Skip to main content
sat_046
sat_046
Bronze III
February 1, 2023
Question

How to introduce delay time between retry attempts for spark streaming job

We are running a spark streaming job which reads data from kafka and writes to RDBMS. We dont want this job to fail easily due to minor fluctuations in cluster health/issue. Write now spark job has configuration to retry 5 attempts before the whole job fails. But all these retries are happening in quick succession , one after another. Is there a way we can put some delay/sleep time between retry attempts for this job?

3 replies

RangaReddy
Gold ⭐️
RangaReddy
Gold ⭐️
February 9, 2023

Hi @sat_046 

 

I don't think we have a specific configuration parameter to handle the task retry attempts with some delay. But we have a parameters to blacklist the node if the task is failed with some no of attempts in the node.

 

References:

1. https://community.cloudera.com/t5/Community-Articles/Configuring-spark-task-maxFailures-amp-spark-blacklist-task/ta-p/335235

2.  https://www.waitingforcode.com/apache-spark/failed-tasks-resubmit/read

 

Translate
VidyaSargur
Community Manager
VidyaSargur
Community Manager
February 13, 2023

@sat_046 Has the reply helped resolve your issue? If so, please mark the appropriate reply as the solution, as it will make it easier for others to find the answer in the future. 

Translate
Regards, Vidya Sargur, Community Manager Was your question answered? Make sure to mark the answer as the accepted solution. If you find a reply useful, say thanks by clicking on the thumbs up button.
RangaReddy
Gold ⭐️
RangaReddy
Gold ⭐️
March 30, 2023

Hi @sat_046 

 

As i mentioned earlier comment, unfortunately it is not possible to delay the tasks. You can find the Spark code when tasks failed.

 

https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala#L879C8-L1002

 

Please accept the solution if you liked my answer.

Translate
Terms & ConditionsAccessibility statement