perry
cf423b3982
1) To eliminate a race condition in which the user could conceivably
...
try to quit or stop the program right at the moment after the it
sends a TSTP to init, move the kill(1, TSTP) to right after we
block most signals, instead of doing it just before.
2) To add TSTP to the signals we ignore. Once we start halting, it
really is not safe to try to stop at all. You can't change your
mind halfway through -- init will already be stopped and a TERM
will have been sent to everything on the system within
milliseconds of starting.
3) Adjust one comment for TNF
2003-04-04 17:43:08 +00:00
wiz
42ec3bd114
Support wtmpx.
2002-08-02 15:05:57 +00:00
wiz
d9deaad608
de__P(), ANSIfy.
2002-08-02 14:59:40 +00:00
tsutsui
4b541a44d0
Drop a trailing space in bootstr[] properly.
2002-07-20 06:44:45 +00:00
cgd
8a986b2e96
convert to use getprogname()
2001-02-19 22:48:57 +00:00
lukem
f022cce90b
use openlog correctly...
2001-01-10 03:41:35 +00:00
hubertf
416a0bca75
Implement a "poweroff" command which is equivalent to "halt -p",
...
but allows powering down the system from an exec(2) call, e.g. via
a user which has "poweroff" as login shell.
The command name is modeled after Solaris.
2000-04-25 14:03:25 +00:00
jdolecek
a519c15410
Ignore SIGPIPE, otherwise we might be stuck when run within a pipe.
...
Fix from bin/5359 by Nathan J Williams.
2000-03-07 20:02:59 +00:00
is
3523dc9c50
Back out the wait-for-slowly-dying-processes change.
1998-10-09 18:50:35 +00:00
augustss
91b9544ee2
Ignore some more signal during the reboot. Without this fix it's impossible
...
to e.g. reboot from a menu in the window manager as the reboot process
gets killed before it is done.
1998-08-29 11:17:20 +00:00
mycroft
b93acf4aaf
const poisoning.
1998-07-26 20:02:36 +00:00
is
a7557bc638
Change poll interval to 1 seconds, for machines booting _really_ fast.
1998-07-06 05:21:07 +00:00
is
25718b8405
Make halt and reboot wait up to 32 seconds after kill(-1, SIGTERM), but
...
check at 3 second intervals if any are left.
This will help slow machines to cleanly shut down X servers (to make the
console visible), databases, or Usenet news servers.
1998-07-03 13:59:56 +00:00
thorpej
5f9840e94e
Add a -p option to halt(8), specifying that the system should be powered
...
down.
1998-06-06 21:05:41 +00:00
mycroft
689db8fcc6
Use warn*(3).
1998-01-20 23:41:57 +00:00
mycroft
bc0a5a65c3
Clean up documentation a bit, and document the `-l' option.
1998-01-20 23:37:27 +00:00
mrg
970a5291cd
fix reboot arguments handling (from cgd).
1997-12-09 05:49:14 +00:00
lukem
be24f789ad
- use .Nm correctly
...
- use __progname instead of rindex(argv[0]) ...
1997-09-15 07:38:32 +00:00
perry
6b1fa5ab98
1) Make WARNS=1 clean
...
2) to do 1), nuke the silly local version of err()
3) nuke "register" declarations.
1997-07-19 23:03:54 +00:00
perry
28753bf1c0
fix pr-2619, from cgd. really obscure bug in exec'ing reboot when single user.
1997-07-19 22:27:20 +00:00
perry
08c34b1604
fix halt when used as login shell, from cgd in pr-2618
1997-07-17 02:52:22 +00:00
mrg
ed46fe3a8e
only log the bootstr if it exists.
1996-09-16 18:09:34 +00:00
mrg
a9f27985cf
use the SunOS derived RB_STRING.
1996-08-10 00:20:58 +00:00
mrg
74693079c6
update for new reboot(2).
1996-08-09 10:32:11 +00:00
mycroft
fd25fd5a19
Add a `-d' option to force a dump. From John Kohl.
1995-10-05 05:36:21 +00:00
cgd
ce5ebe9911
pseudo-merge with Lite; man pages need to be merged seperately, but
...
that's really a port-specific thing. They probably should be updated, too.
1995-03-18 16:11:33 +00:00
cgd
0114e805ce
convert to new RCS Id conventions; reduce my headache
1995-03-18 14:54:19 +00:00
cgd
c9117f53b5
-Wall cleanups from Jim Jegers
1994-12-18 00:15:01 +00:00
mycroft
e9d867ef50
Add RCS identifiers.
1993-08-01 17:54:45 +00:00
cgd
06be60083d
changed "Id" to "Header" for rcsids
1993-03-23 00:22:59 +00:00
cgd
346aa5dd48
added rcs ids to all files
1993-03-22 08:04:00 +00:00
cgd
61f282557f
initial import of 386bsd-0.1 sources
1993-03-21 09:45:37 +00:00