nukeop
Crash on launch due to the GPU process not being usable
Platform: NixOS Linux
Nuclear version: Prelease f5ec0d, specifically the Appimage.
Description of the issue: When attempting to launch Nuclear on NixOS I got the following error:
[nix-shell:~/.cache/nixpkgs-review/pr-169347-1]$ ./results/nuclear/bin/nuclear
main › (node:2) [DEP0005] DeprecationWarning: Buffer() is deprecated due to security and usability issues. Please use the Buffer.alloc(), Buffer.allocUnsafe(), or Buffer.from() methods instead.
(Use `nuclear --trace-deprecation ...` to show where the warning was created)
[2:0421/193633.580170:FATAL:gpu_data_manager_impl_private.cc(415)] GPU process isn't usable. Goodbye.
I have found an issue elsewhere which describes the same problem, see https://github.com/LedgerHQ/ledger-live-desktop/issues/4253#issuecomment-926215043. There someone suggested adding the --no-sandbox flag, which did fix the issue for me. Maybe updating electron will properly fix it as described in said issue?
I maintain the nuclear package in nixpkgs, there we will wrap the binary to always use the --no-sandbox flag by default in the meantime.
This also seems to occur on the .deb package.
libva error: vaGetDriverNameByIndex() failed with unknown libva error, driver_name = (null) libva error: vaGetDriverNameByIndex() failed with unknown libva error, driver_name = (null) libva error: vaGetDriverNameByIndex() failed with unknown libva error, driver_name = (null) [12346:0611/180514.265417:FATAL:gpu_data_manager_impl_private.cc(415)] GPU process isn't usable. Goodbye. fish: Job 1, 'nuclear' terminated by signal SIGTRAP (Trace or breakpoint trap)
But the driver name seems to be null somehow.
Edit: The AppImage also has this error
Duplicate of #1297, can you try running with --no-sandbox? I was told (on Discord) that it helped in this case. Please reopen if it doesn't work.
