Configuration to Detect Database Unavailability

If the database becomes unresponsive or unreachable, the JDBC driver does not automatically notify the EMS server of that situation until the operating system's TCP keep-alive mechanism kicks in, which usually defaults to a delay of 2 hours.

To expedite the notification of such a situation, you can adjust the TCP keep-alive parameters to a shorter delay.

You may also expedite the notification by the JDBC driver to the EMS server of the database becoming unresponsive or unreachable by adjusting the appropriate JDBC driver property. The way to do this varies with the driver vendor.

For example, for the Oracle thin client, add this type of entry to the jre_option property in tibemsd.conf:

jre_option=-Doracle.jdbc.ReadTimeout=30000

This results in EMS server database operations timing out after 30 seconds, leading to a server shutdown when the database is unavailable.

Copyright © Cloud Software Group, Inc. All rights reserved.