They are willing, however, since there is no demand from the customers, it unlikely happen any time soon. As I said, the splitting the network is no go. And solution only one - everyone must support the same protocol, either dual stack or at least IPv4.
IPv6-only nodes currently a dead end, until all customers will have either a dual stack every time or at least IPv6. Most of customers uses IPv4 right now. The IPv6 usage is negligible.
We have had this in the past, the customers very often have had issues with a DNS resolution especially using a native Storj protocol and the high uploads (sometimes even a tiny load could kill their router), so we switched to resolve nodes hostnames to IPv4 satellite side and provides IP of the nodes to the customers. This change significantly increased the speed both for uploads and downloads. So no, hostnames will not be provided to customers anymore, we are not willing to sacrifice the speed.
Unfortunately it’s the only way, they must support either dual stack (and who not - will not be accepted to the network) or at least IPv4 (the current case).
They simple blocks IPv6 traffic transit. So we were forced to disable the support of IPv6 in these countries on our Gateway instances until their conflicts will be resolved, because customers who are able to use IPv6 have had connection timeouts via IPv6 and worked great via IPv4.
I see there is will, when they put a requirement for everybody to have, in a all of nothing thing. Splitting the network is what they think today because they seems to have preferred to not take the time to find a better solution. Saying IPv6 usage is negligible is a typical thing from those who understand little about IPv6. Maybe this reinforces the reasons it has not been something Storj has ever cared about.
Put this on “customer asking” is typically the most ‘noble’ way to not do, as if customers would understand and ask for it broadly.
In fact one of the main issues about IPv6 adoption from people is to keep beleiving that IPv6 is something optional or cosmetic and that can be added later and products in general don’t need it nowadays, specially products that work in the Internet mostly. That says a lot about Storj team it seems !
There is no such think like block IPv6 traffic. I don’t know when they go it. Maybe they tried to use in a certain way and didn’t work as expected and assumed someone was blocking it. It just doesn’t make sense at all really. Maybe was something related to wrongly block of ICMPv6 or something and the lack of proper understanding and interest about IPv6 make it impossible for a minimal troubleshooting, so it creates this thing about IPv6 traffic block.
Don’t waste your time bro. It’s not the first time that those people try to push their IPv6 agenda. The real purpose? IPv6 is much cheaper than IPv4. Lol.
Dude. You keep repeating this mantra but provide no substance. Answer this my question:
Maybe you don’t comment on this because there are no problems and you just want IPv6 just for the heck of it.
This is engineering, you pick solution that works good enough and move on improving your core functionality. If you want to promote IPv6 adoption worldwide storj forums are not the platform for you.
I don’t want to repeat what was already explained. Network splitting is a problem. Needing to do other workarounds is a problem.
They obviously not. DNAT is well known and tested solution. There are 232 ports available per IP. Ask your provider to forward one to you.
Oh, some providers are unwilling? Well, that’s not a technical problem. Same providers may be unwilling to unblock incoming traffic on their IPv6 firewall
I haven’t heard “NAT as security feature” debate for a very long time, nobody thinks that.
The real reason is stated in the beginning of this comment. IPv4 is good enough, storj is not in the business of promoting IPv6, when they have to move to it — they will. Today the reasons are not compelling enough to spend any resources. As simple as that.
Correct. Time, resources, efforts; when existing solution works just fine.
I’m still waiting for the explanation of what’s so wrong with IPv4 and impeding doom of it going away.
What do you find so amusing? Is DNAT not well known or is it not deployed to vast majority of network devices worldwide for the past decades, to the point that even the most cruddy low cost devices seem to have no problem with the implementation?
So, educate me. I’m all ears. Tell me, why should storj drop everything and mess with IPv6 right now?
Specifically. In the list. A is the problem, IPv6 will address it, and increase profits by B%.
Show the cost benefits analisys. What is wrong with IPv4? I don’t know if you noticed but storj works just fine today. And yet you have been smirking and throwing general “IPv4 is dead v6 is the future” sentiments so far. No substance. Point me to your substantiated comment if I missed one.
Plus I wonder how would they implement the /24 rule for v4 to v6. Access to cheap IPs + a large abundance of them would allow them to ingest tons of ingress.
IPv6 means every device has its own public IP. You don’t need to allow any traffic on your router, traffic is already allowed through.
That being said: ISPs are reluctant to use IPv6 for the simple reason that they don’t want every device on the public internet. That comes with its own pros (you get traffic to everything) and its own cons (an infected device being a pivot host).
Right. Once it happens – 99% users with v6 – then switching would be effortless.
I do believe that everything is negotiable. On the other hand, most ISP ToS for residential connections prohibit hosting services in the first place… this is different can of worms.
Any decent provider that cares about residential customers (that are not [supposed to be] hosting services, remember?). Most providers today already block incoming connections on SMTP, SMB, and a few other sensitive ports.
This is wrong, this is what @IsThisOn alluded to. NAT is not a security feature. Your home router block all incoming connections by default, on both ipv4 and ipv6. You need to selectively allow new connections on a specific protocols on a specific networks, both on ipv6 and ipv4, except with ipv4 you also need to enable DNAT, with ipv6 you don’t need to.
Again, this is irrelevant and is not a reason. IN fact, ipv6 security and privacy features are a very welcome addition. Most ISPs want that for their customers, and their customers should want that for themselves.
This is irrelevant in the context of ipv6/ipv4 discussion. You still need a firewall in either case.
No dude, please don’t say that. In is a standard in the broadband services provider industry that any IPv6 going to your home network gets protected by the router by default and you have to allow only what you want towards devices on your LAN. It is not because it is a Public IP that it is unprotected. It is well protected with the same firewall techniques used for IPv4 and easy to allow through.
@ACarneiro please just don’t participate on it if you are bothered instead of trying to censor people to talk things that you don’t agree or like
Storj guys have their points and justifications and people from community have their view and it is not because Storj people said it is not possible hanging on the easy justification that it is not in fact. It is just their point of view based on their current technical knowledge about this issue
There is always a solution for a problem that wants to be solved and there is always an excuse and ‘noble’ justification for the ones that don’t want to.
)