Updated: 2021/Apr/14


RENICE(8)                   System Manager's Manual                  RENICE(8)

NAME
     renice - alter priority of running processes

SYNOPSIS
     renice priority [[-gpu] who ...] ...
     renice -n increment [[-gpu] who ...] ...

DESCRIPTION
     renice alters the scheduling priority of one or more running processes.
     The first argument is the new priority to apply, or if -n is given, the
     change to make (applied additively) to the priority.  This argument may
     be negative.  (The interpretation of priorities is discussed below.)

     The following who parameters name the target processes, as either process
     IDs, process group IDs, or user names.  The -gpu options control the
     interpretation as follows:

     -g      Interpret who parameters as process group ID's.

     -p      Interpret who parameters as process IDs.  This is the default.

     -u      Interpret who parameters as user names.

     Each who parameter is processed separately and updates the priority of
     the processes it names as follows:

     with -g  All processes in the process group are updated to the selected
              priority.  If an increment is used, the increment is added to
              the highest priority found among the members of the process
              group prior to the change.

     with -p  The named process is updated to the selected priority.  If an
              increment is used, the increment is added to the process's
              previous priority.

     with -u  All processes belonging to the specified user are updated to the
              selected priority.  If an increment is used, the increment is
              added to the highest priority found among the processes
              belonging to the user prior to the change.

     In conventional terminology a "high priority" process receives a lot of
     CPU time and a "low priority" process receives relatively little.
     "Niceness" is the inverse concept: a process with a high niceness level
     receives relatively little CPU time.  It is about the process being nice
     to the rest of the system, rather than the system being nice to the
     process.

     The numerical priority values accepted by renice are called priorities
     but are actually nicenesses.  They range from PRIO_MIN (-20) to PRIO_MAX
     (20).  PRIO_MIN is the highest priority, lowest niceness, and receives
     the most CPU time.  PRIO_MAX is the lowest priority, highest niceness,
     and receives the least CPU time.  This is confusing but enshrined in
     historical practice and standards.  If in doubt, check with ps(1):
     processes running with elevated priority (getting more CPU time) include
     `<' in the FLAGS column; processes running with reduced priority (getting
     less CPU time) show `N' for "nice" in FLAGS.  The default priority is 0.

     At priority 20, processes will specifically run only when nothing else
     wants to.

     Users other than the super-user may only alter the priority of processes
     they own, and only by increasing the niceness.  (This prevents overriding
     administrative fiats.)  The super-user may alter the priority of any
     process to any legal value.

FILES
     /etc/passwd  to map user names to user ID's

EXAMPLES
           renice +1 987 -u daemon root -p 32

     changes the priority of process ID's 987 and 32, and all processes owned
     by users daemon and root.

SEE ALSO
     nice(1), prenice(1), getpriority(2), setpriority(2)

HISTORY
     The renice command appeared in 4.0BSD.

BUGS
     Non-super-users cannot increase scheduling priorities of their own
     processes, even if they were the ones that decreased the priorities in
     the first place.

NetBSD 9.99                    October 22, 2020                    NetBSD 9.99