I have 2 servers server A, Server B. I want to know several counters on Server B, so i use the Server A for monitoring it. My question is: Is there a big impact (impact on network, memory, etc) in the both servers for using this tool or
it runs on background?. Aditional, i put this counters with 15 seconds interval (default option)
This is going to depend on what object and counters you collect. If you take all object and counters every 15 seconds you will have an issue. You will need to test test test to determine the impact on your servers
Generally speaking if you are looking to trend/baseline a server I recommend taking samples every 5 minutes for 2 weeks. This will give you enough data to baseline and look for memory leaks.
I personally recommend customer collect data in 300mb circular logs all the time. This give you 20 to 25 days of data using my recommend objects to collect below. If there is an issue the data is there to analyze. You will not have to start
up the data capture and wait for it to happen again. This will reduce time to resolution.
The typically object (all counters in these objects) I recommend collecting are:
• Logical Disk
• Network Interface
• Page File
• Physical Disk
Of course you may collect other object based on the sever role. You can use the below links to help you analyze the data and decide what other object you wish to collect.
My above recommendation write typically between 128k and 256K to the log file every 5 minutes.
I always recommend collecting data local on the box. The only time I collect remotely is when I looking at server hangs because an power cycle can corrupt open data files.
I never recommend logging to CSV files because dynamic counters are not collect after you start logging. If you want an CSV file collect in blg format and use relog.exe to create your CSV.
Microsoft führt eine Onlineumfrage durch, um Ihre Meinung zur Msdn Website zu erfahren. Wenn Sie sich zur Teilnahme entscheiden, wird Ihnen die Onlineumfrage angezeigt, sobald Sie die Msdn Website verlassen.