ctubbsii commented on PR #3789:
URL: https://github.com/apache/accumulo/pull/3789#issuecomment-1740175887

   > I have no issue with this superseding the #3771 and #3776. As long as the 
test becomes stable. (#3771 should be closed in favor of #3776 anyway.)
   
   Okay, I'll go ahead and merge it and close the others, then.
   
   > We may want to consider a specific test that can create / test for the 
condition rather than hoping to hit it by chance in 
`ManagerRepairsDualAssignmentIT` Changing the way killing and testing that it 
has been reported dead may be better as is done in #3776 - but that would mask 
hitting the thrift change.
   
   I had similar thoughts myself. For now, I'm inclined to leave the IT as it 
is, and close the other issues, rather than modify it in a way that would mask 
this issue and create a dedicated test for it. I think this issue basically was 
just an extension of the code reviews for #3737, resulting in us changing our 
mind on a portion of that change prior to a release of it. We're just slightly 
rolling back to the previous status quo, where everything was fine. So, I'm not 
terribly inclined to do much more than just roll that one small change back out.


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to