Backups are a standard sysadmin service. In other words, everyone in an organization, from the CEO to the temporary worker, depends on us to provide backups. People make mistakes, and you're going to need backups of user data, corporate data, and logs. But it's not obvious on Linux systems exactly what you need to include in a backup set. This article guides you as to what you should and should not include in your backups.
Backups are not romantic or exciting, but they're necessary. If you want to know how important backups are to any organization, you need to experience a failure or an outage without a good backup. There's no recovery without a good backup. Sometimes I think that backups are an afterthought because of the many, many times I've seen them neglected, ignored, and failed.
- Backup - A set of directories and files archived to tape, disk, or cloud.
- Back up - The process of creating a backup.
One of the first questions that new system administrators ask is, What should I back up? The answer is pretty easy: Everything you need to and nothing that you don't. Yes, it's a bit sarcastic, but it's 100 percent true. The following is a list of directories that you must regularly back up and one that you should never back up. These directories are in no particular order.
/etc
The /etc directory contains a system's configuration files. It includes the user file, the group file, the password file, networking information, most application configurations, startup files, filesystem information, mount points, and other system-wide configuration files. Without this directory, it will take you a very long time to put the system back to where it was before the outage or failure. Without a good backup of /etc, you will likely have to reinstall your system from media, reinstall all of your applications, and basically start fresh as you would with a new system.
/home
All user data, downloads, documents, pictures, and miscellaneous files are kept in the /home directory under the user's name. For example, my home directory is /home/khess. For a user, there's no more important directory to back up on a system. Users assume that you'll include their personal files in your daily backups. You should do so. Trust my experience on this one. It's been said that good judgment comes from experience, and experience comes from poor judgment. This is true, and I have stories to prove it.
/root
The root user's home directory is an important one to include in a backup. This directory is where you, the system administrator, keep downloads, configurations, scripts, notes, and other essential information that is often unique and irreplaceable.
/var
The /var directory holds databases, web pages, crontabs, log files, and DNS zone files. Files in the /var (variable) directory change and grow on a daily basis. Because of the data it holds, the /var directory will often have its own mount point on its own filesystem and its own disk or disk array.
/usr/local/bin
The /usr/local/bin directory can be empty, or it can contain root-owned scripts or application executables that aren't part of the standard system executables list located in /usr/bin or /usr/sbin. If the directory is empty, I always create a /usr/local/bin/empty file to let me know that it has no files in it rather than thinking that my backup missed an entire directory's files. Yes, I'm OCD. Don't judge me.
[ You might also enjoy: ReaR: Backup and recover your Linux server with confidence ]
/usr/local/sbin
The /usr/local/sbin directory is important for the same reasons as /usr/local/bin. It too is often empty, and I also place a /usr/local/sbin/empty file in it to let me know its status.
/srv
The /srv directory is somewhat of an anomaly. It's used optionally for internet-related services files such as WWW uploads, FTP files, CVS, and so on. I always back it up, but I also place my standard /srv/empty file in it because it may be empty for the entire life of the system. Some sysadmins use it as a file upload/download directory for the aforementioned services, but its use is random and somewhat rare.
/opt
The /opt directory is often used for third-party software or software that a sysadmin wants to keep separate from standard directories. I always install document management systems, Webmin, and other applications to it that I don't want occupying locations such as /usr/local. That's just a personal preference and you not required to do so. I find /opt is a good location to install software that I want to test, such as a new document management system or other application. If it's empty, I create my /opt/empty file there.
Shared directories
A shared directory might belong to a particular group, such as finance or human resources, that has special permissions. You definitely want to back up /finance, /data, or /hr if your system has such directories. Don't create these types of directories until you have a need. Shared directories are important to include in a backup because users store corporate data in them, and there may be no other copies.
NFS-mounted directories
NFS-mounted directories should generally be excluded from your backup scheme. Why? Because these directories are mounted from remote systems. They can be in the same data center, or they could be from a system very far away. Over-the-network backups can be fine if you have a very fast network or a small volume of files. It's my recommendation that you do not back up over the network unless it's necessary. Back up this data from the local system that houses it. You don't want a network glitch to corrupt your backups and render your data useless.
[ Free download: Advanced Linux commands cheat sheet. ]
Wrap up
The two purposes of a backup are to ensure that you have good copies of your data and that it can be restored. Going through the motions of making backups is a waste of resources if the backups aren't usable. I've been there. This is why you must check your backups periodically by restoring a file.
Backups are always important. Do them. Check them. Don't trust that they're being done by someone else.
저자 소개
Ken has used Red Hat Linux since 1996 and has written ebooks, whitepapers, actual books, thousands of exam review questions, and hundreds of articles on open source and other topics. Ken also has 20+ years of experience as an enterprise sysadmin with Unix, Linux, Windows, and Virtualization.
Follow him on Twitter: @kenhess for a continuous feed of Sysadmin topics, film, and random rants.
In the evening after Ken replaces his red hat with his foil hat, he writes and makes films with varying degrees of success and acceptance. He is an award-winning filmmaker who constantly tries to convince everyone of his Renaissance Man status, also with varying degrees of success and acceptance.
유사한 검색 결과
CVE-2026-31431: How Red Hat Advanced Cluster Security and Red Hat Advanced Cluster Management can help
Confidential clusters for Red Hat OpenShift: Developer Preview now available on Microsoft Azure with AMD SEV-SNP
Collaboration In Product Security | Compiler
Keeping Track Of Vulnerabilities With CVEs | Compiler
채널별 검색
오토메이션
기술, 팀, 인프라를 위한 IT 자동화 최신 동향
인공지능
고객이 어디서나 AI 워크로드를 실행할 수 있도록 지원하는 플랫폼 업데이트
오픈 하이브리드 클라우드
하이브리드 클라우드로 더욱 유연한 미래를 구축하는 방법을 알아보세요
보안
환경과 기술 전반에 걸쳐 리스크를 감소하는 방법에 대한 최신 정보
엣지 컴퓨팅
엣지에서의 운영을 단순화하는 플랫폼 업데이트
인프라
세계적으로 인정받은 기업용 Linux 플랫폼에 대한 최신 정보
애플리케이션
복잡한 애플리케이션에 대한 솔루션 더 보기
가상화
온프레미스와 클라우드 환경에서 워크로드를 유연하게 운영하기 위한 엔터프라이즈 가상화의 미래