updating in single-user mode

[Peter Matulis]

--- Eric F Crist <ecrist at secure-computing.net> wrote:

> On Oct 24, 2005, at 11:45 PM, Dimitar Vasilev wrote:
> 
> >> I don't reccommend doing installworld or kernel in multiuser,
> but  
> >> I have never
> >> had any problems doing it on a lightly loaded machine. With
> that  
> >> said what
> >> could bite you is your new kernel not booting or something
> broken in
> >> userland. You will then need console access (serial or local)
> to  
> >> fix it. I
> >> would set up your machine with serial console access and use a 
> 
> >> laptop or
> >> another machine when you reboot.
> >>
> >> Beech
> >> --
> >>
> > I have done it when there is NO activity on the machine. Read  
> > UPDATING first.
> > Reset your securelevel to -1, stop all services except SSH and
> go.
> > It's possible to break your machine though.
> > Then you have to rebuild it again and it's 50/50 to succeed.
> > As advised twice, use serial cables/KVM switches if possible.
> > --
> > Димитър Василев
> > Dimitar Vassilev
> >
> > GnuPG key ID: 0x4B8DB525
> > Keyserver: pgp.mit.edu
> > Key fingerprint: D88A 3B92 DED5 917E 341E D62F 8C51 5FC4 4B8D
> B525
> 
> 
> If this isn't a production machine, try it.  I have been doing
> system  
> updates since 3.4 and not once have I booted into single user mode
> to  
> compile my kernel or userland.  I've even done it as recently as
> two  
> weeks ago.  I don't have a huge userbase, so my system is pretty
> quiet.

I also just finished compiling world and compiling & installing my
kernel in multi-user.  What's the big deal?  I did, though, go into
single to test (boot) the kernel and install world.

Does anyone have a clear understanding of why things can go wrong otherwise?


	

	
		
__________________________________________________________ 
Find your next car at http://autos.yahoo.ca