Github user rdblue commented on a diff in the pull request:
    --- Diff: 
    @@ -62,6 +62,16 @@
       DataWriterFactory<Row> createWriterFactory();
    +  /**
    +   * Returns whether Spark should use the commit coordinator to ensure 
that only one attempt for
    --- End diff --
    Currently, the commit coordinator will only authorize one attempt and only 
authorize another if the authorized attempt fails, so it does ensure that only 
one attempt commits. Do you think the wording here needs to change?
    Instead of documenting the behavior of the commit coordinator here, I'd 
rather point to its docs. Are those written, or is the coordinator an internal 


To unsubscribe, e-mail:
For additional commands, e-mail:

Reply via email to