| 14 Jan 2025 |
 xanderio | Caching the build failure | 18:56:42 |
 Vladimír Čunát | And yes, if one job fails a build step and another one wants to build the same, it is not rebuilt. | 18:56:47 |
| Vladimír Čunát | Yes, that's a feature. | 18:57:01 |
| Vladimír Čunát | Like no, failed stdenv build won't get rebuilt a 40k times (or how many it would be). | 18:57:50 |
| xanderio | I can't find anything in the logs that would point to a build failure, and the cache build faiure was aborted by presumable an admin? | 18:57:50 |
| Vladimír Čunát | https://hydra.nixos.org/log/lylns9mpyc652cn3vf1a85xf88nznlc6-foundationdb-7.3.42.drv | 18:58:17 |
| Vladimír Čunát | No, it was normal failure. | 18:58:22 |
| Vladimír Čunát | It was restarted by someone. | 18:58:37 |
| Vladimír Čunát | (but not yet retried, which is why it isn't finished) | 18:58:54 |
| xanderio | I've look at the log and it just stops, with any kind of error message | 18:59:24 |
| * xanderio is confused by hydra | 19:00:05 |
| Vladimír Čunát | It's being built right now. | 19:00:07 |
| Vladimír Čunát | Failed, someone restarted it and it's being built now. | 19:00:21 |
| xanderio | should this be in this list? https://hydra.nixos.org/job/nixpkgs/trunk/foundationdb.aarch64-linux | 19:00:29 |
| Vladimír Čunát | As the log path is the same, it gets overwritten. | 19:00:35 |
| Vladimír Čunát | Same step being built by a different job. Yeah, it's confusing. | 19:00:55 |
| Vladimír Čunát | This job: https://hydra.nixos.org/build/283836137#tabs-buildsteps | 19:01:14 |
| xanderio | Ok i think I get it now. Thanks for the explaination | 19:01:23 |
| xanderio | And to make things even more confusing the second build step is only display in the trunk-combined job(?). even thougth it's the same path as the trunk job | 19:03:37 |
 hexa | sharing is caring | 19:03:57 |
| hexa | if they build the same derivation it doesn't really matter what jobset you are being pointed at | 19:04:43 |
| Vladimír Čunát | The shared step isn't shown when in progress. (well, it's shown only by one of the sharers) | 19:05:36 |
| Vladimír Čunát | sleepy-brown machine will have issues with kernel updates. | 20:47:13 |
| Vladimír Čunát | The kernels use lots of /tmp space and tend to get scheduled many at once. | 20:47:32 |
| Vladimír Čunát | Then we get out of space and they fail. | 20:47:43 |
| Vladimír Čunát | * Then we get out of space and they fail. | 20:48:57 |
| hexa | hm ok | 20:55:00 |
| hexa | won't get to that today | 20:55:03 |
| hexa | https://grafana.nixos.org/d/rYdddlPWk/node-exporter-full?orgId=1&from=now-24h&to=now&timezone=browser&var-datasource=default&var-job=node&var-node=sleepy-brown.builder.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 | 20:55:37 |
| hexa | https://grafana.nixos.org/d/rYdddlPWk/node-exporter-full?orgId=1&from=2025-01-14T19:39:59.644Z&to=2025-01-14T20:55:13.509Z&timezone=browser&var-datasource=default&var-job=node&var-node=sleepy-brown.builder.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 | 20:55:52 |
