sks-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Sks-devel] sks-network


From: Scott Grayban
Subject: Re: [Sks-devel] sks-network
Date: Sat, 21 May 2011 11:22:10 -0700
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-GB; rv:1.8.1.23) Gecko/20090812 Lightning/0.9.4-Inverse Thunderbird/2.0.0.23 Mnenhy/0.7.5.0

I won't peer with people that do not have a static IP. The whole idea
about peering is to provide stability throughout the network pool. If
this isn't part of the peering rules it should be.

And there is no such thing as semi-static IP's.... it's static or
dynamic, if we are going to explain dns let's be correct about it :)

A decent pipe -- we all know that anything less then a 1mb pipe is just
going to cause issues. First, down/up are never the same speed, down is
usually faster then up and most ISP's don't offer a up faster then
256/512 unless you have your own T1 or faster line. Some newer DSLv2
lines offer a better up/down ratio but that is few and far between
especially in North America.

If people are wanting to peer using a home desktop and they are playing
games/movies that will impact the purpose of peering. Bandwidth will be
sucked dry in no time backing up everything.

If there aren't standards set now what will happen in a couple years ?
Every IT organization has some sort of peering rules and if SKS peering
doesn't them we can't even prove we have a stable pool of servers.
Imagine root dns servers done this way.

I have no intentions of stirring up a bees nest but maybe SKS should
have some standards to enforce. I personally want to be sure I am
peering with stable servers and not some desktop a person uses to play
games on.

BTW my dns does have the LOC record.
>> borgnet.us.             38400   IN      LOC     47 40 5.370 N 117 26
20.800 W 579.00m 1m 10000m 10m


Regards,
Scott Grayban

 /"\
 \ /     ASCII RIBBON
  X        FIGHT BREAST CANCER
 / \


John Clizbe said the following on 05/21/2011 10:30 AM:
> Scott Grayban wrote:
> > You need to have a static IP, no dialup and a decent net pipe. Having
> > IPv6 is a plus as well.
>
> These may be your strict requirements, but I've not ever seen them
> expressed by
> the community and I've been here a while.
>
> The only people for whom static IPs for peers are important are those
> running
> strict firewalls and only allowing traffic from designated peers. 
> About the
> only thing I can think of that is a "strict" requirement is that a
> machine be
> able to be reliably reached through the DNS.  Cable and DSL
> connections are
> subject to changing address. From my experience, the frequency varies
> from when
> the operator does network maintenance in the case of T-W to every time
> the modem
> burps in the case of AT&T.
>
> What's a decent net pipe? I'll agree it not dialup. But what about
> ISDN? One of
> my servers does quite adequately on a 3M/512k DSL connection (It's the
> max AT&T
> has in my area without subscribing to U-Verse.)
>
> We ask folks to use a recent dump, but it's not carved in stone,
> neither is
> knowing where they got it.
>
> There has never been a set format requirement for the email seeking peers.
> There's been a suggestion or two, but I don't recall a vote.
>
> A recommendation I'd like to propose is that each host also have a LOC
> record in
> DNS. This would facilitate physically mapping peers. See
> http://hewgill.com/tools/dnsloc for an example application.
> keyserver.gingerbear.net may be used as input.
>

_______________________________________________
Sks-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/sks-devel




reply via email to

[Prev in Thread] Current Thread [Next in Thread]