| 19 Jun 2026 |
 vcunat | I can see a partial log here
[root@goofy-hopcroft:~]# nix-store -l /nix/store/fdccspjycvm1lm99aimnq5fw7rrbh4ny-firefox-unwrapped-152.0.drv
| 06:04:04 |
| vcunat | But there were multiple aborted attempts of that build on that machine. | 06:04:29 |
| vcunat | (we were restarting the queue-runner a lot, and the build takes hours) | 06:04:40 |
 leona | quite a few builds are 'receiving outputs' for multiple hours now: https://hydra.nixos.org/machines on multiple machines | 08:12:09 |
 Mic92 | I can check again, but yesterday it was genuine compression cpu load it had to do and it was competing with build resources | 08:28:36 |
| Mic92 | we are using the highest zstd compression ratio | 08:28:51 |
| leona |
3h for etc.drv is much..
| 08:31:57 |
| leona | * >3h for etc.drv is much.. | 08:32:05 |
| Mic92 | Has to wait. I am currently working on something else | 08:35:19 |
| Mic92 | Deploying to hydra-staging now to sanity check my scheduler chhanges | 08:52:15 |
| Mic92 | * Deploying to hydra-staging now to sanity check my scheduler changes | 08:58:32 |
 hexa (signing key rotation when) | full disk on mimas o_O | 10:13:14 |
| hexa (signing key rotation when) | https://grafana.nixos.org/d/rYdddlPWk/node-exporter-full?orgId=1&from=now-2d&to=now&timezone=browser&var-datasource=default&var-job=node&var-node=mimas.nixos.org:9100&var-diskdevices=%5Ba-z%5D%2B%7Cnvme%5B0-9%5D%2Bn%5B0-9%5D%2B%7Cmmcblk%5B0-9%5D%2B&refresh=1m&viewPanel=panel-152 | 10:13:56 |
| hexa (signing key rotation when) | what happend last night? | 10:14:02 |
| Mic92 | [root@mimas:~]# systemctl status nix-gc
× nix-gc.service - Nix Garbage Collector
Loaded: loaded (/etc/systemd/system/nix-gc.service; linked; preset: ignored)
Active: failed (Result: exit-code) since Fri 2026-06-19 09:15:03 UTC; 1h 0min ago
Invocation: cde0a9338bf3435c96213d772e8c38b1
TriggeredBy: ● nix-gc.timer
Process: 2752969 ExecStart=/nix/store/6j08xq0n9zlz3jdx7fv97svljmkxci0z-unit-script-nix-gc-start/bin/nix-gc-start (code=exited, status=1/FAILURE)
Main PID: 2752969 (code=exited, status=1/FAILURE)
IP: 0B in, 0B out
IO: 300K read, 0B written
Mem peak: 8M
CPU: 37ms
| 10:15:49 |
| Mic92 | I restarted it | 10:16:58 |
| Mic92 | Jun 19 10:16:56 mimas nix-gc-start[2850168]: error: interrupted by the user | 10:17:08 |
| Mic92 | hexa (signing key rotation when): ^ was this ue? | 10:17:18 |
| Mic92 | * hexa (signing key rotation when): ^ was this you? | 10:17:23 |
| hexa (signing key rotation when) | yep | 10:17:23 |
| hexa (signing key rotation when) | running fast-nix-gc instead | 10:17:44 |
| Mic92 | alright | 10:17:59 |
| hexa (signing key rotation when) | ehhh, the queue-runner came back up | 10:20:02 |
| hexa (signing key rotation when) | oh, socket activation | 10:20:18 |
| Mic92 | oh my god: [root@mimas:~]# du -sh /nix/var/nix/db/db.sqlite
16G /nix/var/nix/db/db.sqlite
| 10:28:55 |
| Mic92 | I was wondering why nix-fast-gc was taking so long in single-threaded mode | 10:29:32 |
| hexa (signing key rotation when) | is this already a size where behaviors get pathological? :P | 10:29:33 |
| Mic92 | no | 10:29:41 |
| Mic92 | but you need the RAM | 10:29:45 |
| Mic92 | I tried to make it efficient | 10:29:49 |
