keith-turner commented on issue #5870:
URL: https://github.com/apache/accumulo/issues/5870#issuecomment-3259570516
The problem is in the ReserveTablets code. It sets the opid and looks for a
lack of location at the same time. We need to set all opid first and then in
another pass after we know the opid was set then scan again for locations. So
the following probably happened
1. Reserve tablets saw no locaton and set the opid
2. The future was set before the opid was set by another thread (TGW thread)
3. The opid is set by reserve tablets
4. Reserve tablets did not see a location and it set the opid successfully,
so it thinks its good. It does not know the future location was concurrently
set after is scanned and before it set the opid.
May need to add a new Repo to the fate operation that runs after
ReserveTablets and in its isReady is scans for locations. The code is
currently doing a scan before setting the opid, it needs to scan after its set.
--
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: notifications-unsubscr...@accumulo.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
