atop
Atop is an ASCII full-screen performance monitor that is capable of reporting the activity of all processes (even if processes have finished during the interval), daily logging of system and process activity for long-term analysis, highlighting overloaded system resources by using colors, etc. At regular intervals, it shows system-level activity related to the CPU, memory, swap, disks, and network layers, and for every active process it shows the CPU utilization, the memory growth, priority, username, state, and exit code.
| Tags | Monitoring Networking Systems Administration |
|---|---|
| Licenses | GPL |
| Operating Systems | POSIX Linux |
| Implementation | C |
Recent releases
- 06 Mar 2008 05:20
Changes: The atop command shows the number of running and sleeping (interruptible/non-interruptible) threads per process and shows the PPID per process. The command atopsar is now able to make summary reports and shows information about the top 3 most active processes per sample. In both commands, several bugs have been fixed.
- 20 Nov 2007 03:16
Changes: This release added colors and (on request) markers in the reports generated by the command atopsar to highlight critical resource consumption. Several major bugs were fixed in atop/atopsar, mainly for larger and busier systems.
- 20 Aug 2007 05:22
Changes: A new command, atopsar, has been added to generate detailed system activity reports (similar to the UNIX command sar) using the compressed logfiles as written by atop. These reports can also be generated for the current situation by specifying an interval for atopsar on the command line.
- 28 Mar 2007 03:56
Changes: Unformatted output can be produced to do further processing via scripts. Steal-percentage and disk-counters are supported for virtual machines. The scripts for daily logging have been improved. More than 65535 processes can be logged per interval now. Standard disk I/O statistics per process will be shown (in Linux kernel 2.6.20 and higher).
- 18 Jan 2007 06:42
Changes: A busy percentage is shown for each network-interface. Colors are used to indicate that a particular hardware resource has (almost) reached a critical busy percentage. The most critical resource is automatically determined to adapt the sorting order for active processes dynamically. Minor bugs were fixed.
- All comments
Recent commentsgreat tool!
atop is a tool, which is not like any others I know. Try it, for it will give you other information about your system then "conventional" tools like top. I use it on our productive systems. At least once a day atop runs on all machines, to observe the difference in processes between two intervals.