Subject: Re: newer G4 models and panic: lockmgr: pid 0, not exclusive
To: Bill Studenmund , Makoto Fujiwara <makoto@ki.nu>
From: Henry B. Hotz <hotz@jpl.nasa.gov>
List: port-macppc
Date: 06/21/2002 17:56:08
At 11:25 AM -0700 6/21/02, Bill Studenmund wrote:
>On Fri, 21 Jun 2002, Makoto Fujiwara wrote:
>
>>  | Machine type G4/800 MP
>>  | PPC 7450
>>  I have cvs checkout today and build kernel:
>>  NetBSD 1.6B (GENERIC) #0: Fri Jun 21 21:57:55 JST 2002
>>
>>  The kernel works upto single user level.
>>
>>  It hungs at 'Setting tty flags.' at the first time.
>
>At, or just after?
>
>I remember we've had problems (back around 1.4 I think) where the step
>after Setting tty flags doesn't announce it's starting, but it's the
>killer. It was frustrating to spend a few hours trying to fix ttyflags,
>when it wasn't the problem. :-)

*ahem*

If you are referring to a certain incident that someone I know 
extremely well ;-) was involved in. . .

*That* problem was a SCSI problem where parallel fsck's would work 
50% of the time depending on which of two disks started fsck first. 
Since several disks had fsck'ed OK before the hang, and the ttyflags 
was the next thing, and Bill had *just* fixed something in ttyflags I 
assumed that there was a connection when there wasn't.

I think this was on a Q840av and it might have been 1.3 instead of 1.4.

In any case Bill's suggestion that you be sure of where it hangs in 
the boot sequence is a good idea.  If I'd done that it would have 
saved us both a lot of time.

>Try adding:
>
>echo "Done."
>
>to the end of ttyflags_start() in /etc/rc.d/ttys.
>
>>  ....
>>  root on wd0a dumps on wd0b
>>  root file system type: ffs
>>  Fri Jun 21 13:21:47 GMT 2002
>>  swapctl: adding /dev/wd0b as swap device at priority 0
>>  starting file system checks:
>>  /dev/rwd0a: file system is clean; not checking
>>  /dev/rwd0g: file system is clean; not checking
>>  /dev/rwd0e: file system is clean; not checking
>>  /dev/rwd0f: file system is clean; not checking
>>  Setting tty flags.
>>   _
>>  (and hungs)
>>  ...
>>  once more the same thing.. it works..
>>  But shortly unable to type in from the console. Also rsh
>  > shell freezes..

-- 
The opinions expressed in this message are mine,
not those of Caltech, JPL, NASA, or the US Government.
Henry.B.Hotz@jpl.nasa.gov, or hbhotz@oxy.edu