runs again
2026-09-04 20:13:13 <iggy> just restart the aports repo every year on Jan 1 (I'm only half joking)
2026-09-05 05:17:09 <ikke> achill: thanks. I've created a new MR that adds moved statements that prevent the memberships from being destroyed
2026-09-05 05:17:13 <ikke> https://gitlab.alpinelinux.org/alpine/infra/gitlab-tf/-/merge_requests/68
2026-09-05 12:18:45 <invoked> if anyone's ever been curious: aports is a 3.1G fossil repo, when exported. https://bpa.st/R5EXK