Enum ClusterNodeState
- Namespace
- Quartz
- Assembly
- Quartz.dll
What the cluster makes of one node's check-in history.
public enum ClusterNodeState
Fields
Alive = 0The node checked in within its own check-in interval: it is doing what a running node does.
Failed = 2The node has been silent past the point at which the cluster stops waiting for it.
The next check-in pass by any node recovers this one's in-flight work and deletes its row, so a node reported failed is normally reported once and then disappears from the listing.
Overdue = 1The node has missed a check-in but has not yet been silent long enough to be declared dead.
Normal under load or across a slow link, and the state a node passes through on its way to Failed. Nothing is recovered from an overdue node.
Remarks
The three values are read off one clock — the observing node's — against the check-in stamps every node writes, so this is what this node believes about the others. The Failed boundary is the same one the recovery sweep applies, so a node this listing calls failed is a node whose work the cluster is about to take over.