I will work on a patch to use libssh instead of starting a separate
ssh process. This might also be able to eliminate a socket in the "-via"
case, but I'll see. :)
I only have a Unix environment - is the "via"
mode supported on Windows or other builds?
-Eric
On Mon, 17 Jan 2011
14:23:41 -0700, Eric Stadtherr wrote:
> Catching the failure to bind
would certainly help (and should probably be done no matter how the
overall issue is addressed), but the result of fixing just that part of
the thread would be that the second viewer either aborts due to the
error (leaving the user to run the exact same command again with
different results) or the code has to loop back around and find another
port. Having confidence in the availability of the local port going in
would seem to be a better, albeit more involved, fix...
>
> -Eric
>
> On Mon, 17 Jan 2011 15:03:27 -0500, Robert Goley wrote:
>
>> Sounds
like the real fix is to pay attention to the failure to bind with the
local port. I know that it displays a warning message. I do not know if
it returns an error code that is easy to capture. Instead of using an
external SSH, a solution could be to use the libssh library like is used
by the KDE project. That is one way we could ensure we caught the result
of the bind port call...
>> On Jan 17, 2011 1:15 PM, "Eric Stadtherr"
wrote:
Links:
------
[1] mailto:estadth...@gmail.com
------------------------------------------------------------------------------
The modern datacenter depends on network connectivity to access resources
and provide services. The best practices for maximizing a physical server's
connectivity to a physical network are well understood - see how these
rules translate into the virtual world?
http://p.sf.net/sfu/oracle-sfdevnlfb
_______________________________________________
Tigervnc-devel mailing list
Tigervnc-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tigervnc-devel
