I have been working with GNS3 (Graphical Network Simulator) in some of my classes. It is a fine tool with which one can build simulated computer networks using various routers and switches. The software program itself is free, however the Cisco IOS images are not included and must be found elsewhere due to copyright issues. This detail is a bit of a pain, but not too bad. Once the program is set up and the appropriate IOS images are loaded, the console functions exactly like whatever router is being simulated. This includes running which ever terminal program is preferred, e.g. hypertermial, putty, or if using the Linux version, x-term, etc.
The advantages to this over something like Cisco’s Packet Tracer program are many. In Packet Tracer, certain functions are locked out and generally there is only one acceptable way to complete any given task. With GNS3, the IOS is fully functional, which means that experimentation and failure are available to play with. Failure is a great way to learn things in any hands on environment. The advantage of virtual failure is that only you know about it.
For real world applications, this means that router and switch configurations can be created, tested and tuned ahead of time then loaded into working devices, saving downtime and potentially handfuls of hair.
A few things about using GNS3, the PC idle tuning is required. Each instance of IOS assumes that the entire processor is available to use, thus starting several routers can wonk a PC’s processor to 100% and windows will never fully recover. Secondly, when starting each router, wait 10 to 20 seconds before starting another one. Again, this has to do with the way IOS uses processors. Also, to save time, store the IOS image as a decompressed file. This saves quite a bit of time on startup. Finally, do not forget to copy running config to startup config. Even though GNS3 says it is saving the router configs, it does not save the running config unless you issue the copy run start command, just like a real router.
Wireshark is a packet protocol analyzer that is free for download and runs on Windows, Linux, BSD, OS X and Solaris. In the evolving broadcasting studio, computer networks are the backbone of the facility. Not just on the office side of the house, but also in the broadcast origination side as well. Today, almost everyone uses some type of computer automation system running on a network. In addition, new technologies such as, AoIP consoles, VoIP phone systems, audio and video routing, remote control, off site monitoring, audio processing, etc continue to develop. Because of this, more and more broadcast engineering work is falling into the computer and networking realm.
Like anything else, networks can fail. Failure modes can originate from both the physical side, e.g. wiring, connectors, patch bays, network interface cards or the software/protocol side. Being able to diagnose problems quickly and take remedial action is important. On the networking side, if a physical problem has been ruled out, then the problem exists with a protocol. That is where Wireshark becomes useful; it takes the guess work out of networking protocol troubleshooting.
Wireshark packet protocol analyzer has the following features (from their website):
Deep inspection of hundreds of protocols, more are in development
Live capture and offline analysis
Standard three-pane packet browser
Versions available for Windows, Linux, OS X, Solaris, FreeBSD, NetBSD, and others OS
Captured network data can be browsed via a GUI, or via the TTY-mode TShark utility
Filtering by protocol, IP address, MAC address, frame type, sequence number, etc
Read/write many different capture file formats: tcpdump (libpcap), Pcap NG, Catapult DCT2000, Cisco Secure IDS iplog, Microsoft Network Monitor, Network General Sniffer® (compressed and uncompressed), Sniffer® Pro, and NetXray®, Network Instruments Observer, NetScreen snoop, Novell LANalyzer, RADCOM WAN/LAN Analyzer, Shomiti/Finisar Surveyor, Tektronix K12xx, Visual Networks Visual UpTime, WildPackets EtherPeek/TokenPeek/AiroPeek, and others
Capture files compressed with gzip can be decompressed on the fly
Live data can be read from Ethernet, IEEE 802.11, PPP/HDLC, ATM, Bluetooth, USB, Token Ring, Frame Relay, FDDI, and others (depending on your platform)
Decryption support for many protocols, including IPsec, ISAKMP, Kerberos, SNMPv3, SSL/TLS, WEP, and WPA/WPA2
Coloring rules can be applied to the packet list for quick, intuitive analysis
Output can be exported to XML, PostScript®, CSV, or plain text
Here is a quick video with some tips and tricks on using Wireshark:
A few things to keep in mind with the physical connection. Connecting a computer to a switchport will establish collision domain between the switchport and the computer which is also called a network segment. The computer NIC will see all traffic on that collision domain and all broadcast traffic on the network or sub network that the switch is attached to. If there is a suspected problem with a particular network segment, the Wireshark computer needs to join that collision domain.
This can be done most simply by installing wireshark on the host in that domain. Alternately, a hub can be used to add another host to the collision domain. Or, if it is a managed switch, there may be a provision to send all traffic on the switch out of one designated port. This is called ‘port mirroring’, ‘port monitoring’, ‘Roving Analysis’ (3Com), or ‘Switched Port Analyzer’ or ‘SPAN’ (Cisco).
A quick tutorial on what to look for when using Wireshark, Part A:
And briefly, that is how it is done. There are many more videos on youtube and elsewhere if interested in learning more.
As data transfer technology progresses, so do cable types. Category 6 UTP copper cable is commonly used today in ethernet installations where 1000BaseT (or gigabit ethernet) systems are required. Cat 6 cable has a certified bandwidth of 250 MHz (500 MHz for Cat6a). Category 6 cable is a newer version of Category 5 and 5e cable wherein the wire pairs are bonded together and there is a separator to keep each pair of wires the same distance apart and in the same relationship to each other. The four twisted pairs in Cat 6 cable are also twisted within the overall cable jacket.
Category 7 cable is much different from its predecessors. It has an overall shield and individual pairs are shielded:
Shields on individual pairs are required to reduce cross talk (FEXT, NEXT). It also requires special shielded connectors called GG45 plugs and jacks. Pinouts and color codes are the same as gigabit ethernet (Category 5e and 6) however, Category 7 (ISO 11801 Class F) jacks and plugs also have to contacts on the corners of the connector or jack. This allows better shielding. A small switch in the jack senses when a category 7 type connector is inserted and switches to the corner contacts, thus keeping jacks and patch panels backwards compatible with Category 5/6 cables.
Category 7 cable is rated for 600 MHz bandwidth (1000 MHz for 7a) which translates to 10 GB ethernet. This was previously the domain of fiber cable. Copper cable has some advantages over fiber; lower propagation delays, requires less complicated equipment, copper is less expensive than fiber and more durable. It is nice to have the flexibility to use copper cable on 10 GB ethernet for runs of 100 meters or less. Longer runs still require fiber.
Category 7 and 7a cable looks remarkably similar to the older Belden multipair “computer cable” pressed into service as audio trunk cable seen so often in older studio installations.
Most broadcast facilities have an engineering department or service and an IT department or service which are separate. There is often a fuzzy line between what machines belong strictly to engineering and what belongs to IT. There are several different systems that have network interfaces but are not generally considered computers and fall squarely in the engineering department. These include such equipment as transmitters, satellite receivers, EAS machines, IP based audio routers and audio consoles and IP audio CODECS. In many cases, windows based automation systems and servers also fall under the responsibility of the engineering department.
As the recent incidents of network intrusions into vulnerable EAS machines shows, after installation, steps must be taken to secure networked equipment from malicious or accidental intrusions. The aforementioned EAS intrusion was bad but it could have been much worse.
Anything with a network interface can be exploited either internally or externally and either by purpose or accident. The threat plain looks like this:
Every unauthorized network access incident falls somewhere on this plain. An unauthorized network intrusion can be as simple as somebody using the wrong computer and gaining access to back end equipment. It can also be the hacker or cracker from a foreign country attempting to breach a fire wall.
Basic network security falls into these categories:
Physical security of machine or server room
Security against internal accidental or malicious use
Security against external intrusion
Protection against malicious software exploitation
The first category is the easiest to understand. Physical security means securing the server room through locking doors and preventing crawl over/under entries. Security cameras and monitoring is also a part of physical security. Something that is often neglected is extended networks that bridge to transmitter sites. Non-maned off site facilities that have network access are a vulnerable point if multiple clients or tower tenants have access to the same room. Locked equipment racks and video cameras are two ways to secure non-maned transmitter sites. Also, when using good quality, managed switches at transmitter sites, switchport security features can be enabled and unused switchports shutdown.
Accidental or malicious internal intrusions can be reduced or eliminated with proper password policies. The first and most important password policy is to always change the default password. There are lists of default router and switch passwords available online. The default passwords for EAS machines and other equipment is published in owner’s manuals and most broadcast engineers know them by heart. Always change the default password, if you do nothing else, do this.
Other password policies include such things as minimum password length, requiring special characters, numbers and both upper and lower case letters. Even taking those steps, passwords are still vulnerable to dictionary attacks. To prevent a dictionary attack, the login attempts should be limited to five or so with a thirty minute freeze out after the attempt limit is reached.
External intrusion can come from a number of different sources. Unsecured WIFI is the easiest way to gain access to a network. Always secure WIFI with WPA or WPA2 AES encrypted pre-shared key. This will keep all but the most determined intruders out. Other external threats can come from man in the middle attacks. IP bridges and WIFI must always be encrypted.
External attacks can also come over the wired network. Most small routers have default network and password settings. I have started moving away from using 192.168 internal networks. Router firewalls and personal software firewalls are effective but not foolproof. Software updates need to be performed regularly to be effective. One recently discovered exploit is UPnP, which is enabled on many home and small office routers. UPnP (Universal Plug-n-Play) SSDP (Simple Service Discovery Protocol) can be exploited of exposed to the public network side of the router. ShieldsUP! by Gibson Research Corporation is a good evaluation tool for router exploits, leaks and phone homes. They also have links to podcasts and youtube videos.
Disabling unused features on routers is a good security policy. Features such as DHCP, DNS, SNMP, CDP, HTTP server, FTP server etc are all vulnerable to exploitation of one form or another. Turning off those protocols that are not in use will eliminate at least a portion of those threats.
Finally, worms, bots, viruses and other malicious software can come from anywhere. Even reputable websites now have drive-bys in linked advertizing banners. Non-windows operating systems are less vulnerable to such programs, but not immune. All windows machines and servers that are in anyway connected to the internet need to have updated antivirus software. Keyloggers can steal passwords and send them to bad places where people have nefarious intent.
There are entire books, standards and upper level classes taught on network security. This less than 1,000 word article barely brushes the surface, as the titles says, these are but a few very basic ways to implement a security policy. It is important for technical managers and engineers to learn about, understand and implement security policies in broadcast facilities or suffer the consequences of complacency.