utmp, wtmp - login records

#include <utmp.h>

The utmp file allows one to discover information about who is currently using the system. There may be more users currently using the system, because not all programs use utmp logging.

Warning: utmp must not be writable, because many system programs (foolishly) depend on its integrity. You risk faked system logfiles and modifications of system files if you leave utmp writable to any user.

The file is a sequence of entries with the following structure declared in the include file (note that this is only one of several definitions around; details depend on the version of libc):

 #define UT_UNKNOWN      0
#define RUN_LVL         1
#define BOOT_TIME       2
#define NEW_TIME        3
#define OLD_TIME        4
#define INIT_PROCESS    5
#define LOGIN_PROCESS   6
#define USER_PROCESS    7
#define DEAD_PROCESS    8
#define ACCOUNTING      9

#define UT_LINESIZE     12
#define UT_NAMESIZE     32
#define UT_HOSTSIZE     256

struct exit_status {
    short int e_termination;    /* process termination status */
    short int e_exit;           /* process exit status */
};

struct utmp {
    short ut_type;              /* type of login */
    pid_t ut_pid;               /* PID of login process */
    char ut_line[UT_LINESIZE];  /* device name of tty - "/dev/" */
    char ut_id[4];              /* init id or abbrev. ttyname */
    char ut_user[UT_NAMESIZE];  /* user name */
    char ut_host[UT_HOSTSIZE];  /* hostname for remote login */
    struct exit_status ut_exit; /* The exit status of a process
                                   marked as DEAD_PROCESS */

    /* The ut_session and ut_tv fields must be the same size when
       compiled 32- and 64-bit.  This allows data files and shared
       memory to be shared between 32- and 64-bit applications */
#if __WORDSIZE == 64 && defined __WORDSIZE_COMPAT32
    int32_t ut_session;         /* Session ID, used for windowing */
    struct {
        int32_t tv_sec;         /* Seconds */
        int32_t tv_usec;        /* Microseconds */
    } ut_tv;                    /* Time entry was made */
#else
     long int ut_session;        /* Session ID, used for windowing */
     struct timeval ut_tv;       /* Time entry was made */
#endif

    int32_t ut_addr_v6[4];       /* IP address of remote host */
    char __unused[20];           /* Reserved for future use */
};

/* Backwards compatibility hacks.  */
#define ut_name ut_user
#ifndef _NO_UT_TIME
#define ut_time ut_tv.tv_sec
#endif
#define ut_xtime ut_tv.tv_sec
#define ut_addr ut_addr_v6[0]
 

This structure gives the name of the special file associated with the user's terminal, the user's login name, and the time of login in the form of time(2). String fields are terminated by '\0' if they are shorter than the size of the field.

The first entries ever created result from init(8) processing inittab(5). Before an entry is processed, though, init(8) cleans up utmp by setting ut_type to DEAD_PROCESS, clearing ut_user, ut_host, and ut_time with null bytes for each record which ut_type is not DEAD_PROCESS or RUN_LVL and where no process with PID ut_pid exists. If no empty record with the needed ut_id can be found, init creates a new one. It sets ut_id from the inittab, ut_pid and ut_time to the current values, and ut_type to INIT_PROCESS.

getty(8) locates the entry by the PID, changes ut_type to LOGIN_PROCESS, changes ut_time, sets ut_line, and waits for connection to be established. login(8), after a user has been authenticated, changes ut_type to USER_PROCESS, changes ut_time, and sets ut_host and ut_addr. Depending on getty(8) and login(8), records may be located by ut_line instead of the preferable ut_pid.

When init(8) finds that a process has exited, it locates its utmp entry by ut_pid, sets ut_type to DEAD_PROCESS, and clears ut_user, ut_host and ut_time with null bytes.

xterm(1) and other terminal emulators directly create a USER_PROCESS record and generate the ut_id by using the last two letters of /dev/ttyp%c or by using p%d for /dev/pts/%d. If they find a DEAD_PROCESS for this ID, they recycle it, otherwise they create a new entry. If they can, they will mark it as DEAD_PROCESS on exiting and it is advised that they null ut_line, ut_time, ut_user, and ut_host as well.

telnetd(8) sets up a LOGIN_PROCESS entry and leaves the rest to login(8) as usual. After the telnet session ends, telnetd(8) cleans up utmp in the described way.

The wtmp file records all logins and logouts. Its format is exactly like utmp except that a null user name indicates a logout on the associated terminal. Furthermore, the terminal name ~ with user name shutdown or reboot indicates a system shutdown or reboot and the pair of terminal names |/} logs the old/new system time when date(1) changes it. wtmp is maintained by login(1), init(1), and some versions of getty(8). Neither of these programs creates the file, so if it is removed, record-keeping is turned off.

Note that on biarch platforms, that is, systems which can run both 32-bit and 64-bit applications (x86-64, ppc64, s390x, etc.), ut_tv is the same size in 32-bit mode as in 64-bit mode. The same goes for ut_session and ut_time if they are present. This allows data files and shared memory to be shared between 32-bit and 64-bit applications. Since ut_tv may not be the same as struct timeval, then instead of the call:

/var/run/utmp
/var/log/wtmp

Linux utmp entries conform neither to v7/BSD nor to System V; they are a mix of the two. v7/BSD has fewer fields; most importantly it lacks ut_type, which causes native v7/BSD-like programs to display (for example) dead or login entries. Further, there is no configuration file which allocates slots to sessions. BSD does so because it lacks ut_id fields. In Linux (as in System V), the ut_id field of a record will never change once it has been set, which reserves that slot without needing a configuration file. Clearing ut_id may result in race conditions leading to corrupted utmp entries and potential security holes. Clearing the above mentioned fields by filling them with null bytes is not required by System V semantics, but it allows to run many programs which assume BSD semantics and which do not modify utmp. Linux uses the BSD conventions for line contents, as documented above.

System V only uses the type field to mark them and logs informative messages such as "new time" in the line field. UT_UNKNOWN seems to be a Linux invention. System V has no ut_host or ut_addr_v6 fields.

Unlike various other systems, where utmp logging can be disabled by removing the file, utmp must always exist on Linux. If you want to disable who(1) then do not make utmp world readable.

Note that the utmp struct from libc5 has changed in libc6. Because of this, binaries using the old libc5 struct will corrupt R /var/run/utmp and/or /var/log/wtmp .

The file format is machine dependent, so it is recommended that it be processed only on the machine architecture where it was created.

Note that on platforms which can run both 32-bit and 64-bit applications (x86-64, ppc64, s390x, etc.), the sizes of the fields of a utmp struct must be the same in 32-bit mode as in 64-bit mode. This is achieved by changing the type of ut_session to R int32_t , and that of ut_tv to a struct with two int32_t fields tv_sec and R tv_usec . (Thus, in order to fill it, first get the time into a real struct timeval, then copy the two fields to R ut_tv .)

This man page is based on the libc5 one, things may work differently now.

ac(1), date(1), last(1), login(1), who(1), getutent(3), updwtmp(3), init(8)