Submit Hint Search The Forums LinksStatsPollsHeadlinesRSS
14,000 hints and counting!

10.3: Use IPv6 by default in Safari Web Browsers

I have been frustrated by Safari using IPv4 by default for websites that have both IPv4 and IPv6 address (most other browsers and IPv6 enabled applications will attempt to use IPv6 before IPv4 if you and the remote server have v6 addresses). Then I noticed that Safari was attempting to connect to servers with an IPv6 address first when accessing secure sites using https: (SSL).

So I turned on the Safari Debug menu. If you go to Debug -> Supported Protocols and turn off http: (Simple Loader), then http: behaves the same way as https: using the CFNetwork loader (correctly).

Sadly this option is not preserved between launches of Safari, and I cannot see a way of setting it in the preferences .plist. Perhaps it can be scripted ?

    •    
  • Currently 2.25 / 5
  You rated: 4 / 5 (8 votes cast)
 
[38,500 views]  

10.3: Use IPv6 by default in Safari | 15 comments | Create New Account
Click here to return to the '10.3: Use IPv6 by default in Safari' hint
The following comments are owned by whoever posted them. This site is not responsible for what they say.
10.3: Use IPv6 by default in Safari
Authored by: sjonke on Jan 22, '04 11:13:26AM

So, why would I care if it was using IPv6 versus IPv4? What does IPv6 do for me? IPv6 is a big mystery to me. If a web site has an IPv4 address why would using IPv6 for it be better?

---
--- What?



[ Reply to This | # ]
10.3: Use IPv6 by default in Safari
Authored by: Crimson Napkin on Jan 22, '04 03:42:11PM

The primary difference between the two is that one, namely IPv6, is 2 numbers MORE than the other, namely IPv4.
The other significant difference is that it sounds quite cool to throw the word around (kind of like... hold on a minute, let me hop over to slashdot...OK, here we go "real-time enterprise", "beowulf cluster", "Jakarta Struts", and "WSDD based dynamic deployment"... damn, I think I grew a chest hair just typing those in.)



[ Reply to This | # ]
10.3: Use IPv6 by default in Safari
Authored by: stephen007 on Jan 22, '04 09:25:49PM

>damn, I think I grew a chest hair just typing those in

And I just busted a gut reading your response :-) LOL.

Stephen



[ Reply to This | # ]
Red Rag
Authored by: NeutronMonk on Jan 23, '04 05:09:39AM

Score:4, Funny



[ Reply to This | # ]
10.3: Use IPv6 by default in Safari
Authored by: raider on Jan 23, '04 01:19:17PM
While your comment was 99% on the mark, and funny - I would point out that Jakarta Struts is widely used, and actually is a real tangible thing that has reasons for using and all of that real world benefit stuff. Unlike IPv6.

Oh, and if Chest hair is a sign of manly-ness, than I AM A GOD!

[ Reply to This | # ]
10.3: Use IPv6 by default in Safari
Authored by: the1truestripes on Jan 22, '04 04:34:16PM

If you have set your own system up to use IPv6 then clearly you would care, and if the other system had gone to the trouble to set up IPv6 they would care.

If you don't care, then well, that's fine.

IPv6 does do some nice things, for example encryption is built into it, if it had be deployed rather then IPv4 nobody would have needed to invent https, or ssh the old easy protocols would have been fine. That is important because it means people don't have to reinvent (normally poorly!) security for each and every network protocol. As another example because it has a very large address space ISPs that provide IPv6 don't charge extra for globally reachable static addresses (i.e. you can run a server, and things like iChat AV work easier, so do P2P type programs). As another, about half of what Rendavus gives you was inspired by IPv6.

On the other hand nothing much about IPv6 is earth shaking. A lot of what was "only in IPv6" got pushed back into IPv4. A lot of it can be worked around. It would be nice if we all switched to IPv6, but the world won't end if IPv4 hangs around forever (which is good, as it will!), or even if IPv4 remains dominant forever - which may happen.

[ Reply to This | # ]

10.3: Use IPv6 by default in Safari
Authored by: u502sou on Jul 01, '07 03:29:03AM

[why use ipv6 when ipv4 is available, too?]

In an ideal world, you'd have a point. However, for me the question is the other way round: why use ipv4 when ipv6 is available, too? My ipv6 provider is cheaper than my ipv4 provider, so I want to use ipv6 as much as possisble.

Also, there are operational benefits - I don't fill the IP-NAT gateway's translation table when I use the directly-routed ipv6 - my ipv6 provider gives me a network, so the Mac has its own global address.



[ Reply to This | # ]
Turn off IPv6?
Authored by: pdxguy on Jan 22, '04 09:29:13PM

I've not yet upgraded to 10.3 - still running the ultra-stable 10.2.6, but does anyone know how to turn IPv6 off? I don't run it and don't need; IPv4 works just fine thank you.

//brian

---
Nothing will ever be attempted, if all possible objections must first be overcome.
-- Samuel Johnson



[ Reply to This | # ]
Turn off IPv6?
Authored by: foobar104 on Jan 22, '04 10:53:31PM

Close your eyes and say "IPv6 is hereby off" three times.

This procedure will have exactly the same effect on your computer as any other attempt to "turn off" IPv6.



[ Reply to This | # ]
Re: IPv4 "just fine"
Authored by: bmerlin on Jan 23, '04 01:49:44AM

It's people like you who stymie innovation in the computer realm. This "good enough" mentality is why people write crap HTML--IE can render it, so it's "good enough".

While the benefits of IPv6 may never overtly present themselves to you, its behind-the-scenes, technical improvements will greatly benefit the internet as a whole. The same goes for (X)HTML and numerous other "new" (in quotes because most aren't new, they are just ignored due to mentalities like your own) and evolving standards. Unfortunately, through apathetic, "I don't need that" attitudes, people like yourself keep them from being widely implemented--thus reducing their impact.



[ Reply to This | # ]
Re: IPv4 "just fine"
Authored by: pdxguy on Jan 23, '04 06:48:47AM
bmerlin: I think your remarks are uncalled for. I posted a comment asking how to turn-off IPv6. I would expect comments along the lines of "you can't turn it off", or "your system will not boot if you do", or "here's how...", or even "you shouldn't do it because..". But your jump to the broad conclusion that I'm stymying innovation because I ask this question is not reasonable or productive IMHO.

You say the technical improvements (of IPv6) will greatly benefit the internet, and by abstraction myself, but you fail to list them. Tell me what they are and how I will benefit from using them; I may decide to use IPv6 but you fail to provide me any convincing reason to do so.

I think you would do best to understand me and what I do much, much more before jumping to any conclusions, let alone ones about my mentality. Just the tip of the iceberg so to speak, is that I compile and install from source the latest versions of Apache, php, mySQL, Cyrus-SASL, Mozilla, Sendmail, and many others including Darwin as well. And when I code webpages I do use XHTML and CSS 2 as well. I mention this information, although I believe it to be irrelevant to my question, to show again that your ad hominem attack is without merit.

If you don't like what I said, please present a good argument about why what I said is wrong and/or suggest something better. To make unsubstantiated and denigrating remarks about me, based on one question, is again uncalled for.

//brian

---
Nothing will ever be attempted, if all possible objections must first be overcome.
-- Samuel Johnson

[ Reply to This | # ]

Re: IPv4 "just fine"
Authored by: jecwobble on Jan 23, '04 09:49:23AM

At the risk of encouraging more 'flamage,' - Well put. For the most part, I frequent this website because of it's educational and useful collective knowledge. I visit no blogs and the like, because frankly, I don't need to know a stranger's heated opinions.

And thank you, too, for throwing in a new (to me) adjective that I had to crack open a dictionary for. See, educational.



[ Reply to This | # ]
Re: IPv4 "just fine"
Authored by: bobw on Jan 23, '04 11:14:46AM

A little Googling answers your questions;

http://www.ipv6.org/

http://en.wikipedia.org/wiki/IPv4

http://www.potaroo.net/ispcolumn/2003-07-v4-address-lifetime/ale.html

http://www.arin.net/policy/ipv4.html



[ Reply to This | # ]
Turn off IPv6?
Authored by: krykertano on Jan 23, '04 07:21:04AM
Why would you want to turn off IPv6? There's virtually no overhead involved--it certainly doesn't bog down your computer, anyway. If anything, I'm no network engineer, but I've heard the routing algorithms are supposed to be more efficient under IPv6 than IPv4--though I don't think it'll affect anything on the computer end.

[ Reply to This | # ]
Turn off IPv6?
Authored by: the1truestripes on Jan 23, '04 12:01:24PM
I've heard the routing algorithms are supposed to be more efficient under IPv6 than IPv4

IPv6 might be more efficient for two reasons, one is there is a lot more address space so each ISP gets very huge blocks of space, enough that you can probably just say "if the top X bits look like this then the packet goes to ISP Y" (one bit masking, and one table lookup, in a "not too big table" even). Two is IPv6 in theory makes it easyish to change IP addresses, so IP address space in v6 is not portable (because the costs of being "not portable" have been significantly reduced).

With IPv4 there is no easy test you can do as an ISP to see who an IP address belongs to. You end up with giant tables and lots of work doing lookups in them (unless someone has a 30 bit CAM...). In theory this will not be needed in IPv6.

Two things gum up the works. Multihomed hosts which are computers with internet service from more then one ISP. Many of these hosts want to use one IP address for all the connections so that things connecting to them (say web browsers) can get the best route to the address rather then randomly picking an address and maybe sending packets to another ISP when they actually share ISPs with the web server! Or picking an address at random and getting a timeout because that ISP has a hardware failure (multi homed hosts normally have more then one ISP both for more bandwidth and for increased reliability). The other is people may force the "not portable" issue because while it is easier to change IPv6 addresses then IPv4 it is still easier to not change the address and insist that is someone else's problem.

Only time will tell if IPv6 can stay more easily routable. Of corse if it won't it is more trouble then IPv4 to route. There isn't really a better way to have more addresses though. Unless maybe you buy into PIP (one of the other proposals for "IPng", which is what IPv6 was called while it was still in the design stage as opposed to the long slow deployment stage)

[ Reply to This | # ]