add env var for nccl debug

Signed-off-by: Nick Hill <[email protected]>

Author: njhill

.buildkite/test-pipeline.yaml

@@ -156,7 +156,7 @@ steps:
   - PP_SIZE=2 torchrun --nproc-per-node=4 distributed/test_torchrun_example.py
   # test with internal dp
   - python3 ../examples/offline_inference/data_parallel.py --enforce-eager
-  - TP_SIZE=2 DP_SIZE=2 pytest -v -s v1/test_async_llm_dp.py
+  - NCCL_DEBUG=INFO TP_SIZE=2 DP_SIZE=2 pytest -v -s v1/test_async_llm_dp.py
   - TP_SIZE=2 DP_SIZE=2 pytest -v -s v1/test_external_lb_dp.py
   - TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/test_internal_lb_dp.py
   - TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/test_hybrid_lb_dp.py