PS(1)\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 User Commands\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PS(1)<\/p>\n
NAME<\/p>\n
ps – report a snapshot of the current processes.<\/p>\n
SYNOPSIS<\/p>\n
ps [options]<\/p>\n
DESCRIPTION<\/p>\n
ps displays information about a selection of the active processes.
\nIf you want a repetitive update of the selection and the displayed
\ninformation, use top(1) instead.<\/p>\n
This version of ps accepts several kinds of options:<\/p>\n
1\u00a0\u00a0 UNIX options, which may be grouped and must be preceded by a
\ndash.
\n2\u00a0\u00a0 BSD options, which may be grouped and must not be used with a
\ndash.
\n3\u00a0\u00a0 GNU long options, which are preceded by two dashes.<\/p>\n
Options of different types may be freely mixed, but conflicts can
\nappear.\u00a0 There are some synonymous options, which are functionally
\nidentical, due to the many standards and ps implementations that this
\nps is compatible with.<\/p>\n
Note that “ps -aux” is distinct from “ps aux”.\u00a0 The POSIX and UNIX
\nstandards require that “ps -aux” print all processes owned by a user
\nnamed “x”, as well as printing all processes that would be selected
\nby the -a option.\u00a0 If the user named “x” does not exist, this ps may
\ninterpret the command as “ps aux” instead and print a warning.\u00a0 This
\nbehavior is intended to aid in transitioning old scripts and habits.
\nIt is fragile, subject to change, and thus should not be relied upon.<\/p>\n
By default, ps selects all processes with the same effective user ID
\n(euid=EUID) as the current user and associated with the same terminal
\nas the invoker.\u00a0 It displays the process ID (pid=PID), the terminal
\nassociated with the process (tname=TTY), the cumulated CPU time in
\n[DD-]hh:mm:ss format (time=TIME), and the executable name (ucmd=CMD).
\nOutput is unsorted by default.<\/p>\n
The use of BSD-style options will add process state (stat=STAT) to
\nthe default display and show the command args (args=COMMAND) instead
\nof the executable name.\u00a0 You can override this with the PS_FORMAT
\nenvironment variable. The use of BSD-style options will also change
\nthe process selection to include processes on other terminals (TTYs)
\nthat are owned by you; alternately, this may be described as setting
\nthe selection to be the set of all processes filtered to exclude
\nprocesses owned by other users or not on a terminal.\u00a0 These effects
\nare not considered when options are described as being “identical”
\nbelow, so -M will be considered identical to Z and so on.<\/p>\n
Except as described below, process selection options are additive.
\nThe default selection is discarded, and then the selected processes
\nare added to the set of processes to be displayed.\u00a0 A process will
\nthus be shown if it meets any of the given selection criteria.<\/p>\n
EXAMPLES<\/p>\n
To see every process on the system using standard syntax:
\nps -e
\nps -ef
\nps -eF
\nps -ely<\/p>\n
To see every process on the system using BSD syntax:
\nps ax
\nps axu<\/p>\n
To print a process tree:
\nps -ejH
\nps axjf<\/p>\n
To get info about threads:
\nps -eLf
\nps axms<\/p>\n
To get security info:
\nps -eo euser,ruser,suser,fuser,f,comm,label
\nps axZ
\nps -eM<\/p>\n
To see every process running as root (real & effective ID) in user
\nformat:
\nps -U root -u root u<\/p>\n
To see every process with a user-defined format:
\nps -eo pid,tid,class,rtprio,ni,pri,psr,pcpu,stat,wchan:14,comm
\nps axo stat,euid,ruid,tty,tpgid,sess,pgrp,ppid,pid,pcpu,comm
\nps -Ao pid,tt,user,fname,tmout,f,wchan<\/p>\n
Print only the process IDs of syslogd:
\nps -C syslogd -o pid=<\/p>\n
Print only the name of PID 42:
\nps -q 42 -o comm=<\/p>\n
SIMPLE PROCESS SELECTION<\/p>\n
a\u00a0\u00a0\u00a0\u00a0\u00a0 Lift the BSD-style “only yourself” restriction, which is
\nimposed upon the set of all processes when some BSD-style
\n(without “-“) options are used or when the ps personality
\nsetting is BSD-like.\u00a0 The set of processes selected in this
\nmanner is in addition to the set of processes selected by
\nother means.\u00a0 An alternate description is that this option
\ncauses ps to list all processes with a terminal (tty), or to
\nlist all processes when used together with the x option.<\/p>\n
-A\u00a0\u00a0\u00a0\u00a0 Select all processes.\u00a0 Identical to -e.<\/p>\n
-a\u00a0\u00a0\u00a0\u00a0 Select all processes except both session leaders (see
\ngetsid(2)) and processes not associated with a terminal.<\/p>\n
-d\u00a0\u00a0\u00a0\u00a0 Select all processes except session leaders.<\/p>\n
–deselect
\nSelect all processes except those that fulfill the specified
\nconditions (negates the selection).\u00a0 Identical to -N.<\/p>\n
-e\u00a0\u00a0\u00a0\u00a0 Select all processes.\u00a0 Identical to -A.<\/p>\n
g\u00a0\u00a0\u00a0\u00a0\u00a0 Really all, even session leaders.\u00a0 This flag is obsolete and
\nmay be discontinued in a future release.\u00a0 It is normally
\nimplied by the a flag, and is only useful when operating in
\nthe sunos4 personality.<\/p>\n
-N\u00a0\u00a0\u00a0\u00a0 Select all processes except those that fulfill the specified
\nconditions (negates the selection).\u00a0 Identical to –deselect.<\/p>\n
T\u00a0\u00a0\u00a0\u00a0\u00a0 Select all processes associated with this terminal.\u00a0 Identical
\nto the t option without any argument.<\/p>\n
r\u00a0\u00a0\u00a0\u00a0\u00a0 Restrict the selection to only running processes.<\/p>\n
x\u00a0\u00a0\u00a0\u00a0\u00a0 Lift the BSD-style “must have a tty” restriction, which is
\nimposed upon the set of all processes when some BSD-style
\n(without “-“) options are used or when the ps personality
\nsetting is BSD-like.\u00a0 The set of processes selected in this
\nmanner is in addition to the set of processes selected by
\nother means.\u00a0 An alternate description is that this option
\ncauses ps to list all processes owned by you (same EUID as
\nps), or to list all processes when used together with the a
\noption.<\/p>\n
PROCESS SELECTION BY LIST<\/p>\n
These options accept a single argument in the form of a
\nblank-separated or comma-separated list.\u00a0 They can be used multiple
\ntimes.\u00a0 For example: ps -p “1 2” -p 3,4<\/p>\n
-123\u00a0\u00a0 Identical to –pid 123.<\/p>\n
123\u00a0\u00a0\u00a0 Identical to –pid 123.<\/p>\n
-C cmdlist
\nSelect by command name.\u00a0 This selects the processes whose
\nexecutable name is given in cmdlist.<\/p>\n
-G grplist
\nSelect by real group ID (RGID) or name.\u00a0 This selects the
\nprocesses whose real group name or ID is in the grplist list.
\nThe real group ID identifies the group of the user who created
\nthe process, see getgid(2).<\/p>\n
-g grplist
\nSelect by session OR by effective group name.\u00a0 Selection by
\nsession is specified by many standards, but selection by
\neffective group is the logical behavior that several other
\noperating systems use.\u00a0 This ps will select by session when
\nthe list is completely numeric (as sessions are).\u00a0 Group ID
\nnumbers will work only when some group names are also
\nspecified.\u00a0 See the -s and –group options.<\/p>\n
–Group grplist
\nSelect by real group ID (RGID) or name.\u00a0 Identical to -G.<\/p>\n
–group grplist
\nSelect by effective group ID (EGID) or name.\u00a0 This selects the
\nprocesses whose effective group name or ID is in grplist.\u00a0 The
\neffective group ID describes the group whose file access
\npermissions are used by the process (see getegid(2)).\u00a0 The -g
\noption is often an alternative to –group.<\/p>\n
p pidlist
\nSelect by process ID.\u00a0 Identical to -p and –pid.<\/p>\n
-p pidlist
\nSelect by PID.\u00a0 This selects the processes whose process ID
\nnumbers appear in pidlist.\u00a0 Identical to p and –pid.<\/p>\n
–pid pidlist
\nSelect by process ID.\u00a0 Identical to -p and p.<\/p>\n
–ppid pidlist
\nSelect by parent process ID.\u00a0 This selects the processes with
\na parent process ID in pidlist.\u00a0 That is, it selects processes
\nthat are children of those listed in pidlist.<\/p>\n
q pidlist
\nSelect by process ID (quick mode).\u00a0 Identical to -q and
\n–quick-pid.<\/p>\n
-q pidlist
\nSelect by PID (quick mode).\u00a0 This selects the processes whose
\nprocess ID numbers appear in pidlist.\u00a0 With this option ps
\nreads the necessary info only for the pids listed in the
\npidlist and doesn’t apply additional filtering rules. The
\norder of pids is unsorted and preserved. No additional
\nselection options, sorting and forest type listings are
\nallowed in this mode.\u00a0 Identical to q and –quick-pid.<\/p>\n
–quick-pid pidlist
\nSelect by process ID (quick mode).\u00a0 Identical to -q and q.<\/p>\n
-s sesslist
\nSelect by session ID.\u00a0 This selects the processes with a
\nsession ID specified in sesslist.<\/p>\n
–sid sesslist
\nSelect by session ID.\u00a0 Identical to -s.<\/p>\n
t ttylist
\nSelect by tty.\u00a0 Nearly identical to -t and –tty, but can also
\nbe used with an empty ttylist to indicate the terminal
\nassociated with ps.\u00a0 Using the T option is considered cleaner
\nthan using t with an empty ttylist.<\/p>\n
-t ttylist
\nSelect by tty.\u00a0 This selects the processes associated with the
\nterminals given in ttylist.\u00a0 Terminals (ttys, or screens for
\ntext output) can be specified in several forms: \/dev\/ttyS1,
\nttyS1, S1.\u00a0 A plain “-” may be used to select processes not
\nattached to any terminal.<\/p>\n
–tty ttylist
\nSelect by terminal.\u00a0 Identical to -t and t.<\/p>\n
U userlist
\nSelect by effective user ID (EUID) or name.\u00a0 This selects the
\nprocesses whose effective user name or ID is in userlist.\u00a0 The
\neffective user ID describes the user whose file access
\npermissions are used by the process (see geteuid(2)).
\nIdentical to -u and –user.<\/p>\n
-U userlist
\nSelect by real user ID (RUID) or name.\u00a0 It selects the
\nprocesses whose real user name or ID is in the userlist list.
\nThe real user ID identifies the user who created the process,
\nsee getuid(2).<\/p>\n
-u userlist
\nSelect by effective user ID (EUID) or name.\u00a0 This selects the
\nprocesses whose effective user name or ID is in userlist.<\/p>\n
The effective user ID describes the user whose file access
\npermissions are used by the process (see geteuid(2)).
\nIdentical to U and –user.<\/p>\n
–User userlist
\nSelect by real user ID (RUID) or name.\u00a0 Identical to -U.<\/p>\n
–user userlist
\nSelect by effective user ID (EUID) or name.\u00a0 Identical to -u
\nand U.<\/p>\n
OUTPUT FORMAT CONTROL<\/p>\n
These options are used to choose the information displayed by ps.
\nThe output may differ by personality.<\/p>\n
-c\u00a0\u00a0\u00a0\u00a0 Show different scheduler information for the -l option.<\/p>\n
–context
\nDisplay security context format (for SELinux).<\/p>\n
-f\u00a0\u00a0\u00a0\u00a0 Do full-format listing. This option can be combined with many
\nother UNIX-style options to add additional columns.\u00a0 It also
\ncauses the command arguments to be printed.\u00a0 When used with
\n-L, the NLWP (number of threads) and LWP (thread ID) columns
\nwill be added.\u00a0 See the c option, the format keyword args, and
\nthe format keyword comm.<\/p>\n
-F\u00a0\u00a0\u00a0\u00a0 Extra full format.\u00a0 See the -f option, which -F implies.<\/p>\n
–format format
\nuser-defined format.\u00a0 Identical to -o and o.<\/p>\n
j\u00a0\u00a0\u00a0\u00a0\u00a0 BSD job control format.<\/p>\n
-j\u00a0\u00a0\u00a0\u00a0 Jobs format.<\/p>\n
l\u00a0\u00a0\u00a0\u00a0\u00a0 Display BSD long format.<\/p>\n
-l\u00a0\u00a0\u00a0\u00a0 Long format.\u00a0 The -y option is often useful with this.<\/p>\n
-M\u00a0\u00a0\u00a0\u00a0 Add a column of security data.\u00a0 Identical to Z (for SELinux).<\/p>\n
O format
\nis preloaded o (overloaded).\u00a0 The BSD O option can act like -O
\n(user-defined output format with some common fields
\npredefined) or can be used to specify sort order.\u00a0 Heuristics
\nare used to determine the behavior of this option.\u00a0 To ensure
\nthat the desired behavior is obtained (sorting or formatting),
\nspecify the option in some other way (e.g.\u00a0 with -O or
\n–sort).\u00a0 When used as a formatting option, it is identical to
\n-O, with the BSD personality.<\/p>\n
-O format
\nLike -o, but preloaded with some default columns.\u00a0 Identical
\nto -o pid,format,state,tname,time,command or -o pid,format,
\ntname,time,cmd, see -o below.<\/p>\n
o format
\nSpecify user-defined format.\u00a0 Identical to -o and –format.<\/p>\n
-o format
\nUser-defined format.\u00a0 format is a single argument in the form
\nof a blank-separated or comma-separated list, which offers a
\nway to specify individual output columns.\u00a0 The recognized
\nkeywords are described in the STANDARD FORMAT SPECIFIERS
\nsection below.\u00a0 Headers may be renamed (ps -o pid,
\nruser=RealUser -o comm=Command) as desired.\u00a0 If all column
\nheaders are empty (ps -o pid= -o comm=) then the header line
\nwill not be output.\u00a0 Column width will increase as needed for
\nwide headers; this may be used to widen up columns such as
\nWCHAN (ps -o pid,wchan=WIDE-WCHAN-COLUMN -o comm).\u00a0 Explicit
\nwidth control (ps opid,wchan:42,cmd) is offered too.\u00a0 The
\nbehavior of ps -o pid=X,comm=Y varies with personality; output
\nmay be one column named “X,comm=Y” or two columns named “X”
\nand “Y”.\u00a0 Use multiple -o options when in doubt.\u00a0 Use the
\nPS_FORMAT environment variable to specify a default as
\ndesired; DefSysV and DefBSD are macros that may be used to
\nchoose the default UNIX or BSD columns.<\/p>\n
s\u00a0\u00a0\u00a0\u00a0\u00a0 Display signal format.<\/p>\n
u\u00a0\u00a0\u00a0\u00a0\u00a0 Display user-oriented format.<\/p>\n
v\u00a0\u00a0\u00a0\u00a0\u00a0 Display virtual memory format.<\/p>\n
X\u00a0\u00a0\u00a0\u00a0\u00a0 Register format.<\/p>\n
-y\u00a0\u00a0\u00a0\u00a0 Do not show flags; show rss in place of addr.\u00a0 This option can
\nonly be used with -l.<\/p>\n
Z\u00a0\u00a0\u00a0\u00a0\u00a0 Add a column of security data.\u00a0 Identical to -M (for SELinux).<\/p>\n
OUTPUT MODIFIERS<\/p>\n
c\u00a0\u00a0\u00a0\u00a0\u00a0 Show the true command name.\u00a0 This is derived from the name of
\nthe executable file, rather than from the argv value.\u00a0 Command
\narguments and any modifications to them are thus not shown.
\nThis option effectively turns the args format keyword into the
\ncomm format keyword; it is useful with the -f format option
\nand with the various BSD-style format options, which all
\nnormally display the command arguments.\u00a0 See the -f option,
\nthe format keyword args, and the format keyword comm.<\/p>\n
–cols n
\nSet screen width.<\/p>\n
–columns n
\nSet screen width.<\/p>\n
–cumulative
\nInclude some dead child process data (as a sum with the
\nparent).<\/p>\n
e\u00a0\u00a0\u00a0\u00a0\u00a0 Show the environment after the command.<\/p>\n
f\u00a0\u00a0\u00a0\u00a0\u00a0 ASCII art process hierarchy (forest).<\/p>\n
–forest
\nASCII art process tree.<\/p>\n
h\u00a0\u00a0\u00a0\u00a0\u00a0 No header.\u00a0 (or, one header per screen in the BSD
\npersonality).\u00a0 The h option is problematic.\u00a0 Standard BSD ps
\nuses this option to print a header on each page of output, but
\nolder Linux ps uses this option to totally disable the header.
\nThis version of ps follows the Linux usage of not printing the
\nheader unless the BSD personality has been selected, in which
\ncase it prints a header on each page of output.\u00a0 Regardless of
\nthe current personality, you can use the long options
\n–headers and –no-headers to enable printing headers each
\npage or disable headers entirely, respectively.<\/p>\n
-H\u00a0\u00a0\u00a0\u00a0 Show process hierarchy (forest).<\/p>\n
–headers
\nRepeat header lines, one per page of output.<\/p>\n
k spec Specify sorting order.\u00a0 Sorting syntax is
\n[+|-]key[,[+|-]key[,…]].\u00a0 Choose a multi-letter key from the
\nSTANDARD FORMAT SPECIFIERS section.\u00a0 The “+” is optional since
\ndefault direction is increasing numerical or lexicographic
\norder.\u00a0 Identical to –sort.<\/p>\n
Examples:
\nps jaxkuid,-ppid,+pid
\nps axk comm o comm,args
\nps kstart_time -ef<\/p>\n
–lines n
\nSet screen height.<\/p>\n
-n namelist
\nSet namelist file.\u00a0 Identical to N.\u00a0 The namelist file is
\nneeded for a proper WCHAN display, and must match the current
\nLinux kernel exactly for correct output.\u00a0 Without this option,
\nthe default search path for the namelist is:<\/p>\n
$PS_SYSMAP
\n$PS_SYSTEM_MAP
\n\/proc\/*\/wchan
\n\/boot\/System.map-$(uname -r)
\n\/boot\/System.map
\n\/lib\/modules\/$(uname -r)\/System.map
\n\/usr\/src\/linux\/System.map
\n\/System.map<\/p>\n
n\u00a0\u00a0\u00a0\u00a0\u00a0 Numeric output for WCHAN and USER (including all types of UID
\nand GID).<\/p>\n
N namelist
\nSpecify namelist file.\u00a0 Identical to -n, see -n above.<\/p>\n
–no-headers
\nPrint no header line at all.\u00a0 –no-heading is an alias for
\nthis option.<\/p>\n
O order
\nSorting order (overloaded).\u00a0 The BSD O option can act like -O
\n(user-defined output format with some common fields
\npredefined) or can be used to specify sort order.\u00a0 Heuristics
\nare used to determine the behavior of this option.\u00a0 To ensure
\nthat the desired behavior is obtained (sorting or formatting),
\nspecify the option in some other way (e.g.\u00a0 with -O or
\n–sort).<\/p>\n
For sorting, obsolete BSD O option syntax is
\nO[+|-]k1[,[+|-]k2[,…]].\u00a0 It orders the processes listing
\naccording to the multilevel sort specified by the sequence of
\none-letter short keys k1,k2, …\u00a0 described in the OBSOLETE
\nSORT KEYS section below.\u00a0 The “+” is currently optional,
\nmerely re-iterating the default direction on a key, but may
\nhelp to distinguish an O sort from an O format.\u00a0 The “-”
\nreverses direction only on the key it precedes.<\/p>\n
–rows n
\nSet screen height.<\/p>\n
S\u00a0\u00a0\u00a0\u00a0\u00a0 Sum up some information, such as CPU usage, from dead child
\nprocesses into their parent.\u00a0 This is useful for examining a
\nsystem where a parent process repeatedly forks off short-lived
\nchildren to do work.<\/p>\n
–sort spec
\nSpecify sorting order.\u00a0 Sorting syntax is
\n[+|-]key[,[+|-]key[,…]].\u00a0 Choose a multi-letter key from the
\nSTANDARD FORMAT SPECIFIERS section.\u00a0 The “+” is optional since
\ndefault direction is increasing numerical or lexicographic
\norder.\u00a0 Identical to k.\u00a0 For example: ps jax –sort=uid,-ppid,
\n+pid<\/p>\n
w\u00a0\u00a0\u00a0\u00a0\u00a0 Wide output.\u00a0 Use this option twice for unlimited width.<\/p>\n
-w\u00a0\u00a0\u00a0\u00a0 Wide output.\u00a0 Use this option twice for unlimited width.<\/p>\n
–width n
\nSet screen width.<\/p>\n
THREAD DISPLAY<\/p>\n
H\u00a0\u00a0\u00a0\u00a0\u00a0 Show threads as if they were processes.<\/p>\n
-L\u00a0\u00a0\u00a0\u00a0 Show threads, possibly with LWP and NLWP columns.<\/p>\n
m\u00a0\u00a0\u00a0\u00a0\u00a0 Show threads after processes.<\/p>\n
-m\u00a0\u00a0\u00a0\u00a0 Show threads after processes.<\/p>\n
-T\u00a0\u00a0\u00a0\u00a0 Show threads, possibly with SPID column.<\/p>\n
OTHER INFORMATION<\/p>\n
–help section
\nPrint a help message.\u00a0 The section argument can be one of
\nsimple, list, output, threads, misc or all.\u00a0 The argument can
\nbe shortened to one of the underlined letters as in:
\ns|l|o|t|m|a.<\/p>\n
–info Print debugging info.<\/p>\n
L\u00a0\u00a0\u00a0\u00a0\u00a0 List all format specifiers.<\/p>\n
V\u00a0\u00a0\u00a0\u00a0\u00a0 Print the procps-ng version.<\/p>\n
-V\u00a0\u00a0\u00a0\u00a0 Print the procps-ng version.<\/p>\n
–version
\nPrint the procps-ng version.<\/p>\n
NOTES<\/p>\n
This ps works by reading the virtual files in \/proc.\u00a0 This ps does
\nnot need to be setuid kmem or have any privileges to run.\u00a0 Do not
\ngive this ps any special permissions.<\/p>\n
This ps needs access to namelist data for proper WCHAN display.\u00a0 For
\nkernels prior to 2.6, the System.map file must be installed.<\/p>\n
CPU usage is currently expressed as the percentage of time spent
\nrunning during the entire lifetime of a process.\u00a0 This is not ideal,
\nand it does not conform to the standards that ps otherwise conforms
\nto.\u00a0 CPU usage is unlikely to add up to exactly 100%.<\/p>\n
The SIZE and RSS fields don’t count some parts of a process including
\nthe page tables, kernel stack, struct thread_info, and struct
\ntask_struct.\u00a0 This is usually at least 20 KiB of memory that is
\nalways resident.\u00a0 SIZE is the virtual size of the process (code+data+
\nstack).<\/p>\n
Processes marked <defunct> are dead processes (so-called “zombies”)
\nthat remain because their parent has not destroyed them properly.
\nThese processes will be destroyed by init(8) if the parent process
\nexits.<\/p>\n
If the length of the username is greater than the length of the
\ndisplay column, the numeric user ID is displayed instead.<\/p>\n
Commands options such as ps -aux are not recommended as it is a
\nconfusion of two different standards.\u00a0 According to the POSIX and
\nUNIX standards, the above command asks to display all processes with
\na TTY (generally the commands users are running) plus all processes
\nowned by a user named “x”.\u00a0 If that user doesn’t exist, then ps will
\nassume you really meant “ps aux”.<\/p>\n
PROCESS FLAGS<\/p>\n
The sum of these values is displayed in the “F” column, which is
\nprovided by the flags output specifier:<\/p>\n
1\u00a0\u00a0\u00a0 forked but didn’t exec
\n4\u00a0\u00a0\u00a0 used super-user privileges<\/p>\n
PROCESS STATE CODES<\/p>\n
Here are the different values that the s, stat and state output
\nspecifiers (header “STAT” or “S”) will display to describe the state
\nof a process:<\/p>\n
D\u00a0\u00a0\u00a0 uninterruptible sleep (usually IO)
\nR\u00a0\u00a0\u00a0 running or runnable (on run queue)
\nS\u00a0\u00a0\u00a0 interruptible sleep (waiting for an event to complete)
\nT\u00a0\u00a0\u00a0 stopped by job control signal
\nt\u00a0\u00a0\u00a0 stopped by debugger during the tracing
\nW\u00a0\u00a0\u00a0 paging (not valid since the 2.6.xx kernel)
\nX\u00a0\u00a0\u00a0 dead (should never be seen)
\nZ\u00a0\u00a0\u00a0 defunct (“zombie”) process, terminated but not reaped by
\nits parent<\/p>\n
For BSD formats and when the stat keyword is used, additional
\ncharacters may be displayed:<\/p>\n
<\u00a0\u00a0\u00a0 high-priority (not nice to other users)
\nN\u00a0\u00a0\u00a0 low-priority (nice to other users)
\nL\u00a0\u00a0\u00a0 has pages locked into memory (for real-time and custom
\nIO)
\ns\u00a0\u00a0\u00a0 is a session leader
\nl\u00a0\u00a0\u00a0 is multi-threaded (using CLONE_THREAD, like NPTL
\npthreads do)
\n+\u00a0\u00a0\u00a0 is in the foreground process group<\/p>\n
OBSOLETE SORT KEYS<\/p>\n
These keys are used by the BSD O option (when it is used for
\nsorting).\u00a0 The GNU –sort option doesn’t use these keys, but the
\nspecifiers described below in the STANDARD FORMAT SPECIFIERS section.
\nNote that the values used in sorting are the internal values ps uses
\nand not the “cooked” values used in some of the output format fields
\n(e.g.\u00a0 sorting on tty will sort into device number, not according to
\nthe terminal name displayed).\u00a0 Pipe ps output into the sort(1)
\ncommand if you want to sort the cooked values.<\/p>\n
KEY\u00a0\u00a0 LONG\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 DESCRIPTION
\nc\u00a0\u00a0\u00a0\u00a0 cmd\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 simple name of executable
\nC\u00a0\u00a0\u00a0\u00a0 pcpu\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 cpu utilization
\nf\u00a0\u00a0\u00a0\u00a0 flags\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 flags as in long format F field
\ng\u00a0\u00a0\u00a0\u00a0 pgrp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 process group ID
\nG\u00a0\u00a0\u00a0\u00a0 tpgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 controlling tty process group ID
\nj\u00a0\u00a0\u00a0\u00a0 cutime\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 cumulative user time
\nJ\u00a0\u00a0\u00a0\u00a0 cstime\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 cumulative system time
\nk\u00a0\u00a0\u00a0\u00a0 utime\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 user time
\nm\u00a0\u00a0\u00a0\u00a0 min_flt\u00a0\u00a0\u00a0\u00a0\u00a0 number of minor page faults
\nM\u00a0\u00a0\u00a0\u00a0 maj_flt\u00a0\u00a0\u00a0\u00a0\u00a0 number of major page faults
\nn\u00a0\u00a0\u00a0\u00a0 cmin_flt\u00a0\u00a0\u00a0\u00a0 cumulative minor page faults
\nN\u00a0\u00a0\u00a0\u00a0 cmaj_flt\u00a0\u00a0\u00a0\u00a0 cumulative major page faults
\no\u00a0\u00a0\u00a0\u00a0 session\u00a0\u00a0\u00a0\u00a0\u00a0 session ID
\np\u00a0\u00a0\u00a0\u00a0 pid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 process ID
\nP\u00a0\u00a0\u00a0\u00a0 ppid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 parent process ID
\nr\u00a0\u00a0\u00a0\u00a0 rss\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 resident set size
\nR\u00a0\u00a0\u00a0\u00a0 resident\u00a0\u00a0\u00a0\u00a0 resident pages
\ns\u00a0\u00a0\u00a0\u00a0 size\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 memory size in kilobytes
\nS\u00a0\u00a0\u00a0\u00a0 share\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 amount of shared pages
\nt\u00a0\u00a0\u00a0\u00a0 tty\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 the device number of the controlling tty
\nT\u00a0\u00a0\u00a0\u00a0 start_time\u00a0\u00a0 time process was started
\nU\u00a0\u00a0\u00a0\u00a0 uid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 user ID number
\nu\u00a0\u00a0\u00a0\u00a0 user\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 user name
\nv\u00a0\u00a0\u00a0\u00a0 vsize\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 total VM size in KiB
\ny\u00a0\u00a0\u00a0\u00a0 priority\u00a0\u00a0\u00a0\u00a0 kernel scheduling priority<\/p>\n
AIX FORMAT DESCRIPTORS<\/p>\n
This ps supports AIX format descriptors, which work somewhat like the
\nformatting codes of printf(1) and printf(3).\u00a0 For example, the normal
\ndefault output can be produced with this: ps -eo “%p %y %x %c”.\u00a0 The
\nNORMAL codes are described in the next section.<\/p>\n
CODE\u00a0\u00a0 NORMAL\u00a0\u00a0 HEADER
\n%C\u00a0\u00a0\u00a0\u00a0 pcpu\u00a0\u00a0\u00a0\u00a0 %CPU
\n%G\u00a0\u00a0\u00a0\u00a0 group\u00a0\u00a0\u00a0 GROUP
\n%P\u00a0\u00a0\u00a0\u00a0 ppid\u00a0\u00a0\u00a0\u00a0 PPID
\n%U\u00a0\u00a0\u00a0\u00a0 user\u00a0\u00a0\u00a0\u00a0 USER
\n%a\u00a0\u00a0\u00a0\u00a0 args\u00a0\u00a0\u00a0\u00a0 COMMAND
\n%c\u00a0\u00a0\u00a0\u00a0 comm\u00a0\u00a0\u00a0\u00a0 COMMAND
\n%g\u00a0\u00a0\u00a0\u00a0 rgroup\u00a0\u00a0 RGROUP
\n%n\u00a0\u00a0\u00a0\u00a0 nice\u00a0\u00a0\u00a0\u00a0 NI
\n%p\u00a0\u00a0\u00a0\u00a0 pid\u00a0\u00a0\u00a0\u00a0\u00a0 PID
\n%r\u00a0\u00a0\u00a0\u00a0 pgid\u00a0\u00a0\u00a0\u00a0 PGID
\n%t\u00a0\u00a0\u00a0\u00a0 etime\u00a0\u00a0\u00a0 ELAPSED
\n%u\u00a0\u00a0\u00a0\u00a0 ruser\u00a0\u00a0\u00a0 RUSER
\n%x\u00a0\u00a0\u00a0\u00a0 time\u00a0\u00a0\u00a0\u00a0 TIME
\n%y\u00a0\u00a0\u00a0\u00a0 tty\u00a0\u00a0\u00a0\u00a0\u00a0 TTY
\n%z\u00a0\u00a0\u00a0\u00a0 vsz\u00a0\u00a0\u00a0\u00a0\u00a0 VSZ<\/p>\n
STANDARD FORMAT SPECIFIERS<\/p>\n
Here are the different keywords that may be used to control the
\noutput format (e.g. with option -o) or to sort the selected processes
\nwith the GNU-style –sort option.<\/p>\n
For example: ps -eo pid,user,args –sort user<\/p>\n
This version of ps tries to recognize most of the keywords used in
\nother implementations of ps.<\/p>\n
The following user-defined format specifiers may contain spaces:
\nargs, cmd, comm, command, fname, ucmd, ucomm,
\nlstart, bsdstart, start.<\/p>\n
Some keywords may not be available for sorting.<\/p>\n
CODE\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 HEADER\u00a0\u00a0\u00a0 DESCRIPTION<\/p>\n
%cpu\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 %CPU\u00a0\u00a0\u00a0\u00a0\u00a0 cpu utilization of the process in “##.#”
\nformat.\u00a0 Currently, it is the CPU time used
\ndivided by the time the process has been
\nrunning (cputime\/realtime ratio), expressed as
\na percentage.\u00a0 It will not add up to 100%
\nunless you are lucky.\u00a0 (alias pcpu).<\/p>\n
%mem\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 %MEM\u00a0\u00a0\u00a0\u00a0\u00a0 ratio of the process’s resident set size\u00a0 to
\nthe physical memory on the machine, expressed
\nas a percentage.\u00a0 (alias pmem).<\/p>\n
args\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 COMMAND\u00a0\u00a0 command with all its arguments as a string.
\nModifications to the arguments may be shown.
\nThe output in this column may contain spaces.
\nA process marked <defunct> is partly dead,
\nwaiting to be fully destroyed by its parent.
\nSometimes the process args will be unavailable;
\nwhen this happens, ps will instead print the
\nexecutable name in brackets.\u00a0 (alias
\ncmd, command).\u00a0 See also the comm format
\nkeyword, the -f option, and the c option.
\nWhen specified last, this column will extend to
\nthe edge of the display.\u00a0 If ps can not
\ndetermine display width, as when output is
\nredirected (piped) into a file or another
\ncommand, the output width is undefined (it may
\nbe 80, unlimited, determined by the TERM
\nvariable, and so on).\u00a0 The COLUMNS environment
\nvariable or –cols option may be used to
\nexactly determine the width in this case.\u00a0 The
\nw or -w option may be also be used to adjust
\nwidth.<\/p>\n
blocked\u00a0\u00a0\u00a0\u00a0 BLOCKED\u00a0\u00a0 mask of the blocked signals, see signal(7).
\nAccording to the width of the field, a 32 or
\n64-bit mask in hexadecimal format is displayed.
\n(alias sig_block, sigmask).<\/p>\n
bsdstart\u00a0\u00a0\u00a0 START\u00a0\u00a0\u00a0\u00a0 time the command started.\u00a0 If the process was
\nstarted less than 24 hours ago, the output
\nformat is ” HH:MM”, else it is ” Mmm:SS” (where
\nMmm is the three letters of the month).\u00a0 See
\nalso lstart, start, start_time, and stime.<\/p>\n
bsdtime\u00a0\u00a0\u00a0\u00a0 TIME\u00a0\u00a0\u00a0\u00a0\u00a0 accumulated cpu time, user + system.\u00a0 The
\ndisplay format is usually “MMM:SS”, but can be
\nshifted to the right if the process used more
\nthan 999 minutes of cpu time.<\/p>\n
c\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 C\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 processor utilization. Currently, this is the
\ninteger value of the percent usage over the
\nlifetime of the process.\u00a0 (see %cpu).<\/p>\n
caught\u00a0\u00a0\u00a0\u00a0\u00a0 CAUGHT\u00a0\u00a0\u00a0 mask of the caught signals, see signal(7).
\nAccording to the width of the field, a 32 or 64
\nbits mask in hexadecimal format is displayed.
\n(alias sig_catch, sigcatch).<\/p>\n
cgroup\u00a0\u00a0\u00a0\u00a0\u00a0 CGROUP\u00a0\u00a0\u00a0 display control groups to which the process
\nbelongs.<\/p>\n
class\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 CLS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 scheduling class of the process.\u00a0 (alias
\npolicy, cls).\u00a0 Field’s possible values are:<\/p>\n
–\u00a0\u00a0 not reported
\nTS\u00a0 SCHED_OTHER
\nFF\u00a0 SCHED_FIFO
\nRR\u00a0 SCHED_RR
\nB\u00a0\u00a0 SCHED_BATCH
\nISO SCHED_ISO
\nIDL SCHED_IDLE
\n?\u00a0\u00a0 unknown value<\/p>\n
cls\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 CLS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 scheduling class of the process.\u00a0 (alias
\npolicy, cls).\u00a0 Field’s possible values are:<\/p>\n
–\u00a0\u00a0 not reported
\nTS\u00a0 SCHED_OTHER
\nFF\u00a0 SCHED_FIFO
\nRR\u00a0 SCHED_RR
\nB\u00a0\u00a0 SCHED_BATCH
\nISO SCHED_ISO
\nIDL SCHED_IDLE
\n?\u00a0\u00a0 unknown value<\/p>\n
cmd\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 CMD\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see args.\u00a0 (alias args, command).<\/p>\n
comm\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 COMMAND\u00a0\u00a0 command name (only the executable name).
\nModifications to the command name will not be
\nshown.\u00a0 A process marked <defunct> is partly
\ndead, waiting to be fully destroyed by its
\nparent.\u00a0 The output in this column may contain
\nspaces.\u00a0 (alias ucmd, ucomm).\u00a0 See also the
\nargs format keyword, the -f option, and the c
\noption.
\nWhen specified last, this column will extend to
\nthe edge of the display.\u00a0 If ps can not
\ndetermine display width, as when output is
\nredirected (piped) into a file or another
\ncommand, the output width is undefined (it may
\nbe 80, unlimited, determined by the TERM
\nvariable, and so on).\u00a0 The COLUMNS environment
\nvariable or –cols option may be used to
\nexactly determine the width in this case.\u00a0 The
\nw or -w option may be also be used to adjust
\nwidth.<\/p>\n
command\u00a0\u00a0\u00a0\u00a0 COMMAND\u00a0\u00a0 See args.\u00a0 (alias args, command).<\/p>\n
cp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 CP\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 per-mill (tenths of a percent) CPU usage.\u00a0 (see
\n%cpu).<\/p>\n
cputime\u00a0\u00a0\u00a0\u00a0 TIME\u00a0\u00a0\u00a0\u00a0\u00a0 cumulative CPU time, “[DD-]hh:mm:ss” format.
\n(alias time).<\/p>\n
drs\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 DRS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 data resident set size, the amount of physical
\nmemory devoted to other than executable code.<\/p>\n
egid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 EGID\u00a0\u00a0\u00a0\u00a0\u00a0 effective group ID number of the process as a
\ndecimal integer.\u00a0 (alias gid).<\/p>\n
egroup\u00a0\u00a0\u00a0\u00a0\u00a0 EGROUP\u00a0\u00a0\u00a0 effective group ID of the process.\u00a0 This will
\nbe the textual group ID, if it can be obtained
\nand the field width permits, or a decimal
\nrepresentation otherwise.\u00a0 (alias group).<\/p>\n
eip\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 EIP\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 instruction pointer.<\/p>\n
esp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 ESP\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 stack pointer.<\/p>\n
etime\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 ELAPSED\u00a0\u00a0 elapsed time since the process was started, in
\nthe form [[DD-]hh:]mm:ss.<\/p>\n
etimes\u00a0\u00a0\u00a0\u00a0\u00a0 ELAPSED\u00a0\u00a0 elapsed time since the process was started, in
\nseconds.<\/p>\n
euid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 EUID\u00a0\u00a0\u00a0\u00a0\u00a0 effective user ID (alias uid).<\/p>\n
euser\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 EUSER\u00a0\u00a0\u00a0\u00a0 effective user name.\u00a0 This will be the textual
\nuser ID, if it can be obtained and the field
\nwidth permits, or a decimal representation
\notherwise.\u00a0 The n option can be used to force
\nthe decimal representation.\u00a0 (alias
\nuname, user).<\/p>\n
f\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 F\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 flags associated with the process, see the
\nPROCESS FLAGS section.\u00a0 (alias flag, flags).<\/p>\n
fgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 FGID\u00a0\u00a0\u00a0\u00a0\u00a0 filesystem access group ID.\u00a0 (alias fsgid).<\/p>\n
fgroup\u00a0\u00a0\u00a0\u00a0\u00a0 FGROUP\u00a0\u00a0\u00a0 filesystem access group ID.\u00a0 This will be the
\ntextual group ID, if it can be obtained and the
\nfield width permits, or a decimal
\nrepresentation otherwise.\u00a0 (alias fsgroup).<\/p>\n
flag\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 F\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see f.\u00a0 (alias f, flags).<\/p>\n
flags\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 F\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see f.\u00a0 (alias f, flag).<\/p>\n
fname\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 COMMAND\u00a0\u00a0 first 8 bytes of the base name of the process’s
\nexecutable file.\u00a0 The output in this column may
\ncontain spaces.<\/p>\n
fuid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 FUID\u00a0\u00a0\u00a0\u00a0\u00a0 filesystem access user ID.\u00a0 (alias fsuid).<\/p>\n
fuser\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 FUSER\u00a0\u00a0\u00a0\u00a0 filesystem access user ID.\u00a0 This will be the
\ntextual user ID, if it can be obtained and the
\nfield width permits, or a decimal
\nrepresentation otherwise.<\/p>\n
gid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 GID\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see egid.\u00a0 (alias egid).<\/p>\n
group\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 GROUP\u00a0\u00a0\u00a0\u00a0 see egroup.\u00a0 (alias egroup).<\/p>\n
ignored\u00a0\u00a0\u00a0\u00a0 IGNORED\u00a0\u00a0 mask of the ignored signals, see signal(7).
\nAccording to the width of the field, a 32 or 64
\nbits mask in hexadecimal format is displayed.
\n(alias sig_ignore, sigignore).<\/p>\n
ipcns\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 IPCNS\u00a0\u00a0\u00a0\u00a0 Unique inode number describing the namespace
\nthe process belongs to. See namespaces(7).<\/p>\n
label\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 LABEL\u00a0\u00a0\u00a0\u00a0 security label, most commonly used for SELinux
\ncontext data.\u00a0 This is for the Mandatory Access
\nControl (“MAC”) found on high-security systems.<\/p>\n
lstart\u00a0\u00a0\u00a0\u00a0\u00a0 STARTED\u00a0\u00a0 time the command started.\u00a0 See also
\nbsdstart, start, start_time, and stime.<\/p>\n
lsession\u00a0\u00a0\u00a0 SESSION\u00a0\u00a0 displays the login session identifier of a
\nprocess, if systemd support has been included.<\/p>\n
lwp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 LWP\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 light weight process (thread) ID of the
\ndispatchable entity (alias spid, tid).\u00a0 See tid
\nfor additional information.<\/p>\n
machine\u00a0\u00a0\u00a0\u00a0 MACHINE\u00a0\u00a0 displays the machine name for processes
\nassigned to VM or container, if systemd support
\nhas been included.<\/p>\n
maj_flt\u00a0\u00a0\u00a0\u00a0 MAJFLT\u00a0\u00a0\u00a0 The number of major page faults that have
\noccurred with this process.<\/p>\n
min_flt\u00a0\u00a0\u00a0\u00a0 MINFLT\u00a0\u00a0\u00a0 The number of minor page faults that have
\noccurred with this process.<\/p>\n
mntns\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 MNTNS\u00a0\u00a0\u00a0\u00a0 Unique inode number describing the namespace
\nthe process belongs to. See namespaces(7).<\/p>\n
netns\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 NETNS\u00a0\u00a0\u00a0\u00a0 Unique inode number describing the namespace
\nthe process belongs to. See namespaces(7).<\/p>\n
ni\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 NI\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 nice value. This ranges from 19 (nicest) to -20
\n(not nice to others), see nice(1).\u00a0 (alias
\nnice).<\/p>\n
nice\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 NI\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see ni.(alias ni).<\/p>\n
nlwp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 NLWP\u00a0\u00a0\u00a0\u00a0\u00a0 number of lwps (threads) in the process.
\n(alias thcount).<\/p>\n
nwchan\u00a0\u00a0\u00a0\u00a0\u00a0 WCHAN\u00a0\u00a0\u00a0\u00a0 address of the kernel function where the
\nprocess is sleeping (use wchan if you want the
\nkernel function name).\u00a0 Running tasks will
\ndisplay a dash (‘-‘) in this column.<\/p>\n
ouid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 OWNER\u00a0\u00a0\u00a0\u00a0 displays the Unix user identifier of the owner
\nof the session of a process, if systemd support
\nhas been included.<\/p>\n
pcpu\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 %CPU\u00a0\u00a0\u00a0\u00a0\u00a0 see %cpu.\u00a0 (alias %cpu).<\/p>\n
pending\u00a0\u00a0\u00a0\u00a0 PENDING\u00a0\u00a0 mask of the pending signals. See signal(7).
\nSignals pending on the process are distinct
\nfrom signals pending on individual threads.
\nUse the m option or the -m option to see both.
\nAccording to the width of the field, a 32 or 64
\nbits mask in hexadecimal format is displayed.
\n(alias sig).<\/p>\n
pgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PGID\u00a0\u00a0\u00a0\u00a0\u00a0 process group ID or, equivalently, the process
\nID of the process group leader.\u00a0 (alias pgrp).<\/p>\n
pgrp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PGRP\u00a0\u00a0\u00a0\u00a0\u00a0 see pgid.\u00a0 (alias pgid).<\/p>\n
pid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PID\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 a number representing the process ID (alias
\ntgid).<\/p>\n
pidns\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PIDNS\u00a0\u00a0\u00a0\u00a0 Unique inode number describing the namespace
\nthe process belongs to. See namespaces(7).<\/p>\n
pmem\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 %MEM\u00a0\u00a0\u00a0\u00a0\u00a0 see %mem.\u00a0 (alias %mem).<\/p>\n
policy\u00a0\u00a0\u00a0\u00a0\u00a0 POL\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 scheduling class of the process.\u00a0 (alias
\nclass, cls).\u00a0 Possible values are:<\/p>\n
–\u00a0\u00a0 not reported
\nTS\u00a0 SCHED_OTHER
\nFF\u00a0 SCHED_FIFO
\nRR\u00a0 SCHED_RR
\nB\u00a0\u00a0 SCHED_BATCH
\nISO SCHED_ISO
\nIDL SCHED_IDLE
\n?\u00a0\u00a0 unknown value<\/p>\n
ppid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PPID\u00a0\u00a0\u00a0\u00a0\u00a0 parent process ID.<\/p>\n
pri\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PRI\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 priority of the process.\u00a0 Higher number means
\nlower priority.<\/p>\n
psr\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PSR\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 processor that process is currently assigned
\nto.<\/p>\n
rgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 RGID\u00a0\u00a0\u00a0\u00a0\u00a0 real group ID.<\/p>\n
rgroup\u00a0\u00a0\u00a0\u00a0\u00a0 RGROUP\u00a0\u00a0\u00a0 real group name.\u00a0 This will be the textual
\ngroup ID, if it can be obtained and the field
\nwidth permits, or a decimal representation
\notherwise.<\/p>\n
rss\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 RSS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 resident set size, the non-swapped physical
\nmemory that a task has used (in kiloBytes).
\n(alias rssize, rsz).<\/p>\n
rssize\u00a0\u00a0\u00a0\u00a0\u00a0 RSS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see rss.\u00a0 (alias rss, rsz).<\/p>\n
rsz\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 RSZ\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see rss.\u00a0 (alias rss, rssize).<\/p>\n
rtprio\u00a0\u00a0\u00a0\u00a0\u00a0 RTPRIO\u00a0\u00a0\u00a0 realtime priority.<\/p>\n
ruid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 RUID\u00a0\u00a0\u00a0\u00a0\u00a0 real user ID.<\/p>\n
ruser\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 RUSER\u00a0\u00a0\u00a0\u00a0 real user ID.\u00a0 This will be the textual user
\nID, if it can be obtained and the field width
\npermits, or a decimal representation otherwise.<\/p>\n
s\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 S\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 minimal state display (one character).\u00a0 See
\nsection PROCESS STATE CODES for the different
\nvalues.\u00a0 See also stat if you want additional
\ninformation displayed.\u00a0 (alias state).<\/p>\n
sched\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SCH\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 scheduling policy of the process.\u00a0 The policies
\nSCHED_OTHER (SCHED_NORMAL), SCHED_FIFO,
\nSCHED_RR, SCHED_BATCH, SCHED_ISO, and
\nSCHED_IDLE are respectively displayed as 0, 1,
\n2, 3, 4, and 5.<\/p>\n
seat\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SEAT\u00a0\u00a0\u00a0\u00a0\u00a0 displays the identifier associated with all
\nhardware devices assigned to a specific
\nworkplace, if systemd support has been
\nincluded.<\/p>\n
sess\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SESS\u00a0\u00a0\u00a0\u00a0\u00a0 session ID or, equivalently, the process ID of
\nthe session leader.\u00a0 (alias session, sid).<\/p>\n
sgi_p\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 P\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 processor that the process is currently
\nexecuting on.\u00a0 Displays “*” if the process is
\nnot currently running or runnable.<\/p>\n
sgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SGID\u00a0\u00a0\u00a0\u00a0\u00a0 saved group ID.\u00a0 (alias svgid).<\/p>\n
sgroup\u00a0\u00a0\u00a0\u00a0\u00a0 SGROUP\u00a0\u00a0\u00a0 saved group name.\u00a0 This will be the textual
\ngroup ID, if it can be obtained and the field
\nwidth permits, or a decimal representation
\notherwise.<\/p>\n
sid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SID\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see sess.\u00a0 (alias sess, session).<\/p>\n
sig\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PENDING\u00a0\u00a0 see pending.\u00a0 (alias pending, sig_pend).<\/p>\n
sigcatch\u00a0\u00a0\u00a0 CAUGHT\u00a0\u00a0\u00a0 see caught.\u00a0 (alias caught, sig_catch).<\/p>\n
sigignore\u00a0\u00a0 IGNORED\u00a0\u00a0 see ignored.\u00a0 (alias ignored, sig_ignore).<\/p>\n
sigmask\u00a0\u00a0\u00a0\u00a0 BLOCKED\u00a0\u00a0 see blocked.\u00a0 (alias blocked, sig_block).<\/p>\n
size\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SIZE\u00a0\u00a0\u00a0\u00a0\u00a0 approximate amount of swap space that would be
\nrequired if the process were to dirty all
\nwritable pages and then be swapped out.\u00a0 This
\nnumber is very rough!<\/p>\n
slice\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SLICE\u00a0\u00a0\u00a0\u00a0 displays the slice unit which a process belongs
\nto, if systemd support has been included.<\/p>\n
spid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SPID\u00a0\u00a0\u00a0\u00a0\u00a0 see lwp.\u00a0 (alias lwp, tid).<\/p>\n
stackp\u00a0\u00a0\u00a0\u00a0\u00a0 STACKP\u00a0\u00a0\u00a0 address of the bottom (start) of stack for the
\nprocess.<\/p>\n
start\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 STARTED\u00a0\u00a0 time the command started.\u00a0 If the process was
\nstarted less than 24 hours ago, the output
\nformat is “HH:MM:SS”, else it is ”\u00a0 Mmm dd”
\n(where Mmm is a three-letter month name).\u00a0 See
\nalso lstart, bsdstart, start_time, and stime.<\/p>\n
start_time\u00a0 START\u00a0\u00a0\u00a0\u00a0 starting time or date of the process.\u00a0 Only the
\nyear will be displayed if the process was not
\nstarted the same year ps was invoked, or
\n“MmmDD” if it was not started the same day, or
\n“HH:MM” otherwise.\u00a0 See also
\nbsdstart, start, lstart, and stime.<\/p>\n
stat\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 STAT\u00a0\u00a0\u00a0\u00a0\u00a0 multi-character process state.\u00a0 See section
\nPROCESS STATE CODES for the different values
\nmeaning.\u00a0 See also s and state if you just want
\nthe first character displayed.<\/p>\n
state\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 S\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see s. (alias s).<\/p>\n
suid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SUID\u00a0\u00a0\u00a0\u00a0\u00a0 saved user ID.\u00a0 (alias svuid).<\/p>\n
supgid\u00a0\u00a0\u00a0\u00a0\u00a0 SUPGID\u00a0\u00a0\u00a0 group ids of supplementary groups, if any.\u00a0 See
\ngetgroups(2).<\/p>\n
supgrp\u00a0\u00a0\u00a0\u00a0\u00a0 SUPGRP\u00a0\u00a0\u00a0 group names of supplementary groups, if any.
\nSee getgroups(2).<\/p>\n
suser\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SUSER\u00a0\u00a0\u00a0\u00a0 saved user name.\u00a0 This will be the textual user
\nID, if it can be obtained and the field width
\npermits, or a decimal representation otherwise.
\n(alias svuser).<\/p>\n
svgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SVGID\u00a0\u00a0\u00a0\u00a0 see sgid.\u00a0 (alias sgid).<\/p>\n
svuid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SVUID\u00a0\u00a0\u00a0\u00a0 see suid.\u00a0 (alias suid).<\/p>\n
sz\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 SZ\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 size in physical pages of the core image of the
\nprocess.\u00a0 This includes text, data, and stack
\nspace.\u00a0 Device mappings are currently excluded;
\nthis is subject to change.\u00a0 See vsz and rss.<\/p>\n
tgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TGID\u00a0\u00a0\u00a0\u00a0\u00a0 a number representing the thread group to which
\na task belongs (alias pid).\u00a0 It is the process
\nID of the thread group leader.<\/p>\n
thcount\u00a0\u00a0\u00a0\u00a0 THCNT\u00a0\u00a0\u00a0\u00a0 see nlwp.\u00a0 (alias nlwp).\u00a0 number of kernel
\nthreads owned by the process.<\/p>\n
tid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TID\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 the unique number representing a dispatchable
\nentity (alias lwp, spid).\u00a0 This value may also
\nappear as: a process ID (pid); a process group
\nID (pgrp); a session ID for the session leader
\n(sid); a thread group ID for the thread group
\nleader (tgid); and a tty process group ID for
\nthe process group leader (tpgid).<\/p>\n
time\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TIME\u00a0\u00a0\u00a0\u00a0\u00a0 cumulative CPU time, “[DD-]HH:MM:SS” format.
\n(alias cputime).<\/p>\n
tname\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TTY\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 controlling tty (terminal).\u00a0 (alias tt, tty).<\/p>\n
tpgid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TPGID\u00a0\u00a0\u00a0\u00a0 ID of the foreground process group on the tty
\n(terminal) that the process is connected to, or
\n-1 if the process is not connected to a tty.<\/p>\n
trs\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TRS\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 text resident set size, the amount of physical
\nmemory devoted to executable code.<\/p>\n
tt\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TT\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 controlling tty (terminal).\u00a0 (alias
\ntname, tty).<\/p>\n
tty\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 TT\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 controlling tty (terminal).\u00a0 (alias tname, tt).<\/p>\n
ucmd\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 CMD\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see comm.\u00a0 (alias comm, ucomm).<\/p>\n
ucomm\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 COMMAND\u00a0\u00a0 see comm.\u00a0 (alias comm, ucmd).<\/p>\n
uid\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 UID\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see euid.\u00a0 (alias euid).<\/p>\n
uname\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 USER\u00a0\u00a0\u00a0\u00a0\u00a0 see euser.\u00a0 (alias euser, user).<\/p>\n
unit\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 UNIT\u00a0\u00a0\u00a0\u00a0\u00a0 displays unit which a process belongs to, if
\nsystemd support has been included.<\/p>\n
user\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 USER\u00a0\u00a0\u00a0\u00a0\u00a0 see euser.\u00a0 (alias euser, uname).<\/p>\n
userns\u00a0\u00a0\u00a0\u00a0\u00a0 USERNS\u00a0\u00a0\u00a0 Unique inode number describing the namespace
\nthe process belongs to. See namespaces(7).<\/p>\n
utsns\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 UTSNS\u00a0\u00a0\u00a0\u00a0 Unique inode number describing the namespace
\nthe process belongs to. See namespaces(7).<\/p>\n
uunit\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 UUNIT\u00a0\u00a0\u00a0\u00a0 displays user unit which a process belongs to,
\nif systemd support has been included.<\/p>\n
vsize\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 VSZ\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 see vsz.\u00a0 (alias vsz).<\/p>\n
vsz\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 VSZ\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 virtual memory size of the process in KiB
\n(1024-byte units).\u00a0 Device mappings are
\ncurrently excluded; this is subject to change.
\n(alias vsize).<\/p>\n
wchan\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 WCHAN\u00a0\u00a0\u00a0\u00a0 name of the kernel function in which the
\nprocess is sleeping, a “-” if the process is
\nrunning, or a “*” if the process is
\nmulti-threaded and ps is not displaying
\nthreads.<\/p>\n
ENVIRONMENT VARIABLES<\/p>\n
The following environment variables could affect ps:<\/p>\n
COLUMNS
\nOverride default display width.<\/p>\n
LINES
\nOverride default display height.<\/p>\n
PS_PERSONALITY
\nSet to one of posix, old, linux, bsd, sun, digital…\u00a0 (see
\nsection PERSONALITY below).<\/p>\n
CMD_ENV
\nSet to one of posix, old, linux, bsd, sun, digital…\u00a0 (see
\nsection PERSONALITY below).<\/p>\n
I_WANT_A_BROKEN_PS
\nForce obsolete command line interpretation.<\/p>\n
LC_TIME
\nDate format.<\/p>\n
PS_COLORS
\nNot currently supported.<\/p>\n
PS_FORMAT
\nDefault output format override. You may set this to a format
\nstring of the type used for the -o option.\u00a0 The DefSysV and DefBSD
\nvalues are particularly useful.<\/p>\n
PS_SYSMAP
\nDefault namelist (System.map) location.<\/p>\n
PS_SYSTEM_MAP
\nDefault namelist (System.map) location.<\/p>\n
POSIXLY_CORRECT
\nDon’t find excuses to ignore bad “features”.<\/p>\n
POSIX2
\nWhen set to “on”, acts as POSIXLY_CORRECT.<\/p>\n
UNIX95
\nDon’t find excuses to ignore bad “features”.<\/p>\n
_XPG
\nCancel CMD_ENV=irix non-standard behavior.<\/p>\n
In general, it is a bad idea to set these variables.\u00a0 The one
\nexception is CMD_ENV or PS_PERSONALITY, which could be set to Linux
\nfor normal systems.\u00a0 Without that setting, ps follows the useless and
\nbad parts of the Unix98 standard.<\/p>\n
PERSONALITY<\/p>\n
390\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like the OS\/390 OpenEdition ps<\/p>\n
aix\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like AIX ps
\nbsd\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like FreeBSD ps (totally non-standard)
\ncompaq\u00a0\u00a0\u00a0\u00a0 like Digital Unix ps
\ndebian\u00a0\u00a0\u00a0\u00a0 like the old Debian ps
\ndigital\u00a0\u00a0\u00a0 like Tru64 (was Digital Unix, was OSF\/1) ps
\ngnu\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like the old Debian ps
\nhp\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like HP-UX ps
\nhpux\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like HP-UX ps
\nirix\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like Irix ps
\nlinux\u00a0\u00a0\u00a0\u00a0\u00a0 ***** recommended *****
\nold\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like the original Linux ps (totally non-standard)
\nos390\u00a0\u00a0\u00a0\u00a0\u00a0 like OS\/390 Open Edition ps
\nposix\u00a0\u00a0\u00a0\u00a0\u00a0 standard
\ns390\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like OS\/390 Open Edition ps
\nsco\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like SCO ps
\nsgi\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 like Irix ps
\nsolaris2\u00a0\u00a0 like Solaris 2+ (SunOS 5) ps
\nsunos4\u00a0\u00a0\u00a0\u00a0 like SunOS 4 (Solaris 1) ps (totally non-standard)
\nsvr4\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 standard
\nsysv\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 standard
\ntru64\u00a0\u00a0\u00a0\u00a0\u00a0 like Tru64 (was Digital Unix, was OSF\/1) ps
\nunix\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 standard
\nunix95\u00a0\u00a0\u00a0\u00a0 standard
\nunix98\u00a0\u00a0\u00a0\u00a0 standard<\/p>\n
SEE ALSO<\/p>\n
pgrep(1), pstree(1), top(1), proc(5).<\/p>\n
STANDARDS<\/p>\n
This ps conforms to:<\/p>\n
1\u00a0\u00a0 Version 2 of the Single Unix Specification
\n2\u00a0\u00a0 The Open Group Technical Standard Base Specifications, Issue 6
\n3\u00a0\u00a0 IEEE Std 1003.1, 2004 Edition
\n4\u00a0\u00a0 X\/Open System Interfaces Extension [UP XSI]
\n5\u00a0\u00a0 ISO\/IEC 9945:2003<\/p>\n
AUTHOR<\/p>\n
ps was originally written by Branko Lankester \u27e8lankeste@fwi.uva.nl\u27e9.
\nMichael K. Johnson \u27e8johnsonm@redhat.com\u27e9 re-wrote it significantly to
\nuse the proc filesystem, changing a few things in the process.
\nMichael Shields \u27e8mjshield@nyx.cs.du.edu\u27e9 added the pid-list feature.
\nCharles Blake \u27e8cblake@bbn.com\u27e9 added multi-level sorting, the
\ndirent-style library, the device name-to-number mmaped database, the
\napproximate binary search directly on System.map, and many code and
\ndocumentation cleanups.\u00a0 David Mossberger-Tang wrote the generic BFD
\nsupport for psupdate.\u00a0 Albert Cahalan \u27e8albert@users.sf.net\u27e9 rewrote
\nps for full Unix98 and BSD support, along with some ugly hacks for
\nobsolete and foreign syntax.<\/p>\n
Please send bug reports to \u27e8procps@freelists.org\u27e9.\u00a0 No subscription
\nis required or suggested.<\/p>\n
COLOPHON<\/p>\n
This page is part of the procps-ng (\/proc filesystem utilities)
\nproject.\u00a0 Information about the project can be found at
\n\u27e8http:\/\/sourceforge.net\/projects\/procps-ng\/\u27e9.\u00a0 If you have a bug
\nreport for this manual page, see
\n\u27e8http:\/\/sourceforge.net\/p\/procps-ng\/tickets\/\u27e9.\u00a0 This page was obtained
\nfrom the project’s upstream Git repository
\n(https:\/\/gitorious.org\/procps) on 2014-12-30.\u00a0 If you discover any
\nrendering problems in this HTML version of the page, or you believe
\nthere is a better or more up-to-date source for the page, or you have
\ncorrections or improvements to the information in this COLOPHON
\n(which is not part of the original manual page), send a mail to
\nman-pages@man7.org<\/p>\n
procps-ng\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 July 2014\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PS(1)<\/p>\n","protected":false},"excerpt":{"rendered":"
PS(1)\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 User Commands\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 PS(1) NAME ps – report a snapshot of the current processes. SYNOPSIS ps [options] DESCRIPTION ps displays information about a selection of the active processes. If you want a repetitive update of the selection and the displayed information, use top(1) instead. This version of ps accepts several kinds of options: 1\u00a0\u00a0 UNIX […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_sitemap_exclude":false,"_sitemap_priority":"","_sitemap_frequency":"","footnotes":""},"categories":[8],"tags":[],"class_list":["post-337","post","type-post","status-publish","format-standard","hentry","category-shell"],"a3_pvc":{"activated":false,"total_views":0,"today_views":0},"_links":{"self":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/posts\/337","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=337"}],"version-history":[{"count":0,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=\/wp\/v2\/posts\/337\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=337"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=337"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.linuxboxen.dk\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=337"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}