| 14 Mar 2026 |
 matthewcroughan | https://github.com/nix-community/disko/pull/1234 | 19:26:04 |
| matthewcroughan | this nocache utility doesn't exist in nixpkgs | 19:26:11 |
| matthewcroughan | But it does exist on Github | 19:26:21 |
| matthewcroughan | So I can only assume a clanker was asked about this issue, and the PR above is the result | 19:26:32 |
|  amadaluzia joined the room. | 19:28:41 |
 Mic92 | That was 12 months ago. Don't think lassulus was using any such tools at this time and llms would have been properly doing a better job at actually testing the code | 19:29:18 |
| matthewcroughan | 12 months ago? | 19:29:44 |
| matthewcroughan | the PR is from 15 mins ago | 19:29:48 |
| matthewcroughan | commit d2fd33640c8207f9bb50da1d6b88ba7ee64c6153 (HEAD -> master)
Author: lassulus <git@lassul.us>
Date: Sat Mar 14 20:13:40 2026 +0100
make-disk-image: use nocache for smaller memory footprint
diff --git a/lib/make-disk-image.nix b/lib/make-disk-image.nix
index a9c732c..80dc561 100644
--- a/lib/make-disk-image.nix
+++ b/lib/make-disk-image.nix
@@ -95,6 +95,7 @@ let
util-linux
findutils
kmod
+ nocache
]
++ cfg.extraDependencies;
preVM = ''
@@ -165,7 +166,7 @@ let
P=${toString cfg.copyNixStoreThreads}
''
}
- xargs -P "$P" -I {} cp --recursive {} "$rootMountPoint"/nix/store < ${closureInfo}/store-paths
+ xargs -P "$P" -I {} nocache cp --recursive {} "$rootMountPoint"/nix/store < ${closureInfo}/store-paths
${systemToInstall.config.system.build.nixos-install}/bin/nixos-install --root "$rootMountPoint" --system ${systemToInstall.config.system.build.toplevel} --keep-going --no-channel-copy -v --no-root-password --option binary-caches ""
umount -Rv "$rootMountPoint"
| 19:31:44 |
| matthewcroughan | The PR now says rsync instead of nocache, will try that | 19:31:57 |
| matthewcroughan | What is from 12 months ago though, are we talking about two different things? | 19:32:10 |
| Mic92 | @matthewcroughan:defenestrate.it: if this the attitude you bring up against us, I might just as well block as well so I don't have to read your slop anymore | 19:32:22 |
| matthewcroughan | I'm sorry but what is the offense for? | 19:32:33 |
| matthewcroughan | I asked if something was clankered, why get upset? | 19:32:40 |
 bandithedoge | are you always so damn confrontational? | 19:32:56 |
| matthewcroughan | Where is the confrontation above? | 19:33:13 |
| matthewcroughan | I asked if the PR was made by an LLM, because it seemed like it was | 19:33:26 |
| matthewcroughan | Is that in and of itself, offensive? | 19:33:41 |
| matthewcroughan | If not, then where did I mis-step? | 19:34:33 |
 lassulus | no, It was the clanker | 19:35:09 |
| lassulus | stop arguing :D | 19:35:12 |
| lassulus | I'm on my phone and didn't check | 19:35:18 |
| Mic92 | You criticized a work in progress pr that tries to solve one of your problems. | 19:35:43 |
| matthewcroughan | No, I've been debugging for 4 hours, the cause of an issue, successfully. | 19:36:00 |
| matthewcroughan | I'll just leave the reproducer on a PR, you can solve it yourself. Thanks. | 19:36:35 |
| Mic92 | I did debug this issue before and back than with zfs but arc back than looked normal while something consumed memory | 19:38:42 |
| matthewcroughan | It is not arc, it's kernel changes | 19:38:55 |
| matthewcroughan | it only happens with 6.12+ | 19:39:02 |
| lassulus | not sure it's changes, because we had this happen in the past as well | 19:39:09 |
| matthewcroughan | well it's some part of the kernel that gets poked often enough to cause the issue to occur across time | 19:39:27 |
