Closed SuperSandro2000 closed 6 months ago
That was only evaluation though, not sure why the checks fail:
Running phase: checkPhase
Running 2 test suites...
Test suite golden-tests: RUNNING...
### Failure in: 0:golden test standard for old-style messages with log from file
test/Golden.hs:124
No running builds
### Failure in: 1:golden test standard for json messages with log from file
test/Golden.hs:124
No running builds
Cases: 4 Tried: 4 Errors: 0 Failures: 2
Test suite golden-tests: FAIL
Test suite logged to: dist/test/nix-output-monitor-2.1.1-golden-tests.log
Test suite unit-tests: RUNNING...
Cases: 6 Tried: 6 Errors: 0 Failures: 0
Test suite unit-tests: PASS
Test suite logged to: dist/test/nix-output-monitor-2.1.1-unit-tests.log
1 of 2 test suites (1 of 2 test cases) passed.
I don’t think this makes sense. The golden tests depend on the hashes of the derivations in questions. So those tests can’t work on other platforms.
I suggest disabling these checks on all systems other than x86_64-linux.
I suggest disabling these checks on all systems other than x86_64-linux.
Changed that. Not sure how to convince cabal to not run the tests though.
I suggest
doCheck = system == "x86_64-linux";
directly in the line above preCheck
.
changed that
Great, thanks!
Tested by deploying the config on an aarch64-linux without an x86_64-linux remote builders (in fact without any remote builders at all).