Conversation
|
I'm not sure why this change is actually necessary. Did you have problems with downloading from downloads.jetbrains.com? From what I see jetbrains.com also just links to downloads. to your side-comment: modern nixpkgs-review downloads the GitHub eval report, when you have the |
Yes, it seems to be considerably less reliable on my ISP than download-cf.jetbrains.com.
If I understand correctly, most of the arguments listed there do not apply in this case. JetBrains packages do not seem to get cached by nix (because they are unfree?). The connections are not unencrypted, neither is any FTP involved. And the concern is not with the load on JetBrains's servers but with availability & reliability.
Ah, thanks a lot! A bit sad that I have to open a PR before I can test this, but it's a major improvement indeed. Btw, might there be a way to cache the eval stage locally somehow? Maybe make it go through the newer |
Some people report that downloading without a mirror can have issues (see here). Also, in sanctioned countries |
|
I'm loathe to merge this due to the mirror system. Would #410186 work for your use case? |
I guess it does, thanks, that's really helpful. Still, maybe consider including |
|
Since |
Replaced
https://download.jetbrains.com/withmirror://jetbrains/, which aliases to the previous url as well ashttps://download-cf.jetbrains.com/. Update according nix files and internal python scripts. Made sure that the scripts are working and the download hashes match.Things done
nix.conf? (See Nix manual)sandbox = relaxedsandbox = truenix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)nixpkgs-review eats up to 10+GiBs of RAM and takes 15 minutes to finish even though my change impacted only fixed-output derivations.
This is a major pain point whenever I try to contribute to nixpkgs.
Add a 👍 reaction to pull requests you find important.