AT&T Uverse Filtering Outbound to my VPS?

[walesmd]

Sometime Friday evening / Saturday morning I completely lost the ability to connect to one of my servers from within my network; thankfully, I've yet to ramp up beyond a development and testing phase so the only customer affected was myself (I'm a glass half-full kind of guy).

The facts:

• All activity is attempted by IP; DNS is a non-issue.
• Prior to "Event X" I could ping/ssh this box, now I can't.
• My whole home network is affected (WiFi/Ethernet & Windows/OSX/Linux).
• I can ping "down box" from one of my other boxes.
• I can ping/ssh any of my other servers perfectly fine (albeit, different service provider)
• The service provider of this box has been able to ping/ssh - don't see any issues.
• I've assigned a new IP to this box - no luck.
• I've reloaded the box, from Ubuntu to CentOS - no luck.

Any ideas? I'm really not looking forward to this phone call, having to explain highly advanced technical terms like "ping" and "timeout".

 

[Aldryic C'boas]

Mind pastebin'ing a traceroute from your location to the IP(s) in question?

 

[Mun]

Are you connecting via IP or dns? If DNS try IP.

Have you put in a ticket with AT&T? If not go do it.

This actually happens every once in a while where a route just simply doesn't work. (like an accidental null for a network do to a DDOS or something)

Have you trace routed it?

Mun

 

[walesmd]

 

[Aldryic C'boas]

Welp, it looks like you can complete a trace, but are otherwise dead in the water from your Macbook.  If I had to take a guess, I'd say it looks like IPTables or similar is running on the VPS, and your modem/router's current IP somehow got added to the blacklist.  If you don't have any type of firewall running on the VM, I would suggest getting with the provider to see if they may have mistakenly nulled your 'home' IP (doesn't look to be the case though, since you can still trace in just fine).

 

[Mun]

Could also be blocking port 22 for some reason. Try changing it to something else (its also a good practice)

In debian: nano /etc/ssh/sshd_config

Mun

 

[walesmd]

 

[mikho]

The exact same situation happened me once with my IPXCore VPS.

I couldn't connect to the VPS from home, no ping... nothing.

From other places it was no problem.

After submitting a ticket, it was the nodes IPTABLES that blocked my home IP, some sort of monitoring system triggered on my at the moment bad connection from home (ISP had some stability problems and my ssh connection reconnected alot).

Perhaps you have encountered the same situation?

 

[walesmd]

Thanks @mikho, I'm fairly certain that is the problem since I can connect to any number of other servers that are similarly configured.