Version: spymemcached-2.10.2.jar
We see constant increase in memory with size of LinkedBlockingQueue growing
beyond 600K.
We also see a lot of CheckedOperationTimeoutException from
net.spy.memcached.internal.OperationFuture.get(OperationFuture.java:166)
It looks like the operations in the queue are not cleared when the operation
timesout.
We do this in finally block of MemcachedClient.asyncGet
if (future != null && !future.isDone()) {
future.cancel(true);
}
does future.cancel(true) interrupt the thread preventing the operation from
being removed from the queue? Should we use future.cancel(false) instead? If
future.cancel(false) is used, can this lead to thread leak if the operation
didn't complete?
Original issue reported on code.google.com by vinothku...@gmail.com on 4 Jul 2014 at 12:12
Original issue reported on code.google.com by
vinothku...@gmail.com
on 4 Jul 2014 at 12:12