CRJAZ2284W The database performance measurement took 2559 ms. A value greater than 2500 ms is considered slow and might indicate a problem with the database.
|
|
Accepted answer
The diagnostic is running a series of simple database operations and timing them. These include a time for ping, read, and write to the database. The tool measures the time for the server to communicate with the database and execute a simple SQL statement. This doesn't include the time for the browser/thick client to communicate with the server. Use the web client, create a health check widget. This widget runs a ping test against the database. This captures the roundtrip time from the server to the database and back. It doesn't run an SQL statement against the database. This test will isolate the network time from the database time. If the ping time is acceptable, then you'll need to look at the database load. If the diagnostic behaves differently under load then when the system is not under load, the database server may be the bottleneck. If this is the case the queries, etc, aren't completing because the database is busy processing other requests, and can't get to yours, or because there is contention for resources blocking the requests. Thembi Sithebe selected this answer as the correct answer
|
Your answer
Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.
Comments
Adding info about your database vendor might bring out the experience...
Were it DB2, my first suggestion would be to execute db2top -d DB and look at the Locks page ( U ) and or the sessions page ( l ) and/or the overall performance of the operating system ( top, iostat ).
DB type is the minimum we need. Database needs reasonable network connection and disc access. Talk to your do says admins.