New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
It's not about many IPs in one /64, but about using DIFFERENT IPs from DIFFERENT /64s, as ONE /64 will be recognized by many services as ONE user (especially by google)
And technical: You cannot use DHCPv6/SLAAC with any network smaller than /64. So if you want to use multiple VLANs/Networks, you need to statically assign IPs.
/64 is the new /30
$10/m for a /64! What's next, $50/m to be able to create processes on the server?
You're right deadbeef. 1$/process
Coming up on LET:
"How to squeeze multiple applications in one process?"
"Learn how to safely bypass the process fee aff link here"
"OpenVZ VPS with no overselling of processes"
dear @elgs
and others as well
what i wrote wasnt correct
the link is ip6.im
I wish I understood half of your ipv-mathematical talks
$10 a month is shocking for a /64. In my opinion, the provider is gouging the customer. $10 might be acceptable as a one-time setup fee, but honestly, IPv6 should be provided at no cost. It should be treated as a competitive advantage against other providers who have not figured out IPv6 yet. If the provider gouges customers for IPv6 like that, can the provider be trusted for other support and services?
@patrick7 - Please reserve a /32 for me. Justification: I am planning to setup the ISP for the Andromeda Galaxy. The only problem is the latency sucks. I am still waiting for the echo reply to come back from my first ping packet. :-p
Let's see if $10/mo makes any sense ... there are 2^64 IPv6 /64 blocks ... that means the entire IPv6 space would be worth $2,213,609,288 or so trillions per year. Meanwhile, the gross world product lags behind at only about $78 trillions (albeit in 1990 PPP terms, so probably more in today's money).
So, no, an IPv6 /64 is not worth $10/mo according to any possible scarcity theory.
Thus, IPv6 /64 should only cost in proportion to administrative burdens, which aren't close to $10/mo either (at least not if you get space handed down from an ISP who is also providing you with colocation, dedi, or VPS service). If you're paying for the space yourself, you need to get at least /48 anyways, making each /64 within that only 1/65,536th of the cost.
You will probably need a larger block.
Because we don't want to waste IP space. We also don't want to have to allocate router resources for blocks that won't even be used.
If you have a justified reason for needing more than a /64, we will likely grant the request. However, we aren't just going to give out space with no rationale.
Wasn't the point of IPv6 that there are so many that you can't "waste" them?
Thanks for the question, but this is our policy and I think it is a reasonable one.
Initially, yes. But, since then, they've tried to so hard to waste them that the space doesn't look so gigantic anymore.
/64 to a subnet? SLAAC is useless hype. DHCP always worked fine.
Now that gives you only 2^64 "subnets" with 2^64 "endpoints" each rather than a single space with 2^128 "endpoints". An absolutely monumental example of waste, given than even the largest sites aren't going to have more than a few tens of thousands of servers. Big compute clusters don't even need/want every node on the public Internet.
And, of course, sites need to be allowed to have a potentially large number of subnets ...
Bottom line, anything can be wasted pretty easily.
Right, 2^64 is an insane number for my network. Even though one could say I should see it as a single address out of a huge 2^64 pool. If the design didn't add burden to the routing system, it at least added burden to my brain to memorize the long addresses. I don't know. What I said might be totally insane. Maybe it is just the best decision. It's just 2^64 is too many for my network.
I see most of /64 ipv6 blocks are unused still most people want ipv6 with their servers.
I do not know why they do so.
Thinking when 1 IPv6 gets blocked, the whole /64 is blocked it's absurd to give less than that.
It's understandable people wants IPv6 with their servers since IPv6 is the future. Moving forward, there might be providers providing IPv6 services. If we don't have IPv6, we will have no chance to use these services.
That absolutely makes sense.
I'm not a hundred percent sure of that information though. It was something I read somewhere and I can't seem to find it.
But nevertheless, this may be old but read this article about how it is not a waste to use /64 subnets
Oh sure, those are solid arguments for using /64s -now that's it's a matter of convention that /64s be used-, and it would be insane to try and do things differently today.
A /64 IPv6 in a /48 is like a single IPv4 in a /16. Why do you throw out 3 IPv4 with less than /16 but only /64 with ways more than /48?