gjoseph92
diff --git a/‎dask_profiling_coiled/run_profile.py‎
Lines changed: 6 additions & 6 deletions b/‎dask_profiling_coiled/run_profile.py‎
Lines changed: 6 additions & 6 deletions
diff --git a/‎environment.yml‎
Lines changed: 1 addition & 2 deletions b/‎environment.yml‎
Lines changed: 1 addition & 2 deletions
diff --git a/‎make-coiled-env.sh‎
Lines changed: 21 additions & 1 deletion b/‎make-coiled-env.sh‎
Lines changed: 21 additions & 1 deletion
diff --git a/‎results/cython-shuffle-gc-debug-noprofiling.html‎
Lines changed: 88 additions & 0 deletions b/‎results/cython-shuffle-gc-debug-noprofiling.html‎
Lines changed: 88 additions & 0 deletions
diff --git a/‎results/cython-shuffle-gc-debug-noprofiling.json‎
Lines changed: 1 addition & 0 deletions b/‎results/cython-shuffle-gc-debug-noprofiling.json‎
Lines changed: 1 addition & 0 deletions
@@ -91,20 +91,20 @@ def main():
     # print("Disabling GC on scheduler")
     # client.run_on_scheduler(disable_gc)
 
-    # def enable_gc_debug():
-    #     import gc
+    def enable_gc_debug():
+        import gc
 
-    #     gc.set_debug(gc.DEBUG_STATS | gc.DEBUG_COLLECTABLE | gc.DEBUG_UNCOLLECTABLE)
+        gc.set_debug(gc.DEBUG_STATS | gc.DEBUG_COLLECTABLE | gc.DEBUG_UNCOLLECTABLE)
 
-    # print("Enabling GC debug logging on scheduler")
-    # client.run_on_scheduler(enable_gc_debug)
+    print("Enabling GC debug logging on scheduler")
+    client.run_on_scheduler(enable_gc_debug)
 
     print("Here we go!")
 
     # This is key---otherwise we're uploading ~300MiB of graph to the scheduler
     dask.config.set({"optimization.fuse.active": False})
 
-    test_name = "cython-shuffle-gc-noprofiling-env"
+    test_name = "cython-shuffle-gc-debug-noprofiling"
     with (
         distributed.performance_report(f"results/{test_name}.html"),
         pyspy_on_scheduler(
 
@@ -28,5 +28,4 @@ dependencies:
       # - git+https://github.com/gjoseph92/scheduler-profilers.git  # TODO this conflicts with --install-option for distributed, using postBuild instead
       # - git+https://github.com/gjoseph92/dask-noop.git
 variables:
-  DASK_DISTRIBUTED__WORKER__PROFILE__INTERVAL: 2h
-  DASK_DISTRIBUTED__WORKER__PROFILE__CYCLE: 10h
+  DASK_CONFIG: dask.yaml
@@ -1,10 +1,30 @@
 #!/bin/bash
 
-# Install py-spy separately so it doesn't conflict with Cythonized distributed
+# Install py-spy separately so it doesn't conflict with Cythonized distributed.
+# Also add dask config.
+
+# HACK: Coiled offers no easy way to add auxiliary data files---or a dask config---in software environments,
+# so we generate a post-build shell script that has the contents of `dask.yaml` within itself, and writes
+# those contents out when executed.
+OUT_CONFIG_PATH="~/.config/dask/dask.yaml"
+YAML_CONTENTS=$(<dask.yaml)
 cat > postbuild.sh <<EOF
 #!/bin/bash
 
 python3 -m pip install git+https://github.com/gjoseph92/scheduler-profilers.git@8d59e7f8b2ab59e22f0937557fefe388eac6ea61
+
+OUT_CONFIG_PATH=$OUT_CONFIG_PATH
+# ^ NOTE: no quotes, so ~ expands (https://stackoverflow.com/a/32277036)
+mkdir -p \$(dirname \$OUT_CONFIG_PATH)
+
+cat > \$OUT_CONFIG_PATH <<INNER_EOF
+$YAML_CONTENTS
+INNER_EOF
+
+echo "export DASK_CONFIG=\$OUT_CONFIG_PATH" >> ~/.bashrc
+
+echo "Wrote dask config to \$OUT_CONFIG_PATH:"
+cat \$OUT_CONFIG_PATH
 EOF
 coiled env create -n profiling --conda environment.yml --post-build postbuild.sh
 rm postbuild.sh