[EMAIL PROTECTED] wrote:
>Can you post the case number?
>
http://developer.java.sun.com/developer/bugParade/bugs/4466587.html
Is the current version. My case was rejected during evaluation, as a
duplicate of a (already closed) case that reflected the problem in
jdk1.3.X . Now they have noticed that the problem is still there in
jdk1.4, and they have closed it again. It seems to be due to a problem
in glibc.
But, as I said, the fact that there is an easy workaround (reducing
maximum native stack size to 2048 K or below) makes it a not very
harmful bug.
>
>
>-------- Original Message --------
>Subject: Re: GenIC Segmentation Fault (19-Sep-2001 13:34)
>From: [EMAIL PROTECTED]
>To: [EMAIL PROTECTED]
>
>>[EMAIL PROTECTED] wrote:
>>
>>>Since on www.JavaSoft.com the only "official supported" JVM for Linux is 1.
>>>
>>3.1, we still have the problem with GenIC gives Segmentation Fault, also
>>EJBServer
>>
>>>gives Segmentation fault when startet not empty.
>>>
>>>So we did some more tries and found out, that if we change all calls to "
>>>
>>java" against "java -classic" in the JOnAS scripts (EJBServer, GenIC etc.),
>>all
>>
>>>works fine! From my point of view, it seems like there is one of two
>>>
>>failures:
>>
>>>a) JOnAS does something totally weird that not all VMs can do - I don't
>>>
>>believe this.
>>
>>>b) Sun JDK 1.3.1-b24 for LINUX Hotspot VM has a bug.
>>>
>>This is old, but still can be useful: all hostpot jdk for linux do have
>>a problem (at least with modern kernels). You need to reduce the stack
>>size before starting java (something like "ulimit -s 2000" will do).
>>This is the only way to avoid the segmentation fault after the memory
>>grows. I wonder why they don't just put a test and call ulimit in the
>>java wrapper.
>>
>>>I will open a case on javasoft.com (I'll post the case number as soon as I
>>>
>>get it), and I hope that those of you wanting to have JOnAS run on LINUX 1.
>>3.1
>>
>>>hotspot VM will vote for it! :-)
>>>
>>I already did it. It has not been solved in jdk1.3.1_01 (I hope it is in
>>jdk1.4...). Still, the workaround is simple, and it works like a charm
>>for me.
>>
>>>
>>>Thank you
>>>Markus
>>>
>>>>We have resolved the problem. The problem was rmic which is the JDK Stub
>>>>compiler. A previous version comes with Linux and existed in /usr/bin.
>>>>What I did was go through the JDK/bin directory and remove any program
>>>>
>>that
>>
>>>>existed in /usr/bin. Once this program was removed the example could be
>>>>compiled. JDK1.3.1-b24 still gives a segmentation fault, but JDK1.3.0_02
>>>>seems to work fine.
>>>>
>>>>Thanks for all your help,
>>>>
>>>>
>>
>>
>>
>>To: [EMAIL PROTECTED]
>>
>
>
>To: [EMAIL PROTECTED]
>Cc: [EMAIL PROTECTED]
>
>----
>To unsubscribe, send email to [EMAIL PROTECTED] and
>include in the body of the message "unsubscribe jonas-users".
>For general help, send email to [EMAIL PROTECTED] and
>include in the body of the message "help".
>
----
To unsubscribe, send email to [EMAIL PROTECTED] and
include in the body of the message "unsubscribe jonas-users".
For general help, send email to [EMAIL PROTECTED] and
include in the body of the message "help".
