Re: [infinispan-dev] a nice HTML5 console for Infinispan & a question on MBean names...
On 27 March 2013 13:21, Tristan Tarrant <ttarrant(a)redhat.com> wrote:
On 03/22/2013 12:47 PM, James Strachan wrote:
> Many thanks!
>
> BTW does this exception ring any bells:
> ERROR | qtp581815510-700 | ISPN019003: Interpreter error
> java.lang.NullPointerException
> at org.infinispan.cli.interpreter.Interpreter.execute(Interpreter.java:161)
> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
> at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> at java.lang.reflect.Method.invoke(Method.java:601)
> at org.infinispan.jmx.ResourceDMBean.invoke(ResourceDMBean.java:287)
> at
com.sun.jmx.interceptor.DefaultMBeanServerInterceptor.invoke(DefaultMBeanServerInterceptor.java:819)
> at com.sun.jmx.mbeanserver.JmxMBeanServer.invoke(JmxMBeanServer.java:791)
That looks like a parser error has resulted in a null statement being
added. Odd it should have caused a parser error. Can you give me an
example of statement you issued ?
Everything I try I get the same response. e.g. any of these...
get
get 'a'
get 5
get key
get key 5
I tried anything I could find on this page:
https://community.jboss.org/wiki/InfinispanCLI
On the hawtio server I emailed you off-list - can you enter any CLi
statements without getting the NPE?
It'd be nice to get a nicer parser exception message than an NPE BTW
:). "Incorrect character at 8,1, expected 'x'" or something would
help; then we could highlight the dodgy bit of the CLI command in the
UI
> Incidentally do the session IDs timeout after a while? Seems like
I
> saw that once or twice (though could be a hawtio bug :). If so I'm
> wondering if it might be easier to have a new method
> executeWithTemporarySessionId(cacheName, sql) or something like that
> so we don't have to keep track of when a session ID is stale or have
> to do 2 operations when we don't mind about the session ID (i.e. use
> one session ID per operation)?.
Yes, sessions timeout. There is a "Ping" command you can issue which
"refreshes" a session. As for allowing a temporary session, you would
lose some of the state information, so I'm not really sure about that.
What state is there between commands? If the commands are basically a
"get" or a "put" and they execute immediately on the cache;
there's
not much state to lose right?
Am just thinking of the complexity of having to keep pinging to keep a
session ID open that may still time out (since the UI can't be sure
its in the same keep-alive cycle as the server).
From hawtio am not aware of any state we need to keep across commands
yet (though I'm an infinispan newbie).
--
James
-------
Red Hat
Email: jstracha(a)redhat.com
Web: http://fusesource.com
Twitter: jstrachan, fusenews
Blog: http://macstrac.blogspot.com/
Open Source Integration