| 29 May 2021 |
 cdepillabout | In reply to @maralorn:maralorn.de
Anyways, I think we three don‘t even have the bump to front right and I actually don‘t think we need it. It would be convenient a situation like the following: We are trying to fix up the mergeable job, and there are only one or two failed child jobs. You believe that the jobs have "incorrectly" failed, so you restart them and bump them to the front. They finish building in a minute or so and you can then go ahead with merging haskell-updates into master. You could basically do it all in one sitting, without having to wait a few hours for the build to start. | 03:12:07 |
| cdepillabout | So basically a situation where you are literally sitting at your computer and waiting to take some action based on the outcome of a job. | 03:13:18 |
 joe (he/him) | cdepillabout: I'll release exact-real with a "fix" tonight\ | 10:23:33 |
| joe (he/him) | * cdepillabout: I'll release exact-real with a "fix" tonight | 10:23:36 |
| joe (he/him) | The fix is changing the arbitrary instance for CReal to generate less problematic numbers https://github.com/expipiplus1/exact-real/pull/38 | 10:24:01 |
| joe (he/him) | * The fix is changing the arbitrary instance for CReal to generate less problematic numbers https://github.com/expipiplus1/exact-real/pull/38 LOL | 10:24:09 |
| joe (he/him) | * The fix is changing the arbitrary instance for CReal to generate less problematic numbers https://github.com/expipiplus1/exact-real/pull/38 lol | 10:24:13 |
| joe (he/him) | (that's not laughing out loud, it's actually me throwing my hands up) | 10:24:33 |
 sterni (he/him) | still haven't accumulated the necessary hydra karma to get darwin builds it seems | 11:31:53 |
| sterni (he/him) | well any minute now :p | 11:31:58 |
|  justinrestivo changed their display name from justinrestivo to oh caml >>=. | 12:20:58 |
| justinrestivo changed their profile picture. | 12:22:01 |
| justinrestivo changed their display name from oh caml >>= to justinrestivo. | 12:22:29 |
| justinrestivo changed their profile picture. | 12:24:00 |
| joe (he/him) | err, I seem to be building ghc on nixos-unstable. Is this expected or have I configured something very strange? | 13:12:13 |
 maralorn | In reply to @joe:monoid.al
err, I seem to be building ghc on nixos-unstable. Is this expected or have I configured something very strange? That shouldn‘t be happening. | 13:13:39 |
| maralorn | The only explanation would be that someone broke ghc on master, which would be a huge mishap. | 13:14:07 |
| maralorn | joe (he/him): So I really hope it’s something you did. | 13:14:20 |
| maralorn | Which ghc version is it building? | 13:14:57 |
| joe (he/him) | err, whichever is built by 8.6.5-binary | 13:15:14 |
| sterni (he/him) | joe (he/him): which GHC? | 13:15:20 |
| joe (he/him) | I think however that I might have actually been using a very old nixpkgs | 13:15:24 |
| sterni (he/him) | and which platform | 13:15:26 |
| joe (he/him) | perhaps it was GCd on hydra | 13:15:35 |
| sterni (he/him) | maybe you have an overlay? | 13:15:53 |
| maralorn | There is no GC on hydra. | 13:15:55 |
| maralorn | 8.6.5 sounds like you are building 8.8.4? Should be cached. | 13:16:32 |
| joe (he/him) | specifically: /nix/store/q8qprh8ynbbmh0dgpz4ds82yz21ij8j8-ghc-8.10.4.drv | 13:16:53 |
| joe (he/him) | I'll investigate, I don't think I have any overlay which could be causing this. | 13:17:08 |
| sterni (he/him) | 884 and 8104 are cached | 13:17:20 |
