From 591b73c7866a06ff69ed06b8fa3b48e7cef05073 Mon Sep 17 00:00:00 2001
From: Duncan Spani <duncanspani@google.com>
Date: Wed, 13 May 2026 16:10:02 -0700
Subject: [PATCH] Remove 'exit 0' from nccl-test script

Having 'exit 0' at the end of the script masks any errors generated and allows pods to exit with "success"/"completed" status instead of failing.
---
 gpudirect-rdma/nccl-test-a4x-max-jobset.yaml | 2 --
 1 file changed, 2 deletions(-)

diff --git a/gpudirect-rdma/nccl-test-a4x-max-jobset.yaml b/gpudirect-rdma/nccl-test-a4x-max-jobset.yaml
index e5adf8883..7ffbb1aae 100644
--- a/gpudirect-rdma/nccl-test-a4x-max-jobset.yaml
+++ b/gpudirect-rdma/nccl-test-a4x-max-jobset.yaml
@@ -159,8 +159,6 @@ spec:
                       sleep 5
                     done
                   fi
-
-                  exit 0
                 volumeMounts:
                 - name: nvidia
                   mountPath: /usr/local/nvidia