三丰云免费服务器实测：博客+备份脚本跑半年，这些坑要提前排
之前一直写“零事故”，今天换个角度——把我会踩的坑先摘出来。折腾云服务器半年以上，最大的收获不是学会了多少命令，而是知道哪里会疼。三丰云这个 免费云服务器 ，我用它跑 Typecho + Nginx + 自动备份脚本，资源常年很轻松，但有几个设置不做好，后期真的很麻烦。 开机自启：别指望它能自己醒 我第一次部署完，重启机器发现博客挂了。Nginx 没设开机自启，Typecho 依赖的 PHP-FPM 也没跟着起。Linux 服务管理很直接， systemctl 或 chkconfig 都要手动配。 现在的做法是写个启动脚本，加入 rc.local 或者用 systemd 服务文件。每次开机，服务自动拉起，不用 SSH 进去手动敲命令。这一步做了，后续省掉很多“为什么挂了”的困惑。 备份清理：磁盘会涨，记得定期删 我观测到的情况是，磁盘每周涨几百 MB。原因很简单：Nginx…
After months of testing, it becomes clear that the biggest lesson learned is which aspects are prone to issues. Using the free Three Feng Cloud server, I ran Typecho with Nginx and an automatic backup script. While the resources were typically light, neglecting key settings led to significant headaches later on.
Upon restarting the machine, the blog failed to load because Nginx lacked an open-source startup script and Typecho's PHP-FPM was not automatically started. Linux service management is straightforward, requiring manual configuration using systemctl or chkconfig. A remedy is to write a startup script and integrate it into rc.local or a systemd service file. This ensures services automatically launch at startup, eliminating the need to manually enter commands each time.
Backup and cleanup are crucial. My observations revealed a weekly increase of a few hundred MB in disk space, primarily due to Nginx logs, Typecho image uploads, and database backup files. To prevent full disk space, I implemented a retention policy: keeping the last 7 days of backups, while cleaning up older weekly backups. The `find` command with the `-mtime` parameter, combined with a cron job, automates this process. As a result, the disk space remains ample, providing peace of mind.
Reminders are essential for renewal. Similar to the 7-day renewal mentioned in the previous article, I set up reminders on my phone to check weekly reminders. Apart from renewal, reminders also cover instance restarts and certificate updates. Without reminders, it's easy to forget important tasks; forgetting can lead to instance termination.
The data side shows stable CPU usage between 0.2 and 0.5, with memory usage staying below half of a 1-core, 1G machine. Running a Typecho blog on this platform comfortably accommodates daily traffic of up to 5M, without experiencing limitations or downgrades. The logs and traffic statistics are accurate, with no hidden charges.
Despite the manual maintenance approach, I found value in using this platform. Configuring open-source startup scripts saves time on manual launches. Automated backup and cleanup scripts maintain disk space, and calendar reminders prevent instances from being recycled. The resources provide sufficient capacity for running a Typecho blog smoothly without any lag.
There are no complex billing structures, only a clear 7-day cycle. The instance demonstrates strong self-healing capabilities, allowing manual reactivation if it fails. It serves as an excellent sandbox environment for learning Linux operations. It's ideal for students, self-taught developers, or those wanting to set up a Typecho blog, run small tools, or practice Linux commands.
However, it's not suitable for high-traffic or long-term storage of critical data. It is a lightweight experimental environment, not a production-level high availability solution. To explore further, search for Three Feng Cloud's official website.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
