Skip to Main Content
Forums

Database Software

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

Interested in getting your voice heard by members of the Developer Marketing team at Oracle? Check out this post for AppDev or this post for AI focus group information.

No log for crsd, cssd

User_R5HLTJan 15 2017 — edited Jan 15 2017

Hi

I am using GI 12.1.0.2.

GRID_HOME=/u01/app/12.1.0.2/grid

but no log in $GRID_HOME/log/hostname/crsd or cssd or evmd

I have only one alert.log in /u01/app/grid/diag/crs/hostname/crs/trace

How is it possible ?

Thanks

Comments

843829
nry2006 wrote:
On an application running in Windows server 2008 environment with two quad core processors and 16GB of RAM

I ran a multi-threaded process with maximum heap size set to --xMx12000M which was returning a CPU utilization of 12% which was too low although the process was running ok. When increased this setting by 1000M , the CPU utilization went up to 100% which would be the intended figure for this application.
Don't attempt to measure your applications performance by CPU utilization.

Measure how much work (whatever it is doing) it does at different heap settings.
843829
In this case the performance (in terms of time to process) is directly related to CPU utilization. I expect CPU utilization to be maximized as the threads are mainly computationally bound.otherwise the process time to run dramatically increases.
843829
nry2006 wrote:
In this case the performance (in terms of time to process) is directly related to CPU utilization. I expect CPU utilization to be maximized as the threads are mainly computationally bound.otherwise the process time to run dramatically increases.
From CPU utilization at 100% all you know is a lot of CPU is being consumed by something (some thread or several threads).
How would you know it is not the JVM doing background stuff (like GC).
843829
In this case the performance (in terms of time to process) is directly related to CPU utilization. I expect CPU utilization to be maximized as the threads are mainly computationally bound.otherwise the process time to run dramatically increases.
843829
From CPU utilization at 100% all you know is a lot of CPU is being consumed by something (some thread or several threads).
How would you know it is not the JVM doing background stuff (like GC).
843829
I know as each thread has a numerical task to perform which does not require it creating or destroying any object instances - so I don't believe gc is an issue, although I expect memory management is an issue depending on whether you provide sufficient heap space - hence my original query.
843829
try to verify garbage collector log, because the only diffrence between having the problem and not is the increase of the heap. So garbage collector has to work harder.

give us the content of GC log when the problem occur.
1 - 7
Locked Post
New comments cannot be posted to this locked post.

Post Details

Locked on Feb 12 2017
Added on Jan 15 2017
#performance-availability, #real-application-clusters
1 comment
783 views