tgravescs commented on issue #27240: [SPARK-30512] Added a dedicated boss event 
loop group
URL: https://github.com/apache/spark/pull/27240#issuecomment-577874153
 
 
   thanks for the explanation, that makes sense then to keep the chunk fetch 
thread pool.
   
   I'm still wondering about the sasl timeout error.  Without the separate boss 
thread it was up to the worker thread to accept it and do sasl communication. 
if that was timing out then it couldn't get a worker thread in that time 
(30seconds by default).  But couldn't that be the same thing with a boss 
thread. Now a boss thread will accept connection but it won't actually be 
processed until a worker thread can handle it, which seems like it could still 
be passed the 30 seconds, the same as if there wasn't. boss thread.
   
   I thought that timeout was for the send and response from server (in this 
case with the auth response) that would start the negotiation. So just 
accepting it wouldn't get a response.
   
   Does the boss thread block on handing a connection to the worker or does it 
have a queueing mechanism it can keep accepting while others wait on workers?

----------------------------------------------------------------
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