Details
-
Task
-
Status: Closed
-
Major
-
Resolution: Invalid
-
None
-
None
-
None
Description
Following a discussion with Stack, the way ReplicationSource manages the list that holds all the entries read from a HLog may be going against GC instead of hand in hand with it. We need to profile the current way it's done VS creating new lists every time we tail.
Attachments
Issue Links
- relates to
-
HBASE-2223 Handle 10min+ network partitions between clusters
- Closed