Index index by Group index by Distribution index by Vendor index by creation date index by Name Mirrors Help Search

sysvinit-2.85-38.4 RPM for i586

From SuSE Linux 9.3 for i386 / suse / i586

Name: sysvinit Distribution: SuSE Linux 9.3 (i586)
Version: 2.85 Vendor: SUSE LINUX Products GmbH, Nuernberg, Germany
Release: 38.4 Build date: Mon May 30 18:23:16 2005
Group: System/Base Build host: g217.suse.de
Size: 811053 Source RPM: sysvinit-2.85-38.4.src.rpm
Packager: http://www.suse.de/feedback
Summary: SysV-Style init
System V style init programs by Miquel van Smoorenburg that control the
booting and shutdown of your system. These support a number of system
runlevels, each one associated with a specific set of utilities. For
example, the normal system runlevel is 3, which starts a getty on
virtual consoles tty1-tty6. Runlevel 5 starts xdm. Runlevel 0 shuts
down the system. See the individual man pages for inittab, initscript,
halt, init, killproc, killall5, powerd, reboot, runlevel, shutdown, and
telinit for more information.



Authors:
--------
    Miquel van Smoorenburg <miquels@drinkel.ow.nl>
    Florian La Roche <florian@suse.de>
    Werner Fink <werner@suse.de>

Provides

Requires

Copyright

GPL, Other License(s), see package

Signatures

internal MD5: 21e66f48dbcfdb56adb3342ef25b1d59
GPG

Changelog

* Thu May 19 2005 - werner@suse.de
  - More fixes for startpar: do not forget active jobs and correct
    exit status handling for the case of stopping jobs.
* Tue Apr 05 2005 - werner@suse.de
  - startpar: fix endless loop and race condition (bug #74256)
* Wed Mar 09 2005 - mls@suse.de
  - startpar: fix splash progress bar if "make" mode is enabled
* Thu Jan 27 2005 - coolo@suse.de
  - catch the case the clock moves back in time during boot
* Tue Jan 18 2005 - coolo@suse.de
  - adding patch by Takashi to sort the services before inserting them
    in the queue to allow predictable boot orders
  - adding patch by me to update the limit more often in case the load
    is too high (30s is just too long)
* Mon Dec 13 2004 - werner@suse.de
  - Update to killproc 2.07 which includes the new ignore file
    feature which works like a pid file but the pid found therein
    will be used to ignore all processes and session with that id.
* Fri Dec 10 2004 - werner@suse.de
  - Fix bottle neck in startpar calculation of number of parallel
    processes. Now the number of really active processes are used.
* Wed Oct 13 2004 - werner@suse.de
  - Do the real fix for the busy loop problem
  - Add a dynamic load balance based on the current process load
    and the current memory usage (bug #45191).
* Mon Oct 04 2004 - mls@suse.de
  - don't call /sbin/splash if proc was unmounted
* Thu Sep 23 2004 - mls@suse.de
  - call splash with $pos:$delta for animated progress bar
  - serialize calls to splash program
* Mon Sep 20 2004 - mls@suse.de
  - change bootsplash 'S' option handling to new bootsplash design
  - also set progressbar on the fly
* Mon Sep 20 2004 - werner@suse.de
  - Add some more information about the new make like scheme of
    startpar into its own manual page.
* Thu Sep 09 2004 - werner@suse.de
  - startproc: get wait status of already terminated parents on very
    fast systems with high load during boot.
* Thu Sep 02 2004 - werner@suse.de
  - Switch to Takashi's startpar version which is fully compatible
    to previous version.
  - Fix startpar to not to loop around if global timeout is expired.
  - Fix struct timeval asignment in startpar for big endian.
  - Add simple loadavg and memory check in startpar to make it
    compatible with UP system with less memory.
* Tue Jun 01 2004 - werner@suse.de
  - Remove any other workaround for hanging write
* Tue Jun 01 2004 - werner@suse.de
  - Remove ioctl workaround in blogd due it does not work
* Tue Jun 01 2004 - werner@suse.de
  - Change signal behaviour of blogd (try to fix bug #40474)
  - Add ioctl workaround to blogd for IA64 serial console
  - Do not overwrite environment of /sbin/init with comandline
* Fri May 14 2004 - werner@suse.de
  - Catch negative return values of safein which may fix bug #34840
* Thu Mar 25 2004 - werner@suse.de
  - showconsole:
    * Do not trap into recursive symlink pointers (bug #36918)
    * Do not scan directories in /dev/ with no terminals
* Mon Mar 22 2004 - werner@suse.de
  - Fix cut&paste error
* Mon Mar 22 2004 - werner@suse.de
  - Support udev within blogd
* Wed Mar 17 2004 - werner@suse.de
  - Make clear descritption of -k option of checkproc and fix
    LSB behaviour of killproc (bug #35851)
* Fri Mar 12 2004 - werner@suse.de
  - Update of sysvinit 2.85 due better support of kernel 2.6
  - Ensure that startpar close the allocated pty's (bug #35763)
* Thu Mar 04 2004 - werner@suse.de
  - Command back: use unsigned int for device numbers which works
    on both 32 and 64bit architectures and with old kernels
* Wed Mar 03 2004 - werner@suse.de
  - Use largest int type for device numbers
* Fri Feb 27 2004 - werner@suse.de
  - Be sure not to fail if startproc is called on fast systems
* Thu Feb 26 2004 - werner@suse.de
  - Add waiting routine on inode/dev compare of startproc binary
    and the forked off service (may fix bug #35008)
* Mon Feb 23 2004 - werner@suse.de
  - Wait on write if kernel returns with EAGAIN
* Sun Feb 22 2004 - schwab@suse.de
  - Fix use of uninitialized variable.
* Tue Jan 13 2004 - werner@suse.de
  - Update to showconsole 1.07 now with warn message if /var is ro
  - Fix bug #33798: remove group write flag from package docs
* Sat Oct 18 2003 - kukuk@suse.de
  - Move mingetty in extra package
* Wed Oct 08 2003 - werner@suse.de
  - Use struct ut_tv on 64 bit also for utmp and not only for wtmp
* Wed Oct 08 2003 - werner@suse.de
  - Redo the rename of barrier define to mem_barrier.
* Tue Oct 07 2003 - werner@suse.de
  - Use the not document struct ut_tv on 64 bit for utmp (bug #32086)
  - blogd: Small optimization at ring buffer handling
* Tue Sep 30 2003 - kukuk@suse.de
  - Rename barrier define to mem_barrier (to avoid clash with kernel
    headers)
* Tue Sep 09 2003 - werner@suse.de
  - Reenable fdatasync in blogd because linuxpthreads are fixed now
* Mon Sep 08 2003 - werner@suse.de
  - blogd: Be sure that _REENTRANT is defined for libconsole.c
* Mon Sep 08 2003 - werner@suse.de
  - blogd: Simply use locking information on broadcast lock, use
    thread join lock only if the writing thread is going to sleep.
* Fri Sep 05 2003 - werner@suse.de
  - Add simple program to check for serial lines on /dev/console to
    replace the wrong serial check during boot with newer kernels
* Tue Sep 02 2003 - werner@suse.de
  - Add workaround for blogd crash on x86_64 (bug #29750,29249,29545)
    * Be paranoid and use SysV signal handling to ensure restart
    * Be paranoid and check always the value of FILE pointer flog
* Wed Aug 06 2003 - werner@suse.de
  - Change detection of serial versus terminal lines (bug #28490)
* Thu Jul 24 2003 - uli@suse.de
  - fixed to build on s390x
* Thu Jul 03 2003 - werner@suse.de
  - Make flags transparent in startproc.c
* Thu Jul 03 2003 - werner@suse.de
  - Make killproc compliant with newest LSB spec (bug #24909)
* Tue Jun 24 2003 - werner@suse.de
  - Ditto
* Mon Jun 23 2003 - werner@suse.de
  - Make it build even on s390/s390x
* Mon Jun 23 2003 - kukuk@suse.de
  - Remove not packaged files from buildroot.
* Mon Jun 23 2003 - kukuk@suse.de
  - Remove compat link for old SPARC kernel
* Tue Jun 17 2003 - werner@suse.de
  - Use BuildRoot
  - Add startproc (0.42), thanks to Michael
  - showconsole: check for environment variable NOTIOCGDEV
* Fri Jun 06 2003 - mfabian@suse.de
  - also set the scroll region to the full window (ESC [ r) when
  --noclear is not set.
* Thu Jun 05 2003 - mfabian@suse.de
  - if the option "--noclear" is not set, don't write a full reset
    because this leaves the unicode mode again if the terminal was in
    unicode mode and it also undos the ESC sequences in CONSOLE_MAGIC
    which are needed for some languages/console-fonts.
    Just put the cursor the home position (ESC [ H) and erase
    everything below the cursor (ESC [ J).
* Thu May 15 2003 - werner@suse.de
  - New showconsole/blogd version 1.06
    * Threaded I/O stuff
* Tue Apr 29 2003 - werner@suse.de
  - New showconsole/blogd version 1.05
    * Do not overwrite ring buffer memory in case of the fifo
    * Avoid not needed new line in case of carriage return.
    * Read at least 3 seconds at blogd close and flush all I/O
* Mon Apr 14 2003 - werner@suse.de
  - Fix script handling of killproc for several kernel versions
    (bug #25767)
* Tue Mar 04 2003 - werner@suse.de
  - Correct exit codes of killproc and update manual page about this
* Fri Nov 22 2002 - werner@suse.de
  - Be less memory consuming with showconsole 1.04
* Thu Nov 21 2002 - werner@suse.de
  - New showconsole 1.03
    * be able to scan dirs under /dev/ in fallback case
* Thu Nov 14 2002 - werner@suse.de
  - Update to mingetty 0.9.6s
    * Be sure to get controlling tty
    * Avoid overruns in string handling
* Wed Nov 13 2002 - werner@suse.de
  - Update to showconsole 1.02 (a cleanup version)
  - Update to mingetty 0.9.5s (s stand for SuSE)
    * New old option for avoiding vcs/vcsa hangup, terminal
    reset, and the usage of the glibc for updating wtmp.
    * Now the default terminal type can be set at compile
    time as an make option.
* Wed Nov 13 2002 - werner@suse.de
  - Update to killproc 2.05:
    * Change usleep which now calls sched_yield() for 0 microseconds
* Tue Nov 12 2002 - werner@suse.de
  - Update to killproc 2.04:
    * Better symlink handling
    * New dialog prompt handling for services
    * New chroot option.
* Mon Nov 11 2002 - ro@suse.de
  - fixed deprecated multiline string literal
* Mon Nov 04 2002 - werner@suse.de
  - Use usleep(1) in startproc/killproc/killall5 to force the
    kernel to run the scheduler.
* Fri Aug 16 2002 - werner@suse.de
  - Add PreReq (bug #18005)
  - Not only blogger and blogd but also set and showconsole together
    with their manual pages.
* Thu Aug 08 2002 - ihno@suse.de
  - adding blogger and blogd to s390/s390x
* Fri Aug 02 2002 - froh@suse.de
  - removed blogger from s390&s390x package lists
  - changed 'ifarch s390' and 'ifarch s390x' to 'ifarch s390 s390x'
* Tue Jul 16 2002 - werner@suse.de
  - Expand possible pid string length from 22 to 255
* Fri Jul 05 2002 - werner@suse.de
  - Re-enable SIGINT in case or re-reading inittab (bug #16469)
* Thu Jul 04 2002 - werner@suse.de
  - Update to killproc 2.03
  - Fix pointer for ioctl in libconsole of showconsole
* Tue Apr 30 2002 - werner@suse.de
  - Even if no blogd is used on S390x, the libblogger is required
* Mon Apr 29 2002 - werner@suse.de
  - Avoid trouble with stupid gcc 3.1 parser and inlined macros.
* Thu Apr 25 2002 - werner@suse.de
  - changes on mingetty (new version 0.9.4c)
    * Add support for other TTY devices than virtual console
    * Add nohost patch from Anders ??
* Mon Apr 22 2002 - sf@suse.de
  - changed path to libcrypt in Makefile to also compile
    on lib64-archs
* Tue Feb 12 2002 - ro@suse.de
  - fix owner/group
* Tue Jan 15 2002 - werner@suse.de
  - Different usage messages for startproc and start_daemon (#12692)
* Fri Dec 14 2001 - werner@suse.de
  - Skip spinner of fsck/e2fsck in boot logging file
* Mon Oct 01 2001 - werner@suse.de
  - Build and install utmpdump
* Mon Oct 01 2001 - werner@suse.de
  - Move to SysVinit 2.82 and migrate our patches to 2.82
* Tue Sep 25 2001 - werner@suse.de
  - Fix race on waiting on console I/O: just try open log file first
* Thu Sep 20 2001 - werner@suse.de
  - Ignore interrupted waitpid(2) calls
* Wed Sep 19 2001 - werner@suse.de
  - Add forgotten configuration of powerd
* Wed Sep 19 2001 - werner@suse.de
  - Update to powerd 2.0.2 because this version initialize the
    serial interface which seems to be required for kernel 2.4.x
* Fri Sep 07 2001 - werner@suse.de
  - startproc: close boot logging FIFO before executing a daemon.
* Wed Sep 05 2001 - werner@suse.de
  - blogd: really read the console a half second (bug #9899)
* Tue Sep 04 2001 - werner@suse.de
  - Add waitpid() loop nodead lock patch of Andrea
  - Add a check in the waitpid() loop to exit on failure
* Thu Aug 30 2001 - werner@suse.de
  - Make blogd/showconsole more robustly against errors
  - Add some fixes of bug #9898
* Fri Aug 24 2001 - werner@suse.de
  - Activate 64 bit file handling interface (large file support)
* Thu Aug 23 2001 - werner@suse.de
  - Do not reset terminal device of /dev/console if on a serial line
* Tue Jun 26 2001 - cstein@suse.de
  - added a patch for the shutdown messages (see bug #8076)
    (also changed some other strings in shutdown.c)
* Mon Jun 25 2001 - bk@suse.de
  - back out read return 0 patch and don't build showconsole on s390x
* Mon Jun 25 2001 - bk@suse.de
  - don't include showconsole programs in filelist on s390x
* Fri Jun 22 2001 - bk@suse.de
  - blogd: when read return 0, this means EOF - don't loop in this case.
* Tue Jun 12 2001 - bk@suse.de
  - don't try to build showconsole on s390x(not supported by kernel)
* Wed May 23 2001 - werner@suse.de
  - Do not eat up argv and environ at boot logging in startproc
* Wed May 23 2001 - werner@suse.de
  - New killproc version 2.01: now we're able again to find deamons
    based on scripts.
  - Make fifo /dev/blog after installation if not exist.
* Tue May 08 2001 - werner@suse.de
  - Remove previous barrier() calls but use them on empty loops.
* Wed May 02 2001 - werner@suse.de
  - Call umask only once for all init processes (pid 1 and others)
  - Add some barrier() against officious compiler
  - Rename cpp macro TEST to SYSVINIT_TEST
* Tue Apr 24 2001 - werner@suse.de
  - Add default umask 022 at boot time
* Tue Apr 17 2001 - werner@suse.de
  - Few minor changes to killproc and co.
* Thu Apr 12 2001 - werner@suse.de
  - Disable boot logging and boot message redirection at common
    runlevels if TIOCGDEV is missed.
* Wed Apr 11 2001 - werner@suse.de
  - blogd: complain about missing ioctl TIOCGDEV at compile
    and run time
* Wed Apr 04 2001 - werner@suse.de
  - After fork reset signal handling of startproc to default for
    common signals
* Thu Mar 22 2001 - werner@suse.de
  - Startproc/killproc: handle by RPM moved and deleted exe links
* Wed Mar 21 2001 - werner@suse.de
  - Add patch for rlstat() of schowconsole (Andreas Schwab)
* Mon Mar 19 2001 - werner@suse.de
  - Startproc/killproc: handle deleted exe links
* Fri Mar 09 2001 - werner@suse.de
  - Add forgotten usleep to file list
* Wed Feb 14 2001 - werner@suse.de
  - Use sysvinit 2.78.4
  - Spilt dif into several patches
  - Use static on the most functions of init.c
  - Integrate showconsole
  - bzip tar's
* Mon Feb 05 2001 - werner@suse.de
  - Update of killproc to version 2.00
    * LSB conform exit values
    * Use boot logging provided by showconsole-0.9.tar.gz of aaa_base
    * New program usleep
* Mon Jan 15 2001 - werner@suse.de
  - Make strncmp of exe link and fullname working upto MAXNAMLEN
* Tue Jan 09 2001 - werner@suse.de
  - Use int instead of kdev_t in bootlogd.c
* Wed Nov 29 2000 - werner@suse.de
  - Make tty settings for sulogin working, and enable job control.
  - Workaround vor sigfholder (may kernel bug)
* Tue Nov 28 2000 - werner@suse.de
  - New and fixed killproc version 1.17
  - Change some man pages of sysvinit to fit our new boot scheme
  - Enable tty settings for sulogin, should be checked on
    serial console (BTW: SHOULD we DO that?).
* Thu Nov 16 2000 - ro@suse.de
  - fixed install for killproc
* Wed Nov 15 2000 - werner@suse.de
  - New killproc version (1.16)
    * base name allowed
    * protected the parent of the parent
    * provide pidofproc (a link to checkproc for a verbose checkproc)
    * provide stat_daemon (a link to statproc)
* Mon Oct 23 2000 - werner@suse.de
  - changes on mingetty (new version 0.9.4b)
    * On ro files systems chmod/chown shouldn't force an exit.
    * Therefore we need a warn message.
    * Due to the lack of a revoke system call we signal
    all file holders including the mmaping processes.
* Wed Jun 28 2000 - bk@suse.de
  - enabled reboot on s390
* Sun Jun 25 2000 - bk@suse.de
  - removed some not needed files on s390
* Tue May 30 2000 - werner@suse.de
  - Correct path of powerstatus (/var/run/powerstatus) of powerd
* Wed May 24 2000 - garloff@suse.de
  - Added minimal docu for swsusp.
* Tue May 23 2000 - werner@suse.de
  - New killproc version 1.15 reading exe link for kernel 2.2
* Wed May 17 2000 - garloff@suse.de
  - Added software-suspend support to init.
  - login option of mingetty limits logname to 42 chars.
* Wed May 10 2000 - garloff@suse.de
  - Working patch for mingetty to support devfs.
* Wed May 10 2000 - garloff@suse.de
  - Preliminary patch for devfs.
* Wed May 10 2000 - garloff@suse.de
  - Make it more secure by passing the logname as one arg and by
    checking for a leading '-'. The manpage tells the sysadmin
    about possible risks now.
* Tue May 09 2000 - garloff@suse.de
  - Added new options --login and --logopts to mingetty, so you may
    create an inittab entry using ssh for login or similar
* Tue Mar 07 2000 - werner@suse.de
  - Ignore SIGTERM for syslogd because this daemon use that signal
    to control the parent process.
  - Make read/write of /etc/ioctrl.save EINTR safe
  - Use defined special character given from <termios.h>
  - No flow control (-ixon), ignore break (ignbrk), and make nl/cr
    more usable (sane).
  - Set the terminal line immediately
* Mon Mar 06 2000 - werner@suse.de
  - Avoid script handling for kernel threads
* Sun Mar 05 2000 - werner@suse.de
  - If open fails -1 is returned (close bug #2200)
* Fri Mar 03 2000 - kukuk@suse.de
  - Create /usr/bin/shutdown link for UltraSPARC kernel
* Mon Feb 21 2000 - werner@suse.de
  - New version of killproc (1.1.4): fix of script handling
  - New version of sysvinit (2.78) with some bug fixes
* Thu Dec 16 1999 - werner@suse.de
  - New version of killproc 1.13
    * Bug fix in handling hard links for daemons using pid files
    * Bug fix: PROC pointers should not be volatile
    * Shorten nonsleeps for none daemon programs
* Tue Nov 09 1999 - werner@suse.de
  - Avoid trouble with hard links of binaries
* Sun Nov 07 1999 - kukuk@suse.de
  - mingetty: Add 1900 to tm_year, it contains the years since 1900.
* Wed Nov 03 1999 - werner@suse.de
  - New killproc 1.12
    * Fix option -q of startproc
    * Fix Bug in wait4 handling of startproc
    * Fix script detection: do not handle startproc, killproc, and
    checkproc as script interpreters
* Mon Nov 01 1999 - werner@suse.de
  - Add a patch from Marius
* Fri Oct 29 1999 - werner@suse.de
  - powerd should tell init if status is OK again
* Thu Sep 30 1999 - werner@suse.de
  - New killproc version 1.11
    * Some small speedups
    * Minor spell correction in the manual pages
    * Allow login onto tty's
  - Use RPM_OPT_FLAGS throughout
* Mon Sep 13 1999 - bs@suse.de
  - ran old prepare_spec on spec file to switch to new prepare_spec.
* Tue Sep 07 1999 - werner@suse.de
  - Enable mingetty resetting the terminal
* Fri Aug 06 1999 - werner@suse.de
  - New killproc-1.10
    * startproc which is able to sleep
    * better zombie handling
  - killproc doesn't ignore zombies
  - checkproc ignores zombies without `-z' by default
  - startproc checks zombie state of child while sleeping
* Tue Jul 06 1999 - werner@suse.de
  - Fix pts problem of shutdown (dowall.c)
* Thu Jun 24 1999 - werner@suse.de
  - Add lastb and its manual page
* Wed Jun 23 1999 - werner@suse.de
  - Integrate newest mingetty patch of jurix
  - Update killproc version 1.9
    * be more sufficient in setting user and/or group id
  - Update to sysvinit 2.76
    * Integrate newest sysvinit patch of jurix into
    our patch. The command last is part of sysvinit.
* Tue Jun 22 1999 - werner@suse.de
  - Update to new killproc version 1.8
    * startproc knows about setsid to set up a new process session
    for the new task (option -s)
    * startproc can change the effective uid and/or gid for a process
    (option -u uid and/or -g gid)
  - Make spec file more handy for newer versions.
* Mon May 03 1999 - werner@suse.de
  - Enable killproc-1.7.dif in specs file
* Mon May 03 1999 - werner@suse.de
  - killproc shouldn't complain a disappeared job/thread
* Mon Apr 19 1999 - werner@suse.de
  - mingetty: reset not only /dev/tty<line> but /dev/vcs<line> and
    /dev/vcsa<line> also
* Fri Feb 26 1999 - werner@suse.de
  - New killproc vesion 1.7
    * read of /proc/ is secure against EINTR
    * handle script names ala `bash -x /usr/bin/myscript -y'
* Mon Jan 18 1999 - werner@suse.de
  - Make killproc more smart, e.g. faster
  - Make startproc alpah compatible
* Thu Dec 10 1998 - ro@suse.de
  - added last manpage
  - link init static
* Tue Dec 08 1998 - ro@suse.de
  - use last from sysvinit-2.75
* Mon Nov 30 1998 - werner@suse.de
  - Close some races, unknowns, and difficulties in killproc tools
* Tue Nov 24 1998 - werner@suse.de
  - Don't blame boot script with killproc having same name as the
    controlled daemon ... new kernel doesn't need this.
* Mon Nov 16 1998 - ro@suse.de
  - install man-page for powerd
* Fri Nov 13 1998 - werner@suse.de
  - Integrate powerd-2.0 for a replacment of the obsolate old one
* Mon Nov 02 1998 - ro@suse.de
  - update to killproc-1.6
* Thu Oct 22 1998 - ro@suse.de
  - update to killproc-1.5  (take 2)
* Thu Oct 08 1998 - ro@suse.de
  - killproc: libinit.c output of "can't open pid file" only "#if DEBUG"
* Tue Oct 06 1998 - ro@suse.de
  - updated killproc to get rid of a memleak
* Mon Oct 05 1998 - ro@suse.de
  - new version: killproc-1.4
* Fri Jul 03 1998 - werner@suse.de
  - added killproc-1.2, a replacement of the old killproc
* Wed May 06 1998 - florian@suse.de
  - added an option to killproc to also kill process groups:
    killproc [-SIG] [-g] prog
* Tue Nov 04 1997 - mantel@suse.de
  - creation of dev/initctl fixed
* Thu Oct 23 1997 - ro@suse.de
  - ready for autobuild

Files

 Εύ


Generated by rpm2html 1.8.1

Daniel Veillard, Thu May 15 04:04:26 2008