tree 0cc68a2c90a10f1bc10f75e49b9f521d502ea3dc
parent 1759eb535ff2dbd3dea3fa8626eba0fe70ecb113
author Francisco Guerrero <francisco.guerrero@apple.com> 1665437712 -0700
committer Yifan Cai <yifan_cai@apple.com> 1666047421 -0700

CASSANDRASC-44 Refactor health check to use vertx timer

Vertx API offers a periodic timer that integrates with it's internal thead pooling
mechanism. In this commit, we utilize vertx's periodic timer in favor of using a
`Executors.newSingleThreadScheduledExecutor()` on each delegate.

Another benefit of this approach is that if the cluster topology changes, i.e.
node replacement, cluster expansion / shrink, then the health checks will be performed
against the actual nodes in the cluster, assuming we receive an updated view of the
cluster when invoking the `Configuration#getInstancesConfig()#instances()` method.
We no longer need to worry about decommissioning the single thread executors running
on each delegate.

patch by Francisco Guerrero; reviewed by Yifan Cai, Dinesh Joshi for CASSANDRASC-44