tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel Gibby <dgi...@resumerabbit.com>
Subject Re: IBM JRE 1.4.1 dumping core on login with http/1.0 browsers
Date Thu, 23 Oct 2003 21:42:33 GMT
I mentioned putting it in the apache script because apache starts up the 
mod_jk or mod_jk2 connectors.
I saw someone else mention to put it in the apache script, so I did it 
to be safe. I think I had put it in my tomcat script and still had the 
problem. Then I put it in my apache script and haven't had the problem 
since I put it in a couple of days ago.

Daniel

George Payne wrote:

> Daniel, et al, thanks for the info.  I have put this in my tomcat 
> script, but I thought it was a kitchen-sink approach, since I only 
> thought this only applied to RH 9.0 with the NPTL thread library.
>
> But now the water is muddier since I found this
> http://speakeasy.rpmfind.net/linux/RPM/redhat/updates/8.0/i586/kernel-smp-2.4.20-20.8.i586.html

> which mentions NPTL in reference to the (8.0) kernel I have.
>
> Anyone know if this applies to RH 8.0 with the latest smp kernel?
>
> I noticed you also mentioned putting it in the apache startup script.  
> Is this important, since apache does not (I thought) share threads 
> with the jvm?
>
>
>
> At 03:56 PM 10/23/2003, you wrote:
>
>> Search for Java IBM LD_ASSUME_KERNEL in google. That may be your 
>> problem.
>>
>> Then set an environment variable in your apache and tomcat startup 
>> scripts:
>> export LD_ASSUME_KERNEL=2.2.5
>>
>> I hope that helps. It seems to have solved my crashes.
>>
>> Daniel Gibby
>>
>> George Payne wrote:
>>
>>> I'm having problems with the IBM 1.4.1 jvm dumping core every few 
>>> days under fairly light use with tomcat 4.1.27.
>>>
>>> I'm looking for general advice (though specific advice would be even 
>>> better).
>>>
>>> Should I be looking for a new JVM?  Or my programming bug or misconfig?
>>>
>>> This problem appears to be directly or indirectly related to the use 
>>> of old browsers.  Each time it has happened (twice now), the logs 
>>> (below) show that it was being accessed via http/1.0, which is 
>>> fairly rare in the log files.  It occurs right after or right before 
>>> authentication.
>>>
>>> More info:
>>>
>>> RH 8, 2.4.20-20.8smp
>>> apache 2.0/mod_ssl
>>> mod_jk
>>>
>>> **** 15/03 ssl_access_log snippet:
>>> 128.104.50.121 - - [15/Oct/2003:23:16:14 -0400] "GET / HTTP/1.0" 200 
>>> 268
>>>
>>> 128.104.50.121 - - [15/Oct/2003:23:16:18 -0400] "GET 
>>> /apptrack/s/myappinfo.jsp HTTP/1.0" 302 0
>>> 128.104.50.121 - - [15/Oct/2003:23:16:20 -0400] "GET 
>>> /apptrack/login.jsp;jsessionid=D82BFF7D0F918CDB106C1A072EC2AB3D 
>>> HTTP/1.0" 200 4
>>> 085
>>> 128.104.50.121 - - [15/Oct/2003:23:16:22 -0400] "GET 
>>> /apptrack/images/applicanttracking_r2_c1.gif HTTP/1.0" 200 9669
>>> 128.104.50.121 - - [15/Oct/2003:23:16:22 -0400] "GET 
>>> /apptrack/images/applicanttracking_leftimg.jpg HTTP/1.0" 200 13065
>>> (***CRASH***)
>>> 128.104.50.121 - - [15/Oct/2003:23:16:42 -0400] "POST 
>>> /apptrack/j_security_check HTTP/1.0" 500 1085
>>> 128.104.50.121 - - [15/Oct/2003:23:16:55 -0400] "GET 
>>> /apptrack/login.jsp;jsessionid=D82BFF7D0F918CDB106C1A072EC2AB3D 
>>> HTTP/1.0" 500 1
>>> 085
>>>
>>> **** 22/03 ssl_access_log snippet:
>>> 149.199.60.202 - - [22/Oct/2003:18:28:27 -0400] "GET / HTTP/1.0" 200 
>>> 268
>>> 149.199.60.202 - - [22/Oct/2003:18:28:28 -0400] "GET 
>>> /apptrack/s/myappinfo.jsp HTTP/1.0" 302 0
>>> 149.199.60.202 - - [22/Oct/2003:18:28:28 -0400] "GET 
>>> /apptrack/login.jsp;jsessionid=E7F6276FD5958C2D224B0C182A47EB1F 
>>> HTTP/1.0" 200 4
>>> 085
>>> 149.199.60.203 - - [22/Oct/2003:18:28:29 -0400] "GET 
>>> /apptrack/images/applicanttracking_leftimg.jpg HTTP/1.0" 304 0
>>> 149.199.60.202 - - [22/Oct/2003:18:28:30 -0400] "GET 
>>> /apptrack/images/applicanttracking_r2_c1.gif HTTP/1.0" 304 0
>>> 149.199.60.203 - - [22/Oct/2003:18:28:45 -0400] "POST 
>>> /apptrack/j_security_check HTTP/1.0" 302 0
>>> (***CRASH***)
>>> 149.199.60.203 - - [22/Oct/2003:18:28:45 -0400] "GET 
>>> /apptrack/s/myappinfo.jsp HTTP/1.0" 500 1085
>>> 149.199.60.203 - - [22/Oct/2003:18:28:50 -0400] "GET 
>>> /apptrack/login.jsp;jsessionid=E7F6276FD5958C2D224B0C182A47EB1F 
>>> HTTP/1.0" 500 1
>>> 085
>>>
>>> ********Sections of core dump file*************
>>> NULL 
>>> ------------------------------------------------------------------------ 
>>>
>>> 0SECTION       TITLE subcomponent dump routine
>>> NULL           ===============================
>>> 1TISIGINFO     signal 11 received
>>> 1TIDATETIME    Date:                 2003/10/22 at 18:28:45
>>> 1TIFILENAME    Javacore filename:
>>> /home/tomcat/javacore.20031022.182845.3741.txt
>>> NULL 
>>> ------------------------------------------------------------------------ 
>>>
>>> 0SECTION       XHPI subcomponent dump routine
>>> NULL           ==============================
>>> 1HPTIME        Wed Oct 22 18:28:45 2003
>>> 1HPSIGRECV     SIGSEGV received in ?? at 0x41c55e85 in 
>>> /opt/IBMJava2-141/jre/bin/libjitc.so. Processing terminated.
>>> 1HPFULLVERSION J2RE 1.4.1 IBM build cxia32141-20030522
>>> NULL
>>> 1HPOPENV       Operating Environment
>>> NULL           ---------------------
>>> 2HPHOSTNAME    Host             : law5.(none)
>>> 2HPOSLEVEL     OS Level         : 2.4.20-20.8smp.#1 SMP Mon Aug 18 
>>> 14:39:22 EDT 2003
>>> 2HPLIBCVER     glibc Version    : 2.3.2
>>> 2HPCPUS        Processors -
>>> 3HPARCH          Architecture     : (not implemented)
>>> 3HPNUMCPUS       How Many         : (not implemented)
>>> 3HPCPUSENABLED   Enabled          : 4
>>>
>>> **snip **
>>>
>>> 1XMEXCPINFO    Exception Info
>>> NULL           --------------
>>> 2XMEXCPINFO    JVM Exception 0x2 (subcode 0x0) occurred in thread 
>>> "Ajp13Processor[8009][7]" (TID:0x10068790)
>>> NULL
>>> 2XMNATIVESTACK Native stack at exception generation:
>>> 3XMSTACKINFO   Program Name                                 Entry 
>>> Name                       Statement ID
>>> 3XMSTACKINFO
>>> NULL
>>> NULL
>>> NULL
>>> 1XMTHDINFO     Thread Info
>>> NULL           -----------
>>> NULL
>>> 2XMFULLTHDDUMP Full thread dump Classic VM (J2RE 1.4.1 IBM build 
>>> cxia32141-20030522, native threads):
>>> 3XMTHREADINFO      "Ajp13Processor[8009][13]" (TID:0x10068490, 
>>> sys_thread_t:0x891A140, state:R, native ID:0x7801F) prio=5
>>> 4XESTACKTRACE          at 
>>> java.net.SocketInputStream.socketRead0(Native Method)
>>> 4XESTACKTRACE          at 
>>> java.net.SocketInputStream.read(SocketInputStream.java(Compiled Code))
>>> 4XESTACKTRACE          at 
>>> org.apache.ajp.Ajp13.readN(Ajp13.java(Compiled Code))
>>> 4XESTACKTRACE          at 
>>> org.apache.ajp.Ajp13.receive(Ajp13.java(Compiled Code))
>>> 4XESTACKTRACE          at 
>>> org.apache.ajp.Ajp13.receiveNextRequest(Ajp13.java:274)
>>> 4XESTACKTRACE          at 
>>> org.apache.ajp.tomcat4.Ajp13Processor.process(Ajp13Processor.java:396)
>>> 4XESTACKTRACE          at 
>>> org.apache.ajp.tomcat4.Ajp13Processor.run(Ajp13Processor.java:585)
>>> 4XESTACKTRACE          at java.lang.Thread.run(Thread.java:568)
>>>
>>>
>>> George Payne
>>> Law ITC
>>> University of Virginia School of Law
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
>>> For additional commands, e-mail: tomcat-user-help@jakarta.apache.org
>>
>>
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: tomcat-user-help@jakarta.apache.org
>>
>
>
> George Payne
> Law ITC
> University of Virginia School of Law
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tomcat-user-help@jakarta.apache.org




---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tomcat-user-help@jakarta.apache.org


Mime
View raw message