Linux has many advantages in terms of functionality, price, and performance. However, as an open operating system, it inevitably has some security risks. This article will tell you how to solve these hidden dangers and provide a secure operation platform for applications. It is also the most basic, common, and effective method.
Linux is a Unix-like operating system. Theoretically, the design of Unix itself has no major security defects. For many years, the vast majority of security problems found on Unix operating systems mainly exist in individual programs. Therefore, most Unix vendors claim to be able to solve these problems and provide secure Unix operating systems. But Linux is somewhat different, because it does not belong to a vendor, and no vendor claims to provide security assurance for it. Therefore, users only have to solve security problems by themselves.
Linux is an open system that allows you to find many ready-made programs and tools on the network. This facilitates both users and hackers, because they can also easily find programs and tools to sneak into the Linux system, or steal important information from the Linux system. However, as long as we carefully set various Linux system functions and add the necessary security measures, hackers will be able to win.
In general, security settings for Linux systems include canceling unnecessary services, restricting remote access, hiding important information, fixing security vulnerabilities, using security tools, and regular security checks. This article teaches you ten ways to improve Linux system security. Although the number of moves is not big, the moves work. You might as well try.
1st tips: Cancel unnecessary services
In earlier versions of Unix, each different network service had a service program running in the background. Later versions used a uniform/etc/inetd server program. Inetd is short for Internetdaemon. It monitors multiple network ports at the same time. Once it receives connection information from the outside, it executes the corresponding TCP or UDP network service.
Due to the unified command of inetd, Most TCP or UDP services in Linux are set in the/etc/inetd. conf file. Therefore, the first step to cancel unnecessary services is to check the/etc/inetd. conf file and add the "#" sign before the unwanted services.
In general, all services except http, smtp, telnet, and ftp should be canceled, such as the simple File Transfer Protocol tftp, the imap/ipop transport protocol used for network mail storage and receiving, the gopher for data searching, and the daytime and time used for time synchronization.
There are also some services that report system status, such as finger, efinger, systat, and netstat. Although it is very useful for system error detection and user searching, it also provides a convenient portal for hackers. For example, hackers can use the finger service to find users' phones, directories, and other important information. Therefore, many Linux systems cancel all or partially cancel these services to enhance system security.
In addition to setting system service items using/etc/Inetd. conf, inetd also uses the/etc/services file to find the ports used by various services. Therefore, you must carefully check the port settings in the file to avoid security vulnerabilities.
In Linux, there are two different service-type states: one is a service that is executed only when necessary, such as the finger service, and the other is a service that is continuously executed and never paused. This type of service starts to run when the system starts. Therefore, you cannot stop the service by modifying inetd, but you can only modify/etc/rc. d/rc [n]. d/file or use Runleveleditor to modify it. NFS servers that provide file services and news that provide NNTP news services belong to such services. If not necessary, it is best to cancel these services.
2nd tips: Restrict System Access
Before entering the Linux system, all users need to log on, that is, users need to enter the user account and password. Only after they pass system verification can users enter the system.
Like other Unix operating systems, Linux typically stores passwords in the/etc/passwd file after encryption. All users in Linux can read the/etc/passwd file. Although the password stored in the file has been encrypted, it is still not safe. Generally, users can use the ready-made password cracking tool to guess the password. The safer method is to set the shadow file/etc/shadow and only allow users with special permissions to read the file.
In Linux, to use a shadow file, you must recompile all the utilities to support the shadow file. This method is troublesome. A simple method is to use the plug-in verification module (PAM ). Many Linux systems use Linux tool PAM, which is an identity authentication mechanism that can be used to dynamically change the authentication methods and requirements without re-compiling other utilities. This is because PAM uses a closed package to hide all authentication-related logic in the module, so it is the best helper for using shadow files.
In addition, PAM has many security features: it can rewrite the traditional DES encryption method to other more powerful encryption methods to ensure that user passwords are not easily decrypted; it can set the upper limit for each user to use computer resources; it can even set the user's computer time and location.
Linux administrators can install and set PAM in just a few hours to greatly improve the security of the Linux system and block many attacks outside the system.