SNMP discovery settings

The defaultSNMPAutoRetries and defaultSNMPAutoTimeout settings control SNMP polling during the first phase of the discovery process.

The defaultSNMPAutoRetries is the number of times that the SNMP poller should try to reach a system after the initial attempt fails. The default value is 3. The defaultSNMPAutoTimeout is the amount of time that the SNMP poller should wait before it considers an SNMP request to have failed or timed out. The default value is 1000 milliseconds. The defaultRetries and defaultTimeout settings control the SNMP poller during the third phase of the discovery process, when the systems are fully probed and the topology is created.

The defaultRetries is the number of times that the SNMP poller should try to reach a system after the initial attempt fails. The default value is 5. The defaultTimeout is the amount of time that the SNMP poller should wait before it considers an SNMP request to have failed or timed out. The default value is 1000 milliseconds.

VMware recommends that you adjust the timeout and retries settings as follows:

• If random systems are experiencing excessive timeouts, a slow system response time is the most probable cause. In this case, increase the time-out value by 50 percent and decrease the number of retries by 1.

• If several systems within the same subnet are experiencing excessive timeouts, a high-packet loss rate is the most probable cause. In this case, increase the number of retries by 1 and decrease the time-out value by 50 percent.

  Do not increase the value of the timeout beyond 8,000 milliseconds. The timeout doubles with each retry. Therefore, an initial value of 8,000 milliseconds with the default number of retries (5) results in a cumulative timeout interval of 504,000 milliseconds (more than 8 minutes) during the discovery process. In addition, because increasing the number of retries by 1 effectively doubles the total timeout, do not increase both the timeout and the number of retries.