Details
-
Bug
-
Status: Patch Available
-
Major
-
Resolution: Unresolved
-
2.0.6, 2.1.9
-
None
-
None
Description
My HBase 2.0/2.1 cluster experienced severe performance drop after running reversed scans for weeks plus 1~2 major compaction. When debug into regionserver logs, it appears the bucket cache is full for every ~100 ms and its eviction can't work properly because ref count has become negative:
(every ~100 ms)
Free started because "Full!"
This block AAA is still referred by -1 readers. Can not be freed now
This block BBB is still referred by -2 readers. Can not be freed now
This block CCC is still referred by -3 readers. Can not be freed now
...
This block ... is still referred by (negative number) readers. Can not be freed now
There have been patch HBASE-22480 by binlijin addressing this ref count problem in 2.2+. I'd like to backport it to 2.1 (understand that 2.0 is EOL)
Attachments
Attachments
Issue Links
- relates to
-
HBASE-22480 Get block from BlockCache once and return this block to BlockCache twice make ref count error.
- Resolved
- links to