This is the mail archive of the
mauve-discuss@sources.redhat.com
mailing list for the Mauve project.
Re: Problems with port 30000 in .net.ServerSocketTest test_params testcase
- From: Brian Jones <cbj at gnu dot org>
- To: "Steve Murry" <Steve dot Murry at sas dot com>
- Cc: <mauve-discuss at sources dot redhat dot com>
- Date: 31 Jan 2003 23:00:56 -0500
- Subject: Re: Problems with port 30000 in .net.ServerSocketTest test_params testcase
- References: <F2E670D5036BE14E89473A3FAEDE6ACE400C41@merc18.na.sas.com>
"Steve Murry" <Steve.Murry@sas.com> writes:
> Many of the .net testcases use hard-coded port numbers when testing
> the Socket and ServerSocket classes. Most of these port numbers
> have not caused us any problems, with the exception of 30000. On
> Compaq systems this is the port of an AFS management server, and
> therefore any use of port 30000 causes the testcase to fail. The
> first test_params() testcase in .net.ServerSocketTest creates a
> ServerSocket bound to port 30000, which is then used to test various
> ServerSocket methods. Would it be possible to alter this test so
> that it either uses a different hard-coded port or specifies port 0,
> which will allow the system to assign it a free port? With the
> latter approach, the getLocalPort method will retrieve the port
> number that was chosen.
>
>
> Thanks,
> Steve Murry
> SAS Institute
>
Steve, we can change the port but this brings up the point it would be
nice to have a way of specifying things like this outside of the
testcase so it can be changed when needed.
--
Brian Jones <cbj@gnu.org>