Re: Trying to track down intermittent classserver problems

Patrick Wright wrote:
> Hi Gregg
> 
> On Fri, Jul 25, 2008 at 4:59 PM, Gregg Wonderly <[email protected]> wrote:
>>> What I'm interested in is how to track down the cause of this if it
>>> recurs. We were more or less at a loss.
>> What operating system was this on?  Windows has interesting behavior
>> problems when you exceed its builtin limit on number of total socket
>> connections possible.  You could track, with netstat, the number of sockets
>> active and see a consistent bordering number to when this failure occurs to
>> see if there is an OS limit related to total sockets active.
> 
> Thanks for the reply. The OS on all systems (client, and both
> LUS/codebase servers) are home-grown, 32bit, 2.4x kernel Linux build.
> When this happens next, I'll track the count of open sockets on both
> sides of the communication to see if it points towards the problem.

The issue that Dennis brought up is another consideration.  It might be that 
there is a DNS related problem, or a connectivity issue related to which 
interface the traffic is originating on.

Gregg Wonderly

--------------------------------------------------------------------------
Getting Started:     http://www.jini.org/wiki/Category:Getting_Started
Community Web Site:  http://jini.org
jini-users Archive:  http://archives.java.sun.com/archives/jini-users.html
Unsubscribing:       email "signoff JINI-USERS"  to [email protected]