[ 
https://issues.apache.org/jira/browse/ROCKETMQ-97?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Jaskey Lam updated ROCKETMQ-97:
-------------------------------
    Description: 
For sendAsync, when excpetion occurs, the developer will have nothing to 
distiguish what kind of exception it is unless read the remark or stacktrace.

Actually, for some broker problems, say service_not_availble or system_error, 
the devs should do some special actions. This can be got from send in sync way 
since the response will return to the caller.

But now, in async send, with no response code , we can't distinguish them.

Even more, for some situation, the queue sent to  is needed , so if possible, 
there should be an ext fields where collect as many message as possible to let 
developer to descide their callback behavior. Which is neither exist in sync 
nor async ways.

  was:
For sendAsync, when excpetion occurs, the developer will have nothing to 
distiguish what kind of exception it is unless read the remark or stacktrace.

Actually, for some broker problems, say service_not_availble or system_error, 
the devs should do some special actions.

But for now, with no response code , we can't distinguish them.

 Even more, for some situation, the queue sent to  is needed , so if possible, 
there should be an ext fields where collect as many message as possible to let 
developer to descide their callback behavior.


> SendCallback should bring response code and extMessags in OnException
> ---------------------------------------------------------------------
>
>                 Key: ROCKETMQ-97
>                 URL: https://issues.apache.org/jira/browse/ROCKETMQ-97
>             Project: Apache RocketMQ
>          Issue Type: Improvement
>          Components: rocketmq-client
>            Reporter: Jaskey Lam
>            Assignee: Xiaorui Wang
>
> For sendAsync, when excpetion occurs, the developer will have nothing to 
> distiguish what kind of exception it is unless read the remark or stacktrace.
> Actually, for some broker problems, say service_not_availble or system_error, 
> the devs should do some special actions. This can be got from send in sync 
> way since the response will return to the caller.
> But now, in async send, with no response code , we can't distinguish them.
> Even more, for some situation, the queue sent to  is needed , so if possible, 
> there should be an ext fields where collect as many message as possible to 
> let developer to descide their callback behavior. Which is neither exist in 
> sync nor async ways.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Reply via email to