failed_node_exception: failed_node_exception

What does it mean?

During a cluster-wide operation (e.g. search, stats, management command), a specific node in the cluster could not process or respond to the request. The coordinating node wraps the failure from that node in this exception.

Why does it happen?

The node crashed, dropped off the network, could not respond due to overload/GC pauses, disk/resource problems, or a shard error on that node. It can also appear transiently during cluster reconfiguration (nodes leaving/joining).

How to fix it

Use caused_by to determine on which node and why the failure occurred. Check node health: GET _cat/nodes, GET _nodes/stats (CPU, memory, GC, disk). If the node is under resource pressure, review the load/heap settings; verify network connectivity. If the node left the cluster, wait for it to rejoin and for shard allocation to recover (GET _cluster/health).