![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to vmstat.1 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Power 6/32 Unix Tahoe 4.2BSD] / cci / usr / src / man / man1|
Return to vmstat.1 CVS log | Up to [Power 6/32 Unix Tahoe 4.2BSD] / cci / usr / src / man / man1 |
1.1 root 1: .TH VMSTAT 1 "26 April 1981"
2: .de s1
3: .if n .sp
4: .if t .sp .1i
5: ..
6: .de t1
7: .if n .ta 5n
8: .if t .ta 1i
9: ..
10: .UC 4
11: .SH NAME
12: vmstat \- report virtual memory statistics
13: .SH SYNOPSIS
14: .B vmstat
15: [
1.1.1.2 ! root 16: .B \-01rfs
1.1 root 17: ]
18: [ interval [ count ] ]
19: .SH DESCRIPTION
20: .I Vmstat
21: delves into the system and normally reports certain statistics kept about
22: process, virtual memory, disk, trap and cpu activity.
1.1.1.2 ! root 23: .PP
! 24: The POWER 6/32 can support up to 32 disk drives.
! 25: .I Vmstat
! 26: can only display 16 drives at one time.
! 27: The
! 28: .B \-0
! 29: argument will display the first 16 mounted disk drives, and the
! 30: .B \-1
! 31: argument will display the second 16 mounted disk drives.
! 32: These options take effect only if the system has more
! 33: than 16 mounted drives.
! 34: The
! 35: .B \-0
! 36: argument is the default.
! 37: Any time during the display, you can type in
! 38: a 0, 1 or r to change the display.
! 39: .PP
! 40: The
! 41: .B \-r
! 42: argument, switches between displaying the first and second
! 43: 16 mounted disk drives, i.e.,
! 44: it switches back and forth between the 0 and 1 options.
! 45: This option takes effect only if the system has more
! 46: than 16 mounted drives.
! 47: The duration for each set of drives will be [interval]/2.
! 48: .PP
! 49: If given an
1.1 root 50: .B \-f
1.1.1.2 ! root 51: argument,
! 52: .I vmstat
! 53: reports on the number of
1.1 root 54: .I forks
55: and
56: .I vforks
1.1.1.2 ! root 57: since system startup and the number
! 58: of pages of virtual memory involved in each
1.1 root 59: kind of fork.
60: If given a
61: .B \-s
62: argument, it instead prints the contents
63: of the
64: .I sum
1.1.1.2 ! root 65: structure, giving the total number
! 66: of several kinds of paging related
1.1 root 67: events which have occurred since boot.
68: .PP
69: If none of these options are given,
70: .I vmstat
1.1.1.2 ! root 71: will report in the first line a summary
! 72: of the virtual memory activity
1.1 root 73: since the system has been booted.
74: If
75: .I interval
1.1.1.2 ! root 76: is specified, then successive lines
! 77: are summaries over the last
1.1 root 78: .I interval
79: seconds.
1.1.1.2 ! root 80: ``vmstat 5'' will print what the system
! 81: is doing every five seconds. This is a good choice
! 82: of printing interval since this is how often
! 83: some of the statistics are sampled in
! 84: the system. Others vary every
! 85: second. Running the output for a while will make it
! 86: apparent which statistics are recomputed every second.
1.1 root 87: If a
88: .I count
89: is given, the statistics are repeated
90: .I count
91: times.
92: The format fields are:
93: .PP
94: Procs: information about numbers of processes in various states.
95: .s1
96: .t1
97: .nf
98: r in run queue
99: b blocked for resources (i/o, paging, etc.)
100: w runnable or short sleeper (< 20 secs) but swapped
101: .fi
102: .s1
1.1.1.2 ! root 103: Memory: information about the usage of
! 104: virtual and real memory. Virtual pages
! 105: are considered active if they belong to processes which
1.1 root 106: are running or have run in the last 20 seconds.
107: A ``page'' here is 1024 bytes.
108: .s1
109: .t1
110: .nf
111: avm active virtual pages
112: fre size of the free list
113: .fi
114: .s1
115: Page: information about page faults and paging activity.
1.1.1.2 ! root 116: These are averaged each five seconds,
! 117: and given in units per second.
1.1 root 118: .s1
119: .t1
120: .nf
1.1.1.2 ! root 121: re page reclaims less the free lists in the swap
! 122: devices and the file system
! 123: at pages of free lists in the swap devices and the file system
1.1 root 124: pi pages paged in
125: po pages paged out
126: fr pages freed per second
127: de anticipated short term memory shortfall
128: sr pages scanned by clock algorithm, per-second
129: .fi
130: .s1
1.1.1.2 ! root 131: x/f/s: Disk operations per second (this
! 132: field is system dependent).
! 133: Typically paging will be split across
! 134: several of the available drives.
1.1 root 135: The number under each of these is the unit number.
136: .s1
137: Faults: trap/interrupt rate averages per second over last 5 seconds.
138: .s1
139: .t1
140: .nf
141: in (non clock) device interrupts per second
142: sy system calls per second
143: cs cpu context switch rate (switches/sec)
144: .fi
145: .s1
146: Cpu: breakdown of percentage usage of CPU time
147: .s1
148: .nf
149: us user time for normal and low priority processes
150: sy system time
151: id cpu idle
152: .SH FILES
153: /dev/kmem, /vmunix
154: .SH SEE ALSO
155: The sections starting with ``Interpreting system activity'' in
156: .IR "Installing and Operating 4.2bsd" .
157: .SH AUTHORS
158: William Joy and Ozalp Babaoglu
159: .SH BUGS
160: There should be a screen oriented program which combines
161: .I vmstat
162: and
163: .IR ps (1)
164: in real time
165: as well as reporting on other system activity.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.