Memcache Server Unreachable - Client locks up

[GoogleCodeExporter]

What version of the product are you using? On what operating system?
2.2 / Windows XP

Tell me more...
In testing I ran a client against 4 memcached servers using the Ketema 
connection factory and the default buffer and queue sizes.  I went in 
killed one of the memcached servers and the client immediately became 
unresponsive and generated an endless loop of the following log entries:

009-01-30 13:28:59.625 INFO net.spy.memcached.MemcachedConnection:  
Reconnecting due to exception on {QA sa=/192.168.181.229:11212, #Rops=0, 
#Wops=1978, #iq=0, topRop=null, 
topWop=net.spy.memcached.protocol.ascii.StoreOperationImpl@748a5d, 
toWrite=0, interested=0}
java.net.ConnectException: Connection refused: no further information
    at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method)
    at 
sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:574)
    at 
net.spy.memcached.MemcachedConnection.handleIO(MemcachedConnection.java:252
)
    at 
net.spy.memcached.MemcachedConnection.handleIO(MemcachedConnection.java:184
)
    at net.spy.memcached.MemcachedClient.run(MemcachedClient.java:1298)
2009-01-30 13:28:59.625 WARN net.spy.memcached.MemcachedConnection:  
Closing, and reopening {QA sa=/192.168.181.229:11212, #Rops=0, #Wops=1978, 
#iq=0, topRop=null, 
topWop=net.spy.memcached.protocol.ascii.StoreOperationImpl@748a5d, 
toWrite=0, interested=0}, attempt 11.
2009-01-30 13:29:00.828 INFO net.spy.memcached.MemcachedConnection:  
Reconnecting {QA sa=/192.168.181.229:11212, #Rops=0, #Wops=1980, #iq=0, 
topRop=null, 
topWop=net.spy.memcached.protocol.ascii.StoreOperationImpl@748a5d, 
toWrite=0, interested=0}
2009-01-30 13:29:01.828 INFO net.spy.memcached.MemcachedConnection:  
Connection state changed for sun.nio.ch.SelectionKeyImpl@667f17
2009-01-30 13:29:01.843 INFO net.spy.memcached.MemcachedConnection:  
Reconnecting due to exception on {QA sa=/192.168.181.229:11212, #Rops=0, 
#Wops=1982, #iq=0, topRop=null, 
topWop=net.spy.memcached.protocol.ascii.StoreOperationImpl@748a5d, 
toWrite=0, interested=0}
java.net.ConnectException: Connection refused: no further information
    at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method)

I would expect retries to occur over a certain period of time and then 
eventually an operation timeout.  However, in this case the client just 
went in an endless loop.

Original issue reported on code.google.com by chase.wo...@gmail.com on 30 Jan 2009 at 8:47

[GoogleCodeExporter]

Hi , I have downloaded memcached 2.2.jar and spy jar available on the home 
page. I am
trying first sample example, however i am getting Connection refused exception. 
We
are running this program using JDK 1.6 and on Windows XP machine. Could you 
please
help us to resolve the issue.

Original comment by rje...@gmail.com on 10 Feb 2009 at 7:05

[GoogleCodeExporter]

rjee83:  Your comment is unrelated to this bug.

Original comment by dsalli...@gmail.com on 10 Feb 2009 at 7:45

[GoogleCodeExporter]

I am running into the exact same issue on: 

spymemcached 2.2
Mac OS X10.5.6
Java HotSpot(TM) Client VM (build 1.5.0_16-133, mixed mode, sharing)

Original comment by jamesggo...@gmail.com on 26 Mar 2009 at 9:09

[GoogleCodeExporter]

This is by design.  You can change the behavior by specifying a different 
FailureMode.

Original comment by dsalli...@gmail.com on 2 Apr 2009 at 3:56

• Changed state: Fixed