Details
-
Bug
-
Resolution: Fixed
-
1.6.0 beta4
-
None
-
Operating System: All
Platform: All
Description
During rebalance of a node which has evicted keys from RAM, I see thousands of the following in the logs:
INFO REPORT <5632.225.0> 2010-07-19 14:33:13
===============================================================================
Message from memcached: Queued a background fetch, now at 1
Found match in report number 378885
INFO REPORT <5632.225.0> 2010-07-19 14:33:13
===============================================================================
Message from memcached: Fetching `00167812' in the background
These messages also continue even after the rebalance is stopped, so it seems like some backfill thread is still running.
I think we don't have any hope of being able to transfer or replicate any vbucket with a substantial number of items paged to disk if we're going to pull items up one at a time in the backfill thread, nor if we log for each item.