home *** CD-ROM | disk | FTP | other *** search
Wrap
Newsgroups: comp.os.linux Path: sparky!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!agate!spool.mu.edu!umn.edu!student.tc.umn.edu!lash0002 From: lash0002@student.tc.umn.edu (tim) Subject: Re: Problems with sysv init sources Message-ID: <1992Dec15.225459.26039@news2.cis.umn.edu> Keywords: init Sender: news@news2.cis.umn.edu (Usenet News Administration) Nntp-Posting-Host: student.tc.umn.edu Organization: University of Minnesota References: <1992Dec15.134909.28520@mintaka.lcs.mit.edu> Date: Tue, 15 Dec 1992 22:54:59 GMT Lines: 53 In article <1992Dec15.134909.28520@mintaka.lcs.mit.edu> wyvern@gnu.ai.mit.edu (The Wyvern) writes: >I recently compiled and installed the sources for the sysv-style init >from tsx-11 (plus the patch). It seems to be having problems >with runlevels though. If I have initdefault set to, say, 3, >and a couple lines like this: > >rc:3:wait:/etc/rc3 >xd:3:once:/usr/X386/bin/xdm > >Init gives me something like: >INIT: Warning: No processes for this runlevel >INIT: Entering single user mode >/ root# >So I end up having to have all the entries in my inittab be either >boot or bootwait, since those are the only ones which get executed. >Right now my inittab looks like this: > >id:3:initdefault: >si::sysinit:/etc/bcheckrc >rc::bootwait:/etc/rc >r3::bootwait:/etc/rc3 >xd::boot:/usr/X386/bin/xdm >#c1:1234:respawn:/etc/getty 9600 tty1 >#c2:23:respawn:/etc/getty 9600 tty2 >#c3:23:respawn:/etc/getty 9600 tty3 >#c4:23:respawn:/etc/getty 9600 tty4 >pf::powerwait:/etc/shutdown -f now ># init > >This is highly annoying - it kind of defeats the purpose of an init >with runlevels. Does anyone have an idea of what might be wrong? > I have had similar problems. I am not sure that i have really found the problem or a good solution, but what I've done is: At the end of the function StartEmIfNeeded(), there is a line that says lastlevel = runlevel; I've changed it to: if(sys_level == NORMAL_LEVEL) lastlevel = runlevel; It appeared to me that during the sysinit and boot stages lastlevel was being set so that entries with an action field of wait or once would appear to have already been executed. This seems to fix it. I am not sure if it really is a fix, or if it is if it is a good one..i may look at it more later. I've sent a note to Miquel asking him about it a few days ago but have not heard from him yet. If anyone else has anything to say about this I'd appreciate hearing about it. tim lash0002@student.tc.umn.edu