← Monitoring resources
Server troubleshooting
Find the cause of the problem on your server with the exact commands to run, then set an alert so you hear about it earlier next time.
Linux
High load average on a Linux serverWhat a high Linux load average means, how to compare it to your CPU count, and how to tell CPU pressure from processes stuck waiting on disk.High CPU usage on a Linux serverFind which process is using all the CPU on a Linux server, tell user from system time, and fix common causes like runaway workers and miners.High iowait on a Linux serverHigh iowait means the CPU is idle waiting on storage. Find the busy device with iostat, the process with iotop, and fix slow disks or heavy writers.The Linux OOM killer is killing processesConfirm the Linux OOM killer ended a process, see which process used the memory, and fix it with limits, swap, tuning or more RAM.High swap usage on LinuxTell harmless swap use from active swapping, find which processes are swapped out, and fix memory pressure on a Linux server.Linux disk is fullFind which directories fill a Linux disk with df and du, and safely free space from logs, package caches, old kernels and Docker.No space left on device, but the disk is not fullFix “No space left on device” when df shows free space: check inode usage with df -i, find directories with millions of files and clean them.df says the disk is full, du disagreesWhy df reports more used space than du on Linux: deleted files still held open by a process. Find them with lsof and release the space.systemd journal logs are filling the diskCheck how much space the systemd journal uses, vacuum it safely with journalctl, and set a permanent size limit in journald.conf.The /boot partition is fullFree space on a full /boot partition by removing old kernels safely with apt, and fix broken upgrades caused by a full /boot.A systemd service has failedFind failed systemd units, read why a service failed with systemctl status and journalctl, and fix common causes like bad config and exit codes.“Too many open files” on LinuxFix “Too many open files” (EMFILE) on Linux: check a process’s file descriptor usage and raise limits with systemd LimitNOFILE or limits.conf.Too many TIME_WAIT connectionsUnderstand TCP TIME_WAIT on Linux, count connections by state with ss, and fix port exhaustion with keep-alive and connection pooling.Network packet drops and errors on LinuxCheck RX/TX errors and dropped packets on a Linux interface with ip -s link and ethtool, and fix ring buffers, duplex and saturation issues.Zombie processes on LinuxWhat zombie (defunct) processes are, how to find their parent with ps, and how to clear them by fixing or restarting the parent process.High CPU steal time on a virtual serverCPU steal time is CPU your VM wanted but the hypervisor gave to others. Measure it with top or vmstat and decide whether to resize or move hosts.CPU temperature is too high on LinuxRead CPU temperature on Linux with lm-sensors or /sys/class/thermal, spot thermal throttling, and fix cooling or load causes.Why did my Linux server reboot?Find out why a Linux server rebooted unexpectedly using last, journalctl -b -1, kernel panics, OOM, unattended-upgrades and provider maintenance.Find what is using a port on LinuxFind the process listening on a port in Linux with ss, lsof or fuser, and fix “address already in use” errors when a service will not start.A crypto miner is running on my Linux serverSigns of a crypto miner on a Linux server — 100% CPU, random process names, binaries in /tmp — and how to contain, remove and find how it got in.Pending security updates on a Linux serverList pending security updates on Ubuntu, Debian and RHEL-family servers, apply them safely, and know when a reboot is required.A Docker container keeps restartingDiagnose a Docker container stuck in a restart loop with docker logs and docker inspect: exit codes, OOM kills, bad commands and failing health checks.Docker is using too much disk spaceSee what Docker uses on disk with docker system df, safely prune images, build cache and volumes, and cap container log size.MySQL is using 100% CPUFind which queries make MySQL or MariaDB use 100% CPU with SHOW PROCESSLIST and the slow query log, then fix them with indexes and tuning.nginx returns 502 Bad GatewayFix nginx 502 Bad Gateway errors: read the nginx error log, check the upstream (PHP-FPM, Node, Gunicorn) is running and listening, and fix sockets and timeouts.The SSL certificate has expiredCheck when an SSL certificate expires with openssl, renew Let’s Encrypt certificates with certbot, and fix renewals that silently fail.A cron job is not runningDebug a cron job that does not run or fails silently: check the cron service and logs, PATH and environment differences, permissions and % escaping.
Windows
High CPU usage on Windows ServerFind which process is using high CPU on Windows Server with PowerShell and Get-Counter, and fix common causes like Defender scans, svchost and updates.Windows Server C: drive is fullFind what fills the C: drive on Windows Server with PowerShell and free space safely: WinSxS cleanup, IIS logs, temp files and old update files.High memory usage on Windows ServerFind which processes use the most memory on Windows Server with PowerShell, understand SQL Server memory, and fix leaks and paging.Find what is using a port on WindowsFind the process listening on a port in Windows with PowerShell Get-NetTCPConnection or netstat -ano, and fix “port already in use” errors.
Hear about it before your users do.
CPU, memory, disk, load and service alerts from a one-line agent, with an AI assistant that investigates incidents.