This is a great story. Almost wish the author had dug a little further in and discovered something like Clifford Stoll in The Cuckoo's Egg, but a nice writeup nontheless.
Makes me wonder how many partly implemented but ignored protocols like this exist.
The article mentions Ascension Island, a small island in the south atlantic. There's a really great spy novel that takes places there, Ascension by Oliver Harris.
> It never really took off though, and even back in its early days it saw barely any use. Over the years it just deteriorated further, and today it's basically completely dead.
It's actually not completely dead... It's just (almost) completely non-public.
You can subscribe to services to get number porting information where the interface is basically e164.arpa queries to a private nameserver over a VPN. I don't know the details, the cost was high enough that it didn't make sense for my employer to pursue it.
It's also not uncommon for telecom providers to use it for their own internal routing because enough telecom software did in fact develop support for it despite lack of public implementations.
As someone who's been in the VoIP industry for over 20 years it makes me sad to think of what could have been if both ENUM and IPv6 were more widely adopted. For many years you could reach me via email, SIP, or Jabber with the same identifier and if there were effective support for ENUM in the USA my work phone number would have been able to connect you to any of them, directly and with G.722 HD voice long before it eventually came to modern cellular networks.
There was a startup called WUPHF that used some telecom wizardry to multicast messages between providers given a single identifier, I'm not sure what happened to it...
My guess is that this is actually what happened here: Some carrier or VoIP software leaking its DNS queries to the public internet because the service is using is only set up to respond for e.g. NANP (+1) number ranges and the rest is falling through to the public (and largely unused) ENUM.
I know opinions about Mitnick are quite divided. But I enjoyed reading "Ghost in the Wires" (https://www.goodreads.com/book/show/10256723-ghost-in-the-wi...). It's giving this 90s-era hacker vibe that's kinda fun (even if the truth might be stretched a bit).
I love the line near the end of the article: "So in the end, I was down 10€ in domain fees, there was sadly no bug bounty (I thankfully didn't get my door kicked in at least)."
Makes me cringe, imagining what that would be like.
Given the author appears to be a 19 year old German girl, UK bureaucrats proxying admin duties for offshore territories are unlikely to be motivated to start an international extradition for what amounts to a helpful tip.
If you see my previous comment, hacker News people are all for big governments trampling over the rights of software developers in the name of national security. Give it a few hours and see the sentiment change.
It's funny how such holes can remain for years, and no one notices until someone stumbles upon them. It's interesting that no serious organization wanted to address the issue until it was discovered that the military was involved.
It's a shame the author wasn't rewarded but at least the story can now be told over a beer.
Man, you really got lucky they didn’t throw you in jail. Anything related to national defense is pretty scary. Do you think you might get some kind of reward for exposing this vulnerability?
I don't want the miss the young girls skills but she is no security researcher but in the eyes for her target simple a hacker.
And I would say too that she got lucky. It would not be the first time that inexperienced young persona have to suffer the consequences even of their well intended actions.
The article covers it, but to reiterate. Yes, The idea is to have a mechanism to map phone numbers to hosts.
The normal path for voip phones given a phone number is to end up at a sip trunk provider go over the traditional phone network (which at this point probably routes over the internet anyway) hit another sip trunk and end up at the receiving sip phone.
This provides a method to bypass the traditional phone network and go directly over the internet. The sip phone looks up the host responsible for that phone number and directly connects. The sip providers would be responsible for maintaining this number to host mapping in dns.
When you think about it DNS is really just a big distributed phone book, a key value store to look up numbers based on names. The reverse records are a method to look up names based on numbers using that same distributed architecture.
There is also an interesting legacy architecture interaction here, traditional phones can only enter numbers. Cell phones could use dns names directly(but don't) or we could use ip addresses as a sort of modern phone number(but don't), The whole world was connected via phone numbers and that is now how we expect phones to operate.
It is a shame they did not actually set up a SIP server and see if any of those requests turned into actual call terminations.
There is another schema called TRIP [1] - telephony routing over ip that uses a number format "1234*1455" designed to be entered on a standard phone keypad. When I registered my ITAD (internet telephony administrative domain, the RHS of a TRIP number) I was lucky enough to get one that matches my local dialling code!
I enjoyed reading this much more than anything I read here recently. In particular I like how it absolutely shows that somethings just… fall through the cracks!
Their homepage, https://lina.sh/, reminds me of the early days of the internet. Are webrings back in fashion? Or is this just a group of old school folks keeping the nostalgia alive?
Now a question daunted on me. Assuming away strategies like store-and-decrypt-later-in-a-quantum-future, could a MITM really eavesdrop, as in, aren’t such ARPA-routed phone calls encrypted?
I recall about 15 years ago Google had this project about one phone number for the person, things got all tidied away and could never find out much more about it.
I’d forgotten the keywords of power to find it again. I liked the idea.
> skipping the expensive phone network and re-routing calls over the cheap internet instead.
I find this such a bad idea, if I want to use the cheap internet for a call I would use an internet based voice call system directly, rather than messing around with telephone numbers and I guess potentially accidentally doing an expensive phone call (same with RCS/SMS vs. just using an internet-based chat directly).
People with POTS brains trying to make them do Internet things leads to a neverending stream of bad... no actually, terrible ideas. Look at H.323 for example, something that could only be created by someone who has no clue how the Internet (with firewalls, NAT, etc) works, and that's just the tip of the iceberg for the ITU-T standards stream.
Newbie question:A malicious actor could simply have MITM'ed the calls the blog states and obtaining the voice traffic. Is that possible because the author had the certificates( since holding the ownership of the domain)?
Although you can run SIP over TLS (for signalling) and use SRTP for media (key exchange done over the aforementioned signalling channel), in practice most SIP is over unencrypted UDP and media is unencrypted RTP.
If you control both endpoints and they support it you can configure them to use encryption, but even then implementation qualities vary widely (just because you enable SIP over TLS doesn’t mean they’ll actually verify the certificates for example - giving you at best opportunistic encryption), and I bet a lot of the implementations also have bugs/vulnerabilities.
If security is needed, it is often implemented by way of running the whole thing over private links (which can be secured with IPSec or any other VPN technology). In fact that’s presumably what’s happening, but misconfigured equipment making those ENUM lookups would allow the attacker to steer the traffic away from the secure link and towards an endpoint they control over the public internet.
84 comments
[ 0.14 ms ] story [ 81.5 ms ] threadMakes me wonder how many partly implemented but ignored protocols like this exist.
It's actually not completely dead... It's just (almost) completely non-public.
You can subscribe to services to get number porting information where the interface is basically e164.arpa queries to a private nameserver over a VPN. I don't know the details, the cost was high enough that it didn't make sense for my employer to pursue it.
As someone who's been in the VoIP industry for over 20 years it makes me sad to think of what could have been if both ENUM and IPv6 were more widely adopted. For many years you could reach me via email, SIP, or Jabber with the same identifier and if there were effective support for ENUM in the USA my work phone number would have been able to connect you to any of them, directly and with G.722 HD voice long before it eventually came to modern cellular networks.
https://www.youtube.com/watch?v=yL1z1ZHD0K4
> So I had accidentally logged hundreds of thousands of phone numbers and timestamps for calls going to military bases.
That's quite a jump to conclusion right there.
Of course Diego Garcia doesn't have the ability to accept or process asylum applications.
It's a very British trap.
I love the line near the end of the article: "So in the end, I was down 10€ in domain fees, there was sadly no bug bounty (I thankfully didn't get my door kicked in at least)."
Makes me cringe, imagining what that would be like.
you know there’s hope for our future.
Uhm...
https://news.ycombinator.com/item?id=49365768
The most I can see is you making a silly American-centric remark, and people gently correcting you that there are other places on earth.
It's a shame the author wasn't rewarded but at least the story can now be told over a beer.
I don't want the miss the young girls skills but she is no security researcher but in the eyes for her target simple a hacker. And I would say too that she got lucky. It would not be the first time that inexperienced young persona have to suffer the consequences even of their well intended actions.
Who knows! Maybe it’s working just fine, but the e164.arpa record is pointing to the public prosecutor’s teapot or something.
Is this related to Softphone / VOIP in any way?
(We just use ours for receive only -> straight to voicemail)
The normal path for voip phones given a phone number is to end up at a sip trunk provider go over the traditional phone network (which at this point probably routes over the internet anyway) hit another sip trunk and end up at the receiving sip phone.
This provides a method to bypass the traditional phone network and go directly over the internet. The sip phone looks up the host responsible for that phone number and directly connects. The sip providers would be responsible for maintaining this number to host mapping in dns.
When you think about it DNS is really just a big distributed phone book, a key value store to look up numbers based on names. The reverse records are a method to look up names based on numbers using that same distributed architecture.
There is also an interesting legacy architecture interaction here, traditional phones can only enter numbers. Cell phones could use dns names directly(but don't) or we could use ip addresses as a sort of modern phone number(but don't), The whole world was connected via phone numbers and that is now how we expect phones to operate.
There is another schema called TRIP [1] - telephony routing over ip that uses a number format "1234*1455" designed to be entered on a standard phone keypad. When I registered my ITAD (internet telephony administrative domain, the RHS of a TRIP number) I was lucky enough to get one that matches my local dialling code!
https://tripresurgence.org/trip/history/ [1]
loads of webrings also started -- 90s internet is back, just look for it <3
I would not at all be surprised seeing that domain being abandoned again at a renewal in the near future.
I’d forgotten the keywords of power to find it again. I liked the idea.
I’m assuming this is is it again.
[1] https://en.wikipedia.org/wiki/Google_Voice
I remember because it seemed quite a nice idea, almost obvious.
I find this such a bad idea, if I want to use the cheap internet for a call I would use an internet based voice call system directly, rather than messing around with telephone numbers and I guess potentially accidentally doing an expensive phone call (same with RCS/SMS vs. just using an internet-based chat directly).
Having said that, SIP is barely better.
If you control both endpoints and they support it you can configure them to use encryption, but even then implementation qualities vary widely (just because you enable SIP over TLS doesn’t mean they’ll actually verify the certificates for example - giving you at best opportunistic encryption), and I bet a lot of the implementations also have bugs/vulnerabilities.
If security is needed, it is often implemented by way of running the whole thing over private links (which can be secured with IPSec or any other VPN technology). In fact that’s presumably what’s happening, but misconfigured equipment making those ENUM lookups would allow the attacker to steer the traffic away from the secure link and towards an endpoint they control over the public internet.