Github user HyukjinKwon commented on the issue:

    https://github.com/apache/spark/pull/21060
  
    I am okay if there's a specific reason. I think this is the point - if 
there's a specific reason, that should be mentioned and explained ahead. 
Actually, I (and @srowen did as well IIUC) asked this many times, see above.
    
    I would have investigated or would have just said that I am okay with 
reverting. I don't usually get in the way if there's a specific reason. It 
would be great if we can have more open talks next time. 
    
    > for the 2.3.x backport, add a config that so it is possible to turn this 
off in production, if somebody actually has their job failed because of this? 
It's a small delta from what this PR already does, and that should alleviate 
the concerns @gatorsmile has.
    
    I am personally fine with reverting or adding a configuration if that's 
what you guys feel strongly; however, I should say it sounds unusual to have a 
config to control this behaviour in branch-2.3 alone and it sounds less worth. 
The case you mention sounds really unlikely and I wonder if that makes sense 
tho. It's also experimental as you all said.
    
    Also, I should note that I have been confused about the backporting policy 
and the bunch of configurations to control each behaviour. If that's just 
concerns to be addressed, that's fine but sounds what people must follow so 
far. If this is true, I feel sure this should be documented. I feel sure we 
shouldn't have such overhead next time. I am pretty sure this isn't the first 
time.


---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org

Reply via email to