FIX: IIS Performance Degrades with a Bad Connection

Microsoft Internet Information Server (IIS) version 3.0 may experience degrading performance when there is a bad connection in the ODBC connection pool and the backend database server for the bad connection is not available for connecting.

With connection pooling enabled, whenever IIS users try to connect to the same backend database server, the ODBC driver manager tests the existing connection in the pool before reusing the connection. If there is a bad connection in the pool and the backend database server for the bad connection is unavailable, the ODBC driver manager keeps testing the bad connection, and IIS essentially creates a thread for each attempted connection. The testing connection process takes time, and the IIS threads are also being generated continuously, which causes the performance to degrade.

Microsoft has confirmed this to be a problem in Microsoft Open Database Connectivity version 3.0. This problem has been corrected in U.S. Service Pack 1 for Microsoft Open Database Connectivity, version 3.0. For more information, contact your primary support provider.

A new registry key and ODBC API have been introduced to avoid testing the connection continuously. The new registry key is under: The Retry Wait registry key is a value of REG_SZ and the default value is 120 (in second). The value for the Retry Wait registry key can be set by calling ODBCSetTryWaitValue ODBC API. The ODBCSetTryWaitValue API takes a DWORD argument.

If there is a bad connection in the pool, the ODBC driver manager will attempt to connect before the connection can be reused for the first time. If the connection still fails, the ODBC driver manager returns the error and marks the connection with the time. From that point until the RetryWait value expires, the ODBC driver manager returns a failure without testing the connection.