[Xymon] IPv6 (again)

[Henrik Størner]

On 22-12-2011 10:49, Steven Carr wrote:
> I notice from the archives this was last brought up in August
> [http://lists.xymon.com/pipermail/xymon/2011-August/032217.html] has
> there been any move forward with IPv6 support in the xymonnet tool? If
> not are there any timescales/milestones of when this is likely to be
> implemented?

Funny you should ask ...

> It is starting to become more prevalent that we monitor our IPv6
> infrastructure and I'd rather not have to move lock, stock and barrel to
> a different monitoring solution.

As I wrote when releasing 4.3.6 and 4.3.7, "Work is now beginning on the 
4.4 series, with the focus currently being on IPv6 support."

This is still the main focus for 4.4, and I have been working the past 
week with constructing a generic communications library that can handle 
both IPv4 and IPv6. This is almost complete - i.e., it works, but I need 
to work out a couple of minor bugs - and I'll start re-doing the network 
I/O pieces of Xymon to use this library instead of each tool carrying 
it's own bunch of network code.

So there is actual code in the works, and it will definitely be the main 
feature for 4.4.

A pleasant side-effect of this is that it turned out to be rather easy 
to implement SSL support as part of this library, in a way that is 
nearly transparent to the application layer code. That means Xymon 4.4 
will also have full support for clients talking to the Xymon server 
using SSL; I know this is an issue in some environments.

As for when this will be ready ... difficult to say. If all goes well, I 
expect to have something up and running around February. This is clearly 
some major surgery on critical parts of Xymon, so it will need some 
thorough testing - I'll have to run it for some time to make sure the 
good ol' IPv4 stuff isn't broken, and then we can start testing the IPv6 
features.

One problem for me is that I don't have anything running IPv6 at work. I 
have it at home and can test the basic bits, but I am looking for people 
who would be willing to test this on a larger scale than one or two 
hosts. If you would like to help out with that, drop me a note and I'll 
be happy to send you some early implementations to try out for me.


Regards,
Henrik