[xgboost] Fix flaky train_small test (ray-project#20529)

Xgboosts train_small timed out because of a CPU borrowing feature related to placement groups. The root bug will be fixed in the coming weeks, but this PR makes the release test consistently pass by requesting 0 CPUs for the remote wrapper script.
jeffreyftang · Nov 18, 2021 · 140a180 · 140a180
1 parent 65a023e
commit 140a180
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/release/lightgbm_tests/workloads/train_small.py b/release/lightgbm_tests/workloads/train_small.py
@@ -38,7 +38,7 @@
 
     start = time.time()
 
-    @ray.remote
+    @ray.remote(num_cpus=0)
     def train():
         os.environ["TEST_OUTPUT_JSON"] = output
         os.environ["TEST_STATE_JSON"] = state

diff --git a/release/xgboost_tests/workloads/train_small.py b/release/xgboost_tests/workloads/train_small.py
@@ -38,7 +38,7 @@
 
     start = time.time()
 
-    @ray.remote
+    @ray.remote(num_cpus=0)
     def train():
         os.environ["TEST_OUTPUT_JSON"] = output
         os.environ["TEST_STATE_JSON"] = state