NAME

     pidfile_open, pidfile_write, pidfile_close, pidfile_remove, pidfile_fileno — library for PID
     files handling

LIBRARY

     Utility functions from BSD systems (libbsd, -lbsd)

SYNOPSIS

     #include <libutil.h>
     (See libbsd(7) for include usage.)

     struct pidfh *
     pidfile_open(const char *path, mode_t mode, pid_t *pidptr);

     int
     pidfile_write(struct pidfh *pfh);

     int
     pidfile_close(struct pidfh *pfh);

     int
     pidfile_remove(struct pidfh *pfh);

     int
     pidfile_fileno(struct pidfh *pfh);

DESCRIPTION

     The pidfile family of functions allows daemons to handle PID files.  It uses flopen(3bsd) to
     lock a pidfile and detect already running daemons.

     The pidfile_open() function opens (or creates) a file specified by the path argument and
     locks it.  If pidptr argument is not NULL and file can not be locked, the function will use
     it to store a PID of an already running daemon or -1 in case daemon did not write its PID
     yet.  The function does not write process' PID into the file here, so it can be used before
     fork()ing and exit with a proper error message when needed.  If the path argument is NULL,
     /var/run/⟨progname⟩.pid file will be used.  The pidfile_open() function sets the O_CLOEXEC
     close-on-exec flag when opening the pidfile.

     The pidfile_write() function writes process' PID into a previously opened file.  The file is
     truncated before write, so calling the pidfile_write() function multiple times is supported.

     The pidfile_close() function closes a pidfile.  It should be used after daemon fork()s to
     start a child process.

     The pidfile_remove() function closes and removes a pidfile.

     The pidfile_fileno() function returns the file descriptor for the open pidfile.

RETURN VALUES

     The pidfile_open() function returns a valid pointer to a pidfh structure on success, or NULL
     if an error occurs.  If an error occurs, errno will be set.

     The pidfile_write(), pidfile_close(), and pidfile_remove() functions return the value 0 if
     successful; otherwise the value -1 is returned and the global variable errno is set to
     indicate the error.

     The pidfile_fileno() function returns the low-level file descriptor.  It returns -1 and sets
     errno if a NULL pidfh is specified, or if the pidfile is no longer open.

EXAMPLES

     The following example shows in which order these functions should be used.  Note that it is
     safe to pass NULL to pidfile_write(), pidfile_remove(), pidfile_close() and pidfile_fileno()
     functions.

     struct pidfh *pfh;
     pid_t otherpid, childpid;

     pfh = pidfile_open("/var/run/daemon.pid", 0600, &otherpid);
     if (pfh == NULL) {
             if (errno == EEXIST) {
                     errx(EXIT_FAILURE, "Daemon already running, pid: %jd.",
                         (intmax_t)otherpid);
             }
             /* If we cannot create pidfile from other reasons, only warn. */
             warn("Cannot open or create pidfile");
             /*
              * Even though pfh is NULL we can continue, as the other pidfile_*
              * function can handle such situation by doing nothing except setting
              * errno to EINVAL.
              */
     }

     if (daemon(0, 0) == -1) {
             warn("Cannot daemonize");
             pidfile_remove(pfh);
             exit(EXIT_FAILURE);
     }

     pidfile_write(pfh);

     for (;;) {
             /* Do work. */
             childpid = fork();
             switch (childpid) {
             case -1:
                     syslog(LOG_ERR, "Cannot fork(): %s.", strerror(errno));
                     break;
             case 0:
                     pidfile_close(pfh);
                     /* Do child work. */
                     break;
             default:
                     syslog(LOG_INFO, "Child %jd started.", (intmax_t)childpid);
                     break;
             }
     }

     pidfile_remove(pfh);
     exit(EXIT_SUCCESS);

ERRORS

     The pidfile_open() function will fail if:

     [EEXIST]           Some process already holds the lock on the given pidfile, meaning that a
                        daemon is already running.  If pidptr argument is not NULL the function
                        will use it to store a PID of an already running daemon or -1 in case
                        daemon did not write its PID yet.

     [ENAMETOOLONG]     Specified pidfile's name is too long.

     [EINVAL]           Some process already holds the lock on the given pidfile, but PID read
                        from there is invalid.

     The pidfile_open() function may also fail and set errno for any errors specified for the
     fstat(2), open(2), and read(2) calls.

     The pidfile_write() function will fail if:

     [EINVAL]           Improper function use.  Probably called before pidfile_open().

     The pidfile_write() function may also fail and set errno for any errors specified for the
     fstat(2), ftruncate(2), and write(2) calls.

     The pidfile_close() function may fail and set errno for any errors specified for the
     close(2) and fstat(2) calls.

     The pidfile_remove() function will fail if:

     [EINVAL]           Improper function use.  Probably called not from the process which made
                        pidfile_write().

     The pidfile_remove() function may also fail and set errno for any errors specified for the
     close(2), fstat(2), write(2), and unlink(2) system calls and the flopen(3bsd) library
     function.

     The pidfile_fileno() function will fail if:

     [EINVAL]           Improper function use.  Probably called not from the process which used
                        pidfile_open().

SEE ALSO

     open(2), daemon(3), flopen(3bsd)

AUTHORS

     The pidfile functionality is based on ideas from John-Mark Gurney <jmg@FreeBSD.org>.

     The code and manual page was written by Pawel Jakub Dawidek <pjd@FreeBSD.org>.