fp.

On a contract job clearing out a data center doing routine stuff like taking inventory and audits before we decommission hardware. The issue is there is one node that keeps coming back that isn't in the documentation. ip is in the 46.28.x.x range Its not in the facilities registry though. Ran it through RIPE and ARIN to find nothing.

The latency is what is getting me though. 0.4 round trip every time. Tested from multiple machines including a phone on LTE to get the same response time. That should theoretically mean I am right next to the machine which doesn't make sense across three different connections.

Checked the physical hardware and it's nothing I've ever seen before. Not standard 1U or 2U ports maybe proprietary. serial format is:

CC-[4 digits]-[2 digits]-[6 alphanumeric]

CC prefix doesn't math Cisco, Ibm, Dec, 3com or anything. went back through the facility's historical logs. node appears in their earliest available records, which go back to 1994. facility was built in 1997.

has anyone seen a CC- serial prefix before? or have an explanation for the latency consistency?

Comments

mmastrac•19h ago

CyberChron. If you don't need to know, don't ask.

And you're also assuming that all the pings are being returned by this box.

cbarrick•18h ago

Well now I gotta ask. What's up with CyberChron?

The only thing I can find on Google is a website straight out of 1999 and lawsuit from 1995. They're obviously a US military contractor, but that's all I can tell.

devmor•19h ago

Is it in what looks like a luggage/waterproof case? If so, that’s milspec networking hardware.

bombcar•19h ago

Could the latency consistency be something designed to make it difficult to pinpoint its location? It sounds like you found the hardware and are just wondering what it is?

Well ain't this place a geographical oddity! 0.4ms from everywhere!

VladVladikoff•19h ago

Wouldn’t that mean all upstream providers are all in on it? Participating in the charade and returning ICMP packets preemptively?

bombcar•11h ago

Unless it’s some equipment that exists at multiple endpoints at the same address.

VladVladikoff•19h ago

Lots of ASNs in 46.28.0.0/16 What’s the actual netblock?

protocolture•18h ago

>The latency is what is getting me though. 0.4 round trip every time. Tested from multiple machines including a phone on LTE to get the same response time. That should theoretically mean I am right next to the machine which doesn't make sense across three different connections.

It means your 3 different connections have decent connectivity to whatever host currently responds to ping for that IP. You cant really derive much more than that from a ping. If it has been there since 1994 it might have been decommed and the IP reassigned. I would suggest a scream test to be honest, especially if you have orders to remove it anyway, seeing if the pings stop responding when you remove the power or networking will tell you more.

bananamogul•18h ago

Would looking at the device's MAC address (which you can get from arp) help? That would give you at least the manufacturer of the network interface.

This is assuming you're on the same subnet.

userbinator•18h ago

The latency is what is getting me though. 0.4 round trip every time. Tested from multiple machines including a phone on LTE to get the same response time.

https://en.wikipedia.org/wiki/Anycast

Animats•18h ago

Is this the hardware you have? [1]

That's some kind of encryption box. It has a "zeroize" button, to clear the keys in an emergency. It might have something that forces uniform latency to make traffic analysis more difficult. Some cryptosystems are totally synchronous, and send random bits at a constant rate when there's no data.

[1] https://www.artisantg.com/TestMeasurement/89462-1/Cyberchron...

mikeyouse•17h ago

Perhaps a CC-Mail server?

https://en.wikipedia.org/wiki/Cc:Mail

If so, you should be able to telnet to that IP on port 3264 [https://www.ietf.org/rfc/rfc1700]

jdw64•17h ago

I think this doesn't make sense. Is there actually a realistic mechanism that makes this possible, or is there a gap in my hardware knowledge?

protocolture•17h ago

Wild speculation:

The IP is currently assigned to some router in a nearby carrier network that has decent connectivity to everything. Potentially the mobile carrier.

The OP happens to be connected to wifi at the DC. Tests across the mobile carrier network didnt take place thanks to route preference over wifi.

Something in carrier land is responding to the IP. It might be assigned in a stupid fashion, like every PE router is responding to it. Or anycast or something.

The OP has the ip address assigned to his laptop and phone. His own devices are responding to the pings.

OP might be using a different ping client that he isnt used to, and is taking 40ms as .4 ms. He doesn't really give us the measurement, just the value.

OP might be using a VPN and the VPN gateway router is responding to this ip with pings for whatever reason. Or the VPN client software has assigned this ip to his laptop.

geocar•16h ago

Yes: This is how Anycast works.

The same IP block is announced from multiple geographic locations, and so IP traffic will be routed to the nearest.

jdw64•15h ago

So there's something called Anycast. Thank you. But does that work on LTE as well?

userbinator•15h ago

Yes, as well as DSL lines, 56k modems, and whatever else you can use to connect to the Internet.

mianos•17h ago

CVC3000, they should send some guys in dark suits and glasses to pick it up.

DANmode•6h ago

Is that humor, to you?

geocar•16h ago

Can you see if the media is carrying 802.1Q traffic tagged 986?

866-RON-0-FEZ•14h ago

A picture would be nice and avoid a lengthy game of Hacker News Charades.

ale42•14h ago

0.4 what latency? 0.4 s or 0.4 ms, or something else? Because if it's 0.4 ms, that seems to be very local, and possibly even impossible on LTE because of network latency. I'm not even sure 5G manages to get under 1 ms of latency.

elevation•5h ago

I had the same thought: the device OP used for LTE testing may have been still connected to a local link and was routing packets over that instead.

zamadatix•13h ago

0.4 ms over LTE doesn't make sense, even if the device (or another device with the same IP) was at the base of the tower you were connected to. If you figure out what's going wrong with that test it might make a lot more sense.