Please wait while we load the content.
Category articles197 articles
Learn VPS tasks in the order you need them, from signing in and transferring files to configuring the system, deploying apps, restoring backups, and troubleshooting.
Enabling automatic security updates on a VPS does not mean vulnerability patches have actually taken effect. This article covers unattended-upgrades, apt-daily.timer, /var/run/reboot-required, the risks of automatic reboots, maintenance windows, and a post-update checklist on Ubuntu/Debian, suitable for small websites and production VPS patch planning.
Your VPS can ping IPs but can't resolve domain names, and the problem is usually in the DNS resolver, /etc/resolv.conf, systemd-resolved, DHCP-provided DNS, or cache state. This article lays out a DNS troubleshooting checklist in the order of ping IP, dig, resolvectl, 127.0.0.53, logs, and temporary fixes.
When a VPS website suddenly shows 502 Bad Gateway or 504 Gateway Timeout, it is usually not a domain issue. More often, Nginx cannot reach the backend, PHP-FPM/Gunicorn has crashed, Unix socket permissions are wrong, the process pool is exhausted, or the upstream response times out. This article troubleshoots layer by layer through logs, ports, sockets, service status, timeouts, and resource usage.
When apt update or apt install fails on an Ubuntu/Debian VPS, it directly blocks everything that follows: Docker, Nginx, PHP, monitoring, and more. This article lays out a safe repair checklist in order: DNS, software sources, GPG keys, dpkg locks, half-installed packages, dependency conflicts, and disk/inode.
When MySQL or MariaDB won't connect on a VPS, there's usually more than one cause: the service isn't running, 3306 isn't listening, bind-address only binds to localhost, a firewall/security group is blocking it, user privileges are wrong, the disk is full, or InnoDB fails to start. This article troubleshoots layer by layer: local connections, port listening, logs, remote access, and database state.
Docker on a VPS can easily fill up the system disk: overlay2 layers, stopped containers, old images, build cache, json-file logs, and orphaned volumes can all take up space. This article walks through troubleshooting in order: df, docker system df, log rotation, safe prune, and volume protection, so a single command doesn't accidentally delete your data.
When PostgreSQL on a VPS won't accept connections, common causes include the service not starting, 5432 not listening, listen_addresses bound only to localhost, pg_hba.conf not permitting the client, UFW/security group blocking, mismatched authentication methods, or startup failures caused by disk or shared memory. This article troubleshoots layer by layer: local connections, service status, logs, port listening, authentication configuration, and remote access.
When Redis won't connect on a VPS, common causes include the service not starting, 6379 not listening, bind restricted to localhost, protected-mode blocking, password/ACL mismatch, firewall or security group blocking, and persistence file permissions causing startup failure. This article troubleshoots layer by layer: local ping, service status, logs, listening address, authentication, and secure exposure.
On a VPS, after changing Nginx configuration, reload fails, the site returns 404/502, or HTTPS doesn't take effect. Common causes include configuration syntax errors, server_name mismatch, include file load order, wrong root/alias usage, incorrect proxy_pass syntax, wrong certificate paths, or ports already in use. This article troubleshoots in the order of nginx -t, systemctl, error.log, site configuration, and rollback.
