this is from 1.4M on a 3000/400.

panic: insmntque into dying filesystem
Stopped in cvs at       cpu_Debugger+0x4:       ret     zero,(ra)
db> trace
cpu_Debugger() at cpu_Debugger+0x4
panic() at panic+0xfc             
insmntque() at insmntque+0x44
getnewvnode() at getnewvnode+0x34c
nfs_nget() at nfs_nget+0x124      
nfs_lookup() at nfs_lookup+0x1124
lookup() at lookup+0x40c         
namei() at namei+0x424  
change_dir() at change_dir+0x2c
sys_chdir() at sys_chdir+0x44  
syscall() at syscall+0x1dc   
XentSys() at XentSys+0x50 
--- syscall (12, netbsd.sys_chdir) ---
--- user mode ---
db>

I've got a flaky network cable, so during periods of heavy NFS, (like
cvs updates,) the alpha loses track of the server (a pmax running 1.4)
for brief periods.  I'll update my kernel (after cvs updating on a
_different_ machine) and see if that fixes anything.  :)

-- 
  Aaron J. Grier | "Not your ordinary poofy goof." | agrier@poofygoof.com