I am currently having severe stability problems with a 1.3K kernel supped on 
2nd March 1999 with an equivalent userland. Since I haven't seen anything 
about stability problems on other ports I'm assuming this is port specific.

The system becomes unstable and eventually hangs under the following 
conditions:

1. ping -fs 3000 running from another machine and a find /dir on a NFS file 
system which the RiscPC is serving from the same machine doing the ping. Under
these conditions all runs well for approximately 10 seconds when the machine 
stops responding the the ping and the find stops. At this point the console
cursor has become corrupt and a few seconds later the machine hangs.

2. When any process consumes a large amount of VM the system hangs but without 
the console cursor corruption.

I have also had hangs when running daily/weekly and monthly cron scripts but 
these do not always cause a failure.

The machine is a RiscPC with 200MHz StrongARM processor. 48MB RAM + 2MB VRAM, 
Acorn SCSI card (with CDROM drive attached but not mounted) Acorn EtherLan600A 
Network card using NetBSD NE2000 driver.

I am unable to get any stack backtrace as the machine hangs and will not enter 
DDB.

I have had similar problems with late 1.3I kernels but was unable to create a 
test case which always reproduced the problem.

Does anyone have any idea what's going on?

Mike