| 5 Dec 2025 |
 aloisw | In reply to @raitobezarius:matrix.org
have you tried a lower value? I have not, but I think the problem is not so much that the size is excessive, but that the checkpointer is never alone so the WAL grows without bound. | 06:26:19 |
| aloisw | In reply to @raitobezarius:matrix.org
but i feel like the fact that lix is blocked by the potential event that the WAL contains a record relevant to it is a mistake given our usage of flock to mark the future happening of a store path Thanks for the hint, I will investigate whether this has any influence on the situation. | 06:27:54 |
| aloisw | In reply to @raitobezarius:matrix.org
maybe we can improve things by initiating checkpoints ourselves at key points… SQLite does that because of autocheckpoint, but the problem is that these are non-blocking (PASSIVE) checkpoints. I will investigate tomorrow whether RESTART makes the situation better; it should prevent unbounded WAL growth at the expense of some concurrency, but maybe it's still a net win if the write transactions aren't slowed down so much. | 06:30:51 |
| aloisw | In reply to @raitobezarius:matrix.org
whereas for writes, it seems it'd be good if we could have multiple WAL so that once one is committed, the other can be still filled? wal2 when | 06:31:04 |
| 13 May 2024 |
|  zrsk joined the room. | 13:37:30 |
|  dariof4 joined the room. | 14:16:05 |
|  abbe joined the room. | 14:42:22 |
 mjm | lix is now in nixos-unstable, what's the best way to use it in a NixOS config? | 15:47:41 |
 K900 | nix.package = pkgs.lix | 15:47:54 |
| K900 | Probably | 15:47:58 |
| K900 | Or just keep using the module | 15:48:02 |
 puck | you can use nix.package = pkgs.lix; if you want; but using the overlay will guarantee more updates (tho less binary cache) | 15:48:44 |
| mjm | well, the module/overlay doesn't easily support using the one from nixpkgs | 15:48:44 |
| mjm | i do want more binary cache | 15:49:21 |
| mjm | yeah i guess i'll just do nix.package then | 15:50:43 |
 raitobezarius | to have binary cache and HEAD, we will probably need to track the channels and bolt a lix on there and rebuild with our own CI then push in our binary cache | 17:33:17 |
 Charles | What's lix using as its binary cache? Attic? | 17:34:36 |
|  Arian left the room. | 17:37:44 |
 @tc424:glasgow.social | Last I heard they were using garage as the back end, and think about attic | 17:39:11 |
| @tc424:glasgow.social | I don't know if that means they're just pushing directly into garage's s3 interface for the moment | 17:39:36 |
| @tc424:glasgow.social | * Last I heard they were using garage as the back end, and thinking about attic | 17:39:57 |
| raitobezarius | In reply to @charles:computer.surgery
What's lix using as its binary cache? Attic? garage without attic | 17:41:16 |
| raitobezarius | it's homemade just nix push thingie | 17:41:24 |
| raitobezarius | and nix sign | 17:41:25 |
 Qyriad | yep | 17:44:26 |
| @tc424:glasgow.social | I've just given up on attic due to weird non-deterministic auth problems that nobody else seems to have experienced | 17:44:31 |
| @tc424:glasgow.social | Still half convinced I'm "holding it wrong" somehow | 17:44:49 |
| Charles | Yeah I experienced lots of jank with attic | 17:44:56 |
| Charles | I think it turned out to be my fault though, for the big issue | 17:45:11 |
| Charles | But sometimes during pushing the connection just dies for no reason | 17:45:29 |
