| 17 Aug 2021 |
 baloo | https://buildkite.com/grahamc/r13y-dot-com/builds/854 | 15:10:51 |
| baloo | so essentially we ran out of disk space | 15:11:24 |
| baloo | I'm confused, how can it be that large | 15:13:17 |
| baloo | buildkite says it uploaded a new index.html, but I can't see it | 15:15:52 |
 atemu12 | https://github.com/grahamc/r13y.com/issues/28 | 15:15:58 |
| baloo | fair fair, but what I don't understand is why it does not produce output for the minimal iso since last thursday | 15:17:04 |
| baloo | https://buildkite.com/grahamc/r13y-dot-com/builds/854#9b7f3b86-36e5-4c7f-92cd-c4442bc7f7b6/41-47 | 15:18:16 |
| baloo | https://r13y.com/iso_minimal/index.html iso_minimal got uploaded | 15:18:42 |
 @grahamc:nixos.org | I had an outage and didn’t fix it until
This morning 😬 | 15:54:30 |
 j4m3s | I'm confused, why is there multiple XXX-bash-44 on the failed iso_minimal ? I know the unchecked is because it failed, I'm asking why there'd be multiple bash derivations ? There is multiple use of bash with different inputs in nixpkgs ? | 15:56:56 |
 toonn | Maybe it's bash from difference stages of the stdenv? | 16:00:22 |
 Linux Hackerman | Those are the patches applied to bash | 17:06:58 |
| Linux Hackerman | Because bash's release model is a bit weird. | 17:08:13 |
| Linux Hackerman | http://ftp.gnu.org/gnu/bash/bash-4.4-patches/ | 17:08:16 |
| Linux Hackerman | That is, those are source drvs | 17:08:32 |
| baloo | In reply to @grahamc:nixos.org
I had an outage and didn’t fix it until
This morning 😬 I don't understand the discrepancy between the log exposed in buildkite and the page not showing up on r13y. Even with your outage :D | 17:29:58 |
| baloo | I might be missing something | 17:30:05 |
| @grahamc:nixos.org | ah, the builds execute on one set of hosts that have no privileges | 17:30:23 |
| baloo | ha well.. it's now updated ... | 17:30:30 |
| @grahamc:nixos.org | then the upload to r13y requires a different task executing on a host which does have privileges | 17:30:42 |
| baloo | HA | 17:30:48 |
| baloo | I thought it was just a cloudfront in front of an S3 | 17:31:34 |
| @grahamc:nixos.org | yeah it is but the machine running the r13y task has no s3 creds | 17:33:07 |
| @grahamc:nixos.org | (or more specifically, no vault creds, and vault is used to get s3 creds | 17:33:16 |
| @grahamc:nixos.org | * (or more specifically, no vault creds, and vault is used to get s3 creds) | 17:33:17 |
| baloo | that fix might not have been enough. All considered https://github.com/grahamc/r13y.com/commit/602f61c3f17485752da422575ec8c41252dc3fce | 17:33:40 |
| baloo | https://buildkite.com/grahamc/r13y-dot-com/builds/854#9b7f3b86-36e5-4c7f-92cd-c4442bc7f7b6/41-45 | 17:34:08 |
| @grahamc:nixos.org | how about issuing cloudfront expirations | 17:34:20 |
| baloo | that's the machine you're talking about right? the task executed at 13:22 UTC today, and it was not showing up at 15:18 | 17:34:37 |
| @grahamc:nixos.org | oh interesting | 17:34:51 |
