How about IPv6?

[Speedy Lan]

[Lightning had shown the list of fixes that he has prepared for the upcoming version 0.4.4 in this post here: viewtopic.php?f=40&t=17456&p=96100&sid=c39639c747b81bfa13585da4a58bc202#p96095 ]

What a list!
I don't want to hyjack this topic, but I how did you find out on all these things that they needed fixing?
You have your own to-do-list, maybe? Because I cannot think of this many bugs been reported.

One more thing I would like to ask is about IPv6.
This is starting to become an urgent thing and I would like to know if freesco can handle IPv6.
I guess it would need a new firewall, or at least a separate set of rules.


Perhaps I should post this as a new topic.
[which I did]

[Lightning]

how did you find out on all these things that they needed fixing?

Well I monitor this forum VERY closely for any type of issue and I perform my own obscure testing of the system in every conceivable mode. But most of the time as I am working on a new feature I just happen across something that doesn't work right.

As for IPv6, this is a subject that has been on my mind for some time. To which there is only one answer. As long as FREESCO is using a 2.0.x kernel and lib5 IPv6 can not be supported and FREESCO will no longer be on a floppy if any other kernel and library is used.

So the issue really is that the above statement is not really an answer to your question. To which it is not intended as an answer, but rather a statement of fact. The problem really is with the amount of time it will take to basically redesign and recompile the entire system. The current system is a compilation of around ten years of programming almost exclusively by myself with the exception of the 02x series and I am not entirely certain that I want to take on such a task. There has been great contributions by justdave in creating special binaries and modifying the kernel from time to time along with several others in more limited ways, but that is not really the issue in this discussion.

However with the above statements made, justdave and I have discussed making the next step with FREESCO and moving it to other boot media with a newer kernel and lib6 and justdave stated that he could devote some time in the endeavor. Which this is one of the reason that 0.4.4 will be released with a CD iso along with a floppy image. As things stand at this very moment, I have not come to any type of actual decision on this subject and probably won't until after 0.4.4 is released.

[Speedy Lan]

Thanks for your explanation.
I expected this somehow and a new kernel will set off so many other things, e.g. all those things that could never fit on a floppy or could not run with the current kernel.
I wish I had the time and the energy to delve into this myself, but as I have seen over the past years, that has not been the case and it does not seem likely that it will change too.
e.g. here: viewtopic.php?f=15&t=17367
However, perhaps the new kernel idea will attract more people that can help in some way.
Or there are already similar things made for those newer kernels that can be of use.

And IF a new version with a higher kernel would be made, what would happen with the current version?
Keep supporting 2 versions? That is a lot of work.

But let's see what the future will bring.

Greetings,

Speedy Lan

[Lightning]

Keep supporting 2 versions? That is a lot of work.

Hehe well we currently support both the 03x and 04x series and I don't think that would change even if there was a new version. I am however doubtful that there are very many people who actually understand the amount of time that has been put into FREESCO or can even conceive of the amount of time to create a new version with a lib6 library and a 2.6.x kernel along with a completely new firewall.

[phillipsjk256]

I think IPv6 is different enough, you would have to re-work the firewall anyway.

One idea I had, but did not really want to discuss it without some initial testing, is that it may be possible to use GNU/Hurd instead of a linux kernel. It is supposed to be highly modular. It may even be possible to emulate IFWADM (is that the correct term?) such that you don't have to change any rules for IPv4 functionality.

I see no inherent reason a slightly different protocol should need so much extra overhead.

[Lightning]

IFWADM (is that the correct term?)

IPFWADM is the correct term.
As for a hurd kernel, I have no idea. I do however know that IPv6 was just a concept in the Linux kernel when the 2.0.x kernel was stopped and it wasn't until the 2.2.x kernel that it became an optional component.

As for changes to the system, I am one man with a full time job and family. Don't get me wrong as I do get help from several other people on specific applications. But past that it is just myself with regards to changes to the OS and Howler with regards to FREESCOsoft and this forum. So I really recommend that you look into a hurd kernel or even a newer Linux kernel and or it's capabilities as I have looked at even a change to just the next Linux kernel version in the amount of work involved. If you decide it is worth while and you pursue it then you could easily become the proud owner of an OS in the same manner to which I did when the original author of FREESCO retired as I have considered.