Introscope to Database Connectivity Lost.

Document ID : KB000031324
Last Modified Date : 14/02/2018
Show Technical Document Details


 Although PostgreSQL is a popular database widely used, and integrates with APM, some issues may occur.



 What to do when Introscope cannot communicate to the PostgreSQL database.



 Any APM network, which has PostgreSQL as a database.



 On the APM Status Console in the Investigator, database access errors may look like the screen below:




 The same problem shows on the IntroscopeEnterpriseManager.log file as this:

 8/06/15 03:06:27.838 PM DRT [WARN] [CheckDatabase.Thread1] [] database inaccessible; will check later
 8/06/15 03:06:46.127 PM DRT [WARN] [CheckDatabase.Thread1] [] database inaccessible; will check later
 8/06/15 03:07:04.448 PM DRT [WARN] [CheckDatabase.Thread1] [] database inaccessible; will check later
 8/06/15 03:07:21.617 PM DRT [WARN] [CheckDatabase.Thread1] [] database inaccessible; will check later

 A close look at the PostgreSQL logs shows the following:


 Login information to the database for APM is kept in the file tess-db-cfg.xml, therefore it has be configured to point to the correct database, with proper login  credentials. Once, the database server is configured properly in this file, three possibilities could be the cause of this problem.

 1. network connectivity errors
 2. user login password problems
 3. database corruption.

 The first one can be verified by using the Ping command and Telnet to the PostgreSQL port 5432; the second, by resetting the database password; and the third, by  rebuilding the database or creating a new one.

In the particular case that facilitated this document, an "ADMIN" database password reset did not work.

The solution was to create a new database user, with same permissions to all tables like admin and change the tess-db-cfg.xml file to show the new user, with the correct password. At that point, Introscope started to connect to the PostgreSQL database and resume normal operation.