I originally posted this at http://www.318.com/TechJournal Ever get an email from yourself that you didnâ€™t send? Ever get spam from someone that you canâ€™t reply to? Using the settings of an email program, it is possible to pretend to be anyone that you would like. If you want to send email from firstname.lastname@example.org then that is entirely possible. Finding the address of who actually sent email is easy, but ensuring the identity of the sender is not part of standard email. This is where the protocols for PGP, Pretty Good Privacy, and GPG, or GNU Privacy Guard, come into play. GPG and PGP are Open Source suites of applications allowing senders to digitally sign outgoing emails in such a way that it is highly unlikely that anyone else could have sent the message. In order to use their digital signature senders are required to enter a password to send the message. It is also possible to use GPG to encrypt email using a shared password. This allows for forcing a password to both send and receive the message. Encrypting messages ensures both the identity of the sender and the identity of the receiver. Anyone that intercepts a message in transit or finds the message on either system at a later date can open the message without the password to do so. GPG and PGP provide strong encryption measures to ensure privacy over public mediums of messaging. Email is not the only use for this. GPG can also be used to encrypt a file before using transferring it using other methods such as FTP or the web. The commercial version of PGP can also be set up to encrypt certain instant messaging traffic and an entire hard disk.
To setup a template user account, create the directory structure and settings here: /System/Library/User Template
A checksum can be used to determine if a file has been tampered with at a later date. To run a checksum use the following command:
openssl dgst -HASHTYPE path_to_fileHASHTYPE would then be md2, md4, md5, mdc2, rmd160, sha or sha1. Let’s go ahead and do a checksum of our smb.conf file:
openssl dgst -md5 /var/db/smb.confYou should then see output similar to the following:
Below is the presentation I gave at the LayerOne Security conference in May of 2006 in Pasadena, CA. Topic was Mac OS X Server Security. mac_server_security-layerone-charles_edge
Web Application Firewalls, or WAFs, are firewalls for web application. They monitor web traffic and decide whether to allow or deny specific requests. IIS web servers (OWA), Apache, WebObjects, Lasso and other web servers will likely end up working with them, although I’ve only tested IIS and Apache at this point.
Only the MDCs are required to see the Metadata LUN. The rest of the clients receive the extents from the MDCs to write to the “data” LUNs. Yes the clients need to see the Data LUNs, but clients do not need to see Metadata LUNs. Before you can upgrade a client to an MDC though, verify that you can see the Metadata LUNs if you are zoning this way. It’s the most secure way but requires a little extra zoning.
I originally posted this at http://www.318.com/TechJournal â€œWeâ€™re not a high profile target.â€ Weâ€™ve heard it countless times before, but that argument just doesnâ€™t hold up any more. There are malicious applications out there that scan entire chunks of the internet for computers that are vulnerable to specific attacks. Most small businesses hold the position that because they are not a â€œhigh profile targetâ€, such attacks do not represent a threat to them. In terms of modern security, the attitude of â€œWeâ€™re not NASA, and therefore our information is not confidential enough to protectâ€, just doesnâ€™t hold up. The security attacks described in this article are sometimes less about your competition covertly gaining access to your trade secrets or client/job data, and more about random entities exploiting your precious technology resources. In addition to stealing confidential data, Internet hacks can compromise the performance of your technology assets with Bots and other Spyware as well as use up most if not all of your Internet bandwidth. all of these potential symptoms cost business in lost productivity and the direct costs of having to resolve these performance issues. No device thatâ€™s open to the webâ€™s protocols is secure Nearly every router and firewall, from consumer grade to professional grade has the option to create what is called a Demilitarized Zone, or a DMZ. DMZs offer the ability to quickly split an Internet connection to many computers while still moving all incoming traffic into a specific computer. Often, the standard setup is to DMZ a server in a small office that has one server. This is especially common when this server is being used for multiple purposes (such as a web server, FTP server, mail server, etc.). Each one of these services uses a specific port to differentiate incoming requests. For example, web traffic typically uses port 80. When selecting ports coming into a network, it is important to remember that the less traffic that comes into a network, the better. However, when using DMZ, all ports are open, giving attackers a virtually limitless amount of ports to scan, infiltrate, and exploit. Selectively granting access is now a must. Attackers are also using Google to find unsecured stations that accidentally get crawled (a book on hacking with Google was just released ). If one of your systems is compromised by a hacker and used to launch an attack on another computer, then those victims have every right to sue you for damages in court. Another excuse that doesnâ€™t hold up any more is, â€œItâ€™s a Mac, and theyâ€™re secure.â€ Itâ€™s true that Mac OS X has been labeled the â€œmost secureâ€ OS on the market. However, the MOST secure doesnâ€™t mean FULLY secure. Macs are going to become higher profile targets in that more and more attacks can be launched from them, even if there are still fewer people attacking them than Windows. Since nothing thatâ€™s open to the web is secure and most every business relies on open connections to the Internet to remain competitive, Three18 recommends that our clients keep as many copies of everything important in as many locations as they can, as well as having routine security audits and port scans. Rotating redundant offsite backup solutions are critical. The best way to protect your data is to back it up. When evaluating the costs, ask yourself how much money one dayâ€™s data is worth to your company. A week? A month? An hour? Then, make decisions on how often to back up based on the backup cost vs. the cost to recreate the data. Protecting your assets requires a plan for both your perimeter and your data as well as your technology assets. Now having said all of this, the real cost of security is inconvenience. The rule of thumb is that the more security is applied to an environment, proportionally the less convenient access to that environment becomes. More often than not, the cost of 100% security is too high for two reasons: it limits the convenient access of a companyâ€™s data both internally and remotely, which often is required to support a companyâ€™s business logic as applied to technology; and it simply costs too much money to implement. The best analogy is that of the homeowner who chooses to get an alarm system and put high quality locks on all the doors of his/her home, but opts to leave all of the windows on the homeâ€™s first floor without bars. In this case, the home is safe from the typical entry points, but at the price of maintaining a nice view through the windows, the home is vulnerable at the same time. Sometimes less than 100% is good enough. Security, as with most business decisions, is a risk-based decision. Factors of costs, convenience and liability must all be considered to fully understand the implications of business security.
Ever been hacked? Had information stolen? Who do you turn to? What do you do? No matter what the level, a security breech has occurred and action must be taken to ensure a repeat offense doesnâ€™t happen. The first reaction to a security breech is to isolate it and fix it as soon as possible. However, writing to the systems in any way can cause clues to be overwritten. Therefore it is important to discover the identity of the attacker. The more quickly that forensic analysis is performed the more likely that the attacker, vandal or thief will be apprehended. One of the best places to start in analysis is making a copy of the system that hasnâ€™t been written to. For Windows this is done using a program like Ghost. On the Mac platform using Carbon Copy Cloner or the Disk Utility to create an image is a good move. It is best to get a copy of your system as soon after a security incident as possible. On local systems, there are some valuable pieces of information that can be obtained about the identity of the person stealing data. This can be anything from the IP address of the attacker to the name of the drive theyâ€™re transferring data to. On many Operating Systems valuable logs or cached files are overwritten on a routine basis. If a clone is made, it is often best to create a clone, or a replica of the system in its current state, as soon as possible. If itâ€™s a server, then the logs of the server provide good clues as to where to look for the perpetrator. Once again it is helpful to create a clone of the system. However, this is not always possible on production servers. Copying the log files is the next best thing. Firewalls can provide good clues as well. The logging cycles on firewalls typically store data for a shorter period of time than on workstations or servers. Creating a screen shot in PDF format of the firewalls logs or exporting the logs into a text file is a good starting point. Firewalls typically provide good information on what addresses are communicating with a network. This makes them good at specifically determining the identity of the attacker and according to logging levels, the attacks used. No matter what the issue, time is of the essence. Contacting a professional to help is a good idea. Getting the FBI or the LA County District Attorneys office involved can take time and this can cause clues to be damaged, lost or destroyed. IT professionals can also assist in creating a chain of custody on the equipment that can later be used in court when and if the person whoâ€™s invaded your privacy is apprehended and put to trial.
War dialing or wardialing is a method of automatically scanning telephone numbers using a modem, usually dialing every telephone number in a local area to find out where computers or fax machines are available, then attempting to access them by guessing passwords. ToneLoc was a popular wardialing computer program for MS-DOS written in the early to mid-1990s by two programmers known by the pseudonyms Minor Threat and Mucho Maas. The name ToneLoc was short for â€œTone Locatorâ€ and was a word play on the name of the rap artist known as Tone LÅc. The utility was created for the purpose of scanning for dial tones or modem carriers in order to find PBXes, long distance carriers, or other modems. THC-SCAN In the cracking scene of the 1980s, demon dialing was a technique by which a computer would repeatedly dial a number (usually to a crowded modem pool) in an attempt to gain access immediately after another user had hung up. Wardriving is searching for Wi-Fi wireless networks by moving vehicle. It involves using a car or truck and a Wi-Fi-equipped computer, such as a laptop or a PDA, to detect the networks. It was also known (as of 2002) as â€œWiLDingâ€ (Wireless Lan Driving, although this term never gained any popularity and is no longer used), originating in the San Francisco Bay Area with the Bay Area Wireless Users Group (BAWUG). It is similar to using a scanner for radio. Many wardrivers use GPS devices to measure the location of the network find and log it on a website (the most popular is WiGLE). For better range, antennas are built or bought, and vary from omnidirectional to highly directional. Software for wardriving is freely available on the Internet, notably, NetStumbler for Windows, Kismet for Linux, and KisMac for Macintosh. Wardriving was named after wardialing (popularized in the Matthew Broderick movie WarGames) because it also involves searching for computer systems with software that would use a phone modem to dial numbers sequentially and see which ones were connected to a fax machine or computer, or similar device. (Audio commentary on the Wargames DVD says that wardialing was named after the movie and the software did not openly exist before the movie.)