Monitoring other server types (PING)
Previous Topic  Top  next topic

PING is different than the previously discussed server monitoring methods (HTTP, FTP, etc.) in that it does not check the status of a specific service and does not establish a TCP connection, but rather checks the general availability of the server computer by pinging it.

The advantage of this monitoring method is that it is very light weight (no TCP connection overhead), it uses very little bandwidth (PING packets are very small), and it can be used for any type of Internet service.
However, it can only detect if a server computer is running or not. Unlike the other monitoring methods, PING cannot determine if a specific service on that computer is running and functioning correctly.

Please note that some firewalls are configured to filter out PING packets (ICMP protocol), in which case this monitoring method cannot be used (check this by pinging your server IP from a command prompt first).

When polling the status of a server, the following steps are taken:

An ICMP Echo Request (PING) is sent to the server IP address.
If the server responds with a matching ICMP Echo Reply, then the server is considering functional.
If a negative response (for example "destination unreachable") is received from a router somewhere in between Simple Failover and the server, then the server is considered non-functional.
If no response is received within one second, then step 1 is repeated 2 more times, and after that the server is considered non-functional.

For details on the ICMP protocol, please see RFC792.
This and other RFCs can be obtained from http://www.rfc-editor.org/rfcsearch.html.

Please note: this monitoring method is only available with the Professional license type.