[FLUME-2355] Exception in process of Avro sink does not reset the cxnResetExecutor

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 1.4.0
• Component/s: Sinks+Sources
• Assignee: Unassigned
• Reporter: Benjamin Fiorini
• Created: 08/Apr/14 10:23
• Updated: 08/Apr/14 10:23

Description

In the AbstractRpcSink, when an exception occurs during the process, say a timeout, the connection is destroyed. But when the connection is created again, a new cxnResetExecutor is scheduled. The first one being still scheduled, this may result in many jobs reseting the connection far too often.

I suppose the createConnection should take care of shutting down any existing reset job before creating new ones.