Monitoring the Double-Take Availability Server

eG Enterprise offers a specialized Double-Take Availability monitoring model, which monitors the uptime of, the connections to, and rate at which the Double-Take Availability server performs mirroring and replication operations, and proactively alerts administrators to current and potential deviations from desired performance levels.

Figure 1 : Layer model of the Double-Take Availability server

Each layer of Figure 1.2 above is mapped to a wide variety of tests that periodically poll the SNMP MIB of the Double-Take Availability server to capture errors and slowdowns in its functioning. Using the statistics so reported, administrators can infer the following:

• Is the Double-Take Availability server available over the network? If so, how quickly is it responding to requests?
• Are all network interfaces supported by the server operating at normal speeds?
• Is any network interface utilizing bandwidth excessively?
• Is the Double-Take Availability server using too much memory from the reserved memory pool for its operations?
• Was the Double-Take Availability server down recently?
• Has any connection to a target been active for an unusually long time?
• Is a target experiencing any errors in the connections to it?
• Does any connection have too many operations in queue? If so, what type of operations hog the queue - mirroring or replication?
• Have any logins to the Double-Take Availability source and/or target failed? Did any of these login failures occur in the last measurement period?