I don't have a lot of time to chase this down right now, but was 
wondering if anyone else has seen this particular assertion... it's 
happened on two consecutive shutdowns, now.

I'm running a 2.99.10 kernel and in-sync userland from about a week ago 
on i386.

unmounting file systems...
unmounting /amd/boogers/work (boogers:/work)...panic: kernel diagnostic 
assertio
n "np->n_dirgens == NULL" failed: file "../../../../nfs/nfs_node.c", 
line 312
Stopped in pid 5226.1 (reboot) at       netbsd:breakpoint+0x4:  leave
db> trace
breakpoint(c056a398,d3d2bca8,d3d2bc9c,c0343e1e,d3d2bca8) at 
netbsd:breakpoint+0x
4
cpu_Debugger(d3d2bca8,104,d3d2bccc,c0517683,c0607e20) at 
netbsd:cpu_Debugger+0xb

panic(c0607e20,c0555256,c0555225,c055523b,138) at netbsd:panic+0xd4
__assert(c0555256,c055523b,138,c0555225,d3d2bd14) at netbsd:__assert+0x1f
nfs_reclaim(d3d2bd14,1b,0,0,c056f360) at netbsd:nfs_reclaim+0xcb
VOP_RECLAIM(cdd17be8,cc59bcc0,11,c03421cf,c067ab60) at 
netbsd:VOP_RECLAIM+0x37
vclean(cdd17be8,8,cc59bcc0,c037d4ee,cdf5c354) at netbsd:vclean+0x33d
vgonel(cdd17be8,cc59bcc0,d3d2bdac,c03723d3,cdf5c354) at netbsd:vgonel+0x57
vflush(c1f8e000,cdf5c354,2,cc59bcc0,cdf6d88c) at netbsd:vflush+0xab
nfs_unmount(c1f8e000,80000,cc59bcc0,cc59bcc0,0) at netbsd:nfs_unmount+0xdd
dounmount(c1f8e000,80000,cc59bcc0,c0344715,0) at netbsd:dounmount+0x153
vfs_unmountall(cc59bcc0,0,d3d2be5c,c0374110,cc59bcc0) at 
netbsd:vfs_unmountall+0
xb1
vfs_shutdown(282,0,d3d2bf2c,c033b5aa,0) at netbsd:vfs_shutdown+0xa7
cpu_reboot(0,0,d3d2bebc,c032a2f4,0) at netbsd:cpu_reboot+0x45
sys_reboot(cc45ebe0,d3d2bf64,d3d2bf5c,d3d2bf60,c03e515f) at 
netbsd:sys_reboot+0x
9e
syscall_plain() at netbsd:syscall_plain+0x13b
--- syscall (number 208) ---
0x480869db:
db>

I'm happy to send-pr this if someone tells me what other info I should 
include...

+j