On Wed, Jan 31, 2007 at 09:40:44PM +0000, Alan Chandler wrote: > On Wednesday 31 January 2007 20:40, Andrew Sackville-West wrote: > > On Wed, Jan 31, 2007 at 07:52:59PM +0000, Alan Chandler wrote: > > > On Wednesday 31 January 2007 19:12, Andrew Sackville-West wrote: > > > > On Wed, Jan 31, 2007 at 06:58:56PM +0000, Alan Chandler wrote: > > > > > The following text appears several times whilst booting. Its > > > > > relatively new (for the last couple of months). Any idea what > > > > > does it? as it doesn't appear to be when the main filesystems > > > > > are mounted. I am running Debian SID. > > > > > > > > [snipped mount usage info] > > > > > > > > do you have some stuff in your fstab that is out of date? > > > > > > like what? only comments and filesystems that I want mounted - see > > > below > > > > [snipped perfectly good fstab] > > > > just a stab in the dark. shrug. what point in booting does this > > happen? is it before or after pivot-root? > > Its afterwards. > > The first record I have is just after the hardware clock is set > > Wed Jan 31 18:52:45 2007: Will now activate swap. > Wed Jan 31 18:52:45 2007: swapon on /dev/sda2 > Wed Jan 31 18:52:45 2007: swapon on /dev/sdb2 > Wed Jan 31 18:52:45 2007: Done activating swap. > Wed Jan 31 18:52:45 2007: /etc/rcS.d/S10checkroot.sh: line > 379: /etc/init.d/mountvirtfs: No such file or directory I wonder if this is where the problem starts, I suspect the system clock setting below is concurrent with S10checkroot.sh. I took a look at that script and there is a bunch of mount stuff going on in there. The specific line reference (379) is just a call to the do_start subroutine, so that's not much help. But I'm guessing the problem is in there some where. You've got something in that script that is calling mount incorrectly resulting in it producing the usage text instead of doing anything useful. > Wed Jan 31 18:52:45 2007: Setting the system clock.. > Wed Jan 31 18:52:47 2007: System Clock set. Local time: Wed Jan 31 > 18:52:47 GMT 2007. > Wed Jan 31 18:52:47 2007: Usage: mount -V : print > version > Wed Jan 31 18:52:47 2007: mount -h : print this > help > Wed Jan 31 18:52:47 2007: mount : list mounted > filesystems > Wed Jan 31 18:52:47 2007: mount -l : idem, > including volume labels > ... above is likely caused by a bad mount call in S10checkroot.sh > > > Then later some sort of cleanup > > Wed Jan 31 18:52:55 2007: Cleaning /tmp...done. > Wed Jan 31 18:52:56 2007: Cleaning /var/run...done. > Wed Jan 31 18:52:56 2007: Cleaning /var/lock...done. > Wed Jan 31 18:52:56 2007: Usage: mount -V : print > version > ... > either this is s10checkroot.sh still hanging around doing stuff, or its another boot script having the same issue. I looked at S36mountall-bootclean.sh and all it does is call /etc/init.d/bootclean and that doesn't seem to have a mount at all so I doubt its that. Though, that is what is going on on the first three lines there. hth A
Attachment:
signature.asc
Description: Digital signature