Date: Mon, 22 Apr 2002 16:18:19 +0200 From: "Jeroen Massar" <jeroen@unfix.org> To: "'Pim van Pelt'" <pim@ipng.nl> Cc: <itojun@iijlab.net>, "'Robert'" <robert@chalmers.com.au>, "'6bone'" <6bone@ISI.EDU>, "'ipv6users'" <users@ipv6.org>, "'freebsd-stable'" <freebsd-stable@freebsd.org> Subject: RE: A DNS question re 6to6/IPv6 host IN A records. Message-ID: <003d01c1ea08$8e66e330$534510ac@cyan> In-Reply-To: <20020422105456.GK7029@bfib.colo.bit.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
Pim van Pelt [mailto:pim@ipng.nl] wrote: > Hi, > > I agree with Pekka mostly. Having the same IN A/AAAA RRs for the > hostnames in your zonefile can make for awkward situations. > One example might be the NL-BIT6 deployment. We have a C3640 with a > 10 mbps port acting as vlan router for IPv6. It then pushes the traffic > to the AMS-IX. If I am sitting at any IPv6 peer-site, and > ssh/ftp/telnet to my machine at the colo, and it were to have both > protocols reachable via the same name, then I would connect using IPv6 > because this is preferred. ssh -4 purgatory.unfix.org or the 'ssh purgatory.ipv4.unfix.org' trick but I don't have that one in the outside dns apparently ;) > However, I like my pron to transfer fast, so the gigabit IPv4 connection > (yes I have a 1000SX board in my colo-box :) is preferrable over the > turtle-speed IPv6 connection. IMHO you should upgrade that IPv6 connect. Fortunatly 10mbit is still 2mbit more than my inet-uplink is capable of And: --- purgatory.unfix.org ping statistics --- 5 packets transmitted, 5 received, 0% loss, time 4035ms rtt min/avg/max/mdev = 19.342/21.498/24.997/2.005 ms vs: --- purgatory.unfix.org ping6 statistics --- 5 packets transmitted, 5 packets received, 0% packet loss round-trip min/avg/max = 19.9/21.9/27.3 ms Doesn't differ much for me, latency wise. Besides that I don't have that heavy pr0n traffic desirement ;) Btw.. did you see that nice 10/100/1000mbit port on those cute Powerbook G4's ? And they can do IPv6, now I'll only have to find some financial aid and that gbit uplink <grin> > The other point one might make is that IPv6 is often less well > maintained than the IPv4 network. Some tunnel might go down, zebra might > crash (or even IOS) and the connection will be left unattended by many > administrators. This is why I normally make some distinction either by > hostname 'hog.colo.bit.nl IN A' vs 'hog.colo.ipv6.bit.nl IN AAAA' or by > domain name 'hog.colo.bit.nl IN A' vs 'hog.ipng.nl IN AAAA'. Absolutely, but I personally know who to kick when you bring down my IPv6 uplink <evil grin> Also IPng.nl fortunatly has only been down due to scheduled maintainances and not because it 'failed' suddenly. And you probably also remember how the couple of times we saved a box because the IPv4 routing was peeped and we still could reach it over IPv6; Long live native IPv6. This whole story ofcourse all depends on the fact how far one is in the transition process and if you take IPv6 for granted as a 'must-work' service level just like IPv4. Personal taste also comes in mind ofcourse ;) Greets, Jeroen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?003d01c1ea08$8e66e330$534510ac>