1VMSTAT(1) General Commands Manual VMSTAT(1)
2
6 vmstat - report virtual memory statistics
7
9 vmstat [ -fsi ] [ drives ] [ interval [ count ] ]
10
12 Vmstat delves into the system and normally reports certain statistics
13 kept about process, virtual memory, disk, trap and cpu activity. If
14 given a -f argument, it instead reports on the number of forks and
15 vforks since system startup and the number of pages of virtual memory
16 involved in each kind of fork. If given a -s argument, it instead
17 prints the contents of the sum structure, giving the total number of
18 several kinds of paging related events which have occurred since boot.
19 If given a -i argument, it instead reports on the number of interrupts
20 taken by each device since system startup.
21 If none of these options are given, vmstat will report in the first
23 line a summary of the virtual memory activity since the system has been
24 booted. If interval is specified, then successive lines are summaries
25 over the last interval seconds. ``vmstat 5'' will print what the sys‐
26 tem is doing every five seconds; this is a good choice of printing
27 interval since this is how often some of the statistics are sampled in
28 the system; others vary every second, running the output for a while
29 will make it apparent which are recomputed every second. If a count is
30 given, the statistics are repeated count times. The format fields are:
31 Procs: information about numbers of processes in various states.
33 r in run queue
35 b blocked for resources (i/o, paging, etc.)
36 w runnable or short sleeper (< 20 secs) but swapped
37 Memory: information about the usage of virtual and real memory. Vir‐
39 tual pages are considered active if they belong to processes which are
40 running or have run in the last 20 seconds. A ``page'' here is 1024
41 bytes.
42 avm active virtual pages
44 fre size of the free list
45 Page: information about page faults and paging activity. These are
47 averaged each five seconds, and given in units per second.
48 re page reclaims (simulating reference bits)
50 at pages attached (found in free list)
51 pi pages paged in
52 po pages paged out
53 fr pages freed per second
54 de anticipated short term memory shortfall
55 sr pages scanned by clock algorithm, per-second
56 up/hp/rk/ra: Disk operations per second (this field is system depen‐
58 dent). Typically paging will be split across several of the available
59 drives. The number under each of these is the unit number.
60 Faults: trap/interrupt rate averages per second over last 5 seconds.
62 in (non clock) device interrupts per second
64 sy system calls per second
65 cs cpu context switch rate (switches/sec)
66 Cpu: breakdown of percentage usage of CPU time
68 us user time for normal and low priority processes
70 sy system time
71 id cpu idle
72 If more than 4 disk drives are configured in the system, vmstat dis‐
74 plays only the first 4 drives, with priority given to Massbus disk
75 drives (i.e. if both Unibus and Massbus drives are present and the
76 total number of drives exceeds 4, then some number of Unibus drives
77 will not be displayed in favor of the Massbus drives). To force vmstat
78 to display specific drives, their names may be supplied on the command
79 line.
80
82 /dev/kmem, /vmunix
83
85 systat(1), iostat(1)
86 The sections starting with ``Interpreting system activity'' in
88 Installing and Operating 4.2bsd.
894th Berkeley Distribution March 15, 1986 VMSTAT(1)