本帖最后由 fanbells 于 2014-4-16 14:38 编辑
Hbase集群中有一个节点的io特大,查看了hbase日志发现全是下面的日志,问一下是什么原因导致的?
DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction completed; freed=50.89 MB, total=381.31 MB, single=251.01 MB, multi=190.39 MB, memory=0 KB
DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction started; Attempting to free 50.82 MB of total=431.71 MB
DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction completed; freed=50.9 MB, total=386.07 MB, single=255.71 MB, multi=190.52 MB, memory=0 KBDEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction started; Attempting to free 50.83 MB of total=431.72 MB
DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction completed; freed=50.85 MB, total=381.18 MB, single=250.91 MB, multi=190.52 MB, memory=0 KB
DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction started; Attempting to free 50.82 MB of total=431.7 MB
我猜测原因是不是在对表扫描时开启了块缓存,导致一次又一次的触发缓存驱逐过程。
|