[PlanetCCRMA] (Slightly OT) CCRMA ate my machine!!

[Phil Moors]

On Sun, Mar 07, 2004 at 04:51:24PM +1100, Peter Howard wrote:
> Subject: Re: [PlanetCCRMA] (Slightly OT) CCRMA ate my machine!!
> From: Peter Howard <pjh@coastal.net.au>
> To: Phil Moors <pmoors@greenbelt.com>
> Cc: Fernando Pablo Lopez-Lezcano <nando@ccrma.Stanford.EDU>,
>    Planet-CCRMA
> 	 <planetccrma@ccrma.Stanford.EDU>
> X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
> Date: 07 Mar 2004 16:51:24 +1100
> 
> On Sat, 2004-03-06 at 15:43, Peter Howard wrote:
> > On Sat, 2004-03-06 at 12:51, Phil Moors wrote:
> > > On Fri, Mar 05, 2004 at 01:45:33PM +1100, Peter Howard wrote:
> > > > Subject: Re: [PlanetCCRMA] (Slightly OT) CCRMA ate my machine!!
> > > > From: Peter Howard <pjh@coastal.net.au>
> > > > To: Fernando Pablo Lopez-Lezcano <nando@ccrma.Stanford.EDU>
> > > > Cc: Planet-CCRMA <planetccrma@ccrma.Stanford.EDU>
> > > > X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
> > > > Date: 05 Mar 2004 13:45:33 +1100
> > > > 
> > > > On Fri, 2004-03-05 at 13:34, Fernando Pablo Lopez-Lezcano wrote:
> > > > > > OK, not exactly true, but as it happened while playing with audio bits
> > > > > > I'm hoping someone else on the list has experienced something similar.
> > > > > > 
> > > > > > I posted to the list a week or so ago about problems with fluidsynth
> > > > > > connecting to JACK.  With the package updates yesterday I decided to
> > > > > > update my machine (a laptop, Toshiba Satellite TE2000) to see if the
> > > > > > problem was still there.  It was.  I noticed rosegarden had updated. 
> > > > > > Interested in just checking the version, I started it with nothing else
> > > > > > running.  The result? The sequencer crashed.  No big deal I thought. 
> > > > > > Having checked the version (which is all I wanted to do and hadn't
> > > > > > seemed to change) I started up JACK to go back to my "real" problem.  Or
> > > > > > tried to.  Now I was getting an assertion on startup (either by qjackctl
> > > > > > or directly).  I can't tell you what it is, keep reading and you'll get
> > > > > > an idea why :(
> > > > > > 
> > > > > > OK, reboot the machine an see if it goes away.  I don't know as now the
> > > > > > machine won't boot with any of the 4 kernels I have on the machine. 
> > > > > > Each stops with "cannot open initial console".  And I can't get it to go
> > > > > > any further.  Of course I can emergency boot from CD, but nothing looks
> > > > > > out of place and I don't know enough as to where to start looking.
> > > > > > 
> > > > > > Suggestions? Pointers?
> > > > > 
> > > > > Ouch... I'm sorry...... I've never seen this happen. All kernels??
> > > > > 
> > > > > A bit of googling got me this:
> > > > >   http://kerneltrap.org/node/view/1901
> > > > > 
> > > > > When in the boot process do you get that message?
> > > > 
> > > > No.  It just says "cannot open an initial console.", then sits there. 
> > > > No panic.
> > > > 
> > > > > In the thread somebody says:
> > > > > 
> > > > > "make initrd manually after kernel image install and then run
> > > > > grub-install, that should do it."
> > > > > 
> > > > > But I'm not sure what that would fix as I don't understand what is
> > > > > really happening. 
> > > > 
> > > > Of course it may have nothing directly to do with CCRMA stuff . . . I
> > > > did some googling too, but everything that I found was related to
> > > > changing a _specific_ kernel. 
> > > > 
> > > > > What other messages do you see around the "cannot open
> > > > > initial console" message?
> > > > > 
> > > > It's just the normal startup the last message before is the "freeing
> > > > unused kernel memory" one.
> > > >  
> > > 
> > > I just went through something similar with my laptop. My boot would fail
> > > about halfway and drop me to a root prompt with an error message about
> > > not being able to get a controlling tty. Running memtest86 found that my
> > > 64MB addon card had errors all over the place. My laptop is old, so
> > > replacing the memory only cost $30. However, the problem still
> > > persisted. I paved the harddrive and reinstalled FC1 and everything was
> > > good.
> > > 
> > > Thinking about it, I believe that when my memory went bad it also
> > > corrupted a bunch of files on my disk including some of the tty and pts
> > > devices in /dev. I may have been able to restore operability by forcing
> > > a reinstall of the dev package. By reinstalling the entire OS, I'm sure
> > > that everything is okay.
> > 
> > Thanks Phil.  The corruption of /dev was the issue (I suspect that it
> > may have been part of JACK's problem as well).  Once I managed to get
> > the rpm for dev onto the machine (#$%@!#@! rescue boot from FC1 disk 1 
> > doesn't actually mount the CD itself.  Or even create /dev/hdc so you
> > can manually mount it), one rpm -i --force later and the problem was
> > solved. 
> > 
> > I think I'll keep that RPM on my machine handy for now :)
> > 
> > I'll run memtest x86 soon & let you know the result.
> > 
> Ran for an hour and a half this afternoon ; no errors by that stage.
> 

I'd say you're okay on the memory. I had 7 errors in the first hour.

Phil

> -- 
> Peter Howard
> pjh@coastal.net.au
> Jabber Id: pjhacnau@12jabber.com
> AIM Id: pjhacnau