[App_rpt-users] IPV6 and upgrades --was: toy throwing time?

[David McGough]

Hi Everyone,

Every time these discussions about upgrading software come up, I 
re-evaluate the possible gains of a major upgrade.

I'll remind everyone of a major hurdle which can't be side-stepped: The
AllStar project would continue to be a FORK of Asterisk---not code that
becomes part of the Digium distro, like it once was.  You ask why?? Well,
that's due to Digium's draconian contributor license, which doesn't
guarantee the code will remain open source and all developer's must
surrender their code ownership to Digium, to do with as they please.

So, I guess my question becomes: WHY? ...what glaring issues will a major
upgrade resolve?  What issues need to be fixed? What -TRULY- known
problems are present? Why not just FIX those problems, including
back-porting code, if Asterisk needs a patch.

As the overall Asterisk project has marched forward, gazillions of
features have been added...And gazillions of NEW bugs.  Are those features
needed for AllStar? For the most part, I would argue probably not.
AllStar's primary duty isn't as a phone PBX...

One aspect many AllStar users are probably unaware of is that there were
major command syntax changes as of Asterisk 1.6.x...One thing this means
to the user is that the dialplan syntax in the 1.4.x extensions.conf file
isn't compatible with syntax in 1.6.x and later...So, now both styles of
dialplans must be supported, the old 1.4.x variety and the new 1.6.x
style. Yet more confusion.

As for IPV6, yes, native IPV6 compatibility would be nice, but simply NAT
mapping IPV6 address space to IPV4 addresses will and does work too. But,
this is only part of the problem, it doesn't fix other major issues: All
the AllStar registration and node database software must be upgraded to
IPV6 as well! THEN, suddenly, none of the IPV6 users can talk to IPV4
users---and, yes, there will be IPV4 address space in use by ISP's for
years to come. The IPV4 vs. IPV6 migration problem gets more and more
complex to fix, the deeper you dig.

....As another long term solution, the IPV6 code could be back-ported to
the 1.4.x chan_iax2, chan_sip.  Wouldn't this be easier than a full 
upgrade??

Anyhow, as a recap, what is broken that needs to be fixed???? IMO, fixing
core issues in the CURRENT codebase is much better use of developer's
time, since developers are so few and far between.


Just my $0.02

73, David KB4FXC




On Tue, 13 Sep 2016, Stacy wrote:

> That is true, 1.8 does support IPv6.  Some places are already starting 
> to offer you only an IPv6 address with a connection. Mostly mobile hot 
> spots.
> 
> If we don't want Allstar to become something of the past, we should move 
> up to at least 1.8 so that it supports IPv6.
> 
> This code really needs someone else's eyes to go over it, it works but I 
> know there are things that need to be fixed.
> 
> 
> -Stacy
> 
> KG7QIN
> 
>