Quote:
Originally Posted by kanis
Hi,
My J2EE application is deployed in Weblogic 9.2 MP3 in Linux box.
The problem is "api response time in production system is higher that test system( test system has more data and load w.r.t production .)"
we got large amount of page fault in production Garbage collection log , where as in test system (with same load ) page fault is Zero.And this page fault making my production system slow.
In respect of JVM(JrockIt) both the system memory settings(-Xms1024m -Xmx1024m -Xgcprio:throughput) are same.
Can any one let me know is any system/kernel parameter set is missing in production which might cause this large amount of page fault. If any more information is required let me know?
Please help.
|
Well, there are system config files (/etc/security/limits.conf, etc.), that contain parameters for system-wide settings. Are they they same? What file(s) did you modify on test, that you didn't on production?
Quote:
Linux Version in Prod:
Linux version 2.6.18-53.el5 (brewbuilder@hs20-bc1-7.build.redhat.com) (gcc version 4.1.2 20070626 (Red Hat 4.1.2-14)) #1 SMP Wed Oct 10 16:34:19 EDT 2007
Linux Version in Test:
Linux version 2.6.9-55.ELsmp (brewbuilder@ls20-bc2-14.build.redhat.com) (gcc version 3.4.6 20060404 (Red Hat 3.4.6-3)) #1 SMP Fri Apr 20 17:03:35 EDT 2007
|
So you're running two different kernels, but your test box is running the SMP kernel, and your prod box isn't? Have you contacted RedHat support, since you're paying for it, with your RHEL subscription?
There are many files and settings that it could be, but you haven't said how you configured the systems, what kind of hardware is on them, posted any of the code, or given any real details. There are too many variables for us to guess about.
