viable/strict/1791543288: [distributed] Tear down test_init_with_idx explicitly (#200033)
- PyTorch: 2091 events in the last 90 days
- PyTorch: 2062th Release in the last 90 days
- Previous: earlier the same day · viable/strict/1791541947
What happened
Fixes #200032 Summary distributed/test_c10d_nccl.py::ProcessGroupNCCLGroupTest::test_init_with_idx fails intermittently with the NCCL timeout watchdog observing cudaErrorCudartUnloading ("driver shutting down") during interpreter shutdown, followed by abortProcess() -> SIGABRT (exit -6, expected 0). Error signature: abortProcess() -> SIGABRT(6)"> [rank0]: WorkNCCL: CUDA error during end event query: CUDA error: driv…
Summary assembled by rule from the sources below