AngersZhuuuu opened a new pull request #27419: [SPARK-30694][SHUFFLE]If 
exception occured while fetching blocks by ExternalBlockClient, fail early when 
External Shuffle Service is not alive
URL: https://github.com/apache/spark/pull/27419
 
 
   
   ### What changes were proposed in this pull request?
   Like [SPARK-27637](https://github.com/apache/spark/pull/25469), 
   When we fetch data use External Shuffle Service, we use 
ExternalBlockStoreClient to fetch blocks
   from External Shuffle Service, in real huge cluster, we always need to 
offline some bad machine, and remote External Shuffle Service will down and we 
will can't create TransportClient to connect.
   It will throw IOException, but in current way, we will only retry until run 
out of retry times. 
   
   I think we should throw this error early like  
[SPARK-27637](https://github.com/apache/spark/pull/25469)
   
   
   ### Why are the changes needed?
   Old solution not comprehensive. Throw exception early and reduce unnecessary 
retry.
   
   
   ### Does this PR introduce any user-facing change?
   NO
   
   ### How was this patch tested?
   WIP
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to