Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Archived » Memory Analyzer (MAT) » Unsupported segment type 254 at position 366,014,124
Unsupported segment type 254 at position 366,014,124 [message #2358] Tue, 15 July 2008 14:18 Go to next message
Eclipse UserFriend
Originally posted by: christian.mallwitz.xbridge.com

I'm using MAT 0.7.0.20080613 on Windows.

My Windows JDK is:

java version "1.5.0_06"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-b05)
Java HotSpot(TM) Client VM (build 1.5.0_06-b05, mixed mode, sharing)

My "heap dump" environment is:

SunOS 5.9 Generic_122300-25 sun4u sparc SUNW,Ultra-60

java version "1.5.0_05"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_05-b05)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_05-b05, mixed mode)

I have a head dump generated using

jmap -J-Djava.io.tmpdir=/www d64 -heap:format=b {pid}

Please note the -J-Djava.io.tmpdir=/www - we needed that in order to make
jmap create temp files somewhere other than /var/tmp as it was to small.

Jmap finished without error and the heap.bin file doesn't seem to be
corrupted (e.g. it doesn't have 0x00 only after a while)

The 366,014,124th byte into the heap dump is indeed 0xFE (254). There are
more none 0x00 bytes after that.

If you could point me to other heap dump processing tools I would have
them have a go at my heap dump to check if they have the same problem.

I can make the dump available on the web if necessary (47 MB gzipped).

Thanks
Christian
Re: Unsupported segment type 254 at position 366,014,124 [message #2389 is a reply to message #2358] Wed, 16 July 2008 21:21 Go to previous messageGo to next message
Andreas Buchen is currently offline Andreas BuchenFriend
Messages: 113
Registered: July 2009
Senior Member
Hi Christian,

I have seen many corrupted hprof files in the past... To test the
integrity, try using jhat which comes with JDK 6 (bin/jhat <dump>). If
this works, I am more than happy to look at the dump file itself.

Also, if you have the option to use Java 6, try creating a hprof dump
there. Java 5's jmap is known to be not so reliable and sometimes produces
corrupt dumps.

Andreas.
Re: Unsupported segment type 254 at position 366,014,124 [message #2418 is a reply to message #2389] Thu, 17 July 2008 13:10 Go to previous messageGo to next message
Eclipse UserFriend
Originally posted by: christian.mallwitz.xbridge.com

I checked using JDK 6 jhat and it gives me the same error message.

As far as doing a dump using JDK 6: Our system runs on Java 5. Do you
think I can install a JDK 6 along it and use the JDK 6 jmap to safely
create a dump of a JDK 5 VM? Or did you mean to run the VM to be
heap-dumped using JDK 6?

Thanks
Christian
Re: Unsupported segment type 254 at position 366,014,124 [message #2448 is a reply to message #2418] Mon, 21 July 2008 09:26 Go to previous message
Andreas Buchen is currently offline Andreas BuchenFriend
Messages: 113
Registered: July 2009
Senior Member
Hi Christian,

> Or did you mean to run the VM to be heap-dumped using JDK 6?

That's what I meant. This is, of course, not an option for a production
system.

Andreas.
Previous Topic:OQL Preferences
Next Topic:Group by java.lang.ref.Finalizer.referent class name
Goto Forum:
  


Current Time: Sun Dec 21 08:43:58 GMT 2014

Powered by FUDForum. Page generated in 0.01818 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software