Monitoring system load with vmstatEdit
Little trick for monitoring CPU, memory, I/O load during a long running process:
$ vmstat -n -S M -t 1
Options:
-n
: display header line once at start rather than periodically.-S M
: use megabyte (1,048,576) units.-t
: append timestamp on each line.1
: delay 1 second between updates.
Sample output from a mostly idle machine:
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu----- -----timestamp-----
r b swpd free buff cache si so bi bo in cs us sy id wa st CET
2 0 0 123963 656 1760 0 0 1 1 5 1 0 0 100 0 0 2022-02-19 21:40:06
0 0 0 123962 656 1760 0 0 0 0 725 1270 0 0 100 0 0 2022-02-19 21:40:07
0 0 0 123962 656 1760 0 0 0 56 735 1293 0 0 100 0 0 2022-02-19 21:40:08
0 0 0 123967 656 1760 0 0 0 0 893 1436 0 0 100 0 0 2022-02-19 21:40:09
0 0 0 123967 656 1760 0 0 0 36 707 1287 0 0 100 0 0 2022-02-19 21:40:10
0 0 0 123966 656 1760 0 0 0 0 699 1264 0 0 100 0 0 2022-02-19 21:40:11
And here’s a sample from a machine running stress -c 32
:
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu----- -----timestamp-----
r b swpd free buff cache si so bi bo in cs us sy id wa st CET
32 0 0 123933 657 1785 0 0 1 1 5 1 0 0 100 0 0 2022-02-19 21:43:52
32 0 0 123933 657 1785 0 0 0 0 9949 1088 100 0 0 0 0 2022-02-19 21:43:53
32 0 0 123933 657 1785 0 0 0 0 9912 1146 100 0 0 0 0 2022-02-19 21:43:54
32 0 0 123933 657 1785 0 0 0 0 9791 969 100 0 0 0 0 2022-02-19 21:43:55
32 0 0 123933 657 1785 0 0 0 84 9793 973 100 0 0 0 0 2022-02-19 21:43:56
32 0 0 123933 657 1785 0 0 0 0 9814 992 100 0 0 0 0 2022-02-19 21:43:57
32 0 0 123933 657 1785 0 0 0 0 9784 905 100 0 0 0 0 2022-02-19 21:43:58