We’re using Camel’s “master” feature around a timer route, because we have two Kubernetes (k8s) Pods and we only want either one of them to run the timer. We’re using the camel-file implementation.

When we first deploy our (two) pods, it all works fine for quite a while: i.e. only one pod runs the timer. Everything may run fine for days.

But, if k8s terminates a pod and moves it to another node, it too starts to run the timer
The Pods are configured to use the same file-share for camel to coordinate its lock-files, so not sure what could be going on.

Why would the moved pod work differently from the original two?
Any help would be appreciated