Fix occasional hang in COPY FROM.

If an error occurs while doing COPY FROM, it was possible the
Pgpool-II waited forever for a response from backend after COPY end
marker was sent from frontend. Pgpool expected a new message arrives
to socket, but it is possible that the message (in this case an error
message) is already in the backend read buffer. The fix is, check the
buffer is empty or not before reading from the socket.  New test case
(07.copy_hang) is also added.

The bug was found by Bo Peng.

diff --git a/src/protocol/pool_proto_modules.c b/src/protocol/pool_proto_modules.c
index 1fc527d1dce267966a602724090d963fda5a47f3..e3d4bc2e63594fcc5277ce9281205559ef7fa5d3 100644 (file)
--- a/src/protocol/pool_proto_modules.c
+++ b/src/protocol/pool_proto_modules.c
@@ -5,7 +5,7 @@
  * pgpool: a language independent connection pool server for PostgreSQL
  * written by Tatsuo Ishii
  *
- * Copyright (c) 2003-2019     PgPool Global Development Group
+ * Copyright (c) 2003-2021     PgPool Global Development Group
  *
  * Permission to use, copy, modify, and distribute this software and
  * its documentation for any purpose and without fee is hereby
@@ -3006,6 +3006,9 @@ POOL_STATUS CopyDataRows(POOL_CONNECTION *frontend,
                }
        }
 
+       /*
+        * Wait till backend responds
+        */
        if (copyin)
        {
                for (i=0;i<NUM_BACKENDS;i++)
@@ -3014,11 +3017,19 @@ POOL_STATUS CopyDataRows(POOL_CONNECTION *frontend,
                        {
                                pool_flush(CONNECTION(backend, i));
 
-                               if (synchronize(CONNECTION(backend, i)))
-                    ereport(FATAL,
-                        (return_code(2),
-                         errmsg("unable to copy data rows"),
-                             errdetail("failed to synchronize")));
+                               /*
+                                * Check response from the backend.  First check SSL and read
+                                * buffer of the backend. It is possible that there's an error
+                                * message in the buffer if the COPY command went wrong.
+                                * Otherwise wait for data arrival to the backend socket.
+                                */
+                               if (!pool_ssl_pending(CONNECTION(backend, i)) &&
+                                       pool_read_buffer_is_empty(CONNECTION(backend, i)) &&
+                                       synchronize(CONNECTION(backend, i)))
+                                       ereport(FATAL,
+                                                       (return_code(2),
+                                                        errmsg("unable to copy data rows"),
+                                                        errdetail("failed to synchronize")));
                        }
                }
        }

diff --git a/src/test/regression/tests/076.copy_hang/test.sh b/src/test/regression/tests/076.copy_hang/test.sh
new file mode 100755 (executable)
index 0000000..2abe488
--- /dev/null
+++ b/src/test/regression/tests/076.copy_hang/test.sh
@@ -0,0 +1,54 @@
+#!/usr/bin/env bash
+#-------------------------------------------------------------------
+# test script for copy plus error case.
+# It was reported that following sequece of copy command cause psql hang.
+#
+# CREATE TEMP TABLE vistest (a text);  
+# COPY vistest FROM stdin CSV FREEZE;
+# Enter data to be copied followed by a newline.
+# End with a backslash and a period on a line by itself, or an EOF signal.
+# >> p
+# >> g
+# >> \.
+#
+# In the normal case an error should be returned to psql.
+#ERROR:  cannot perform COPY FREEZE because the table was not created or truncated in the current subtransaction
+
+source $TESTLIBS
+TESTDIR=testdir
+PSQL=$PGBIN/psql
+
+rm -fr $TESTDIR
+mkdir $TESTDIR
+cd $TESTDIR
+
+# create test environment
+echo -n "creating test environment..."
+$PGPOOL_SETUP -n 3 || exit 1
+echo "done."
+
+source ./bashrc.ports
+
+export PGPORT=$PGPOOL_PORT
+
+./startall
+wait_for_pgpool_startup
+
+# execute COPY
+
+timeout 10 $PSQL test <<EOF
+CREATE TEMP TABLE vistest (a text);  
+COPY vistest FROM stdin CSV FREEZE;
+p
+g
+\\.
+\\q
+EOF
+
+if [ ! $? -eq 0 ];then
+    echo ...timed out.
+    ./shutdownall
+    exit 1
+fi
+echo ...ok.
+./shutdownall