| 6 Oct 2021 |
 sterni | ah nvm that's an input in the hydra config isn't it | 14:13:47 |
 hexa | it is | 14:18:47 |
 Vladimír Čunát | I think that was temporary due to darwin queue overload. | 14:24:13 |
| Vladimír Čunát | * I suspect that was temporary due to darwin queue overload. | 14:24:28 |
| Vladimír Čunát | ... though that happens regularly. | 14:24:39 |
| 7 Oct 2021 |
| sterni | even… right now | 10:26:53 |
| sterni | darwin seems quite stuck | 10:26:57 |
 lukegb (he/him) | hrm, python3 aiohttp seems broken in staging | 15:47:18 |
| lukegb (he/him) | (and python3 argh) | 15:47:55 |
| sterni | lukegb (he/him): what is the python failure | 15:49:46 |
| sterni | * lukegb (he/him): what is the python3 failure? | 15:49:51 |
| lukegb (he/him) | no, the package argh for python3 | 15:50:02 |
| lukegb (he/him) | not "argh! python3 is broken" | 15:50:08 |
| sterni | ah lol | 15:50:09 |
| lukegb (he/him) | (I realise that was ambiguous) | 15:50:16 |
| sterni | so I haven't broken it :p | 15:50:26 |
| lukegb (he/him) | argh is https://github.com/neithere/argh/issues/148 | 15:51:16 |
 Alyssa Ross | argh | 16:24:30 |
 jonringer | still curious why this broke between patch releases, first time I've seen such a failure | 16:27:49 |
 trofi | The library seems to rely heavily on docstrings to do thing. I found it very fragile as those get discarded (yes?) when python bytecode is optimised. | 18:00:40 |
| lukegb (he/him) | Only at double-optimized "this might break some programs" mode | 18:04:25 |
| 8 Oct 2021 |
|  Ryan Burns joined the room. | 19:08:21 |
| hexa | we might need to revert the python39 bump on staging due to some rather interesting breakages in aiohttp et al | 23:16:58 |
| lukegb (he/him) | yeah | 23:58:20 |
| lukegb (he/him) | something is breaking tests, too | 23:58:33 |
| lukegb (he/him) | but not... consistently? | 23:58:39 |
| lukegb (he/him) | like, the python interpreter appears to sometimes be crashing on shutdown | 23:58:47 |
| 9 Oct 2021 |
| hexa | meh. | 00:00:17 |
| lukegb (he/him) | e.g. https://hydra.nixos.org/build/155606118/nixlog/39 | 00:18:32 |
| lukegb (he/him) | Fatal Python error: _enter_buffered_busy: could not acquire lock for <_io.BufferedWriter name='<stderr>'> at interpreter shutdown, possibly due to daemon threads
Python runtime state: finalizing (tstate=0x441670)
Current thread 0x0000fffff7ff47b0 (most recent call first):
<no Python frame>
/nix/store/cf43j6r16pir93wc97wl10n8ybww6pa8-stdenv-linux/setup: line 1319: 4 Aborted (core dumped) /nix/store/hn2kc3fkzm72acrgj0nmvqcchln7jxrc-nixos-test-driver-dhparams/bin/nixos-test-driver
builder for '/nix/store/gqaa94hihgl54szsjgpv9iaghkpcsyx8-vm-test-run-dhparams.drv' failed with exit code 134
| 00:18:43 |
