|
#1
|
||||
|
||||
|
from computerwire
VeriSign Inc is testing changes to its domain name system services, which could generate tens of millions in revenue a year for itself and partners, and which would impact the way almost every internet user surfs the web. Sil |
|
#2
|
|||
|
|||
|
Sounds like it could pave the way for anti-trust action against Verisign too.
How exactly will it work? Does it mean that names which don't belong to anyone - such as doyouthinkanyonehasthoughtofthisnameyet.com - will resolve to IP addresses used to host VeriSign web sites, instead of not resolving at all? A further complication. Just because a host name doesn't resolve to an IP address, doesn't mean to say that the domain name doesn't exist. Maybe the domain has an authoritative name server, but the owner has chosen not to assign it, or any of its subdomains, to any IP addresses. Is VeriSign able to do anything about that? If they can and do, then that would definitely open a can of worms.
__________________
- |
|
#3
|
||||
|
||||
|
It seems to have gone live, trying to browse non-existant domains redirects to a search page with adverts..
http://www.versign-are-chimps.com which may or may not work depending on your dns servers.
__________________
uk's worst isp |
|
#4
|
||||
|
||||
|
doesnt appear to work for me (at the moment at least) - I really think this kind of thing is pretty similar to domain hijacking..?
Sil |
|
#5
|
||||
|
||||
|
actually - I think it did try and do it - I ended up at a url with
http://sitefinder.verisign.com/lpc?u...are-chimps.com which I presume means it tried to work? Sil |
|
#6
|
|||
|
|||
|
http://www.doyouthinkanyonehasthoughtofthisnameyet.com - yep, it works.
![]() http://www.i-cant-believe-verisign-w...ch-a-thing.com I think this is big stuff. Do you reckon verisign's search portal might nick Google's crown? http://sitefinder.verisign.com/spc?sb=search - I notice that Google is the first non-sponsored result. I think that sponsored links here completely changes the marketplace for domain registration. And it's not really fair on those who have already bought domains, too. Where exactly do verisign get their search results from? My prediction? Be surprised if you haven't seen anti-trust action within three months from now. http://www.antitrust-action-against-verisign.com Do you think they'll get the hint?
__________________
- |
|
#7
|
|||
|
|||
|
http://64.94.110.11
We didn't find: "64.94.110.11"
__________________
- |
|
#8
|
|||
|
|||
|
There have been lots of talk of network administrators getting rather riled by this. Especially at http://slashdot.org/articles/03/09/1...&tid=98&tid=99
This will have the immediate effect of making network trouble-shooting much more difficult. Before, a mis-typed domain name in an email address, web browser, or other network configuration item would result in an obvious error message. You might not have known what to do about it, but at least you knew something was wrong. Now, though, you will have to guess. Every time. So I ran my port scanner against 64.94.110.11 ports 1 through to 1023 (I think my ISP can make an exception to the "no port scanning" rule in this case). The only open ports were 25 (smtp) and 80 (http). Some have pointed out that this will make an important anti-spam check impossible. A common anti-spam measure is to check and make sure the domain name of the sender really exists. (While this is easy to force, every little bit helps.) Since all .COM and .NET domain names now exist, that anti-spam check is useless. Spammers thought of how to get round that one ages ago. Spammers can check that domains exist just as easily as anyone else. This so-called "important" anti-spam check was already useless - VeriSign hasn't changed this.Any network administrator who says otherwise is just trying to bamboozle you with technobabble, in the hope that his employers and their shareholders won't notice that his wages are a complete waste of money. ![]() This isn't about spam. It's bigger than that. It's about being bitten on the bum by a monopoly that we've allowed to grow without noticing it. In the past, we used to talk about Microsoft, AOL and Google monopolies - then VeriSign put it all sharply into perspective on Monday. It makes a mockery of anti-squatting action too. If we knew VeriSign was going to do something like this, then we would all have been much more lenient on the cybersquatters. And we also worry about future generations, who might not remember a time before VeriSign, when there used to be a thing called "freedom of speech". I think the implications of VeriSign's action are huge, they go way beyond spam.
__________________
- |
|
#9
|
||||
|
||||
|
I def think it's an abuse of their position,. it's like they've suddently got dibs on every single domain name that doesn't yet exist for their own marketing efforts,,. which can't be right,.
perhaps ppl will switch to the alternative DNS root tree.. ![]() Sil PS - how does it work,. is it just a DNS trick - but verisign don't own the dns root servers do they? |
|
#10
|
|||
|
|||
PS - how does it work,. is it just a DNS trick - but verisign don't own the dns root servers do they? So, any hostnames made up of a subdomain of .net or .com that isn't owned by anyone will resolve to 64.94.110.11 For example ....
There's a web server at 64.94.110.11. If the Host: field of your browser's HTTP request is anything other than sitefinder.verisign.com, then it returns an HTML redirect page, putting the contents of the Host: field in the query string of the address it redirects to. That's why your address bar changes. Erm - have I explained that okay? Opinion in a mo.
__________________
- |
|
#11
|
|||
|
|||
No, they don't. They're only responsible for the authoritative name servers for the .com and .net top level domains. My attitude has mellowed out a bit now. ![]() Apparently Paxfire did the same thing on .biz earlier this year. But they had their knuckles rapped for it, so they pulled it. ![]() But ICOM has been doing a technically similar thing with .museum for some time. A subdomain of museum that doesn't exist will resolve to IP address 195.7.77.20. There's a web server at this address, which helpfully offers you a list of .museum subdomains that do exist, and links to their home pages. ![]() People say VeriSign have messed up spam filters, by wild-carding .com and .net. But if that was actually true, then ICOM would also be guilty of messing up spam filters when they wild-carded .museum. Oddly enough, no-one complained about that! ![]() So in VeriSign's defence, they could say that's unfair. But VeriSign and ICOM are different. If you put a hostname into the address bar of a web browser, it will try to turn it into an IP address. If it can't, then most browsers will stick .com after it, and try again. Internet Explorer, Netscape, AOL and Mozilla all do this, chances are Opera does too. This gives VeriSign a natural monopoly. To that, VeriSign could say - so what? VeriSign don't make the browsers. It's not VeriSign's fault. Microsoft and AOL could have just as easily made their browsers to try .museum, or another TLD instead. Or just give up on wrong addresses. Or send the wrong address to their favourite search engine - Google, MSN Search, AOL Search, Altavista, or whoever pays them the most for the privilege. Many browsers already do this - but not until after trying the .com TLD.My prediction - Paxfire set the precedent, so VeriSign is likely to be forced to pull out too. But they'll squeal about ICOM .museum. I think it would be a shame if ICOM are forced to pull out. I think ICOM's list of .museum links is genuinely useful - it's not like VeriSign's blatent advertising at all. (But then again, that's what we said about Google when we were comparing it with Altavista a few years ago, and look how powerful Google are now - or were ...) I'm all in favour of alternatives to the ICANN root servers, and I think you'll see more alternative roots being set up in the next few months. But I doubt they'll make a big dent in ICANN's market share for several years yet.
__________________
- |
|
#12
|
||||
|
||||
|
ISC have released patches for its BIND DNS server (used by a mere 80% of DNS servers on the net) that work around this "feature"
http://yro.slashdot.org/yro/03/09/17...tid=187&tid=95 The problem with the mail server is it gives the wrong error back when you attempt to send email through it, meaning that the mail will just fail, so if the primary mx record mailserver domain disappears the mail will not get sent to the secondary mx mailserver. This is bad and defeats the object of having multiple mx records. (ok, its still not *that* bad, just a *little* bad)
__________________
uk's worst isp |
|
#13
|
||||
|
||||
|
Thanks for the explanation squidgy
![]() I still don't like it tho ![]() I guess adding a routing path to 64.94.110.11 - and making it null (somehow?) would remove the interferance ,. not sure how to do it - but it should be possible.. Sil |
|
#14
|
|||
|
|||
I guess adding a routing path to 64.94.110.11 - and making it null (somehow?) would remove the interferance ,. not sure how to do it - but it should be possible.. There's a difference between not being able to convert a hostname into an IP addresses, and not being able to make a TCP connection with whatever port number (usually 80 for the web) of an IP address that's already known. Internet Explorer gives you the same "This page cannot be displayed" error message either way. But other client software tells them apart. For example, SmartFTP says "Host not found" or "Software caused connection abort" And Mozilla says "Please check the name and try again" or "The operation timed out" Obvious flaw - what happens when VeriSign change the IP address? There's probably a way of dealing with that, I guess this BIND patch would block non-existent .com domains, no matter what IP address VeriSign try and point them at.The problem with the mail server is it gives the wrong error back when you attempt to send email through it, meaning that the mail will just fail, so if the primary mx record mailserver domain disappears the mail will not get sent to the secondary mx mailserver. I know how to do MX resolution, but that's where you've lost me. When you say "mail server", do you mean the SMTP MX exchanger that receives all the mail for a particular domain? Or do you mean the SMTP relay, that receives all the outgoing mail from one particular network, and forwards it on to the recipient email address exchangers?Thanks. ![]() About my port scan results - I can see why port 80 is open, that's obviously the web search interface. But I have no idea what VeriSign gains from having port 25 open. Why are they hosting what looks like an SMTP server? And would it simplify the mail problems people are talking about if they didn't host an SMTP server? Thanks.
__________________
- |
|
#15
|
|||
|
|||
|
I'm guessing here now, but I think the SMTP server is something to do with rejecting mail.
Correct me if I'm wrong, but, when a domain doesn't have an MX record, then mail relays will just look for an SMTP server on the A record IP address instead. For non-existent .com domains, this will be 64.94.110.11. If there wasn't an SMTP server hosted on this address, then relay programs might assume that there's supposed to be a mail exchanger there, but perhaps it's switched off or something, and that someone might turn it on again a bit later. So it waits a bit, and tries again later a few times before it gives up. This would mean that anyone who tries sending email to .com domains that don't really exist, might not know about it for a few hours. Or even days. However, the SMTP server helpfully rejects absolutely all mail that you try to upload, with a "550 user domain does not exist" message. So if you send mail to a .com domain that doesn't exist, you should know about it straight away.
__________________
- |
|
#16
|
||||
|
||||
|
well - it's not that diff to stop the ip being addressed (could also do it by adding an entry into the hosts file I expect)
on win2k (don't know how to make it a static route tho) route add 64.94.110.11 MASK 255.255.255.255 192.168.0.0 METRIC 1 I think an entry like 64.94.110.11 0.0.0.0 in the hosts file would prolly do it also? Sil |
|
#17
|
||||
|
||||
|
route add 64.94.110.11 mask 255.255. 255.255 192.168.0.0 -p
should set it up as a static/persistent route You can check whether or not it's taken by checking the registry afterwards. Navigate to : My Computer/HKEY_LOCAL_MACHINE/SYSTEM/CurrentControlSet/Services/Tcpip/Parameters/PersistentRoutes You should also be able to see it listed in the routing table at the bottom of ROUTE PRINT (persistent routes) ![]() If you have Win2k/2k3 Server then you should really either use the "Routing and Remote Access" MMC snap-in or use the netsh command line shell. 'Slo Last edited by Onslo; 18-September-2003 at 06:47. |
|
#18
|
||||
|
||||
|
http://www.petitiononline.com/icanndns/petition.html
Petition against it. Dunno what good it will do, but there I am at #7701
Last edited by NA-RYAN; 18-September-2003 at 14:03. |
|
#19
|
|||
|
|||
|
Pleased by the implication that 7700 peeps have registered already. My own opinion ...
We internet users, who either own domain names or have an interest in the domain name system, wish to object to the Verisign Sitefinder system. We believe that the system: 2. breaks technical standards affecting email services, and other internet systems; 3. is anti-competitive, providing Verisign with 20 million eyeballs per day for "free", while not paying for the domains they are resolving. All other market participants pay at least $6 per domain per year (wholesale); 4. violates trademark rights of domain holders, by typosquatting on their .com and .net domains; and 5. violates the authoritative nature of DNS, turning it instead into a "best guess" system filled with uncertainty, thereby destroying the coherence of the DNS for Verisign's own short-term profit. Still, best of luck to them.
__________________
- |
|
#20
|
||||
|
||||
|
from compwire..
VeriSign Inc, which shocked many in the internet community with the launch of a domain name redirection service last week, has said it will not deactivate the service, despite pleas from ICANN and almost universal opposition. VeriSign Inc's controversial new domain name system service, Site Finder, caused a number of technical problems, including one potentially dangerous security hole, when it abruptly went live last week, according to experts. |
|
#21
|
|||
|
|||
|
Hmm, interesting developments.
I still don't see much ground for proving that VeriSign have done anything wrong, given that no-one minded when other TLD's did something similar. I think the best chance we have against VeriSign is to invoke laws about monopoly and competition. Isn't this how ABC was broken from NBC back in the 1930's? As for Site Finder being a big money spinner - are we sure that advertisers want to be associated with it? How much can VeriSign charge for it? Will people click through? Would you buy anything you saw advertised on site finder? Online advertising has taken a nose-dive over the past year or two, so I don't see how it will be different for Site Finder. But Google has been an exception. Maybe VeriSign plans to copy what Google have done - but I think the Google advertising business is still safe for some time yet.
__________________
- |
|
#22
|
||||
|
||||
|
seems some ISPs are putting on the bind patch to make verisigns thing not work
![]() also some scripts / hosts file changes will block it.. see http://www.spywareinfo.net/sep24,2003#verisign Sil |
![]() |
| Tags |
| None |
| Thread Tools | |
| Display Modes | |
|
|