Is there any ways to check the process usage for specific time interval. I tried with prstat but am not able to obtain the expected result and even with /var/adm/messages also.
Please rephrase the question such that it is clear what information you require from the Operating System. Please give examples stating the precise units of any numbers.
Actually i want to see the already completed processes during a specific interval of time. For ex: want to check the process status of a host during 4pm to 6 pm
I'm afraid it's still unclear what you are trying to achieve.
Can you write a plain text example of the output you would like to get about these completed processes status between 4 to 6pm ?
Actually in my client network in one host some process is occupying more space during a particular time daily , but i cant able to get in log messages;; SO is ther any way to check the processes which already ran during that particular time
Can you write a plain text example of the output you would like to get about these completed processes status between 4 to 6pm ?
Please explain "occupying more space" and "cant able to get in log messages".
some process is occupying more space during the specific time interval. Since am not able to check at that time is it possible to check that processes which occupied more space on the next day.
our issue with this post is "more space" is very generic and can mean several different things to several different people. i'm guessing that you are referring to address space??? again, i am not sure what you are looking for. what are the symptoms of your host(s)?
historical data is best seen through sar if you have set it up. there are other ways of getting host level data but again, there has to be a mechanism in place to capture data. if i were you, i would script something that captures several different outputs about your system every hour or so. you can obviously adjust this to what you see fit. process table, io stats, network stats, etc etc.
If what I understood is what you are clarifying - I would do the following
cron the following command to run every 10 minutes and save the output to a log file
0,10,20,30,40,50 * * * * * ps -u <ur username> -o pid,pcpu,pmem,command >> /tmp/mem_log
The above will log all your processes with cpu % and mem %. You can sort on the mem field and select the top 10 to see the top 10 mem hogging processes.
If you have certain processes spawing during the mentioned time, you need to write a script to check the utilisation of that PID using grep with prstat or sar.from there, you need to drill down to check the child process using ptree and see where the issue goes. If this is related to oracle, you can provide the pid to your DBA and they will know what to fine tune to the oracle parameters
Sorry but no one will succeed to help if you fail answering the questions asked.
What output are you expecting ? You just ignored that question of mine.
What is that "space" you are talking about ?
sorry for the confusion,, its CPU utilization,.Hope you are clear now.
So did you try gowri_g_s' suggestion ?