Details
-
Bug
-
Resolution: Duplicate
-
Major
-
1.7.1, 2.0, 2.0.1, 2.1.0
-
Security Level: Public
-
02/Sep/2013 - 20/Sep/2013
Description
rebalancing out a node when cluster is running out of capacity causes all memcached nodes to crash
it would be nice if rebalance process fails to start before we fill up the space on some of the nodes in the cluster so that all nodes don't crash
in my testing i rebalanced out 6->4 nodes and i actually stopped rebalanced but all memcached processes crashes even on those nodes where there was still some space left
Attachments
Issue Links
- relates to
-
MB-8792 ns_server should protect itself when running out of disk space
- Closed