New versions of this document will be periodically posted to comp.os.linux.answers. They will also be added to the various sites that archive such information, including:

http://www.linuxdoc.org/

The very latest version of this document should also be available in various formats from:

All comments, error reports, additional information and criticism of all sorts should be directed to:

kevin-securityhowto@tummy.com

and

dave@linuxsecurity.com

Note: Please send your feedback to both authors. Also, be sure and include "Linux" "security", or "HOWTO" in your subject to avoid Kevin's spam filter.

No liability for the contents of this document can be accepted. Use the concepts, examples and other content at your own risk. Additionally, this is an early version, possibly with many inaccuracies or errors.

A number of the examples and descriptions use the RedHat(tm) package layout and system setup. Your mileage may vary.

As far as we know, only programs that, under certain terms may be used or evaluated for personal purposes will be described. Most of the programs will be available, complete with source, under GNU terms.

This document is copyrighted (c)1998-2000 Kevin Fenzi and Dave Wreski, and distributed under the following terms:

tjbynum@metalab.unc.edu

In the ever-changing world of global data communications, inexpensive Internet connections, and fast-paced software development, security is becoming more and more of an issue. Security is now a basic requirement because global computing is inherently insecure. As your data goes from point A to point B on the Internet, for example, it may pass through several other points along the way, giving other users the opportunity to intercept, and even alter, it. Even other users on your system may maliciously transform your data into something you did not intend. Unauthorized access to your system may be obtained by intruders, also known as "crackers", who then use advanced knowledge to impersonate you, steal information from you, or even deny you access to your own resources. If you're wondering what the difference is between a "Hacker" and a "Cracker", see Eric Raymond's document, "How to Become A Hacker", available at http://www.catb.org/~esr/faqs/hacker-howto.html.

First, keep in mind that no computer system can ever be completely secure. All you can do is make it increasingly difficult for someone to compromise your system. For the average home Linux user, not much is required to keep the casual cracker at bay. However, for high-profile Linux users (banks, telecommunications companies, etc), much more work is required.

Another factor to take into account is that the more secure your system is, the more intrusive your security becomes. You need to decide where in this balancing act your system will still be usable, and yet secure for your purposes. For instance, you could require everyone dialing into your system to use a call-back modem to call them back at their home number. This is more secure, but if someone is not at home, it makes it difficult for them to login. You could also setup your Linux system with no network or connection to the Internet, but this limits its usefulness.

If you are a medium to large-sized site, you should establish a security policy stating how much security is required by your site and what auditing is in place to check it. You can find a well-known security policy example at http://www.faqs.org/rfcs/rfc2196.html. It has been recently updated, and contains a great framework for establishing a security policy for your company.

Before you attempt to secure your system, you should determine what level of threat you have to protect against, what risks you should or should not take, and how vulnerable your system is as a result. You should analyze your system to know what you're protecting, why you're protecting it, what value it has, and who has responsibility for your data and other assets.

Create a simple, generic policy for your system that your users can readily understand and follow. It should protect the data you're safeguarding as well as the privacy of the users. Some things to consider adding are: who has access to the system (Can my friend use my account?), who's allowed to install software on the system, who owns what data, disaster recovery, and appropriate use of the system.

A generally-accepted security policy starts with the phrase

“ That which is not permitted is prohibited”

This means that unless you grant access to a service for a user, that user shouldn't be using that service until you do grant access. Make sure the policies work on your regular user account. Saying, "Ah, I can't figure out this permissions problem, I'll just do it as root" can lead to security holes that are very obvious, and even ones that haven't been exploited yet.

rfc1244 is a document that describes how to create your own network security policy.

rfc1281 is a document that shows an example security policy with detailed descriptions of each step.

Finally, you might want to look at the COAST policy archive at ftp://coast.cs.purdue.edu/pub/doc/policy to see what some real-life security policies look like.

This document will discuss various means with which you can secure the assets you have worked hard for: your local machine, your data, your users, your network, even your reputation. What would happen to your reputation if an intruder deleted some of your users' data? Or defaced your web site? Or published your company's corporate project plan for next quarter? If you are planning a network installation, there are many factors you must take into account before adding a single machine to your network.

Even if you have a single dial up PPP account, or just a small site, this does not mean intruders won't be interested in your systems. Large, high-profile sites are not the only targets -- many intruders simply want to exploit as many sites as possible, regardless of their size. Additionally, they may use a security hole in your site to gain access to other sites you're connected to.

Intruders have a lot of time on their hands, and can avoid guessing how you've obscured your system just by trying all the possibilities. There are also a number of reasons an intruder may be interested in your systems, which we will discuss later.

This document has been divided into a number of sections. They cover several broad security issues. The first, Section 3, “Physical Security”, covers how you need to protect your physical machine from tampering. The second, Section 4, “Local Security”, describes how to protect your system from tampering by local users. The third, Section 5, “Files and File system Security”, shows you how to setup your file systems and permissions on your files. The next, Section 6, “Password Security and Encryption ”, discusses how to use encryption to better secure your machine and network. Section 7, “Kernel Security” discusses what kernel options you should set or be aware of for a more secure system. Section 8, “Network Security”, describes how to better secure your Linux system from network attacks. Section 9, “Security Preparation (before you go on-line)”, discusses how to prepare your machine(s) before bringing them on-line. Next, Section 10, “What To Do During and After a Breakin”, discusses what to do when you detect a system compromise in progress or detect one that has recently happened. In Section 11, “Security Sources”, some primary security resources are enumerated. The Q and A section Section 13, “Frequently Asked Questions”, answers some frequently-asked questions, and finally a conclusion in Section 14, “Conclusion”

The two main points to realize when reading this document are:

Many modern PC cases include a "locking" feature. Usually this will be a socket on the front of the case that allows you to turn an included key to a locked or unlocked position. Case locks can help prevent someone from stealing your PC, or opening up the case and directly manipulating/stealing your hardware. They can also sometimes prevent someone from rebooting your computer from their own floppy or other hardware.

These case locks do different things according to the support in the motherboard and how the case is constructed. On many PC's they make it so you have to break the case to get the case open. On some others, they will not let you plug in new keyboards or mice. Check your motherboard or case instructions for more information. This can sometimes be a very useful feature, even though the locks are usually very low-quality and can easily be defeated by attackers with locksmithing.

Some machines (most notably SPARC's and macs) have a dongle on the back that, if you put a cable through, attackers would have to cut the cable or break the case to get into it. Just putting a padlock or combo lock through these can be a good deterrent to someone stealing your machine.

The BIOS is the lowest level of software that configures or manipulates your x86-based hardware. LILO and other Linux boot methods access the BIOS to determine how to boot up your Linux machine. Other hardware that Linux runs on has similar software (Open Firmware on Macs and new Suns, Sun boot PROM, etc...). You can use your BIOS to prevent attackers from rebooting your machine and manipulating your Linux system.

Many PC BIOSs let you set a boot password. This doesn't provide all that much security (the BIOS can be reset, or removed if someone can get into the case), but might be a good deterrent (i.e. it will take time and leave traces of tampering). Similarly, on S/Linux (Linux for SPARC(tm) processor machines), your EEPROM can be set to require a boot-up password. This might slow attackers down.

Another risk of trusting BIOS passwords to secure your system is the default password problem. Most BIOS makers don't expect people to open up their computer and disconnect batteries if they forget their password and have equipped their BIOSes with default passwords that work regardless of your chosen password. Some of the more common passwords include:

j262 AWARD_SW AWARD_PW lkwpeter Biostar AMI Award bios BIOS setup cmos AMI!SW1 AMI?SW1 password hewittrand shift + s y x z

I tested an Award BIOS and AWARD_PW worked. These passwords are quite easily available from manufacturers' websites and http://astalavista.box.sk and as such a BIOS password cannot be considered adequate protection from a knowledgeable attacker.

Many x86 BIOSs also allow you to specify various other good security settings. Check your BIOS manual or look at it the next time you boot up. For example, some BIOSs disallow booting from floppy drives and some require passwords to access some BIOS features.

Note: If you have a server machine, and you set up a boot password, your machine will not boot up unattended. Keep in mind that you will need to come in and supply the password in the event of a power failure. ;(

The various Linux boot loaders also can have a boot password set. LILO, for example, has password and restricted settings; password requires password at boot time, whereas restricted requires a boot-time password only if you specify options (such as single) at the LILO prompt.

>From the lilo.conf man page:

password=password
              The per-image option `password=...' (see below) applies to all images.

restricted
              The per-image option `restricted' (see below) applies to all images.

       password=password
              Protect the image by a password.

       restricted
              A password is only required to boot the image if
              parameters are specified  on  the  command  line 
              (e.g. single).

Keep in mind when setting all these passwords that you need to remember them. :) Also remember that these passwords will merely slow the determined attacker. They won't prevent someone from booting from a floppy, and mounting your root partition. If you are using security in conjunction with a boot loader, you might as well disable booting from a floppy in your computer's BIOS, and password-protect the BIOS.

Also keep in mind that the /etc/lilo.conf will need to be mode "600" (readable and writing for root only), or others will be able to read your passwords!

>From the GRUB info page: GRUB provides "password" feature, so that only administrators can start the interactive operations (i.e. editing menu entries and entering the command-line interface). To use this feature, you need to run the command `password' in your configuration file (*note password::), like this:

password --md5 PASSWORD

If this is specified, GRUB disallows any interactive control, until you press the key <p> and enter a correct password. The option `--md5' tells GRUB that `PASSWORD' is in MD5 format. If it is omitted, GRUB assumes the `PASSWORD' is in clear text.

You can encrypt your password with the command `md5crypt' (*note md5crypt::). For example, run the grub shell (*note Invoking the grub shell::), and enter your password:

grub> md5crypt Password: ********** Encrypted: $1$U$JK7xFegdxWH6VuppCUSIb.

Then, cut and paste the encrypted password to your configuration file.

Grub also has a 'lock' command that will allow you to lock a partition if you don't provide the correct password. Simply add 'lock' and the partition will not be accessable until the user supplies a password.

If anyone has security-related information from a different boot loader, we would love to hear it. (grub, silo, milo, linload, etc).

Note: If you have a server machine, and you set up a boot password, your machine will not boot up unattended. Keep in mind that you will need to come in and supply the password in the event of a power failure. ;(

If you wander away from your machine from time to time, it is nice to be able to "lock" your console so that no one can tamper with, or look at, your work. Two programs that do this are: xlock and vlock.

xlock is a X display locker. It should be included in any Linux distributions that support X. Check out the man page for it for more options, but in general you can run xlock from any xterm on your console and it will lock the display and require your password to unlock.

vlock is a simple little program that allows you to lock some or all of the virtual consoles on your Linux box. You can lock just the one you are working in or all of them. If you just lock one, others can come in and use the console; they will just not be able to use your virtual console until you unlock it. vlock ships with RedHat Linux, but your mileage may vary.

Of course locking your console will prevent someone from tampering with your work, but won't prevent them from rebooting your machine or otherwise disrupting your work. It also does not prevent them from accessing your machine from another machine on the network and causing problems.

More importantly, it does not prevent someone from switching out of the X Window System entirely, and going to a normal virtual console login prompt, or to the VC that X11 was started from, and suspending it, thus obtaining your privileges. For this reason, you might consider only using it while under control of xdm.

If you have a webcam or a microphone attached to your system, you should consider if there is some danger of a attacker gaining access to those devices. When not in use, unplugging or removing such devices might be an option. Otherwise you should carefully read and look at any software with provides access to such devices.

The first thing to always note is when your machine was rebooted. Since Linux is a robust and stable OS, the only times your machine should reboot is when you take it down for OS upgrades, hardware swapping, or the like. If your machine has rebooted without you doing it, that may be a sign that an intruder has compromised it. Many of the ways that your machine can be compromised require the intruder to reboot or power off your machine.

Check for signs of tampering on the case and computer area. Although many intruders clean traces of their presence out of logs, it's a good idea to check through them all and note any discrepancy.

It is also a good idea to store log data at a secure location, such as a dedicated log server within your well-protected network. Once a machine has been compromised, log data becomes of little use as it most likely has also been modified by the intruder.

The syslog daemon can be configured to automatically send log data to a central syslog server, but this is typically sent unencrypted, allowing an intruder to view data as it is being transferred. This may reveal information about your network that is not intended to be public. There are syslog daemons available that encrypt the data as it is being sent.

Also be aware that faking syslog messages is easy -- with an exploit program having been published. Syslog even accepts net log entries claiming to come from the local host without indicating their true origin.

Some things to check for in your logs:

We will discuss system log data Section 9.5, “Keep Track of Your System Accounting Data” in the HOWTO.

You should make sure you provide user accounts with only the minimal requirements for the task they need to do. If you provide your son (age 10) with an account, you might want him to only have access to a word processor or drawing program, but be unable to delete data that is not his.

Several good rules of thumb when allowing other people legitimate access to your Linux machine:

Many local user accounts that are used in security compromises have not been used in months or years. Since no one is using them they, provide the ideal attack vehicle.

The most sought-after account on your machine is the root (superuser) account. This account has authority over the entire machine, which may also include authority over other machines on the network. Remember that you should only use the root account for very short, specific tasks, and should mostly run as a normal user. Even small mistakes made while logged in as the root user can cause problems. The less time you are on with root privileges, the safer you will be.

Several tricks to avoid messing up your own box as root:

If you absolutely positively need to allow someone (hopefully very trusted) to have root access to your machine, there are a few tools that can help. sudo allows users to use their password to access a limited set of commands as root. This would allow you to, for instance, let a user be able to eject and mount removable media on your Linux box, but have no other root privileges. sudo also keeps a log of all successful and unsuccessful sudo attempts, allowing you to track down who used what command to do what. For this reason sudo works well even in places where a number of people have root access, because it helps you keep track of changes made.

Although sudo can be used to give specific users specific privileges for specific tasks, it does have several shortcomings. It should be used only for a limited set of tasks, like restarting a server, or adding new users. Any program that offers a shell escape will give root access to a user invoking it via sudo. This includes most editors, for example. Also, a program as innocuous as /bin/cat can be used to overwrite files, which could allow root to be exploited. Consider sudo as a means for accountability, and don't expect it to replace the root user and still be secure.

The umask command can be used to determine the default file creation mode on your system. It is the octal complement of the desired file mode. If files are created without any regard to their permissions settings, the user could inadvertently give read or write permission to someone that should not have this permission. Typical umask settings include 022, 027, and 077 (which is the most restrictive). Normally the umask is set in /etc/profile, so it applies to all users on the system. The resulting permission is calculated as follows: The default permission of user/group/others (7 for directories, 6 for files) is combined with the inverted mask (NOT) using AND on a per-bit-basis.

Example 1:

file, default 6, binary: 110 mask, eg. 2: 010, NOT: 101

resulting permission, AND: 100 (equals 4, r__)

Example 2:

file, default 6, binary: 110 mask, eg. 6: 110, NOT: 001

resulting permission, AND: 000 (equals 0, ___)

Example 3:

directory, default 7, binary: 111 mask, eg. 2: 010, NOT: 101

resulting permission, AND: 101 (equals 5, r_x)

Example 4:

directory, default 7, binary: 111 mask, eg. 6: 110, NOT: 001

resulting permission, AND: 001 (equals 1, __x)

		# Set the user's default umask
		umask 033

Be sure to make root's umask 077, which will disable read, write, and execute permission for other users, unless explicitly changed using chmod. In this case, newly-created directories would have 744 permissions, obtained by subtracting 033 from 777. Newly-created files using the 033 umask would have permissions of 644.

If you are using Red Hat, and adhere to their user and group ID creation scheme (User Private Groups), it is only necessary to use 002 for a umask. This is due to the fact that the default configuration is one user per group.

It's important to ensure that your system files are not open for casual editing by users and groups who shouldn't be doing such system maintenance.

Unix separates access control on files and directories according to three characteristics: owner, group, and other. There is always exactly one owner, any number of members of the group, and everyone else.

A quick explanation of Unix permissions:

Ownership - Which user(s) and group(s) retain(s) control of the permission settings of the node and parent of the node

Permissions - Bits capable of being set or reset to allow certain types of access to it. Permissions for directories may have a different meaning than the same set of permissions on files.

Read:

Write:

Execute:

You - The owner of the file

Group - The group you belong to

Everyone - Anyone on the system that is not the owner or a member of the group

File Example:

        -rw-r--r--  1 kevin  users         114 Aug 28  1997 .zlogin
        1st bit - directory?             (no)
         2nd bit - read by owner?         (yes, by kevin)
          3rd bit - write by owner?        (yes, by kevin)
           4th bit - execute by owner?      (no)
            5th bit - read by group?         (yes, by users)
             6th bit - write by group?        (no)
              7th bit - execute by group?      (no)
               8th bit - read by everyone?      (yes, by everyone)
                9th bit - write by everyone?     (no)
                 10th bit - execute by everyone?  (no)

The following lines are examples of the minimum sets of permissions that are required to perform the access described. You may want to give more permission than what's listed here, but this should describe what these minimum permissions on files do:

-r--------  Allow read access to the file by owner
--w-------  Allows the owner to modify or delete the file
            (Note that anyone with write permission to the directory
             the file is in can overwrite it and thus delete it)
---x------  The owner can execute this program, but not shell scripts, 
	     which still need read permission
---s------  Will execute with effective User ID = to owner
--------s-  Will execute with effective Group ID = to group
-rw------T  No update of "last modified time".  Usually used for swap
	     files
---t------  No effect.  (formerly sticky bit)

Directory Example:

        drwxr-xr-x  3 kevin  users         512 Sep 19 13:47 .public_html/
        1st bit - directory?             (yes, it contains many files)
         2nd bit - read by owner?         (yes, by kevin)
          3rd bit - write by owner?        (yes, by kevin)
           4th bit - execute by owner?      (yes, by kevin)
            5th bit - read by group?         (yes, by users
             6th bit - write by group?        (no)
              7th bit - execute by group?      (yes, by users)
               8th bit - read by everyone?      (yes, by everyone)
                9th bit - write by everyone?     (no)
                 10th bit - execute by everyone?  (yes, by everyone)

The following lines are examples of the minimum sets of permissions that are required to perform the access described. You may want to give more permission than what's listed, but this should describe what these minimum permissions on directories do:

dr--------  The contents can be listed, but file attributes can't be read
d--x------  The directory can be entered, and used in full execution paths
dr-x------  File attributes can be read by owner
d-wx------  Files can be created/deleted, even if the directory
	     isn't the current one
d------x-t  Prevents files from deletion by others with write
	     access. Used on /tmp
d---s--s--  No effect

System configuration files (usually in /etc) are usually mode 640 (-rw-r-----), and owned by root. Depending on your site's security requirements, you might adjust this. Never leave any system files writable by a group or everyone. Some configuration files, including /etc/shadow, should only be readable by root, and directories in /etc should at least not be accessible by others.

Another very good way to detect local (and also network) attacks on your system is to run an integrity checker like Tripwire, Aide or Osiris. These integrety checkers run a number of checksums on all your important binaries and config files and compares them against a database of former, known-good values as a reference. Thus, any changes in the files will be flagged.

It's a good idea to install these sorts of programs onto a floppy, and then physically set the write protect on the floppy. This way intruders can't tamper with the integrety checker itself or change the database. Once you have something like this setup, it's a good idea to run it as part of your normal security administration duties to see if anything has changed.

You can even add a crontab entry to run the checker from your floppy every night and mail you the results in the morning. Something like:

		# set mailto
		MAILTO=kevin
		# run Tripwire
		15 05 * * * root /usr/local/adm/tcheck/tripwire 

will mail you a report each morning at 5:15am.

Integrity checkers can be a godsend to detecting intruders before you would otherwise notice them. Since a lot of files change on the average system, you have to be careful what is cracker activity and what is your own doing.

You can find the freely available unsusported version of Tripwire at http://www.tripwire.org, free of charge. Manuals and support can be purchased.

Aide can be found at http://www.cs.tut.fi/~rammer/aide.html.

Osiris can be found at http://www.shmoo.com/osiris/.

"Trojan Horses" are named after the fabled ploy in Virgil's "Aenid". The idea is that a cracker distributes a program or binary that sounds great, and encourages other people to download it and run it as root. Then the program can compromise their system while they are not paying attention. While they think the binary they just pulled down does one thing (and it might very well), it also compromises their security.

You should take care of what programs you install on your machine. RedHat provides MD5 checksums and PGP signatures on its RPM files so you can verify you are installing the real thing. Other distributions have similar methods. You should never run any unfamiliar binary, for which you don't have the source, as root. Few attackers are willing to release source code to public scrutiny.

Although it can be complex, make sure you are getting the source for a program from its real distribution site. If the program is going to run as root, make sure either you or someone you trust has looked over the source and verified it.

Public-key cryptography, such as that used for PGP, uses one key for encryption, and one key for decryption. Traditional cryptography, however, uses the same key for encryption and decryption; this key must be known to both parties, and thus somehow transferred from one to the other securely.

To alleviate the need to securely transmit the encryption key, public-key encryption uses two separate keys: a public key and a private key. Each person's public key is available by anyone to do the encryption, while at the same time each person keeps his or her private key to decrypt messages encrypted with the correct public key.

There are advantages to both public key and private key cryptography, and you can read about those differences in the RSA Cryptography FAQ, listed at the end of this section.

PGP (Pretty Good Privacy) is well-supported on Linux. Versions 2.6.2 and 5.0 are known to work well. For a good primer on PGP and how to use it, take a look at the PGP FAQ: http://www.pgp.com/service/export/faq/55faq.cgi

Be sure to use the version that is applicable to your country. Due to export restrictions by the US Government, strong-encryption is prohibited from being transferred in electronic form outside the country.

US export controls are now managed by EAR (Export Administration Regulations). They are no longer governed by ITAR.

There is also a step-by-step guide for configuring PGP on Linux available at http://mercury.chem.pitt.edu/~angel/LinuxFocus/English/November1997/article7.html. It was written for the international version of PGP, but is easily adaptable to the United States version. You may also need a patch for some of the latest versions of Linux; the patch is available at ftp://metalab.unc.edu/pub/Linux/apps/crypto.

There is a project maintaining a free re-implementation of pgp with open source. GnuPG is a complete and free replacement for PGP. Because it does not use IDEA or RSA it can be used without any restrictions. GnuPG is in compliance with OpenPGP. See the GNU Privacy Guard web page for more information: http://www.gnupg.org/.

More information on cryptography can be found in the RSA cryptography FAQ, available at http://www.rsa.com/rsalabs/newfaq/. Here you will find information on such terms as "Diffie-Hellman", "public-key cryptography", "digital certificates", etc.

Often users ask about the differences between the various security and encryption protocols, and how to use them. While this isn't an encryption document, it is a good idea to explain briefly what each protocol is, and where to find more information.

Along with CIPE, and other forms of data encryption, there are also several other implementations of IPSEC for Linux. IPSEC is an effort by the IETF to create cryptographically-secure communications at the IP network level, and to provide authentication, integrity, access control, and confidentiality. Information on IPSEC and Internet draft can be found at http://www.ietf.org/html.charters/ipsec-charter.html. You can also find links to other protocols involving key management, and an IPSEC mailing list and archives.

The x-kernel Linux implementation, which is being developed at the University of Arizona, uses an object-based framework for implementing network protocols called x-kernel, and can be found at http://www.cs.arizona.edu/xkernel/hpcc-blue/linux.html. Most simply, the x-kernel is a method of passing messages at the kernel level, which makes for an easier implementation.

Another freely-available IPSEC implementation is the Linux FreeS/WAN IPSEC. Their web page states, “"These services allow you to build secure tunnels through untrusted networks. Everything passing through the untrusted net is encrypted by the IPSEC gateway machine and decrypted by the gateway at the other end. The result is Virtual Private Network or VPN. This is a network which is effectively private even though it includes machines at several different sites connected by the insecure Internet."”

It's available for download from http://www.xs4all.nl/~freeswan/, and has just reached 1.0 at the time of this writing.

As with other forms of cryptography, it is not distributed with the kernel by default due to export restrictions.

ssh and stelnet are suites of programs that allow you to login to remote systems and have a encrypted connection.

openssh is a suite of programs used as a secure replacement for rlogin, rsh and rcp. It uses public-key cryptography to encrypt communications between two hosts, as well as to authenticate users. It can be used to securely login to a remote host or copy data between hosts, while preventing man-in-the-middle attacks (session hijacking) and DNS spoofing. It will perform data compression on your connections, and secure X11 communications between hosts.

There are several ssh implementiations now. The original commercial implementation by Data Fellows can be found at The ssh home page can be found at http://www.datafellows.com.

The excellent Openssh implementation is based on a early version of the datafellows ssh and has been totally reworked to not include any patented or proprietary pieces. It is free and under a BSD license. It can be found at: http://www.openssh.com.

There is also a open source project to re-implement ssh from the ground up called "psst...". For more information see: http://www.net.lut.ac.uk/psst/

You can also use ssh from your Windows workstation to your Linux ssh server. There are several freely available Windows client implementations, including the one at http://guardian.htu.tuwien.ac.at/therapy/ssh/ as well as a commercial implementation from DataFellows, at http://www.datafellows.com.

SSLeay is a free implementation of Netscape's Secure Sockets Layer protocol, developed by Eric Young. It includes several applications, such as Secure telnet, a module for Apache, several databases, as well as several algorithms including DES, IDEA and Blowfish.

Using this library, a secure telnet replacement has been created that does encryption over a telnet connection. Unlike SSH, stelnet uses SSL, the Secure Sockets Layer protocol developed by Netscape. You can find Secure telnet and Secure FTP by starting with the SSLeay FAQ, available at http://www.psy.uq.oz.au/~ftp/Crypto/.

SRP is another secure telnet/ftp implementation. From their web page:

“"The SRP project is developing secure Internet software for free worldwide use. Starting with a fully-secure Telnet and FTP distribution, we hope to supplant weak networked authentication systems with strong replacements that do not sacrifice user-friendliness for security. Security should be the default, not an option!" ”

For more information, go to http://www-cs-students.stanford.edu/~tjw/srp/

Newer versions of the Red Hat Linux and Debian Linux distributions ship with a unified authentication scheme called "PAM". PAM allows you to change your authentication methods and requirements on the fly, and encapsulate all local authentication methods without recompiling any of your binaries. Configuration of PAM is beyond the scope of this document, but be sure to take a look at the PAM web site for more information. http://www.kernel.org/pub/linux/libs/pam/index.html.

Just a few of the things you can do with PAM:

Within a few hours of installing and configuring your system, you can prevent many attacks before they even occur. For example, use PAM to disable the system-wide usage of .rhosts files in user's home directories by adding these lines to /etc/pam.d/rlogin:

		#
		# Disable rsh/rlogin/rexec for users
		#
		login auth required pam_rhosts_auth.so no_rhosts

The primary goal of this software is to provide a facility for secure (against eavesdropping, including traffic analysis, and faked message injection) subnetwork interconnection across an insecure packet network such as the Internet.

CIPE encrypts the data at the network level. Packets traveling between hosts on the network are encrypted. The encryption engine is placed near the driver which sends and receives packets.

This is unlike SSH, which encrypts the data by connection, at the socket level. A logical connection between programs running on different hosts is encrypted.

CIPE can be used in tunnelling, in order to create a Virtual Private Network. Low-level encryption has the advantage that it can be made to work transparently between the two networks connected in the VPN, without any change to application software.

Summarized from the CIPE documentation:

“The IPSEC standards define a set of protocols which can be used (among other things) to build encrypted VPNs. However, IPSEC is a rather heavyweight and complicated protocol set with a lot of options, implementations of the full protocol set are still rarely used and some issues (such as key management) are still not fully resolved. CIPE uses a simpler approach, in which many things which can be parameterized (such as the choice of the actual encryption algorithm used) are an install-time fixed choice. This limits flexibility, but allows for a simple (and therefore efficient, easy to debug...) implementation.”

Further information can be found at http://www.inka.de/~bigred/devel/cipe.html

As with other forms of cryptography, it is not distributed with the kernel by default due to export restrictions.

Kerberos is an authentication system developed by the Athena Project at MIT. When a user logs in, Kerberos authenticates that user (using a password), and provides the user with a way to prove her identity to other servers and hosts scattered around the network.

This authentication is then used by programs such as rlogin to allow the user to login to other hosts without a password (in place of the .rhosts file). This authentication method can also used by the mail system in order to guarantee that mail is delivered to the correct person, as well as to guarantee that the sender is who he claims to be.

Kerberos and the other programs that come with it, prevent users from "spoofing" the system into believing they are someone else. Unfortunately, installing Kerberos is very intrusive, requiring the modification or replacement of numerous standard programs.

You can find more information about kerberos by looking at the kerberos FAQ, and the code can be found at http://nii.isi.edu/info/kerberos/.

[From: Stein, Jennifer G., Clifford Neuman, and Jeffrey L. Schiller. "Kerberos: An Authentication Service for Open Network Systems." USENIX Conference Proceedings, Dallas, Texas, Winter 1998.]

Kerberos should not be your first step in improving security of your host. It is quite involved, and not as widely used as, say, SSH.

Shadow passwords are a means of keeping your encrypted password information secret from normal users. Recent versions of both Red Hat and Debian Linux use shadow passwords by default, but on other systems, encrypted passwords are stored in /etc/passwd file for all to read. Anyone can then run password-guesser programs on them and attempt to determine what they are. Shadow passwords, by contrast, are saved in /etc/shadow, which only privileged users can read. In order to use shadow passwords, you need to make sure all your utilities that need access to password information are recompiled to support them. PAM (above) also allows you to just plug in a shadow module; it doesn't require re-compilation of executables. You can refer to the Shadow-Password HOWTO for further information if necessary. It is available at http://metalab.unc.edu/LDP/HOWTO/Shadow-Password-HOWTO.html It is rather dated now, and will not be required for distributions supporting PAM.

If for some reason your passwd program is not enforcing hard-to-guess passwords, you might want to run a password-cracking program and make sure your users' passwords are secure.

Password cracking programs work on a simple idea: they try every word in the dictionary, and then variations on those words, encrypting each one and checking it against your encrypted password. If they get a match they know what your password is.

There are a number of programs out there...the two most notable of which are "Crack" and "John the Ripper" (http://www.openwall.com/john/) . They will take up a lot of your CPU time, but you should be able to tell if an attacker could get in using them by running them first yourself and notifying users with weak passwords. Note that an attacker would have to use some other hole first in order to read your /etc/passwd file, but such holes are more common than you might think.

Because security is only as strong as the most insecure host, it is worth mentioning that if you have any Windows machines on your network, you should check out L0phtCrack, a Crack implementation for Windows. It's available from http://www.l0pht.com

CFS is a way of encrypting entire directory trees and allowing users to store encrypted files on them. It uses an NFS server running on the local machine. RPMS are available at http://www.zedz.net/redhat/, and more information on how it all works is at ftp://ftp.research.att.com/dist/mab/.

TCFS improves on CFS by adding more integration with the file system, so that it's transparent to users that the file system that is encrypted. More information at: http://www.tcfs.it/.

It also need not be used on entire file systems. It works on directory trees as well.

For 2.0.x kernels, the following options apply. You should see these options during the kernel configuration process. Many of the comments here are from ./linux/Documentation/Configure.help, which is the same document that is referenced while using the Help facility during the make config stage of compiling the kernel.

For 2.2.x kernels, many of the options are the same, but a few new ones have been developed. Many of the comments here are from ./linux/Documentation/Configure.help, which is the same document that is referenced while using the Help facility during the make config stage of compiling the kernel. Only the newly- added options are listed below. Consult the 2.0 description for a list of other necessary options. The most significant change in the 2.2 kernel series is the IP firewalling code. The ipchains program is now used to install IP firewalling, instead of the ipfwadm program used in the 2.0 kernel.

There are a few block and character devices available on Linux that will also help you with security.

The two devices /dev/random and /dev/urandom are provided by the kernel to provide random data at any time.

Both /dev/random and /dev/urandom should be secure enough to use in generating PGP keys, ssh challenges, and other applications where secure random numbers are required. Attackers should be unable to predict the next number given any initial sequence of numbers from these sources. There has been a lot of effort put in to ensuring that the numbers you get from these sources are random in every sense of the word.

The only difference between the two devices, is that /dev/random runs out of random bytes and it makes you wait for more to be accumulated. Note that on some systems, it can block for a long time waiting for new user-generated entropy to be entered into the system. So you have to use care before using /dev/random. (Perhaps the best thing to do is to use it when you're generating sensitive keying information, and you tell the user to pound on the keyboard repeatedly until you print out "OK, enough".)

/dev/random is high quality entropy, generated from measuring the inter-interrupt times etc. It blocks until enough bits of random data are available.

/dev/urandom is similar, but when the store of entropy is running low, it'll return a cryptographically strong hash of what there is. This isn't as secure, but it's enough for most applications.

You might read from the devices using something like:

	root#  head -c 6 /dev/urandom | mimencode

This will print six random characters on the console, suitable for password generation. You can find mimencode in the metamail package.

See /usr/src/linux/drivers/char/random.c for a description of the algorithm.

Thanks to Theodore Y. Ts'o, Jon Lewis, and others from Linux-kernel for helping me (Dave) with this.

One of the most common ways intruders gain access to more systems on your network is by employing a packet sniffer on a already compromised host. This "sniffer" just listens on the Ethernet port for things like passwd and login and su in the packet stream and then logs the traffic after that. This way, attackers gain passwords for systems they are not even attempting to break into. Clear-text passwords are very vulnerable to this attack.

Example: Host A has been compromised. Attacker installs a sniffer. Sniffer picks up admin logging into Host B from Host C. It gets the admins personal password as they login to B. Then, the admin does a su to fix a problem. They now have the root password for Host B. Later the admin lets someone telnet from his account to Host Z on another site. Now the attacker has a password/login on Host Z.

In this day and age, the attacker doesn't even need to compromise a system to do this: they could also bring a laptop or pc into a building and tap into your net.

Using ssh or other encrypted password methods thwarts this attack. Things like APOP for POP accounts also prevents this attack. (Normal POP logins are very vulnerable to this, as is anything that sends clear-text passwords over the network.)

Before you put your Linux system on ANY network the first thing to look at is what services you need to offer. Services that you do not need to offer should be disabled so that you have one less thing to worry about and attackers have one less place to look for a hole.

There are a number of ways to disable services under Linux. You can look at your /etc/inetd.conf file and see what services are being offered by your inetd. Disable any that you do not need by commenting them out (# at the beginning of the line), and then sending your inetd process a SIGHUP.

You can also remove (or comment out) services in your /etc/services file. This will mean that local clients will also be unable to find the service (i.e., if you remove ftp, and try and ftp to a remote site from that machine it will fail with an "unknown service" message). It's usually not worth the trouble to remove services from /etc/services, since it provides no additional security. If a local person wanted to use ftp even though you had commented it out, they would make their own client that used the common FTP port and would still work fine.

Some of the services you might want to leave enabled are:

If you know you are not going to use some particular package, you can also delete it entirely. rpm -e packagename under the Red Hat distribution will erase an entire package. Under Debian dpkg --remove does the same thing.

Additionally, you really want to disable the rsh/rlogin/rcp utilities, including login (used by rlogin), shell (used by rcp), and exec (used by rsh) from being started in /etc/inetd.conf. These protocols are extremely insecure and have been the cause of exploits in the past.

You should check /etc/rc.d/rc[0-9].d (on Red Hat; /etc/rc[0-9].d on Debian), and see if any of the servers started in those directories are not needed. The files in those directories are actually symbolic links to files in the directory /etc/rc.d/init.d (on Red Hat; /etc/init.d on Debian). Renaming the files in the init.d directory disables all the symbolic links that point to that file. If you only wish to disable a service for a particular run level, rename the appropriate symbolic link by replacing the upper-case S with a lower-case s, like this:

       root#  cd /etc/rc6.d
       root#  mv S45dhcpd s45dhcpd

If you have BSD-style rc files, you will want to check /etc/rc* for programs you don't need.

Most Linux distributions ship with tcp_wrappers "wrapping" all your TCP services. A tcp_wrapper (tcpd) is invoked from inetd instead of the real server. tcpd then checks the host that is requesting the service, and either executes the real server, or denies access from that host. tcpd allows you to restrict access to your TCP services. You should make a /etc/hosts.allow and add in only those hosts that need to have access to your machine's services.

If you are a home dial up user, we suggest you deny ALL. tcpd also logs failed attempts to access services, so this can alert you if you are under attack. If you add new services, you should be sure to configure them to use tcp_wrappers if they are TCP-based. For example, a normal dial-up user can prevent outsiders from connecting to his machine, yet still have the ability to retrieve mail, and make network connections to the Internet. To do this, you might add the following to your /etc/hosts.allow:

ALL: 127.

And of course /etc/hosts.deny would contain:

ALL: ALL

which will prevent external connections to your machine, yet still allow you from the inside to connect to servers on the Internet.

Keep in mind that tcp_wrappers only protects services executed from inetd, and a select few others. There very well may be other services running on your machine. You can use netstat -ta to find a list of all the services your machine is offering.

Keeping up-to-date DNS information about all hosts on your network can help to increase security. If an unauthorized host becomes connected to your network, you can recognize it by its lack of a DNS entry. Many services can be configured to not accept connections from hosts that do not have valid DNS entries.

identd is a small program that typically runs out of your inetd server. It keeps track of what user is running what TCP service, and then reports this to whoever requests it.

Many people misunderstand the usefulness of identd, and so disable it or block all off site requests for it. identd is not there to help out remote sites. There is no way of knowing if the data you get from the remote identd is correct or not. There is no authentication in identd requests.

Why would you want to run it then? Because it helps you out, and is another data-point in tracking. If your identd is un compromised, then you know it's telling remote sites the user-name or uid of people using TCP services. If the admin at a remote site comes back to you and tells you user so-and-so was trying to hack into their site, you can easily take action against that user. If you are not running identd, you will have to look at lots and lots of logs, figure out who was on at the time, and in general take a lot more time to track down the user.

The identd that ships with most distributions is more configurable than many people think. You can disable it for specific users (they can make a .noident file), you can log all identd requests (We recommend it), you can even have identd return a uid instead of a user name or even NO-USER.

The Postfix mail server was written by Wietse Venema, author of Postfix and several other staple Internet security products, as an "attempt to provide an alternative to the widely-used Sendmail program. Postfix attempts to be fast, easy to administer, and hopefully secure, while at the same time being sendmail compatible enough to not upset your users."

Further information on postfix can be found at the Postfix home and in the Configuring and Securing Postfix.

There are a number of different software packages out there that do port and service-based scanning of machines or networks. SATAN, ISS, SAINT, and Nessus are some of the more well-known ones. This software connects to the target machine (or all the target machines on a network) on all the ports they can, and try to determine what service is running there. Based on this information, you can tell if the machine is vulnerable to a specific exploit on that server.

SATAN (Security Administrator's Tool for Analyzing Networks) is a port scanner with a web interface. It can be configured to do light, medium, or strong checks on a machine or a network of machines. It's a good idea to get SATAN and scan your machine or network, and fix the problems it finds. Make sure you get the copy of SATAN from metalab or a reputable FTP or web site. There was a Trojan copy of SATAN that was distributed out on the net. http://www.trouble.org/~zen/satan/satan.html. Note that SATAN has not been updated in quite a while, and some of the other tools below might do a better job.

ISS (Internet Security Scanner) is another port-based scanner. It is faster than Satan, and thus might be better for large networks. However, SATAN tends to provide more information.

Abacus is a suite of tools to provide host-based security and intrusion detection. Look at it's home page on the web for more information. http://www.psionic.com/abacus/

SAINT is a updated version of SATAN. It is web-based and has many more up-to-date tests than SATAN. You can find out more about it at: http://www.wwdsi.com/~saint

Nessus is a free security scanner. It has a GTK graphical interface for ease of use. It is also designed with a very nice plug in setup for new port-scanning tests. For more information, take a look at: http://www.nessus.org

One of the most important services you can provide is a mail server. Unfortunately, it is also one of the most vulnerable to attack, simply due to the number of tasks it must perform and the privileges it typically needs.

If you are using sendmail it is very important to keep up on current versions. sendmail has a long long history of security exploits. Always make sure you are running the most recent version from http://www.sendmail.org.

Keep in mind that sendmail does not have to be running in order for you to send mail. If you are a home user, you can disable sendmail entirely, and simply use your mail client to send mail. You might also choose to remove the "-bd" flag from the sendmail startup file, thereby disabling incoming requests for mail. In other words, you can execute sendmail from your startup script using the following instead:

		# /usr/lib/sendmail -q15m

This will cause sendmail to flush the mail queue every fifteen minutes for any messages that could not be successfully delivered on the first attempt.

Many administrators choose not to use sendmail, and instead choose one of the other mail transport agents. You might consider switching over to qmail. qmail was designed with security in mind from the ground up. It's fast, stable, and secure. Qmail can be found at http://www.qmail.org

In direct competition to qmail is "postfix", written by Wietse Venema, the author of tcp_wrappers and other security tools. Formerly called vmailer, and sponsored by IBM, this is also a mail transport agent written from the ground up with security in mind. You can find more information about postfix at http://www.postfix.org

A "Denial of Service" (DoS) attack is one where the attacker tries to make some resource too busy to answer legitimate requests, or to deny legitimate users access to your machine.

Denial of service attacks have increased greatly in recent years. Some of the more popular and recent ones are listed below. Note that new ones show up all the time, so this is just a few examples. Read the Linux security lists and the bugtraq list and archives for more current information.

You can find code for most exploits, and a more in-depth description of how they work, at http://www.rootshell.com using their search engine.

NFS is a very widely-used file sharing protocol. It allows servers running nfsd and mountd to "export" entire file systems to other machines using NFS filesystem support built in to their kernels (or some other client support if they are not Linux machines). mountd keeps track of mounted file systems in /etc/mtab, and can display them with showmount.

Many sites use NFS to serve home directories to users, so that no matter what machine in the cluster they login to, they will have all their home files.

There is some small amount of security allowed in exporting file systems. You can make your nfsd map the remote root user (uid=0) to the nobody user, denying them total access to the files exported. However, since individual users have access to their own (or at least the same uid) files, the remote root user can login or su to their account and have total access to their files. This is only a small hindrance to an attacker that has access to mount your remote file systems.

If you must use NFS, make sure you export to only those machines that you really need to. Never export your entire root directory; export only directories you need to export.

See the NFS HOWTO for more information on NFS, available at http://metalab.unc.edu/mdw/HOWTO/NFS-HOWTO.html

Network Information service (formerly YP) is a means of distributing information to a group of machines. The NIS master holds the information tables and converts them into NIS map files. These maps are then served over the network, allowing NIS client machines to get login, password, home directory and shell information (all the information in a standard /etc/passwd file). This allows users to change their password once and have it take effect on all the machines in the NIS domain.

NIS is not at all secure. It was never meant to be. It was meant to be handy and useful. Anyone that can guess the name of your NIS domain (anywhere on the net) can get a copy of your passwd file, and use "crack" and "John the Ripper" against your users' passwords. Also, it is possible to spoof NIS and do all sorts of nasty tricks. If you must use NIS, make sure you are aware of the dangers.

There is a much more secure replacement for NIS, called NIS+. Check out the NIS HOWTO for more information: http://metalab.unc.edu/mdw/HOWTO/NIS-HOWTO.html

Firewalls are a means of controlling what information is allowed into and out of your local network. Typically the firewall host is connected to the Internet and your local LAN, and the only access from your LAN to the Internet is through the firewall. This way the firewall can control what passes back and forth from the Internet and your LAN.

There are a number of types of firewalls and methods of setting them up. Linux machines make pretty good firewalls. Firewall code can be built right into 2.0 and higher kernels. The user-space tools ipfwadm for 2.0 kernels and ipchains for 2.2 kernels, allows you to change, on the fly, the types of network traffic you allow. You can also log particular types of network traffic.

Firewalls are a very useful and important technique in securing your network. However, never think that because you have a firewall, you don't need to secure the machines behind it. This is a fatal mistake. Check out the very good Firewall-HOWTO at your latest metalab archive for more information on firewalls and Linux. http://metalab.unc.edu/mdw/HOWTO/Firewall-HOWTO.html

More information can also be found in the IP-Masquerade mini-howto: http://metalab.unc.edu/mdw/HOWTO/mini/IP-Masquerade.html

More information on ipfwadm (the tool that lets you change settings on your firewall, can be found at it's home page: http://www.xos.nl/linux/ipfwadm/

If you have no experience with firewalls, and plan to set up one for more than just a simple security policy, the Firewalls book by O'Reilly and Associates or other online firewall document is mandatory reading. Check out http://www.ora.com for more information. The National Institute of Standards and Technology have put together an excellent document on firewalls. Although dated 1995, it is still quite good. You can find it at http://csrc.nist.gov/nistpubs/800-10/main.html. Also of interest:

Linux IP Firewalling Chains is an update to the 2.0 Linux firewalling code for the 2.2 kernel. It has many more features than previous implementations, including:

If you are currently using ipfwadm on your 2.0 kernel, there are scripts available to convert the ipfwadm command format to the format ipchains uses.

Be sure to read the IP Chains HOWTO for further information. It is available at http://www.adelaide.net.au/~rustcorp/ipfwchains/ipfwchains.html

In yet another set of advancements to the kernel IP packet filtering code, netfilter allows users to set up, maintain, and inspect the packet filtering rules in the new 2.4 kernel.

The netfilter subsystem is a complete rewrite of previous packet filtering implementations including ipchains and ipfwadm. Netfilter provides a large number of improvements, and it has now become an even more mature and robust solution for protecting corporate networks.

iptables

is the command-line interface used to manipulate the firewall tables within the kernel.

Netfilter provides a raw framework for manipulating packets as they traverse through various parts of the kernel. Part of this framework includes support for masquerading, standard packet filtering, and now more complete network address translation. It even includes improved support for load balancing requests for a particular service among a group of servers behind the firewall.

The stateful inspection features are especially powerful. Stateful inspection provides the ability to track and control the flow of communication passing through the filter. The ability to keep track of state and context information about a session makes rules simpler and tries to interpret higher-level protocols.

Additionally, small modules can be developed to perform additional specific functions, such as passing packets to programs in userspace for processing then reinjecting back into the normal packet flow. The ability to develop these programs in userspace reduces the level of complexity that was previously associated with having to make changes directly at the kernel level.

Other IP Tables references include:

VPN's are a way to establish a "virtual" network on top of some already-existing network. This virtual network often is encrypted and passes traffic only to and from some known entities that have joined the network. VPNs are often used to connect someone working at home over the public Internet to an internal company network.

If you are running a Linux masquerading firewall and need to pass MS PPTP (Microsoft's VPN point-to-point product) packets, there is a Linux kernel patch out to do just that. See: ip-masq-vpn.

There are several Linux VPN solutions available:

See also the section on IPSEC for pointers and more information.

Discussion of backup methods and storage is beyond the scope of this document, but here are a few words relating to backups and security:

If you have less than 650mb of data to store on a partition, a CD-R copy of your data is a good way to go (as it's hard to tamper with later, and if stored properly can last a long time), you will of course need at least 650MB of space to make the image. Tapes and other re-writable media should be write-protected as soon as your backup is complete, and then verified to prevent tampering. Make sure you store your backups in a secure off-line area. A good backup will ensure that you have a known good point to restore your system from.

A six-tape cycle is easy to maintain. This includes four tapes for during the week, one tape for even Fridays, and one tape for odd Fridays. Perform an incremental backup every day, and a full backup on the appropriate Friday tape. If you make some particularly important changes or add some important data to your system, a full backup might well be in order.

You should do periodic tests of your backups to make sure they are working as you might expect them to. Restores of files and checking against the real data, sizes and listings of backups, and reading old backups should be done on a regular basis.

In the event of an intrusion, you can use your RPM database like you would use tripwire, but only if you can be sure it too hasn't been modified. You should copy the RPM database to a floppy, and keep this copy off-line at all times. The Debian distribution likely has something similar.

The files /var/lib/rpm/fileindex.rpm and /var/lib/rpm/packages.rpm most likely won't fit on a single floppy. But if compressed, each should fit on a seperate floppy.

Now, when your system is compromised, you can use the command:

			root#  rpm -Va

to verify each file on the system. See the rpm man page, as there are a few other options that can be included to make it less verbose. Keep in mind you must also be sure your RPM binary has not been compromised.

This means that every time a new RPM is added to the system, the RPM database will need to be rearchived. You will have to decide the advantages versus drawbacks.

It is very important that the information that comes from syslog not be compromised. Making the files in /var/log readable and writable by only a limited number of users is a good start.

Be sure to keep an eye on what gets written there, especially under the auth facility. Multiple login failures, for example, can indicate an attempted break-in.

Where to look for your log file will depend on your distribution. In a Linux system that conforms to the "Linux Filesystem Standard", such as Red Hat, you will want to look in /var/log and check messages, mail.log, and others.

You can find out where your distribution is logging to by looking at your /etc/syslog.conf file. This is the file that tells syslogd (the system logging daemon) where to log various messages.

You might also want to configure your log-rotating script or daemon to keep logs around longer so you have time to examine them. Take a look at the logrotate package on recent Red Hat distributions. Other distributions likely have a similar process.

If your log files have been tampered with, see if you can determine when the tampering started, and what sort of things appeared to be tampered with. Are there large periods of time that cannot be accounted for? Checking backup tapes (if you have any) for untampered log files is a good idea.

Intruders typically modify log files in order to cover their tracks, but they should still be checked for strange happenings. You may notice the intruder attempting to gain entrance, or exploit a program in order to obtain the root account. You might see log entries before the intruder has time to modify them.

You should also be sure to separate the auth facility from other log data, including attempts to switch users using su, login attempts, and other user accounting information.

If possible, configure syslog to send a copy of the most important data to a secure system. This will prevent an intruder from covering his tracks by deleting his login/su/ftp/etc attempts. See the syslog.conf man page, and refer to the @ option.

There are several more advanced syslogd programs out there. Take a look at http://www.core-sdi.com/ssyslog/ for Secure Syslog. Secure Syslog allows you to encrypt your syslog entries and make sure no one has tampered with them.

Another syslogd with more features is syslog-ng. It allows you a lot more flexibility in your logging and also can has your remote syslog streams to prevent tampering.

Finally, log files are much less useful when no one is reading them. Take some time out every once in a while to look over your log files, and get a feeling for what they look like on a normal day. Knowing this can help make unusual things stand out.

Most Linux users install from a CD-ROM. Due to the fast-paced nature of security fixes, new (fixed) programs are always being released. Before you connect your machine to the network, it's a good idea to check with your distribution's ftp site and get all the updated packages since you received your distribution CD-ROM. Many times these packages contain important security fixes, so it's a good idea to get them installed.

Spotting a security compromise under way can be a tense undertaking. How you react can have large consequences.

If the compromise you are seeing is a physical one, odds are you have spotted someone who has broken into your home, office or lab. You should notify your local authorities. In a lab, you might have spotted someone trying to open a case or reboot a machine. Depending on your authority and procedures, you might ask them to stop, or contact your local security people.

If you have detected a local user trying to compromise your security, the first thing to do is confirm they are in fact who you think they are. Check the site they are logging in from. Is it the site they normally log in from? No? Then use a non-electronic means of getting in touch. For instance, call them on the phone or walk over to their office/house and talk to them. If they agree that they are on, you can ask them to explain what they were doing or tell them to cease doing it. If they are not on, and have no idea what you are talking about, odds are this incident requires further investigation. Look into such incidents , and have lots of information before making any accusations.

If you have detected a network compromise, the first thing to do (if you are able) is to disconnect your network. If they are connected via modem, unplug the modem cable; if they are connected via Ethernet, unplug the Ethernet cable. This will prevent them from doing any further damage, and they will probably see it as a network problem rather than detection.

If you are unable to disconnect the network (if you have a busy site, or you do not have physical control of your machines), the next best step is to use something like tcp_wrappers or ipfwadm to deny access from the intruder's site.

If you can't deny all people from the same site as the intruder, locking the user's account will have to do. Note that locking an account is not an easy thing. You have to keep in mind .rhosts files, FTP access, and a host of possible backdoors.

After you have done one of the above (disconnected the network, denied access from their site, and/or disabled their account), you need to kill all their user processes and log them off.

You should monitor your site well for the next few minutes, as the attacker will try to get back in. Perhaps using a different account, and/or from a different network address.

So you have either detected a compromise that has already happened or you have detected it and locked (hopefully) the offending attacker out of your system. Now what?

The LinuxSecurity.com web site has numerous Linux and open source security references written by the LinuxSecurity staff and people collectively around the world.

CERT is the Computer Emergency Response Team. They often send out alerts of current attacks and fixes. See ftp://ftp.cert.org for more information.

ZEDZ (formerly Replay) (http://www.zedz.net) has archives of many security programs. Since they are outside the US, they don't need to obey US crypto restrictions.

Matt Blaze is the author of CFS and a great security advocate. Matt's archive is available at ftp://ftp.research.att.com/pub/mab

tue.nl is a great security FTP site in the Netherlands. ftp.win.tue.nl

Bugtraq: To subscribe to bugtraq, send mail to listserv@netspace.org containing the message body subscribe bugtraq. (see links above for archives).

CIAC: Send e-mail to majordomo@tholia.llnl.gov. In the BODY (not subject) of the message put (either or both): subscribe ciac-bulletin

Red Hat has a number of mailing lists, the most important of which is the redhat-announce list. You can read about security (and other) fixes as soon as they come out. Send email to redhat-announce-list-request@redhat.com with the Subject Subscribe See https://listman.redhat.com/mailman/listinfo/ for more info and archives.

The Debian project has a security mailing list that covers their security fixes. See http://www.debian.com/security/ for more information.

There are a number of good security books out there. This section lists a few of them. In addition to the security specific books, security is covered in a number of other books on system administration.