!UNVBThoJtlIiVwiDjU:nixos.org

Staging

408 Members
Staging merges | Find currently open staging-next PRs: https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+sort%3Aupdated-desc+head%3Astaging-next+head%3Astaging-next-21.05+is%3Aopen132 Servers

Load older messages

SenderMessageTime
5 Jul 2026
@k900:0upti.meK900Possibly21:16:19
@k900:0upti.meK900But also like21:16:21
@k900:0upti.meK900We need a fix for this now21:16:24
@k900:0upti.meK900Do we just sed it or something21:16:33
@emilazy:matrix.orgemilyI think just sedding it is the way to go. we should try to reproduce all the hashes in Nixpkgs with that though.21:18:25
@emilazy:matrix.orgemilyin case there are other divergences21:18:29
@emilazy:matrix.orgemilywe should also do something invalidate the store paths21:18:38
@emilazy:matrix.orgemilyso that anyone with a now-broken hash gets an error and not a silent mismatching cache hit21:18:50
@emilazy:matrix.orgemily(also, if we do that, then Hydra will reproduce all the hashes for us)21:19:08
@k900:0upti.meK900Yeeeeeeeeeeeeeah21:19:14
@k900:0upti.meK900Maybe we need Hydra to just rebuild FODs21:19:33
@emilazy:matrix.orgemilyI mean we can just change the name21:20:05
@k900:0upti.meK900No like21:20:10
@k900:0upti.meK900Generally21:20:11
@emilazy:matrix.orgemilyright21:20:20
@scrumplex:duckhub.ioScrumplexI'm OOTL when it comes to cache.nixos.org, but does it deduplicate derivations? Otherwise I would be concerned that invalidating store paths might balloon the cache long term21:20:28
@k900:0upti.meK900The outpath won't chang21:20:43
@k900:0upti.meK900Redacted or Malformed Event21:20:44
@emilazy:matrix.orgemilyI don't think a one-time payment of all the pnpm FODs should be a big deal21:20:44
@k900:0upti.meK900Also we have like21:20:50
@emilazy:matrix.orgemilyit will21:20:48
@k900:0upti.meK900100 pnpm FODs21:20:53
@k900:0upti.meK900Or something21:20:54
@emilazy:matrix.orgemilythat's the only way you can invalidate a FOD21:20:56
@k900:0upti.meK900If we change the name yes21:20:59
@k900:0upti.meK900If we make the queue runner force rebuild them no21:21:04
@emilazy:matrix.orgemily okay, but we ought to change the name because downstream users might get their FODs broken by the SQLite update in case our sed isn't good enough 21:21:25
@scrumplex:duckhub.ioScrumplex
In reply to @emilazy:matrix.org
I don't think a one-time payment of all the pnpm FODs should be a big deal
Ah I thought you meant invalidating store paths going forward. One time should be fine :D 		21:21:29
@emilazy:matrix.orgemilyi.e. it's not just Hydra even if we ignore the hypothetical nature of changing Hydra here21:22:03
@k900:0upti.meK900Valid21:22:36

Show newer messages

Back to Room ListRoom Version: 6