Hi Atanas, > However, the usage of this EIP requires modification of the original route > (insertion of an idempotentConsumer instruction before every destination > endpoint) and I would prefer to use an alternative approach (if it's > possible).
Why would modifying the route (e.g. inserting idempotent consumers) be a problem? > Does the exchange keep information about which route endpoints have > successfully processed it? Maybe it's the exchange's unit of execution > holding this info? Take a look at the message history of the exchange: http://camel.apache.org/message-history.html Also as you can read on the above page you can also use the Java AP of the UnitOfWork: http://camel.apache.org/maven/current/camel-core/apidocs/org/apache/camel/spi/TracedRouteNodes.html. E.g: exchange.getUnitOfWork().getTracedRouteNodes().getNodes(). But I think using an idempotent consumer is easier. Kind regards, Richard On Tue, Jan 21, 2014 at 8:26 PM, Atanas Shindov <atanas.shin...@gmail.com>wrote: > Hi Richard, > > thanks for your prompt reply. > > It looks like the Idempotent Consumer might help, especially if in it's > File/JDBC based implementations the message ids are durably persisted (it > seems quite logical to be so), so that they are available after a JVM > restart for example. > > However, the usage of this EIP requires modification of the original route > (insertion of an idempotentConsumer instruction before every destination > endpoint) and I would prefer to use an alternative approach (if it's > possible). > > Does the exchange keep information about which route endpoints have > successfully processed it? Maybe it's the exchange's unit of execution > holding this info? > > Thanks again for your help. > > Best regards, > Atanas > > > > -- > View this message in context: > http://camel.465427.n5.nabble.com/Avoiding-duplicate-messages-tp5746278p5746304.html > Sent from the Camel - Users mailing list archive at Nabble.com. >
