Submitted On 09-AUG-2001
hui_ye
We see similiar problem with HotSpot server vm in our
application running on both Windows 2000(sp2) and NT(4.0
sp6). Our server application dies on 'Exception in
thread"CompileThread0" java.lang.OutOfMemoryError:
requested xxxx bytes' in 15 hours using HotSpot server VM
even with minimal activity.
Unfortunately, Optimizeit and Jprobe doesn't support server
VM, so we cann't profile the memory usage.
Since workaround suggest to use compiler1, I am wondering
if somebody could share how to switch to compiler1? Thanks
in advance.
Submitted On 03-SEP-2001
mwoodruff
I'm seeing a similar problem with large JSPs under TomCat
under 1.3.1 on Windows 98. Switching to -classic works
around the problem, but runs much slower.
Submitted On 25-OCT-2001
ivelin
We were able to produce the same problem under WebLogic 5.1
sp 10.
We were running our app on two different boxes and the
error occured between midnight and 1:00am on both machines,
even though they were started at different times within a
few hours difference.
The exact error output was:
[wlrun] Exception in thread "CompileThread0"
java.lang.OutOfMemoryError: requested 201326600 bytes
Seems like its a bug related to the one reported, because
the compile thread also tried to allocate 200M.
The available Win 2000 system memory on these machines was
stable around the time of the crash and much less than the
max available memory. The java server was allowed to use up
to 512MB (-Xmx512M).
Submitted On 31-OCT-2001
vedhan
I'm facing the same problem in JBoss. My application fails
in 1.3.1_01 build. This is the dump
Exception in thread "CompileThread0"
java.lang.OutOfMemoryError: requested 268435468 bytes
Unexpected Signal : 11 occurred at PC=0x401ca45b
Function name=is_deoptimized_frame__C5frame
Library=/opt/jdk1.3.1_01/jre/lib/i386/server/libjvm.so
Current Java thread:
****************
Another exception has been detected while we were handling
last error.
Dumping information about last error:
ERROR REPORT FILE = (N/A)
PC = 0x0x401ca45b
SIGNAL = 11
FUNCTION NAME = is_deoptimized_frame__C5frame
LIBRARY NAME
= /opt/jdk1.3.1_01/jre/lib/i386/server/libjvm.so
Please check ERROR REPORT FILE for further information, if
there is any.
Good bye.
Submitted On 26-NOV-2001
MarkWeisser
All I get is this:
#
# HotSpot Virtual Machine Error, Internal Error
# Please report this error at
# http://java.sun.com/cgi-bin/bugreport.cgi
#
# Error ID: 4349254E560E43505000F4
#
# Problematic Thread: prio=5 tid=0x76bed0 nid=0x28d runnable
#
Submitted On 05-DEC-2001
jaykamm
So, where can we get this fix? 1.3.1_02 isn't yet
available for download anywhere at this time, and we're
seeing this issue (with JRun instead of Weblogic) on Win2k
with 1.3.1_01 as well. I've got nine production servers
serving up a major public site, and I just had to restart
two of them in the last hour due to this issue.
Submitted On 05-DEC-2001
DarkFire7
Could we get an explanation on how to use "compiler1"?
Submitted On 18-DEC-2001
Warith
Our server (which has extremely high memory usage) ran for
10 full days with no problems, then crashed with 'Exception
in thread "CompileThread0" java.lang.OutOfMemoryError:
requested 268435468 bytes' during a Full GC (explicitly
invoked with System.gc())
WHEN WILL THE FIX BE AVAILABLE!?
This is an extremely serious bug. Our customers want to
know why our server software sometimes goes down, and
demand that we fix it. But we can't, because it's Java to
blame and there is no workaround, short of dumping HotSpot,
which we need for performance. Sun boasts about the
stability of Java, but it is Java that is now making our
company look bad! Please make the fix for this extremely
critical bug available ASAP!
Submitted On 07-JAN-2002
jaykamm
Indeed, 1.3.1_02 is on the public website now. It'd be
nice if any of the various "Recent Releases" lists included
this, but at least it's public now. =v)
Submitted On 29-JAN-2002
averygross
I'm experiencing the same problem on 1.3.1_02-b02 (even
though it's supposedly fixed in that version)
The server sporadically crashes after printing a message
similar to this one:
Exception in thread "CompileThread0"
java.lang.OutOfMemoryError: requested 12550
1448 bytes
The crashes are seemingly random -- they can occur right
after we start the server or several hours later. Usually,
the out of memory line is the last thing printed to the
output/error stream (we redirect stderr and stdout to the
same file so I can't say for certain which one) but I saw
one that printed the following after the out of memory
message:
Unexpected Signal : 11 occurred at PC=0xfecb2380
Function name=
Library=/usr/j2se/jre/lib/sparc/server/libjvm.so
Current Java thread:
This is getting pretty frustrating. Hotspot 1.3.1_02
appears to be very unstable. Every day since we went live
one or two of the servers in our server farm crash from
this bug, bug #4430074, bug #4339224, bug #4357866, or bug
#4622693.
We are using the server JVM with the following arguments:
-Xverify:none
-Xnoclassgc
-Xms1536m -Xmx1536m
-verbose:gc
-XX:MaxNewSize=384m -XX:NewSize=384m
-XX:+PrintTenuringDistribution -XX:-UseTLE
-XX:SurvivorRatio=40
-XX:+OverrideDefaultLibthread
-XX:MaxPermSize=192m
Submitted On 01-APR-2002
yantra
Is this issue is resolved in jdk1.3.1_03 ?
email me : ashirodkar@yantra.com
Submitted On 03-MAY-2002
gnshm
I cannot find any documentation saying that this bug has
been fixed for jdk 1.3.1.03. Someone help.
gnshm@hotmail.com
Submitted On 09-MAY-2002
dhirendrams
We are facing the same problem frequently and our weblogic
6.0 is crashing and it is throwing java.lang.outofmemory
error before crashing. Does any body have any fix for this.
#
# HotSpot Virtual Machine Error, Internal Error
# Please report this error at
# http://java.sun.com/cgi-bin/bugreport.cgi
#
# Error ID: 4349254E560E43505000F4 01
#
# Problematic Thread: prio=5 tid=0x9e178 nid=0xa runnable
#
Submitted On 18-JUL-2002
msluyter
We're also getting the problem using JDK v 1.3.1_02. This occurs during execution of our junit tests, using BEA Weblogic 6.1:
<Jul 18, 2002 6:09:46 AM CDT> <Error> <Kernel> <ExecuteRequest failed
java.lang.OutOfMemoryError
<<no stack trace available>>
#
# HotSpot Virtual Machine Error, Internal Error
# Please report this error at
# http://java.sun.com/cgi-bin/bugreport.cgi
#
# Error ID: 4349254E560E43505000F4
#
# Problematic Thread: prio=5 tid=0x23e7f8 nid=0x73c runnable
#
Submitted On 25-JUL-2002
avijit_de
I am getting this error in solaris 8 as well. Is there any patch
available for this error
The error came like said below
*******************************
HotSpot Virtual Machine Error, Internal Error
# Please report this error at
# http://java.sun.com/cgi-bin/bugreport.cgi
#
# Error ID: 4349254E560E43505000F4 01
#
# Problematic Thread: prio=5 tid=0xa9700 nid=0xb runnable
Submitted On 06-OCT-2002
oweise
Also having this Problem on Tomcat 4.0.4 with JDK 1.4.1 !!!
After some time Tomcat crashes with:
PANIC: Insufficient memory.
Exception in thread "CompileThread0"
java.lang.OutOfMemoryError: requested 182412 bytes
and with popup
"unknown software exception" (0x80000003)
This problem is very serious and seems not to be fixed at
all - even with the most updated JDK-Version - and needs
some review!
Submitted On 15-OCT-2002
oweise
Seems, my problem can be "workarounded" by using the -server
parameter when starting the vm (which tomcat doesnt use out
of the box - even the jdk1.4-specific version). Server runs
now without errors.
Submitted On 19-NOV-2002
ty
We have similar issues: with -hotspot the server finally ran
out of memory and shown erratic behavior in RMI/JDBC etc.
After switching to -server the problem seem to have gone
away. Any idea why the difference?
PLEASE NOTE: JDK6 is formerly known as Project Mustang
|
