Open shonfeder opened 2 years ago
Can it be related to #1906?
Iiuc, #1906 was merged in after the failure showed up on unstable?
I don't recall doing any work on those tests, so if I appear in git blame it's probably some renaming propagation. Also, as an unintended side-effect, the code in #1906 probably strictly better-performant.
There's ~90 lines of
[warn] In the last 10 seconds, 5.231 (56.2%) were spent in GC. [Heap: 0.02GB free of 1.00GB, max 1.00GB] Consider increasing the JVM heap using `-Xmx` or try a different collector, e.g. `-XX:+UseG1GC`, for better performance.
in the logs before that exception is thrown.
So it could be that the actual bug is somewhere else and only lead to manifest in an OutOfMemoryError
at this specific unit test.
We got another one at https://github.com/informalsystems/apalache/actions/runs/3297326043/jobs/5438010088#step:6:114 on commit https://github.com/informalsystems/apalache/pull/2244/commits/901be0195ad25d78fadbfea487f8b1d8a791f045
It's weird, because the two failing tests are wildly different in what they test.
The Github macOS VMs have 14GB of RAM, we could start by increasing the heap size via -Xmx
as the error message indicates.
A unit test failed in commit https://github.com/informalsystems/apalache/commit/9efb01fb85d9dda0f0730810b4e01e0e6b91fd11 during evaluation of a generate module:
(source in CI logs)
This is a non-deterministic failure, and the build is not broken ,but probably work investigating at some point to see if we hit an interesting edge case.
FYI @thpani and @Kukovec, who have both touched this test (not to say it's your responsibility or fix it, or even that it deserves the attention to fix atm ;) ).