New on LowEndTalk? Please Register and read our Community Rules.
Looking for LEB with /64 ipv6
I am looking for an LEB provider that offers a VPS with a /64 block of native ipv6.
I don't need any other resources, 128 MB RAM and 2 GB disk is fine. I'd largely prefer monthly payments over yearly ones.
The only provider I know that offered /64 with a VPS was StyleXNetworks, I haven't seen them around here in some time though. You could always route an HE /64 Tunnel to your VPS.
The reason I don't think many providers offer it is because it's not supported by SolusVM.
I think BuyVM is working on offering this with Stallion2 but I'm not completely sure of that, but I believe they're still using tunneled IPv6(for the moment anyhow).
We offer also /64 with all KVM including cloud (shared network, not isolated ones).
Edis /48 in Austria for 5$ on time fee
Evorack /64 in UK - but not as good as they were, on autopilot now
AllSimple /64 in uk
XenVZ/OpenITC /64 in UK (Maidenhead only, Newcastle not yet)
Hitme PL /64 with their Xen nodes
Prometeus /64 with their KVM nodes (in case there's still available anything)
Since the cloud launch, many customers migrated so we have stock not only on KVM, but on OVZ too since the Dallas opening. Even xen benefited.
So, we can give almost everything, even if in places there is no stock listed, we have for LET/customers/friends.
My question may sound stupid, I am not used to ipv6 and not trolling,
What can someone do on a vps that requires a ipv6 /64 block?
Earlier discussion about the same: http://lowendtalk.com/discussion/4552/leb-providers-with-an-ipv6-64-or-more/p1
Care to explain?
It's generally the reason why someone wants a whole /64 on their VPS. I don't see any other reason.
@awson that's not really explanation just statement which tells nothing.
/64 is the minimum for auto assignment/SLAAC autoconfigured VPN on IPv6. https://en.wikipedia.org/wiki/SLAAC#Stateless_address_autoconfiguration_.28SLAAC.29
the EUI-64 mechanism for stateless autoconfiguration of IPv6 addresses requires a subnet to have 64 bits. This means most, if not all subnets (except point-to-point links), will have a size of /64 in the future.
There could be also some other trivial but pretty realistic reasons like preventing k-lines at IRC networks because "too many connections globally" (all users from same /64 are usually considered as one user). In case of abuse no one block individual /128s but whole /64 subnet because again - /64 is considered as one end user.
@awson if you have a /64 you can choose the lower bits from your address. So you can do something like xxxx:xxxx:xxxx:xxxx:dead:beaf:1234:5678 or other vanity addresses if you want.
openITC you get 2 * /64 on KVM plan
Nope. Even providers begin to "see the light" now, e.g. the BuyVM's change from smaller subnet or individual addresses to /64s (and they were the quickest to ridicule the idea in the past, seriously thinking someone was going to assign all /64 worth of IPs to an interface).
The reason is simple: the whole world is going to thread "one /64 = one end user".
Because that's what home ISPs with IPv6 provide to each end user. (why? wikipedia://SLAAC).
All bans and ratelimits are going to work per /64 at the minimum. Spam lists, forum flood control, search engine ratelimits, all of them. If a provider keeps all their VPS customers within one /64 (perhaps doling out /112s from it or whatever), they're going to have huge trouble if e.g. one of them starts spamming/flooding or gets banned. RBLs, DNSBLs, bans, there goes the whole /64, no one is going to give you special treatment because you give out smaller subnets. And then have fun explaining to other customers why they are banned at some service or ended up in a spam list.
Much easier to give /64 to each user and limit any bad impact they can cause to just their /64, not affecting other customers in their own /64s.
You got mulitple things going on here
1) Yes, we use HE for now but it's a BGP tunnel with our own allocations. What this means is that whenever fiberhub does get off their butts about V6 there will be no IP swapping
2) Yes, we also provide /64's to both OpenVZ & KVM's. There is a migration script:
We don't bind off a full /64 to you but you can then assign however many (as well as specifically which) IPV6 addresses you want:
Evorack have /64
Prometeus have /56
BuyVm have /64.
Don't go with stylexnetworks. Last time I try it, I cant use the whole /64. I pick an anddress randomly, some can not be use
if you still require one, kindly PM me
, edited thx.
Hetzner's vServers (KVM) include a /64 w/rDNS in Robot, and your contact details registered in RIPE WHOIS, but are above LEB pricing.
Apparently iwstack has /64 allocations.
Yes, for each IPv4, however, that is only for the shared network, isolated networks lack IPv6 altogether. With that exception, all our KVMs have /64 for each IPv4. iwstack or not.
I can provide a full /48, on own RIPE handles and reverse DNS delegated, routed to an IPv6 on the server, subdelegated on RIPE for own assignments (SWIP) on request as well
Austria only, any KVM/Dedicated/Colo qualifies, Network is routed to an IPv6 out of our default /112 assignment, Customer data (Unique customer ID, no personal data unless RIPE handles are provided by customer) is published in RIPE DB.
Atrato/ReTn/HE (only HE customers and EU exchanges)/VIX/AMSIX (TBD Q4'13) blend
One time fee 5EUR/7$ due to the manual work involved.