| 24 Dec 2025 |
|  Chiroptical joined the room. | 17:13:30 |
| Chiroptical | Is there a good place to check for updates on infra being down? | 17:14:12 |
 raitobezarius | large scale downtimes are being broadcasted in https://zulip.afnix.fr/#narrow/channel/3-public/topic/Planned.20downtime/near/16487 | 17:26:25 |
| raitobezarius | but we should put up a status page somewhere and have a better way to communicate those | 17:26:36 |
| raitobezarius | once cache.lix.systems will be up again, we will notify people (or we will migrate if needed) | 17:26:53 |
| raitobezarius | https://zulip.lix.systems/#narrow/channel/7-Infra/topic/x86.20.26.20macOS.20CI.20down/with/3434 is also a good place to check | 17:27:12 |
 EsperLily [she/her] | i just hit this issue myself. i'm rather surprised it would scale up the timeout time so much, the fifth attempt had an 80-second timeout, so the cache being down meant my build stalled for over 2.5 minutes before continuing. is there any way to make the infra at least reject the connection while it's down? | 23:50:07 |
 niko ⚡️ | I know I don’t check Zulip so I have email notifications for relevant channels so I’m probably one of the relatively few people who knew about downtimes ahead of time lol | 17:30:15 |
| 25 Dec 2025 |
 Winter | i’m surprised c.l.s is even defined on anyone’s system? | 01:06:30 |
| Winter | is it only used when you use the installer? | 01:06:43 |
| raitobezarius | The infra is up again | 04:16:37 |
| raitobezarius | In reply to @esperlily:matrix.org
i just hit this issue myself. i'm rather surprised it would scale up the timeout time so much, the fifth attempt had an 80-second timeout, so the cache being down meant my build stalled for over 2.5 minutes before continuing. is there any way to make the infra at least reject the connection while it's down? Are you using IPv4? | 04:16:57 |
| raitobezarius | Yeah, there could be some blackholing somewhere to make this fail faster on the network level for sure | 04:17:18 |
| EsperLily [she/her] | i'm not sure what it's using to try and contact the cache, but I do have my computer set up with an IPv6 address and https://ipv6test.app says i'm using IPv6 to access that site | 04:18:38 |
| raitobezarius | Good to know, I will open issues to see what to do about that, thanks! | 04:19:20 |
| EsperLily [she/her] | that's actually a good question, i have it set up as an extra substituter in my configuration but i don't remember why | 04:31:56 |
| EsperLily [she/her] | was it recommended at one point to use past installation? | 04:32:11 |
| raitobezarius | The installer automatically put it | 04:33:16 |
| EsperLily [she/her] | no i added it to my nix-darwin configuration, in a commit 19 months ago that just says "Add cache settings for cache.lix.systems" | 04:34:18 |
| EsperLily [she/her] | i'm deleting it now of course, but i don't know why i added it in the first place, and searching for it in the current docs just shows it as part of an installer invocation. perhaps 19 months ago there was a different recommendation | 04:36:41 |
| raitobezarius | We definitely had it recommended in the very first days and it was a bad idea given we never populated it meaningfully with cached nightly builds | 04:37:53 |
| EsperLily [she/her] | alright that would explain it then, since i switched during the early days | 04:38:05 |
|  flacks changed their display name from flax to flacks. | 08:42:59 |
|  pentane (DECT CYPT/2978) changed their display name from pentane ⭔ to pentane (DECT CYPT/2978). | 11:23:34 |
|  Nicole/Drezil 🏳️⚧️ (she/her) changed their profile picture. | 13:08:40 |
 Sofie 🏳️⚧️ (she/her) | https://nesbitt.io/2025/12/24/package-managers-keep-using-git-as-a-database.html | 15:19:55 |
 K900 | Except nixpkgs also exists as snapshot tarballs that we host ourselves? | 15:28:04 |
| K900 | Like, "nixpkgs is giving github issues" is a "nixpkgs is a big repo" problem | 15:28:13 |
| K900 | Not a "nixpkgs uses git" problem | 15:28:17 |
 Lotte (it/its)/Cinny (she/her) | git handles big repos more poorly than they need to be
although nixpkgs still needs a significant amount of the repo checked out to function i think
| 15:35:06 |
