Your node is under memory pressure, after the GC there is still 5.7GB in use. In fact it looks like memory usage went up during the GC process.
Can you reduce the memtable size, caches or the number of CF's or increase the JVM size? Also is this happening under heavy load ? Aaron On 6/03/2011, at 10:34 AM, ruslan usifov wrote: > Hello > > I have cluster of 3 cassandra 0.7.3 nodes, and some times nodes a frozen. As > i understand this because GC, in this moment i see > > INFO [ScheduledTasks:1] 2011-03-05 15:21:23,524 GCInspector.java (line 128) > GC for ConcurrentMarkSweep: 18052 ms, -997761672 reclaimed leaving 5796586088 > > How can i reduce this? I installed JNA (i install it correctly: > "CLibrary.java (line 106) JNA mlockall successful" log record in system.log > present on all nodes), and set vm.swappiness = 0, but this doesn't help. Any > suggest? > > PS: Nodes have 12GB of memory with 6GB HeapSize