When monitoring Exchange Servers, you should know which performance aspects are most important. You can use the common counters and threshold values to proactively identify potential issues and help identify the root cause of issues when troubleshooting.
Since these values are general guidelines, it is important to trend and perhaps adjust these values to meet the needs of the specific environment. To determine values that work in a specific environment, you need to document normal operating values to create a baseline. After creating the baseline, you must set thresholds so that when performance metrics are not met, you know that the server is not operating optimally.
The processor is one of the fundamental components that you need to monitor to ensure server health on all Exchange Server roles. Standard counters include the total percentage of processor time, the percentage of user mode processor time, and the percentage of privilege mode processor time.
An additional counter related to processor performance is the processor queue length. A processor queue length greater than the specified threshold value can indicate that there is more work available than the processor can handle. When this number is greater than 10 per processor core, it is a strong indicator that the processor is at capacity, particularly when coupled with high CPU utilization.
Another key performance indicator is the memory counter. Tracking the available memory and how much memory has to be written to the page file can tell you when you need to increase server memory, or reduce server load.
MS Exchange ADAccess Domain Controller
Exchange Server relies heavily on Active Directory Domain Services (AD DS) and Active Directory for information. Therefore, it is essential to measure the response time and connection health.
Monitoring services and logs
It is also important to verify that each of the services on the Exchange Server are running and servicing requests. You can monitor services by polling the service status using the Services management tool, the Get-Services cmdlet, or a third-party monitoring tool. Items logged in the Event logs may also indicate Exchange Server problems. These events are usually classified as Errors or Warnings.
|Processor||_Total\% Processor Time||< 75%|
|_Total\% User Time||< 75%|
|_Total\% Privileged Time||< 75%|
|System||Processor Queue Length||< 5 Per Processor Core|
|Memory||Available Mbytes||> 100 MB|
|Pool Paged Bytes||No Value. Monitor for increase in pool paged bytes, which may indicate a possible memory leak|
|Transition Pages Repurposed/sec||< 100 and spike should be < 1000|
|Page Reads/sec||< 100|
|Pages Input/sec||< 1000|
|Pages Output/sec||< 1000|
|MSExchange ADAccess Domain Controllers||LDAP Read Time||< 50ms and spike should not be higher than 100ms|
|LDAP Search Time||< 50ms and spike should not be higher than 100ms|
|LDAP Searches timed out per minute||< 10 at all times, for all roles, Higher values may indicate issues.|
|Long running LDAP operations/min||< 50ms at all times. Higher value may indicate issues.|