VMSTAT(1) —

NAME

vmstat − report virtual memory statistics (includes NFS extensions)

SYNOPSIS

vmstat [ −fsi ] [ drives ] [ interval [ count ] ]

DESCRIPTION

Vmstat delves into the system and normally reports certain statistics kept about process, virtual memory, disk, trap and cpu activity. If given a −f argument, it instead reports on the number of forks and vforks since system startup and the number of pages of virtual memory involved in each kind of fork. If given a −s argument, it instead prints the contents of the sum structure, giving the total number of several kinds of paging related events which have occurred since boot. If given a −i argument, it instead reports on the number of interrupts taken by each device since system startup.

If none of these options are given, vmstat will report in the first line a summary of the virtual memory activity since the system has been booted. If interval is specified, then successive lines are summaries over the last interval seconds. “vmstat 5” will print what the system is doing every five seconds; this is a good choice of printing interval since this is how often some of the statistics are sampled in the system; others vary every second, running the output for a while will make it apparent which are recomputed every second. If a count is given, the statistics are repeated count times. The format fields are:

Procs: information about numbers of processes in various states.

rin run queue
bblocked for resources (i/o, paging, etc.)
wrunnable or short sleeper (< 20 secs) but swapped

Memory: information about the usage of virtual and real memory. Virtual pages are considered active if they belong to processes which are running or have run in the last 20 seconds. A “page” here is pagesize(1) bytes.

avmactive virtual pages
fresize of the free list

Page: information about page faults and paging activity. These are averaged each five seconds, and given in units per second.

repage reclaims (simulating reference bits)
atpages attached (found in free list)
pipages paged in
popages paged out
frpages freed per second
deanticipated short term memory shortfall
srpages scanned by clock algorithm, per-second

hd: Disk operations per second (this field is system dependent). Typically paging is split across several of the available drives. The number under each of these is the unit number. Faults: trap/interrupt rate averages per second over last 5 seconds.

in(non clock) device interrupts per second
sysystem calls per second
cscpu context switch rate (switches/sec)

Cpu: breakdown of percentage usage of CPU time

ususer time for normal and low priority processes
sysystem time
idcpu idle

To force vmstat to display specific drives, their names may be supplied on the command line.

NOTES

In NFS, vmstat -s returns a different display of the name lookup cache than 4.3BSD.

NFS:

117952 total name lookups (cache hits 80%)
hits 94696, misses 23256, toolong lookup 0, toolong enter 17755,
LRU empty 42, purges 45

4.3BSD:

25890144 total name lookups (cache hits 81% system, 10% per-process)
badhits 66549, falsehits 206464, toolong 99982

The NFS information is vnode-based and does not use namei.

FILES

/dev/kmem, /vmunix

Museum