From c75bed0eebb1f937db02eb98deecd380724f747d Mon Sep 17 00:00:00 2001
From: Jey Kottalam <jey@cs.berkeley.edu>
Date: Fri, 21 Jun 2013 12:13:48 -0400
Subject: Fix reporting of PySpark exceptions

---
 python/pyspark/daemon.py | 22 ++++++++++++++++++----
 python/pyspark/worker.py |  2 +-
 2 files changed, 19 insertions(+), 5 deletions(-)

(limited to 'python')

diff --git a/python/pyspark/daemon.py b/python/pyspark/daemon.py
index 2b5e9b3581..78a2da1e18 100644
--- a/python/pyspark/daemon.py
+++ b/python/pyspark/daemon.py
@@ -21,6 +21,15 @@ def should_exit():
     return exit_flag.value
 
 
+def compute_real_exit_code(exit_code):
+  # SystemExit's code can be integer or string, but os._exit only accepts integers
+  import numbers
+  if isinstance(exit_code, numbers.Integral):
+    return exit_code
+  else:
+    return 1
+
+
 def worker(listen_sock):
     # Redirect stdout to stderr
     os.dup2(2, 1)
@@ -65,10 +74,15 @@ def worker(listen_sock):
                 listen_sock.close()
                 # Handle the client then exit
                 sockfile = sock.makefile()
-                worker_main(sockfile, sockfile)
-                sockfile.close()
-                sock.close()
-                os._exit(0)
+                exit_code = 0
+                try:
+                  worker_main(sockfile, sockfile)
+                except SystemExit as exc:
+                  exit_code = exc.code
+                finally:
+                  sockfile.close()
+                  sock.close()
+                  os._exit(compute_real_exit_code(exit_code))
             else:
                 sock.close()
 
diff --git a/python/pyspark/worker.py b/python/pyspark/worker.py
index f76ee3c236..379bbfd4c2 100644
--- a/python/pyspark/worker.py
+++ b/python/pyspark/worker.py
@@ -55,7 +55,7 @@ def main(infile, outfile):
     except Exception as e:
         write_int(-2, outfile)
         write_with_length(traceback.format_exc(), outfile)
-        raise
+        sys.exit(-1)
     finish_time = time.time()
     report_times(outfile, boot_time, init_time, finish_time)
     # Mark the beginning of the accumulators section of the output
-- 
cgit v1.2.3