Teliax DOA ?

Anyone have experience with Teliax VOIP services? I ask because we tried to sign up and found there was no there there.

They do have a very nice website, and a nice list of features, however,

  • The form under is broken with PHP errors.
  • This form also lists your full credit card number, expiration, and everything else as part of the submit URL, _very_ insecure.
  • The phone number listed on website has returned a busy signal for several days now.
  • The website's "live help" also says "there is one support call queued before you". That call has apparently also been in the queue for several days.
  • The teliax.com domain has only two DNS nameservers and they're both on the same /26 subnet. Neither has valid rDNS. The subnet is registered to a "David Comean" and appears to a single FreeBSD server in a "RockyNET" colo. At least one of the upstream providers is Yipes (well known as a spamhaus).

From all appearances there is no company behind the website. Is this some kind of front to a VC scam?

Fred

Reply to
Fred Friedman
Loading thread data ...

There is no doubt that teliax is a very small company that is still learning how to do basic machine administration. I too noticed all the DNS problems you mentioned and still went with them. They do have a reasonably good deal on pay-per-minute service and DID numbers.

At this point in time, it is relatively hard to find companies that offer DID'S in the SF Bay area and have a sip server within a few ms of the bay area.

-wolfgang

  • I have to admit, I only noticed the CC in URL issue long after I signed up. Yikes.
Reply to
Wolfgang S. Rupprecht

VoipJet has a server on the West Coast:

formatting link
But as you say, they don't offer DID numbers, and supporting only IAX and not SIP one can't even use the free service by Ipkall, which appears to be further North on the same coast, in Washington State :-(

Enzo

Reply to
Enzo Michelangeli

They appear to have fixed it now...

Why? The security is the same as for a POST submission (although, double-charging might more easily occur if the "Back" button takes again the browser to that page). If the destination page, and, for server authentication reasons, the page where the "submit" button is, are secure, SSL protects the card number regardless of the method used to submit the form.

Enzo

Reply to
Enzo Michelangeli

You mean clicking on "continue"? I did it a few days ago when I subscribed, and I didn't notice anything strange. Now I've gone (with Firefox 1.0.4) up to the step 4, stopping there because I don't want to enter fake data, and it appears to work fine.

Yes it is. The URL is not transmitted at all: first a TCP connection is established, then the SSL security is negotiated, and only then does the client send a "GET" specifying the local part of the URL and, in HTTP 1.0 or higher, the hostname in a "Host:" header.

This is true even when there are proxies in the path: see e.g.

formatting link
for a copy of Ari Luotonen's original draft,
formatting link
If that scheme was not followed, the client couldn't possibly validate the server certificate (and that's intentional, in order to defeat possible Man-in-The-Middle attacks by intermediate proxies). The only place where the URL could possibly be logged is the target web server, which is supposed to be secure.

Enzo

Reply to
Enzo Michelangeli

It's a multi-part form. Did you follow it through to the end?

No, this is incorrect. POST data is not sent in the URL, does not get cached by browsers, is not logged by proxy servers, and is encrypted. GET data (transmitted in the URL), in contrast, is not encrypted regardless of whether you use SSL (https) or not.

Fred

Reply to
Fred Friedman

GET arguments in https requests are not logged by (or available to) proxy servers, and they are encrypted from end to end. Stop and think for a moment about how the https protocol works.

The only thing worth worrying about is that they are often cached by browsers. And, I suppose, they appear in raw server log files if the server operator makes those available, in which case they've probably got bigger problems.

miguel

Reply to
Miguel Cruz

You did not follow it through then. If you had you would see that it was down.

Thanks for the correction, and referring URL. I was wrong about the encryption of GET data and apologize. Still not comfortable with the needless use of cacheable credit card info but at least it's not sniffable at the packet level.

Fred

Reply to
Fred Friedman

Because it's encrypted (SSL), proxy servers can't see it anyway (without a lot of effort).

I'm a little annoyed that they send standard (unencrypted) e-mail with much of that same information along with a message that says that it's all encrypted.

--kyler

Reply to
Kyler Laird

Kyler, what is being sent in the welcome email that has security hazzards?

Reply to
teliax

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.