Admin console producing hs_pid errors

Question

Just recently I've discovered that the admin-serv/config directory is producing lots of hs_pid errors
An unexpected exception has been detected in native code outside the VM.
Unexpected Signal : 11 occurred at PC=0xFEEBB384
Function=PR_Write+0x0
Library=/opt/directory/lib/libnspr4.so
Cannot obtain thread information
Dynamic libraries:
0x10000 ns-httpd
0xff380000 /opt/directory/bin/base/jre/lib/sparc/server/libjsig.so
0xff360000 /opt/directory/bin/https/lib/libsmartheap_smp.so
0xff340000 /opt/directory/bin/https/lib/liblibdbm.so
0xff180000 /opt/directory/bin/https/lib/libns-httpd40.so
0xff150000 /opt/directory/bin/https/lib/liblibsi18n.so
0xff130000 /opt/directory/bin/https/lib/libgetprop.so
0xff110000 /opt/directory/bin/https/lib/libnsprwrap.so
0xff0c0000 /opt/directory/lib/libldap50.so
0xff0a0000 /opt/directory/lib/libprldap50.so
0xff070000 /opt/directory/lib/libsasl.so
0xfef00000 /opt/directory/bin/https/lib/libxerces-c.so
etc........
and the console no longer allows you to login.
Stopping the admin-serv and removing the error files and then starting admin-serv backup is a temporary fix but I cannot find any reason for why this has started to happen.
The time they occur and the logs in the directory give me nothing to go on.
The errors just continue looping until the partition fills up.
Is this a bug with the Software, has anyone else seen this?
Regards
James

Answer 1

It may not be a bug, I found a post says:
http://forums.java.sun.com/thread.jspa?threadID=629508&messageID=3618384
===
Re: How to detect JVM crash
Author: zhmur Posts: 110 Registered: 04.05.05
          May 27, 2005 12:48 AM (reply 4 of 5)      
Starting from 1.5.0 jvm automatilly dumps crash info to file named "hs_pid*". But 1.4.2 doesn't have this feature. You can try to setup vprintf hook in VMInitArgs.
===
Gary