If you stop sending to the machine & leave it for a while does it
	come back to life - there was a problem in the way NetBSD handled
	low memory situations - I think a workaround was committed to the
	tree shortly after 1.2 was released, but I'm not sure on this...

	What architecture are you running, and have you tried setting up
	ddb so you can get a traceback of what happens at the hang? 

		David/abs	david@{mono.org,southern.com,mhm-internet.com}

System Manager: Southern Studios Ltd, PO Box 59, London N22 1AR.
Satisfied User: NetBSD, free Un*x {i386,sparc,mac68k,+more} 'www.netbsd.org'.
  System Admin: MHM Internet, 14 Barley Mow Passage, Chiswick, London W4 4PH.
         SysOP: Monochrome, Largest UK Internet BBS - 'telnet mono.org'.

On Wed, 13 Nov 1996, HARAWAT.IN.ORACLE.COM wrote:

> 
> 	Iam using a 8 Mb diskless machine with swap over nfs, when Iam using a 
> applications that receices a lot of packets send by the server at a very high 
> rate the machine hangs consistently, but when I increase the values of low 
> water mark and  high water mark for the waking up of the pager daemon then the 
> application runs for the most of the times, moreover  when I run the same 
> application on a machine with disk ie. the swap is on the hard disk, the 
> machine never hangs. Any idea what could be the reason for this hangs, is it 
> because of some deadlock, what are the extra resources used by the kernel when 
> the swap is over nfs. 
>  
> Harish 
> -----------------------------------------------------------------------------
> 
> Harish Singh Rawat			   E-Mail : harawat@in.oracle.com 
> Oracle Software India Pvt. Ltd.            Voice  : +91 (80) 225 - 6098/6099 
> Indian Development Center ( IDC )          Fax    : +91 (80) 225 - 6207/1591 
> Embassy Point, 150 Infantry Road 
> Bangalore - 560001 India 
> 
>