Capture live CPU, disk, network and memory metrics in one command. 03.07.2026 | reading time: 2 min dstat brings together CPU, disk, network and memory statistics into one live display, letting the administrator spot resource bottlenecks quickly and without juggling multiple tools. Live troubleshooting example Run dstat to sample five times at one second intervals and watch CPU, disk, network and memory: ```bash $ dstat -cndm 1 5 ----total-cpu-usage---- -dsk/total- ----net/total- ---memory-usage- usr sys idl wai| read writ| recv send | used buff cach 3 1 95 1 | 0B 4k | 1k 2k | 512M 20M 300M 4 2 93 1 | 8k 16k | 2k 1k | 514M 20M 300M 2 1 96 1 | 16k 8k | 0k 0k | 513M 20M 300M 3 1 95 1 | 0B 12k | 3k 2k | 515M 20M 300M 1 0 98 1 | 4k 0B | 1k 1k | 512M 20M 300M ``` Useful options and plugins Use `-c`, `-d`, `-n` and `-m` to select CPU, disk, network and memory columns; save output with `--output` for CSV analysis, add `-t` for timestamps and try `--top-cpu` or `--top-io` to see offending processes, or write a small Python plugin to capture a custom metric. When dstat isn't enough For historical collection or very low overhead sampling use `sar` or `vmstat`, for deep disk analysis use `iostat`, and for per-process live views use `htop` or per-connection network views like `iftop` or `nethogs` to complement dstat's aggregated picture. Next steps Mastering dstat speeds up troubleshooting and supports smarter tuning decisions. To dive deeper, consider certifications such as CompTIA Linux+ or LPIC-1 and intensive exam preparation at bitsandbytes.academy. Join Bits & Bytes Academy First class LINUX exam preparation. utilities troubleshooting processes network storage Dieser Artikel wurde mithilfe von KI erstellt.
