[
https://issues.apache.org/jira/browse/FLINK-2624?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14734404#comment-14734404
]
Hilmi Yildirim edited comment on FLINK-2624 at 9/8/15 8:41 AM:
---------------------------------------------------------------
There is a problem with the method restoreState. It is not possible to restore
the state. If a worker node crashes and the connection to RabbitMQ is lost,
then the messages which are not acknowledged by the worker node will be sent to
other consumers. Therefore, it is not possible to restore arbitrary states. The
only way to restore an arbitrary state is to save all received messages
internally. But it is is possible to restore the state of the last snapshot by
reconnecting to the RabbitMQ Queue.
was (Author: hilmiyildirim):
There is a problem with the method restoreState. It is not possible to restore
the state. If a worker node crashes and the connection to RabbitMQ is lost,
then the messages which are not acknowledged by the worker node will be sent to
other consumers. Therefore, it is not possible to restore arbitrary states. The
only way to restore an arbitrary state is to save all received messages
internally. But is is possible to restore the state of the last snapshot by
reconnecting to the RabbitMQ Queue.
> RabbitMQ source / sink should participate in checkpointing
> ----------------------------------------------------------
>
> Key: FLINK-2624
> URL: https://issues.apache.org/jira/browse/FLINK-2624
> Project: Flink
> Issue Type: Bug
> Components: Streaming Connectors
> Affects Versions: 0.10
> Reporter: Stephan Ewen
> Assignee: Hilmi Yildirim
>
> The RabbitMQ connector does not offer any fault tolerance guarantees right
> now, because it does not participate in the checkpointing.
> We should integrate it in a similar was as the {{FlinkKafkaConsumer}} is
> integrated.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
