I would like to have a Camel Route deploying on multiple servers/docker containers to run in active-standby mode. After some studies, I'm targeting to apply RoutePolicy that leverage "leader election" solutions like etcd and consul.
Checked the Camel source code in Github and found that there are EtcdRoutePolicy(https://github.com/apache/camel/blob/master/components/camel-etcd/src/main/java/org/apache/camel/component/etcd/policy/EtcdRoutePolicy.java) and ConsulRoutePolicy (https://github.com/apache/camel/tree/master/components/camel-consul/src/main/java/org/apache/camel/component/consul/policy) available in the latest branch. However, I'm a bit confused on the implementation... If I understand correctly, both of the route policies requires the definition of serviceName/servicePath during initialization, and acquire lock using those 2 attributes during route "doStart". However, it seems more make sense to me to override onStart, in which it has the route object as the argument, thus I could acquire lock using the route-id... Would like to understand if there is any reason those 2 routepolicies leverage doStart instead of onStart? what is the difference between these 2 events in a route lifecycle? -- View this message in context: http://camel.465427.n5.nabble.com/RoutePolicySupport-Difference-between-onStart-and-doStart-tp5784061.html Sent from the Camel - Users mailing list archive at Nabble.com.
