 |
Index for
Section 8 |
|
 |
Alphabetical
listing for D |
|
 |
Bottom of
page |
|
diskusg(8)
NAME
diskusg - Generates disk accounting data by user ID
SYNOPSIS
/usr/sbin/acct/diskusg [-U number] [-s] [-v] [-p pw_filename] [-u
u_filename] [-i ignlist] [file_spec...]
OPTIONS
-U number
Sets the number of internal allocated user structures to the specified
number, one for each user. The default is 1000.
-i ignlist
Ignores the data in the specified ignlist file system. The ignlist
operand specifies a list of file system names, separated with commas or
enclosed by quotation marks.
-p pw_filename
Uses the specified pw_filename file as the password file used to
generate login names. The default password file is /etc/passwd.
-s Combines all records for a single user into a single record. Input
data must be in the same format as the output of the diskusg command.
If you specify the file_spec operand with the -s option, the operand
must be an ASCII file containing data in the appropriate format. If
the file_spec operand is not specified, input data will be taken from
standard input.
-u u_filename
Writes a record to u_filename for each file that has changed its user
ID to user no one. Each record consists of the special file name, the
inode number, and the user ID.
-v Writes a list of all files charged to user no one to standard error.
OPERANDS
file_spec...
Specifies one or more file systems. How you specify a file system
depends on whether it is UFS or AdvFS and whether it is mounted.
To specify a UFS file system that is mounted, enter the name of its
character device special file, for example: /dev/rrz3c.
To specify a UFS file system that is not mounted, enter the name of its
block device special file, for example: /dev/rz3c.
To specify an AdvFS fileset that is mounted, enter the name of the file
domain, a pound-sign(#) character, and the name of the fileset, for
example: root_domain#root.
You cannot specify an AdvFS file system that is not mounted.
DESCRIPTION
The diskusg command generates intermediate disk accounting information
about files in the specified file_spec or from standard input.
The diskusg command obtains user login names from the /etc/passwd file, by
default, and reports one record per user to standard output. The command
usually reads only the inodes of the specified file system.
The diskusg command is usually called from the dodisk shell procedure when
the cron daemon executes commands in the /usr/spool/cron/crontabs/root
file. You can also manually invoke the command.
If the diskusg command is executed from the dodisk command, its output is
directed to the /var/adm/dtmp file, which is used as input to the acctdisk
command to produce a total accounting record. This total accounting record
can be merged with other total accounting records to produce a daily
report. Note that if the -o option is specified with the dodisk command,
the acctdusg command is used instead of the diskusg command. The acctdusg
command provides a more thorough accounting of disk usage than the diskusg
command.
Records output by the diskusg command are in the following format:
user_ID login_name disk_blocks
The output contains the following information:
user_ID
Assigned user number
login_name
User login name
disk_blocks
Total number of disk blocks allocated to the user
NOTES
This command can be used only for local devices.
The command's behavior when it encounters a nonexisting UFS file system is
different from its behavior when it encounters a nonexisting AdvFS fileset.
If the nonexisting file_spec is a UFS file system, the command produces an
error and continue to display output for other file systems and filesets.
If the missing file_spec is an AdvFS fileset, the command exits without
producing any output.
EXAMPLES
1. To manually invoke the diskusg command, you must be logged in as
superuser. For example, you could enter the following command:
# /usr/sbin/acct/diskusg /dev/rrz3a
0 root 63652
1 daemon 84
2 bin 71144
4 adm 976
5 uucp 3324
322 homer 2
521 whistler 2
943 cellini 363
1016 pollock 92
1098 hopper 317
2. To generate daily disk accounting information, add a line similar to
the following to the /usr/spool/cron/crontabs/root file:
0 2 * * 4 /usr/sbin/acct/dodisk
3. The previous example shows a typical, periodically invoked command
that the cron daemon reads and executes. The time period is expressed
by a 6-field entry using the format:
mm hh dayofmonth month wkday command
The previous format shows the following information:
mm The number of minutes past the hour, from 0 to 59
hh The hour of the day in 24-hour clock notation
dayofmonth
The day of the month
month
The month, from 1 to 12
wkday
The day or days of the week, from 0 to 6, where 0 is Sunday and
inclusive days are separated with a - (hyphen)
command
The command that the cron daemon must execute.
Unspecified times must use an * (asterisk) to define an empty field.
4. In the previous example, the dodisk shell procedure runs at 02:00
hours (2) every Thursday (4). Usually, the dodisk shell procedure
calls the diskusg command to redirect its output to a temporary file
and then calls acctdisk to redirect disk usage records from the
temporary file as input to the /var/adm/acct/nite/[filename] file as
output. The file stored in the /var/adm/acct/nite subdirectory is a
permanent binary record of disk usage for the specified period.
FILES
/usr/sbin/acct/diskusg
Specifies the command path.
/usr/include/sys/acct.h, /usr/include/utmp.h
Accounting header files that define formats for writing accounting
files.
/etc/passwd
User database file.
SEE ALSO
Commands: acct(8), acctdisk(8), acctdusg(8), acctmerg(8), dodisk(8),
runacct(8)
Functions: acct(2)
System Administration
|
Index for
Section 8 |
|
|
Alphabetical
listing for D |
|
 |
Top of
page |
|