30 Jul 2024 |
Matthew Kenigsberg | cole-h: how did you track down the URL for the beta update? I wanted to try version 4 (you linked 3) but I'm not sure where to find it. And can't sign in with my apple id in the VM | 22:16:30 |
cole-h | I googled and downloaded it from a random website 😅 | 22:17:01 |
cole-h | Or rather found the URL on a random website | 22:17:28 |
cole-h | I think the magical search terms are something like "macOS 15 beta package installer" | 22:18:17 |
Matthew Kenigsberg | Lol nice | 22:19:18 |
31 Jul 2024 |
| @maralorn:maralorn.de left the room. | 13:13:22 |
abathur | Matthew Kenigsberg cole-h already got a tentative response:
We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:04:01 |
abathur | * Matthew Kenigsberg cole-h already got a tentative response:
We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:04:23 |
cole-h | (Imagining that last line is their response and yours is everything before that) | 15:07:57 |
abathur | the nested quote was from mine, and everything after that was the devrel's response | 15:08:34 |
cole-h | It just shows one whole quote for me (on mobile), but understood | 15:11:13 |
abathur | ah | 15:14:53 |
abathur | * Matthew Kenigsberg cole-h already got a tentative response:
Abathur: We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
Devrel: We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:15:24 |
abathur | * Matthew Kenigsberg cole-h already got a tentative response:
**Abathur: ** We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
Devrel: We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:15:44 |
abathur | * Matthew Kenigsberg cole-h already got a tentative response:
Abathur:
We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
Devrel: We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:16:11 |
abathur | * Matthew Kenigsberg cole-h already got a tentative response:
Abathur:
We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
Devrel: We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:16:23 |
abathur | * Matthew Kenigsberg cole-h already got a tentative response:
Abathur: We're tentatively looking at setting the GID of our group (nixbld) to 331 and the UID of our first build user (_nixbld1) to 331. Default installs create 32 build users, but Nix users can customize this at install time. Since Nix needs one build user for every concurrent build, we're trying to balance between the risk of low UIDs clashing with new macOS users in future updates, and still starting low enough that Nix users with Max/Ultra CPUs configure enough users to put that hardware to good use :)
Devrel: We had a bit of a discussion, and this is probably the most expedient but the number being tossed around was 350.
I’m getting access to the bug again to see if anyone had any better advice.
| 15:16:28 |
abathur | annotated | 15:16:46 |
Matthew Kenigsberg | Well it's nice you're getting a response | 16:47:17 |
| jopejoe1 joined the room. | 20:25:37 |
3 Aug 2024 |
abathur | Matthew Kenigsberg cole-h pushed on the migration script again; new approach that tries to move the existing users to an Nth unused range before trying to actually place them | 06:24:20 |
8 Aug 2024 |
Matthew Kenigsberg | Sorry haven't gotten around to looking yet. Hoping to get to it soon | 04:17:59 |
| Alyssa Ross set a profile picture. | 18:48:55 |
12 Aug 2024 |
Matthew Kenigsberg | Okay finally got to it, seemed to work! | 23:41:50 |
13 Aug 2024 |
Matthew Kenigsberg | @tomberek has been working on how we handle UIDs and was thinking about joining tomorrow, sound okay abathur ? | 17:43:11 |
16 Aug 2024 |
abathur | cole-hMatthew Kenigsberg after y'all had to drop @tomberek had one more thought that may be worth weighing WRT migration
The most-immediate cause of the _nixbldN errors may be the nixbld group holding on to GroupMembership entries for the deleted users.
I think he ran something like dscl . delete /Groups/nixbld GroupMembership _nixbld1 _nixbld2 _nixbld3 _nixbld4 to clean them up, which should mean the daemon won't try assigning builds to the clobbered users.
I'm still chewing on whether that's useful...
- It does give us something like a one-liner to get affected Nix installs working (as long as they had at least 5 build users)
- If something was built into Nix-nix that could fully fix up the users, the above oneliner could be chained with something to update Nix
- But without that feature, I do worry a little that publicizing it as a workaround may also lead people to avoid dealing with the UIDs of the remaining users and leave themselves open to a repeat of the same problem on near-future macOS updates?
| 18:06:19 |
abathur | I also looked at the 2nd install experience and (with the asterisk that this was on the pre-catalina macbook) after installing 2.18.5 once, both the 2.18.5 and 2.24.2 installers failed before the user stage on the check for existing shell profile backups.
That has me leaning towards not shimming the migration script into the installer, or at least saving that for a followup. Even if I shimmed it in before that check, I don't think it sends the right impression if an install that ultimately fails early does indeed still mutate these users (whether or not we're telling those users to fix it that way or not).
| 18:14:21 |
11 Sep 2024 |
| Matthew Kenigsberg left the room. | 14:57:49 |
16 Sep 2024 |
| silentlurker joined the room. | 20:01:17 |
28 Sep 2024 |
| rajudev joined the room. | 22:14:30 |