libpq: Try next host if one of them times out.

If one host in a multi-host connection string times out, move on to
the next specified host instead of giving up entirely.

Takayuki Tsunakawa, reviewed by Michael Paquier.  I added
a minor adjustment to the documentation.

Discussion: http://postgr.es/m/0A3221C70F24FB45833433255569204D1F6F42F5@G01JPEXMBYT05

doc/src/sgml/libpq.sgml

@@ -1041,9 +1041,10 @@ postgresql://%2Fvar%2Flib%2Fpostgresql/dbname
        string). Zero or not specified means wait indefinitely.  It is not
        recommended to use a timeout of less than 2 seconds.
        This timeout applies separately to each connection attempt.
-       For example, if you specify two hosts and both of them are unreachable,
-       and <literal>connect_timeout</> is 5, the total time spent waiting for a
-       connection might be up to 10 seconds.
+       For example, if you specify two hosts and <literal>connect_timeout</>
+       is 5, each host will time out if no connection is made within 5
+       seconds, so the total time spent waiting for a connection might be
+       up to 10 seconds.
       </para>
       </listitem>
      </varlistentry>

src/interfaces/libpq/fe-connect.c

@@ -1720,6 +1720,7 @@ connectDBComplete(PGconn *conn)
 {
 	PostgresPollingStatusType flag = PGRES_POLLING_WRITING;
 	time_t		finish_time = ((time_t) -1);
+	int			timeout = 0;
 
 	if (conn == NULL || conn->status == CONNECTION_BAD)
 		return 0;
@@ -1729,8 +1730,7 @@ connectDBComplete(PGconn *conn)
 	 */
 	if (conn->connect_timeout != NULL)
 	{
-		int			timeout = atoi(conn->connect_timeout);
-
+		timeout = atoi(conn->connect_timeout);
 		if (timeout > 0)
 		{
 			/*
@@ -1745,6 +1745,8 @@ connectDBComplete(PGconn *conn)
 
 	for (;;)
 	{
+		int			ret = 0;
+
 		/*
 		 * Wait, if necessary.  Note that the initial state (just after
 		 * PQconnectStart) is to wait for the socket to select for writing.
@@ -1761,7 +1763,8 @@ connectDBComplete(PGconn *conn)
 				return 1;		/* success! */
 
 			case PGRES_POLLING_READING:
-				if (pqWaitTimed(1, 0, conn, finish_time))
+				ret = pqWaitTimed(1, 0, conn, finish_time);
+				if (ret == -1)
 				{
 					conn->status = CONNECTION_BAD;
 					return 0;
@@ -1769,7 +1772,8 @@ connectDBComplete(PGconn *conn)
 				break;
 
 			case PGRES_POLLING_WRITING:
-				if (pqWaitTimed(0, 1, conn, finish_time))
+				ret = pqWaitTimed(0, 1, conn, finish_time);
+				if (ret == -1)
 				{
 					conn->status = CONNECTION_BAD;
 					return 0;
@@ -1782,6 +1786,23 @@ connectDBComplete(PGconn *conn)
 				return 0;
 		}
 
+		if (ret == 1)	/* connect_timeout elapsed */
+		{
+			/* If there are no more hosts, return (the error message is already set) */
+			if (++conn->whichhost >= conn->nconnhost)
+			{
+				conn->whichhost = 0;
+				conn->status = CONNECTION_BAD;
+				return 0;
+			}
+			/* Attempt connection to the next host, starting the connect_timeout timer */
+			pqDropConnection(conn, true);
+			conn->addr_cur = conn->connhost[conn->whichhost].addrlist;
+			conn->status = CONNECTION_NEEDED;
+			if (conn->connect_timeout != NULL)
+				finish_time = time(NULL) + timeout;
+		}
+
 		/*
 		 * Now try to advance the state machine.
 		 */

src/interfaces/libpq/fe-misc.c

@@ -991,11 +991,9 @@ pqWait(int forRead, int forWrite, PGconn *conn)
 /*
  * pqWaitTimed: wait, but not past finish_time.
  *
- * If finish_time is exceeded then we return failure (EOF).  This is like
- * the response for a kernel exception because we don't want the caller
- * to try to read/write in that case.
- *
  * finish_time = ((time_t) -1) disables the wait limit.
+ *
+ * Returns -1 on failure, 0 if the socket is readable/writable, 1 if it timed out.
  */
 int
 pqWaitTimed(int forRead, int forWrite, PGconn *conn, time_t finish_time)
@@ -1005,13 +1003,13 @@ pqWaitTimed(int forRead, int forWrite, PGconn *conn, time_t finish_time)
 	result = pqSocketCheck(conn, forRead, forWrite, finish_time);
 
 	if (result < 0)
-		return EOF;				/* errorMessage is already set */
+		return -1;				/* errorMessage is already set */
 
 	if (result == 0)
 	{
 		printfPQExpBuffer(&conn->errorMessage,
 						  libpq_gettext("timeout expired\n"));
-		return EOF;
+		return 1;
 	}
 
 	return 0;