I've been following Palo Alto as a networking company for a couple of years now. Their claim is that the days of the port-based firewall are dead and that their application-centric approach is a far better way to enforce your access controls. Take the HTTP protocol for example. HTTP typically runs as a service on port 80, but does that mean that everything running on port 80 is HTTP? As an attacker looking for a way to funnel data out of your organization, why not use the standard HTTP port to send data, since I know you leave it wide open in order for your employees to surf the web. There's nothing to say that I actually have to be running an HTTP server on the other end and there's nothing on my classic firewall to tell any differently. At first, I was admittedly a bit skeptical. I didn't think that you could really tell enough about different applications on the web to be able to separate them out like Palo Alto claims to. Fortunately, Palo Alto reached out to me and provided me with a brand new PA-200 in an attempt to change my mind.
When the PA-200 arrived, it came with everything that I would need to get it up and running. That includes the unit itself, a power supply, a D89 to RJ45 console cable, an ethernet cable, and some instructions and warranty information.
On the front of the unit is four ethernet ports for your devices, a management port, a USB port, a console port, and several status indicator LEDs.
By default, the appliance is configured with ethernet ports 1 and 2 paired as a WAN to LAN link as this is the configuration that the majority of the people who buy it will likely use it for. That said, by following the instructions to connect your computer up to the management port, you can quickly access the user interface that allows you to change this assignment.
This shows the ethernet 1 and 2 interfaces as both being a "virtual wire" and here we can see the virtual wire that connects the two.
From here, we can take a look at the "zones" and see that our two interfaces have been defined as an untrusted (ethernet 1) and trusted (ethernet 2) zone.
To think of this a different way, my cable modem WAN connection (ie. the Internet) goes in my "untrust" zone and my local network (ie. LAN) goes in my "trust" zone. Now all that's left is to set our policy and for ease of management to start with, I set it to allow everything out with a default deny all inbound.
With this configuration I had done enough to be up and running on the device and I immediately started to see data populate the dashboard on the top applications running on my network.
It's color coded based on risk level and the dashboard also provides me a similar view of Top High Risk Applications. Any of these boxes can be clicked on in order to provide additional data about the protocol, sources, destinations, countries, and more.
Now, let me say that while I'm running this on my home internet connection, this thing is a hoss and can do way more than I can throw at it. With their App-ID technology enabled you can throw 100 Mbps of throughput at it no problem. In addition to being an application firewall, it also does standard port-based firewalling, VPN, routing, switching, and so much more. It's so extremely versatile that this thing could easily be placed in a smaller branch office and replace multiple other devices on their network such as a firewall, router, and VPN concentrator. More functionality for less money...who wouldn't want that? In addition to these default capabilities, additional licensing can also be obtained to allow you to do URL filtering, malware detection, and more. Having just gotten this up and running, I'm still exploring the ins and outs of all of the functionality, but it's pretty exciting to have all of this capability in a box that is smaller than the cable modem my ISP provides me. More posts to come on this as I get deeper into the guts of running my new Palo Alto PA-200 !
About a week ago I turned on a new rule on our IPS system that is designed to detect (and block) users who are using TOR to make their activities on our network anonymous. You can say that TOR is about protecting a user's privacy all you want, but I'd argue that while using corporate assets you should have no expectation of privacy (at least in that sense) and that the use of anonymizers on a corporate network can typically be viewed as a sign that you are up to no good. Almost immediately when I turned on this new rule, I began seeing associated events in the IPS console. I decided that the best approach was to contact the user directly as they may be wondering why their Internet connection was no longer working. I reached out to this particular user and explained that if this was the case, then it was because of the new IPS rule. The solution was simple; just reconfigure his browser to no longer use TOR as the proxy. But as I began this process, things started getting weird.
I began by telling the user to look for names like "TOR", "The Onion Router", and "Privoxy" in his Add & Remove Programs. Strange....there was nothing there. Then I asked him to check his Task Manager to look for a running process called "tor.exe" or similar. Again, nothing. I was at a loss. I decided that this was something I needed to get my hands on to figure out so I scheduled some time with the user.
This morning when I sat with the user, I noticed little wrong with his system. He had a few standard applications running, but nothing unusual. I checked his process listing and saw nothing out of the ordinary. I ran Hijack This! and that, too, looked pretty normal. All this, yet in the meantime I continued to see alerts on the IPS system that his computer was using TOR. Even when I was sitting at the console with NO browser activity. So, to make a long story short, here's how I finally figured out what was happening. I checked the IPS system and came up with the source ports for the requests that I was seeing alerts on. I then went on the system and ran a netstat -nao. This listed all network connections on the users system along with the associated process. I checked the list and found the entry that matched the port number I was seeing the alerts on. I then ran the command tasklist /svc /FI "PID eq <process_num>" This provided me with the name of the process that was running with this process ID which it turns out was "iexplore.exe". Wait. Internet explorer isn't even running on this computer. Or is it? Since the default process viewer in the Task Manager is pretty lame, I downloaded the Microsoft Sysinternals Process Monitor. It's a free tool available from Microsoft and provides a ton more information about running processes and allows you to see what they are doing in real time. I used the Process Monitor to view these processes and focused particularly on the flags that were used when they started. What I found was actually pretty startling.
Both of the Internet Explorer processes were started with a special flag that told them to start silently (ie. without the UI) in the background. They also specified a flag similar to this:
--HiddenServiceDir "C:\Documents and Settings\<User_Name>\Application Data\tor\hidden_service" -- HiddenServicePort "55080 127.0.0.1:55080"
Aha! We found our culprit! TOR was running as a hidden service out of the Application Data directory. Once I found this, it was all over. Scanning through the Application Data directory, I also found a file under "Enemvy\ugbie.exe" that was extremely suspect. A later scan via Malwarebytes identified it as a variant of Trojan.ZbotR. I deleted these directories and Malwarebytes found one registry key associated with the ugbie.exe file and deleted it. All is good now and the system is no longer alerting about use of TOR.
So, what's our lesson here? The malware writers are getting sneaky. They've realized that we've created blacklists of their servers and they need to be able to adapt around that. Now, they are using anonymizers, like TOR, to get around these blacklists. Apparently this isn't the first use of TOR in malware either as I read about something called SkyNet that did something similar. In any case, they would have gotten away with it if it weren't for my IPS rule to detect TOR and a fair amount of persistence in finding the root cause. If you're not already detecting this on your network, I think that it's about high time you did it. You can thank me later.
This post is going to be short and sweet as it's something I meant to put up here when I found it sometime back in mid-2011. I'm not even sure if Time Warner is still using these Ubee cable modems for their RoadRunner offering, but I'm sure that there are at least a few people out there who still have them. When you get the modem installed initially, they give you some default credentials. Something like user/user or admin/admin. Using these credentials, you are able to access the device and many of the features that it has to offer you. What you are not able to do is access the menus where you can change how the router is actually configured for internet access, change the master password, or prevent Time Warner from accessing your modem, and subsequently, your network. To fix this, you just need to know the following secret...
The real administrator username that comes configured on these modems when you get them from Time Warner is the last eight digits of the unit's MAC address sans the colons separating out the values. This is unique to your device, but can be found pretty easily by looking at the user interface that you do have access to. The password for this user is "c0nf1gur3m3". Use that and you should be in. Feel free to change the password while you're in there to keep the Time Warner folks out.
One other kinda secret thing to note is that if you do want to change how the router is configured for internet access, you will need to go to http://192.168.0.1/TlModeChange.asp on your router to do so. Once there, you can change it to Bridge mode, NAT mode, Router mode, or NAT Router mode depending on what you are looking to do with it. Hope you enjoyed this simple solution for getting the real administrator access to Time Warner RoadRunner's Ubee cable modem.
***Update: If the above isn't working for you on Time Warner Cable, try one of these suggestions from the comments:
- Username: admin / Password: cableroot
- Username: technician / Password: C0nf1gur3Ubee#
- Username: admin / Password: C0nf1gur3Ubee#