Please enable JavaScript.
Coggle requires JavaScript to display documents.
Utilihive Integration flow-server runtime - Coggle Diagram
Utilihive Integration flow-server runtime
Management API
exposed as Spring boot actuator endpoints <flow-server management root>:<flow-server management port>/actuator/flows
flow sources
FIRST processor is treated as a source
types
normal
are not sharded but scoped on single actor system instances. Normal flow sources are created on every node. Normal flow sources are commonly used for server-like sources, such as REST and SOAP APIs.
singleton
Singleton sources are only created on one node. They are usually needed for polling endpoints such as file consumers.
provisioning
Each actor system has a LocalFlowSourcesManagerActor that manages all the sources in the actor system
The GlobalFlowSourcesProvisionerActor makes sure each LocalFlowSourcesManagerActor is running the desired sources at all times. Whenever a LocalFlowSourcesManagerActor is terminated with active sources, the GlobalFlowSourcesProvisionerActor will spawn short-lived endpoint-less sources with other available LocalFlowSourcesManagerActors to recover any unconfirmed messages from the sources that were shut down to avoid message-loss
actions
flow-server joining cluster
flow deletion
flow update
flow-server leaving cluster
flow deployment
FlowSourceActor
persistent actor
forwards to the sharded integration flow for further processing
coupled with SourceEndpoint
SourceEndpoint is the component responsible for handling the actual transport and protocol of the source
created by SourceEndpointFactory
Standalone flows
can only have a single processor - the source
takes the full responsibility of processing messages and does not delegate to a sharded flow