Can you extract the value of the lines from the IBM PHD java heap dump?

I have a heap dump of a PHD file from IBM jvm and I want to examine the values ​​of some lines. With hprof Sun JVM binary dumps it is possible, but I have not been able to recover this information from the IBM dump.

I tried:

  • Eclipse Memory Analyzer (0.8.0.200906170940) with IBM DTFJ Portal Heap Dump Reader (1.3.0.200903241644)
  • IBM Support Assistant (3.1.0.00) with Java Memory Dump Diagnostics (2.0.0.3b)
  • Running "lines" over a dump file
  • Convert to text dump with com.ibm.jvm.ras.findroots.PrintHeapdump from svcdump.jar found in mdd4j (2.0.0.3b)

In Eclipse Memory Analyzer, I can view a char array inside a String, but all entries in the array are null. I also don't see any primitive attributes for any objects.

IBM hints that they store primitive values ​​at http://www-01.ibm.com/support/docview.wss?rs=180&uid=swg21190476

This is made possible by using the Java DataOutputStream class, which allows you to output primitive Java data types in a portable binary format.

I am using Java 5.0:

2XHOSLEVEL OS Level: AIX 5.3
2XHCPUS Processors -
3XHCPUARCH Architecture: ppc64
3XHNUMCPUS How Many: 8
1CIJAVAVERSION J2RE 5.0 IBM J9 2.3 AIX ppc64-64 build j9vmap6423-20070426
1CIVMVERSION VM build 20070420_12448_BHdSMr
+2


source to share


1 answer


http://www.java-doc.de/tag/ibm/



* .phd contains no values. It is a "portable" format. To see the values, you need a complete system dump, which creates a much larger file.

+2


source







All Articles