db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kristian Waagan <Kristian.Waa...@Sun.COM>
Subject Re: Derby confuse with SWing Application Framework
Date Wed, 02 Jun 2010 08:31:19 GMT
On 31.05.10 10:58, Kristian Waagan wrote:
> On 31.05.10 05:46, Hawkx wrote:
>>> Maybe there is a problem on you machine, or maybe it is a bug in Derby.
>>> It would be helpful if you could do what Bryan suggested - obtain the
>>> stack trace from the Java process when it is hanging.
>>> Besides from what Bryan mentioned, you can also use jstack (or 
>>> visualvm)
>>> to do this.
>> jstack message while blocking:
>
> Hi,
>
> I can't see why this shouldn't work, and I'm afraid I only have some 
> more general comments and suggestions:
>   - should we add a timeout to the socket creation in the various 
> commands? (ping(), sysinfo(), etc)

See also this related issue:
https://issues.apache.org/jira/browse/DERBY-1467 ("It would be useful to 
have a NetworkServerControl start API that took a timeout parameter")


-- 
Kristian

>   - when the application is hanging, are you able to connect to the 
> network server using a separate process, like ij?
>   - what happens if you start the network server stand alone? Is the 
> application then able to complete the sysinfo-call?
>
> As far as I can see, the sysinfo-call is awaiting a response from the 
> server.
> The server process accepting connections (ClientThread) is still doing 
> ServerSocket.accept().
> Could this be a threading problem because of AWT? (i.e., another task 
> is waiting for the event queue, which is stuck in the socket call?)
>
>


Mime
View raw message