| 16 Mar 2026 |
hexa | -sh-3.2# df -h
Filesystem Size Used Avail Capacity iused ifree %iused Mounted on
/dev/disk3s1s1 228Gi 11Gi 42Gi 22% 453k 436M 0% /
devfs 203Ki 203Ki 0Bi 100% 704 0 100% /dev
/dev/disk3s6 228Gi 2.0Gi 42Gi 5% 2 436M 0% /System/Volumes/VM
/dev/disk3s4 228Gi 8.2Gi 42Gi 17% 3.7k 436M 0% /System/Volumes/Preboot
/dev/disk3s2 228Gi 65Mi 42Gi 1% 108 436M 0% /System/Volumes/Update
/dev/disk1s2 500Mi 6.0Mi 483Mi 2% 1 4.9M 0% /System/Volumes/xarts
/dev/disk1s1 500Mi 5.7Mi 483Mi 2% 30 4.9M 0% /System/Volumes/iSCPreboot
/dev/disk1s3 500Mi 964Ki 483Mi 1% 39 4.9M 0% /System/Volumes/Hardware
/dev/disk3s3 228Gi 96Gi 42Gi 70% 1.9M 436M 0% /System/Volumes/Data
/dev/disk3s7 228Gi 68Gi 42Gi 62% 2.6M 436M 1% /nix
map auto_home 0Bi 0Bi 0Bi 100% 0 0 - /System/Volumes/Data/home
/dev/disk4 512Mi 12Mi 500Mi 3% 4 4.3G 0% /private/var/run/agenix.d
| 11:35:36 |
hexa |  Download image.png | 11:36:19 |
hexa | which brings it down to 81.8% | 11:36:30 |
hexa | are we looking at a cow filesystem that incorrectly reports usage data? | 11:37:32 |
K900 | We have another Pipewire bugfix release | 13:57:36 |
K900 | I'm tempted to send it | 13:57:40 |
K900 | Because the darwins are stalling everything out anyway | 13:57:46 |
| 17 Mar 2026 |
emily | APFS is volume-based, like btrfs/ZFS/etc.; the 228 GiB on the Nix store is shared with /System/Volumes/Data, so ~72%? though that's still lower than Prometheus | 00:51:44 |
hexa | so we get ~60 GiB of good use on those machines? | 00:52:39 |
hexa | we should get rid off them in that case | 00:52:49 |
hexa | and replace them with machines that have sensible disk capacity | 00:53:00 |
emily | er sorry yeah it's shared between all of those volumes | 01:02:25 |
emily | and yeah a ~256 GiB drive is not very reasonable for macOS + Nix | 01:02:31 |
emily | at least for a builder node | 01:02:53 |
Vladimír Čunát | Sandro 🐧: I noticed that you maintain python3Packages.mitmproxy-linux which is regressing now (in a weird way): https://hydra.nixos.org/build/323488743/nixlog/3/tail | 14:30:09 |
hexa | cp ${./fix-mitmproxy-linux-redirector-path.diff} tmp.diff
substituteInPlace tmp.diff \
--replace-fail @mitmproxy-linux-redirector@ $out/bin/mitmproxy-linux-redirector
patch -p1 < tmp.diff
patches + replaceVars | 14:42:50 |
hexa | ok, then I guess we should cancel all hetzner macs and get some at oakhost instead | 14:45:23 |
hexa | we're currently spending ~60 EUR for each one and have five | 14:45:41 |
hexa | so the current budget is ~300 EUR | 14:45:48 |
hexa | * so the current budget is ~300 EUR/mo | 14:45:50 |
hexa |  Download image.png | 14:46:25 |
hexa | cool, the interesting options are all out of stock | 14:46:33 |
hexa | I stand corrected, that is not the offender | 14:47:07 |
hexa | ../mitmproxy-rs-*-vendor/aya-build-*/src/lib.rs resolves to nothing | 14:48:14 |
Vladimír Čunát | Out of stock and overall the pricing is much higher? | 14:48:55 |
hexa | kinda, yeah | 14:50:35 |
hexa | but hetzner doesn't offer any new macs anymore, so we're out of options | 14:50:54 |
hexa | we could buy them instead | 14:51:09 |
Vladimír Čunát | We can't... like attach an SSD over USB, right? 😜 | 14:51:49 |
Vladimír Čunát | (don't know what else to do with Apple pricing) | 14:52:04 |