> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  line 65
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line65>
> >
> >     maybe firstCallbackException?
> >     Don't think we need @volatile anymore.
> >     
> >     Minor: Don't need type declaration.
> >     Minor: update documentation.

fixed.


> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  line 117
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line117>
> >
> >     Can you explain/document what we're trying to achieve here? Is this 
> > producer meant to somehow heal itself eventually and stop throwing this 
> > exception?
> >     
> >     If that's the case doesn't that mean that 
> > "callbackExceptionFirstThrown" is a misnomer - it isn't the exception first 
> > thrown anymore.

This means we will throw the exception to the user, instead of failing 
container directly. We clear the exception since it's already reported.


> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  line 152
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line152>
> >
> >     What does this retry refer to? Looks like we just close the producer in 
> > case of failure?

This is legacy doc. Should be removed. fixed.


> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  line 156
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line156>
> >
> >     Minor: use logger.error for consistency.
> >     Minor: capitalize "Closing"

Fixed.


> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  lines 157-158
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line157>
> >
> >     If the producer is meant to heal itself eventually, is this still true?

I fixed the doc. Please take a look again.


> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  line 162
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line162>
> >
> >     Curious why we use format instead of Scala interpolated strings?

legacy logging.


> On Sept. 29, 2016, 7:33 p.m., Prateek Maheshwari wrote:
> > samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala,
> >  line 202
> > <https://reviews.apache.org/r/52403/diff/1/?file=1516363#file1516363line202>
> >
> >     Wondering why we rethrow a previously saved exception instead of 
> > no-oping the flush? Is the user expected to handle this? If so, should 
> > document why & how.
> 
> Prateek Maheshwari wrote:
>     Nvm this, I think the intention is to throw in the next operation after a 
> send error callback.

I added the comments. Please take a look again.


- Xinyu


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/52403/#review150909
-----------------------------------------------------------


On Sept. 29, 2016, 9 p.m., Xinyu Liu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/52403/
> -----------------------------------------------------------
> 
> (Updated Sept. 29, 2016, 9 p.m.)
> 
> 
> Review request for samza, Navina Ramesh and Jagadish Venkatraman.
> 
> 
> Repository: samza
> 
> 
> Description
> -------
> 
> Current the error log happens after produce close and reset the exception in 
> later callbacks, which caused the trouble shooting to be harder in cases of 
> multithreading. We should log error before closing and keep atomic reference 
> of the initial exception.
> 
> 
> Diffs
> -----
> 
>   
> samza-kafka/src/main/scala/org/apache/samza/system/kafka/KafkaSystemProducer.scala
>  5ff6d3caf54ed148aa40c7c752c587e556a4f34a 
> 
> Diff: https://reviews.apache.org/r/52403/diff/
> 
> 
> Testing
> -------
> 
> Tested in jobs deployed in Yarn cluster.
> 
> 
> Thanks,
> 
> Xinyu Liu
> 
>

Reply via email to