Comparison
EvictionvsQuality of service class
Eviction
a healthy workload was terminated because the node it happened to be on ran out of disk, and it had done nothing wrong.
The platform terminating running workloads to relieve pressure on a node — memory, disk or PIDs — choosing victims by QoS class and by how far over their requests they are. It is a deliberate sacrifice rather than a failure, and it is why a workload can die for reasons entirely outside its own behaviour. Disk pressure from unbounded logs or image sprawl is the most common trigger and the easiest to prevent.
Full entry →Quality of service class
the workload with no requests set was the first one killed when the node ran out of memory, and nobody had chosen that.
A classification the platform derives from your requests and limits, which decides the order things are killed under pressure: equal requests and limits are protected, requests below limits are burstable, neither set is expendable. It is not a field you set, it is a consequence of two fields you did set, which is why teams are surprised by their own ranking. Anything whose loss would be an incident should be in the protected class deliberately.
Full entry →