!zghijEASpYQWYFzriI:nixos.org

Hydra

402 Members
113 Servers

Load older messages


SenderMessageTime
26 May 2026
@phanirithvij:matrix.orgphanirithvij changed their display name from loudgolem to phanirithvij.11:42:05
27 May 2026
@nyxkrage:matrix.orgCarsten Kragelund Jørgensen joined the room.21:03:27
@nyxkrage:matrix.orgCarsten Kragelund JørgensenIs there a preferred way to get bulk data from hydra? I'm looking to build out a dataset of build failures, but don't want to be rude and just have a script running 24/7 downloading individual build results.21:34:02
@hexa:lossy.networkhexathere are already various platforms that provide that information21:34:43
@hexa:lossy.networkhexa

https://zh.fail/
https://malob.github.io/nix-review-tools-reports/

and a tool by Fernando Rodrigues which I don't remember the name for that groups by error

21:35:21
@nyxkrage:matrix.orgCarsten Kragelund Jørgensenzhf is essentially what i be looking for, though i am more interested in historical failures stats22:25:29
@hexa:lossy.networkhexainfra can provide you with selective database dumps if you can point to what you need22:32:09
@sigmasquadron:matrix.orgFernando Rodriguesuhh, that's not mine?23:02:02
@hexa:lossy.networkhexaoops, sorry23:02:20
@sigmasquadron:matrix.orgFernando RodriguesI think, I don't remember doing anything with hydra?23:02:21
@whispers:catgirl.cloudwhispers [& it/fae]that was sigmagnificent https://github.com/Sigmanificient/nixpkgs-failure-dashboard23:04:02
@whispers:catgirl.cloudwhispers [& it/fae]* that was sigmagnificient https://github.com/Sigmanificient/nixpkgs-failure-dashboard23:04:07
@hexa:lossy.networkhexa😬23:04:21
@hexa:lossy.networkhexaoff by one sigma, I guess23:04:27
@hexa:lossy.networkhexawe are very interested in a list of packages that have been failing for a long time23:07:56
@hexa:lossy.networkhexaso we can remove them, disable them on certain platforms, save us constant rebuild churn23:08:14
@nyxkrage:matrix.orgCarsten Kragelund JørgensenIdeal would be a dump of the builds from the nixpkgs:unstable builds from like 2021 or so, and including the build logs if possible. But I also wouldn't be opposed to expanding out the scope to do any extra analysis that would help the team. As for transfer of the dumps, I can provide access to an s3 bucket in whatever region would be local, or whatever is easier for you guys.23:14:24
@nyxkrage:matrix.orgCarsten Kragelund JørgensenRedacted or Malformed Event23:14:43
@nyxkrage:matrix.orgCarsten Kragelund Jørgensen* Ideal would be a dump of the failing builds from the nixpkgs:unstable builds from like 2021 or so, and including the build logs if possible. But I also wouldn't be opposed to expanding out the scope to do any extra analysis that would help the team. As for transfer of the dumps, I can provide access to an s3 bucket in whatever region would be local, or whatever is easier for you guys.23:15:02
@hexa:lossy.networkhexabuild logs are not in the database I'm afraid23:25:56
@hexa:lossy.networkhexathey are on cache.nixos.org23:26:01
@nyxkrage:matrix.orgCarsten Kragelund Jørgensenoh they are, i somehow completely missed the different domain when manually inspecting, yeah, i can handle backfilling the logs as needed from the cache after having filtered down to what i personally need then, thats no problem.23:31:51
28 May 2026
@whispers:catgirl.cloudwhispers [& it/fae] * 05:01:17
@adam:robins.wtfadamcstephens joined the room.13:35:07
@adam:robins.wtfadamcstephensHow big are the database dumps?13:39:56
@hexa:lossy.networkhexawell, not full dumps, selective dumps13:46:38
@hexa:lossy.networkhexa

zroot/pg 466G 1.17T 393G /var/lib/postgresql

13:46:51
@adam:robins.wtfadamcstephensselective because there are some user/auth tables?13:48:52
@hexa:lossy.networkhexafor example13:52:29
@hexa:lossy.networkhexaand not all data is needed for every purpose13:52:37

Show newer messages


Back to Room ListRoom Version: 6