Basically, the guy took an application (“pre-approved credit line – just sign here and return!”), cut it up into many pieces, reassembled it with tape, filled it out with a change of address and change of phone number, mailed it in, and got the approved, ready-to-use credit card back in the mail at the new address.

Most people probably don’t tear those things up, let alone shred, incinerate and bury them like I prefer to.  And while I don’t want anyone constantly living in fear that their identity will be stolen, there are some reasonable precautions to take.  After all, it’s not paranoia if they’re really after you.

1) Apparently, all you have to do to throw off the facial recognition software that protects us from identity theft or worse, is smile:

The Indiana Bureau of Motor Vehicles is restricting glasses, hats, scarves — and even smiles — in driver’s license photographs. The new rules imposed last month were deemed necessary so that facial recognition software can spot fraudulent license applications, said BMV spokesman Dennis Rosebrough.

And then he had the gall to spin it as an improvement, since it would be horrible to admit that humans had done a better job:

The new technology represents an advancement of what the BMV already was doing, Rosebrough said. BMV employees always have looked at the old photo of a person to see if it looked like the person seeking a new license.

FAIL.

2) I was at a local video store yesterday, trying to rent a video using Anna Lisa’s account. I gave the cashier her phone number and name, and he said he’d have to call her to verify that it was okay for me to rent on her account. When she didn’t pick up, I offered to call her on my cell phone (in case she wasn’t picking up the call from an unknown number), and the cashier said, “okay, yeah, just ask her if it’s okay and then you can tell me what she said.”

FAIL.

As his team prepares a final judgment on whether he can keep using e-mail, perhaps even in a read-only fashion, several authorities in presidential communication said they believed it was highly unlikely that he would be able to do so.

Diana Owen, who leads the American Studies program at Georgetown University, said presidents were not advised to use e-mail because of security risks and fear that messages could be intercepted.

“They could come up with some bulletproof way of protecting his e-mail and digital correspondence, but anything can be hacked,” said Ms. Owen, who has studied how presidents communicate in the Internet era. “The nature of the president’s job is that others can use e-mail for him.”

Surely there’s some middle ground to keep a President as tech-savvy as Barack Obama from being forced off of e-mail altogether? I mean, this is the guy who announced his VP pick by SMS text message, for crying out loud.

Here are some scenarios to explore:

It’s clear that the President could probably not have a public e-mail address that was directly addressable from any other e-mail account, and that was expected to be read by the President himself. It would be flooded hourly with requests, comments and threats from around the world, and instantly become useless as a form of effective correspondence.

The White House does already have a public e-mail address, comments@whitehouse.gov, but they make it clear that they probably won’t respond, and that you can’t even send graphics or attachments. (How the heck are we supposed to share funny photos of our cats, then!?) I suspect that they have some serious hardware and network capacity dedicated just to receiving and processing mail sent to that address – there’s no way that much or any of it will make it’s way to a laptop on the President’s desk.

Here’s one option: the President could have a private e-mail address that is directly addressable within the executive branch of the U.S. government, using a private domain (e.g. president@whitehouse.gov.private) and set of mail exchangers. This would allow Barack Obama to e-mail with other government officials who have established addresses, taking advantage of the utility of e-mail for more direct conversations and unfiltered access to information. Anyone wishing to reply to his messages could send e-mail like usual, and as long as they were on that private network, it would go through.

When the President wanted to send e-mail outside of this private network of addresses to a regular e-mail address, his message would be intercepted by a correspondence manager who could insure that the message was free of sensitive information, and that it would not be a source of embarrassment for anyone if delivered to the wrong hands.

Here’s a hard part: should that recipient with a regular e-mail address be able to reply? If it’s to some other regular e-mail address, e.g. “barack.obama@whitehouse.gov,” then that just wouldn’t do, because the address would eventually leak out and become just as much a target for a flood of messages as “comments@” might be now.

BUT, a variation on that might work. The President’s outgoing message could be dynamically rewritten to appear to come from an address that had an expiration date, and that used a unique hash token to prevent guessing it – this is common in software like TMDA and other special e-mail submission systems. So instead of

From: barack.obama@whitehouse.gov

it might be

From: barack.obama-JLK23ADSF23423K@whitehouse.gov

and it would only work for, say, up to 70 hours (or a week, or…). The next time the President responds, it comes from a different address, which also only works for 70 hours. In this manner, correspondents of the President with regular e-mail addresses could carry on an e-mail conversation and not worry about the address becoming public in a way that mattered. If someone tried to e-mail the expired address, they’d get a bounce back saying, “sorry, please e-mail comments@whitehouse.gov.”

For someone with a regular e-mail address to initiate a conversation with the President, we’d have to take a slightly different approach – the addresses would essentially need to go on a pre-approved list of correspondents whose messages would be allowed to go to “comments@whitehouse.gov” and be automatically filtered straight on through to the President’s internal (not publicly addressable) account.

So, to initiate an e-mail conversation with the President, you’d either have to be a known associate with a known e-mail address, or you’d have to be an employee of the Federal government with an established e-mail account. Better than nothing, right?

Another pitfall: the President would still generally have to consider any content sent via e-mail, whether it was to a private internal address or not, as up for public scrutiny. Until whomever controls Presidential records passes a law identifying some kinds of Presidential communications as truly private, it’s just the reality of the thing. And even if that did happen, we all know how easy it is for an e-mail that wasn’t meant for you to make it into your hands, so it’s probably just safe to assume that’s happening anyway.

From the article:

For all the perquisites and power afforded the president, the chief executive of the United States is essentially deprived by law and by culture of some of the very tools that other chief executives depend on to survive and to thrive. Mr. Obama, however, seems intent on pulling the office at least partly into the 21st century on that score; aides said he hopes to have a laptop computer on his desk in the Oval Office, making him the first American president to do so.

Well, that’s my brain dump on how it might work to let the President of the United States of America have an e-mail account he can actually use. What do you think?

Recent media reports, including a March 10th article in the Wall Street Journal, show us how much information spy agencies are allowed to legally collect and monitor:

• Recipient and sender address, subject line, timestamp of e-mail messages
• Internet sites visited and searches conducted
• Incoming and outgoing numbers dialed on cell and regular phones, length of calls, where you physically were when a cell phone call happened
• Pretty much everything about your financial transactions

Makes you wonder what’s actually happening beyond the law’s provisions. Again, I’ll generally assume the worst.

For example, earlier this week, a user visited my website without any referring URL information. This means they probably entered the address directly in their browser’s location bar, but it could also mean they followed a bookmark, or are actively trying to hide where they came from. As soon as they got to my site, they started searching for the word “congress” in my content. When I traced the IP address, it went back to a location in McLean, Virginia, which is the home of the Central Intelligence Agency.

So what can we conclude from this? Obviously, a CIA operative was investigating my website because in my ramblings about politics and the government, I’ve clearly come too close to the truth about a cover-up related to U.S. energy policy and the War on Terra, and now they’re coming to take me away, ha-ha.

Fun game, eh?

Some recent search terms people have used in Google or other search engines that cause them to end up on my website and weblog:

• so many people I want to kill (YIKES!)
• how do I remove candle wax from my cat
  (probably because of this…but YIKES!)
• ramada worldwide keith j. pierce
• summary of hope for the flowers by trina paulus (dude, it’s a really short book, just read it!)

And the list goes on.

Don’t worry, I take the privacy of my website visitors very seriously, and I won’t reveal the usage information of any particular user (though my friend at the CIA might prefer otherwise). That said, be careful where you browse, and if you don’t want someone watching you on the other end, use protection.

If you’re interested in having me speak to your organization about computer security, please see my page on speaking requests. My company, Summersault, is available for certain kinds of security consulting services.

This document is not an authoritative or comprehensive one; you should check with the information management policy of your particular institution for steps to follow to secure your system. The author of this document shall not be liable for any damage, direct or indirect, incurred in the following of this advice. If you have experienced a security breach, you should contact an experienced security professional to evaluate recovery options.

Contents

1. Physical Security
2. Network Security
3. Protocols / Services
4. User Security
5. Data Storage Security
6. Passwords
7. System Administration

1. Physical Security

Physical security is the MOST important part of maintaining the security of a computer system, and is often overlooked by careless system administrators who assume their occasional proximity to a system is enough protection. This may be sufficient for some systems, but in most cases, there are more factors to be considered before a system can be called physically safe and secure.

• Is the system located on a sturdy, stable surface as close to the ground as possible?
• Is the system safe from excessive sunlight, wind, dust, water, or extreme hot/cold temperatures?
• Is this system located in a monitored, isolated area that sees little human traffic?
• Is the room/building in which the system is located secured by lock and alarm system to which only a few trusted personnel have access? Are these locks and alarms locked and armed during off-hours?
• Is the terminal of the system secured to prevent someone from casually walking up to the system and using it (even if just for a few seconds)? Are all users logged out from the terminal?
• Are the power and reset switches protected or disabled?
• Are any input devices to the system secured/turned off: are all removable disk drives locked/secured? Are the parallel/serial/infared/USB/SCSI ports secured or removed? Are any attached hard drives physically locked down to the system?

2. Network Security

Network security is the SECOND MOST important part of maintaining a system security. While good physical security can go a long way, if you operate your system in a networked/multi-user environment, the system is many times more susceptible to outside attacks than a standalone system. Network security is also harder to evaluate because it requires a thorough understanding of the various components and layers of your system and all the external services that interact with your system.

• Physical network: is the network connection a secure “pipe” with no danger of unauthorized rewiring? Do only authorized personnel have physical access to the physical network to which the system is attached? Do you know and trust all of the various points where your physical network connection is managed/administered by another person or entity?
• Are the other systems on the same network physically and electronically secure? If your system is reasonably secure but another system on the network is not, your system’s vulnerability is increased greatly.
• Approved Network Traffic
  • Do you know the names, functionality, vendor, and nature of the software on your system that participates in any network activity? Have you checked all the vendors for security patches, and do you regularly receive security updates about patches/vulnerabilities to the software you use in a networked environment?
  • Have you thoroughly tested any and all services that interact with the network to insure that they do not, by default, provide any unauthorized users with useful security information that could be used to attack the system?
  • Do you effectively limit your users` abilities to make sensitive information about the system available over the network?
  • Do you only allow trusted users shell/command line access to your system?
  • Are you aware of any security holes created by certain software packages interacting with each other?
  • Do you keep sufficient logs of all approved network activity?
  • Are you aware of all the software that should be interacting with the network, the port numbers they use, the size and location of their binaries, etc.?
  • Do user accounts that are accessible over the network regularly have their passwords changed?
  • Do you encrypt sensitive data that is transferred over the network?
• Unapproved Network Traffic
  • Do you regularly check for repeated unauthorized attempts to connect to your system over a network? Do you keep sufficient logs of all network activity related to your system?
  • Do you regularly check for unauthorized programs running on your system that could potentially allow a user to connect over the network?
  • Do you monitor for excessive or unusual network activity that comes to your system?

3. Protocols / Services

Once you are past the physical and network layers of your system, the next category of evaluation is perhaps one of the largest; computers are made to compute, and depending the purpose of your system, it will be running many different kinds of software and programs at any point in time. It is likely in most cases that, because all of the software was written by different people with different understandings of security (and because there are always people who know more about security), at least one of those programs has some sort of security hole that could be exploited.

• While it is generally safe to assume that software that comes pre-installed on a new system is reasonably secure, you should always check with software vendors for security patches, release notes, and other relevant information to your particular configuration.
• For any software that you install onto a new system, make sure you are fully aware of the credentials of the vendor, any security patches, existing exploits, and release notes that exist. You should make it a habit to check in with vendors every month or so for new releases that may have security fixes. It’s also a good idea to subscribe to mailing lists for your software, or general mailing lists, that would announce security holes early.
• Misconfiguration is probably the most common cause of someone exploiting a security hole. Most software is written to be reasonably secure, but even the most secure software can be used for unintended purposes if it is poorly configured. Always follow the vendor’s instructions for installing software, and always take notes on any problems you encounter in the configuration process. If a piece of software requires special privileges to be installed or run (e.g. running setuid on a UNIX system), make sure you understand the full implications of having it do so, and any side-effects created in the process. Test your configuration of the software thoroughly; try to break it, try to hack into it, and see if others can do the same.
• If a program accesses sensitive data, make sure that it can only be executed by authorized users, and make sure that any logs or temporary information is stored in a safe place and promptly disposed of; people can do amazing things with the simple information found in a system log file.
• If a piece of software runs as a daemon (i.e. it is constantly running and responds to requests from users locally or over the network), make sure it properly handles buffer overflows, denial of service attacks, and general heavy system load. It’s generally a good idea to have as few services as possible running as daemons, as they allow continuous and typically unmonitored access to your system.
• Be aware of all the services that are supposed to be running on your system, the typical amount of resources (e.g. CPU time, memory, disk space) that they take up. Check for unidentifiable daemons or software, or programs that are unusual in their resource consumption. Remember that most security breaches occur using the existing configuration of a system rather than installing a new one; unless you’re careful, an intruder can manipulate the system to their liking and you won’t notice anything out of the ordinary.
• Run process accounting to keep track of typical software usage patterns of your users.

4. User security

The particulars of user security varies widely with the nature of the system you’re running. In some cases, a system will be an isolated machine performing mostly server functions with very few users who actually log in to the system and use it directly, most of the users thusly being people interacting with the server functions. In other cases, a system might have hundreds of users directly accessing the system simultaneously. Obviously, the degree to which user security is a concern depends largely on the character of your users, but be aware that one user who attempts to breach security, or who has poor security practices, can affect and possibly endanger an entire system.

• Develop a standard method for creating and maintaining user accounts. Develop clear and concise acceptable use policies, and publish them well to your users. Don’t create user accounts for people or organizations whom you have not previously interacted with in some form, or who have been known to have security problems on other systems.
• You should set limits on the amount of resources a user can consume, from number of logins to amount of disk space; make sure that the user cannot cause a security breach or take down the system out of pure stupidity (e.g. a recursive script that creates a 10 M file each time)
• In some cases, you may want to limit the manner in which a user can connect to the system; if you’re providing a terminal login, make sure the terminal itself is secure and reasonably maintained. If you provide direct access via protocols such as telnet, consider running services such as tcp_wrappers or identd that verify the user is connecting from the system they claim to be connecting from.
• Keep accurate logs of user activity; specifically, connection time, connection duration, and the place where they logged in/connected from. In some cases you may want to log more detail with process accounting, user command history, and activity monitoring.
• You should regularly check for irregular user activity; there are many programs available that constantly “patrol” for failed attempts on the part of users to gain administrator privileges, access files that they shouldn’t, or perform other unauthorized tasks.

5. Data storage security

Data and file storage, at first, does not seem to present itself as a security risk; either people have access to files or they don’t! In reality, it turns out that there are many and complicated ways to access the same data on a given system, and a good system administrator should be aware of these schemes.

• Know the file ownership scheme that your system implements; is it group based, user based, role based, or some combination of these? Know the different levels of protection you can apply to files and directories, and be aware of who has access to make changes to these protections.
• Know the general structure of your filesystems, how much is stored where, and who typically accesses what parts of them. Keep logs of disk activity (e.g. significant changes in disk space used) and of any disk problems.
• Make sure that users are only able to access the parts of the system relevant to their use of it; your protection scheme should clearly and easily include a logical and conceptual separation of user and data files from system files.
• Make sure that the file ownership schemes are consistent for various directories (i.e. that the owner of a directory owns all the files in that directory, etc.)
• Insure that users cannot have access to more disk resources than you intend; often user disk quotes are the best solution to this.
• If your filesystems are available via any sort of network or sharing protocol., carefully examine the security of these protocols (see the protocols/services section above). Always check your configuration of these services to make sure that only authorized users and hosts are allowed to access shared data; many configurations by default allow for unauthorized access.
• Always maintain secure backups of a system; the most standard conventional method is to backup files to a tape disk and then to remove that tape from the site to guard against data loss from fire, flooding, etc. In the case of operating systems, it’s a good idea to maintain a known good copy of your operating system’s configuration on a read-only media such as a CD-ROM.
• If you maintain any databases, make sure that the database is accessible only to authorized users, both on the client side (via a data querying tool such as SQLnet) and on the server side (i.e. the actual database files stored on the disk drive of your system). As with other services, make sure any network and sharing of databases is done securely.

6. Passwords

Passwords are the central components in most security schemes; user accounts, sensitive websites, system services are all protected by them. If you know the right passwords, you can gain administrative privileges on a system where you may not even be a user or infiltrate an environment you’ve never even worked with before. They are conventionally accepted as a good way to implement security because they can be incorporated easily into most operating systems and sensitive software, and yet can be made complex enough to be difficult to “crack”, while still being remembered by a user. Their downfall as a security scheme are in their power; one password is all you need to have complete access to an entire system, and passwords CAN be cracked. The best you can do is try to make these two events very unlikely.

• Require unique, complex passwords for all user accounts on your system; it’s not acceptable to have “guest” accounts or other accounts that don’t require any sort of authentication. If an account is not ever used for connection (i.e. that account will never be accessed), remove its ability to login altogether.
• Passwords should contain at least 6 characters and have a combination of letters and numbers, uppercase and lowercase. Passwords should not resemble any word, name, idea, or concept that might appear in any dictionary anywhere in the world. A good example: jY2EHxqy
• Enforce password rotation and expiration; users should never be able to keep a password for more than a few months at a time, as someone could easily (but unnoticeably) brute force hack a password over a long period of time. You should also advise users against using the same password in other places.
• The password file or similar mechanism for storing the passwords should be encrypted, and should not be available to the average user if possible (e.g. via shadowing). If a user can obtain the password file, they can use another system to try to crack the passwords without you noticing.
• Never write passwords down or store them in anything but human memory.
• System passwords should be changed at least once a month, and should not be shared with more people than is necessary.

7. System Administration

Quality system administration techniques can make all the difference in security prevention. There’s not a lot required for most modern systems; many do self-checks and keep the system administrator automatically informed of any suspicious changes. But there are still a few general tips to follow:

• Regularly browse through your system, looking at the contents of system directories, logs, and other files. Note file locations, file sizes. Observe the usage patterns of your machine and your users.
• Run cracking tools (such as “CRACK” and “Satan” in the Unix environment) regularly to check for vulnerabilities in your system configuration
• Manually try to break into your system through different means.
• Be aware of persons or groups who may have intentions of breaking into your system.
• Keep your users advised of your techniques and what you expect of them to maintain security.