Feature #10

define criterion how long a client waits for a remote call

Added by Jörg Plewe over 2 years ago. Updated about 2 years ago.

Status:New Start:09/24/2009
Priority:Normal Due date:
Assigned to:- % Done:

0%
Category:-
Target version:0.5.4

Description

Currently a caller waits for a configurable time (org.bundlebee.remoteservicecall.call.readtimeout, defaults to 5000ms) before it throws an exception. Any number given as a timeout is equally wrong for no a priori assumption can be made about a methods runtime.

IMHO a call should only be prematurely cancelled by exception when something goes wrong (connection loss, remote method throwing exception).

Is a http heartbeat appropriate to detect a remote node fading away?

History

Updated by Jörg Plewe over 2 years ago

Currently, we decided to leave it with a configurable timeout:

org.bundlebee.remoteservicecall.call.readtimeout=60000

Updated by Philipp Haußleiter about 2 years ago

  • Target version set to 0.5.4

Also available in: Atom PDF