What is Uptime?
In addition to slow and fast transactions to measure health, Plumbr also monitors JVM uptime to detect situations where the JVM itself has crashed and is unavailable to service any incoming transactions.
Uptime monitoring is built upon active checks from the Plumbr Server to the Agent that occur once every five seconds. Such checks are called heartbeats. When eight or more sequential heartbeats go unanswered by a JVM, a downtime event is registered.
Such downtime can be either planned downtime (caused by a graceful JVM shutdown) or unplanned downtime (e.g. caused by an OutOfMemoryError or a segmentation fault). For downtime events where Plumbr is able to capture the reason why downtime was triggered, Plumbr also exposes it as the root cause.