Analysis steps

The basic steps in analysis are

  1. Process the options which control the analyzer(s) to be run, the output format or the level of information.
  2. Read the dump file using an appropriate DTFJ ImageFactory. At present there are two possible formats and we employ the factories in the most likely order depending on the name of the dump. Dumps ending .dmp.zip are produced by VMs at 1.5.0 and above whereas dumps ending in .sdff are produced by VMs at the 1.4.2 level. Errors in this step generally prevent any further analysis and are often the result of the data not being correct (typically some other file such as javacore, trace or verbose GC) or the data being corrupt (often due to transmission in ASCII).
  3. Run an analysis script against the image. This script in turn runs analysis routines each written in Java™ and analyzing the image produced in the first step.

© Copyright IBM Corp. 2007, 2008 All Rights Reserved. US Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.