milleruntime commented on issue #1916:
URL: https://github.com/apache/accumulo/issues/1916#issuecomment-843413614


   > I'm not sure it's a race condition with metadata writes. I'm thinking this 
has something to do with a logic error involving WALs and the garbage 
collector. In both SuspendedTabletsIT and in #608, there was a tserver failure 
and WAL recovery.
   
   So you think its more likely to be another recovery bug? Specifically, a 
recovery bug when recovering the metadata table? I know this IT used to have 
issue when crashing a tserver hosting Metadata tablets. If this was fixed in 
#1888 or if it shows up too infrequently in SuspendedTabletsIT, then perhaps we 
need another test, specifically for this bug.


-- 
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]


Reply via email to