| 26 Oct 2021 |
 @grahamc:nixos.org | done | 16:56:19 |
 Janne Heß | thank you as well :) | 16:56:30 |
 Linux Hackerman | grahamc (he/him): while you're at it, could you give my git@sphalerite.org account restart access as well? I already have it on my mayflower account, but would prefer to use github login generally | 16:57:08 |
| @grahamc:nixos.org | done | 16:58:17 |
| Linux Hackerman | Thanks! | 17:24:28 |
| Janne Heß | Good news: Looks like the hibernate test on trunk-combined isn't flakey anymore
Bad news: It fails consistently now | 17:36:38 |
| @grahamc:nixos.org | great! | 17:45:49 |
| Janne Heß | Looks like that's a regression of the improved thread cleanup as well. The test calls create_machine (haven't seen that one before tbh…) and this machine is never registered with the Driver so the Driver doesn't know that it also needs to be killed | 20:29:11 |
 cole-h | In reply to @grahamc:nixos.org
I think cole-h is going to spend some time on ofborg today And spend some time I did -- somehow most of the builders / evaluators fell over (and the singular Mac builder we have was unreachable), so that was fixed first. Then, the aarch64 builders were stuck, so I had to fix that (which really boiled down to just restarting the box, but there were a few more annoying steps involved). Now, things should start to improve. Went from ~450 pending evals to 390, and 3120 (lol) pending aarch64 builds to 3000 in ~5 minutes. | 21:29:50 |
| @grahamc:nixos.org | Thank you a lot Cole | 21:32:05 |
| Janne Heß | Vladimír Čunát: you did also see breakages on ext4, right? | 21:38:08 |
| Janne Heß | currently throwing together a bug report | 21:38:44 |
 Vladimír Čunát | In reply to @janne.hess:helsinki-systems.de
Vladimír Čunát: you did also see breakages on ext4, right? the peertube regression but not jx | 21:39:03 |
| Janne Heß | Redacted or Malformed Event | 21:42:55 |
| Janne Heß | Redacted or Malformed Event | 21:43:09 |
| Janne Heß | aww man. All my messages went to the wrong room, I'm sorry :( | 21:45:35 |
| Janne Heß | Should have been in the staging room | 21:45:55 |
|  piegames joined the room. | 21:48:33 |
 K900 | Are Darwin builds on Hydra still broken? | 22:22:00 |
 Ryan Burns | Most (all?) x86_64-darwin builders are stuck right now, if that's what you mean | 22:23:01 |
| piegames | *The* darwin builder (I just learned that apparently there is only one) | 22:27:10 |
| Ryan Burns | In reply to @piegames:matrix.org
*The* darwin builder (I just learned that apparently there is only one) there are many hydra builders, there is only one ofborg builder | 22:27:47 |
| piegames | Ah, thank you | 22:28:59 |
| K900 | In reply to @r-burns:matrix.org
Most (all?) x86_64-darwin builders are stuck right now, if that's what you mean Yeah, that | 22:36:32 |
| K900 | Has anyone looked into it? | 22:37:01 |
| K900 | It seems to be happening consistently for the past few weeks... | 22:37:21 |
| Ryan Burns | I think grahamc (he/him) is the only one with the power | 22:38:40 |
| Ryan Burns | Supposedly restarting the queue runners regularly is a workaround | 22:39:19 |
| @grahamc:nixos.org | there are a few others but they're less active | 22:41:11 |
| @grahamc:nixos.org | I'll kick it over | 22:41:15 |
