nfs pxebooted box hangs on reboot

[Patrick Welche <prlw1%cam.ac.uk@localhost>]

I don't seem to be able to reboot a -current/i386 box which
pxeboots/nfs mounts another -current/i386 box (both running today's
source) During shutdown all looks promising:

Dec  5 18:19:24 egg shutdown: reboot by prlw1: 
Dec  5 18:19:34 egg syslogd[180]: Exiting on signal 15
nfs send error 49 for 10.0.6.7:/iscsi/i386
syncing disks... done
wd1: detached
...
audio0: detached
md0: detached

But then nothing. Breaking into ddb:

db{0}> ps
PID    LID S CPU     FLAGS       STRUCT LWP *               NAME WAIT
526      1 3   3   9020000           cdfb6340             dhcpcd netio
453      1 3   3         0           cdfb60c0             reboot tstile
186      1 3   3   9020000           cdc25040             dhcpcd wait
1        1 3   2   8020080           cd1235e0               init wait
0       69 3   2       200           cd130520        vmem_rehash
vmem_rehash
0       68 3   1       200           cd1307a0           aiodoned
aiodoned
0       67 3   2       200           cd130a20            ioflush syncer
0       66 3   2       200           cd130ca0           pgdaemon
pgdaemon
0       65 3   2       200           cd12c000              nfsio nfsiod
0       64 3   3       200           cd12c280              nfsio nfsiod
0       63 3   2       200           cd12c500              nfsio nfsiod
0       62 3   1       200           cd12c780              nfsio nfsiod
0       61 3   2       200           cd123ae0            pfpurge pftm
...
and the 4 on-CPU processes are "idle".

BTW trying to kill PID 526, I got

db{0}> kill 0t526
This command is not currently supported.

which seems odd...

db{0}> trace/a cdfb6340
trace: pid 526 lid 1 at 0xce03933c
sleepq_block(1f4,0,cdfb6340,23,31,0,31,0,cdfb6340,4) at
netbsd:sleepq_block+0x124
cv_timedwait(c3498ea0,cc746f40,1f4,c05e571c,c3498dc4,0,14,c3498dc0,cc746f40,c3498dc0)
 at netbsd:cv_timedwait+0x3a
sbwait(c3498e78,0,ce039480,c07130b6,6,0,fffffffd,23,0,1) at netbsd:sbwait+0x54
soreceive(c3498dc0,ce039514,ce039490,ce039518,0,ce0394b8,7,c349b540,8280,14) at 
netbsd:soreceive+0x3e8
nfs_receive(cdc11a80,ce039514,ce039518,cdfb6340,0,ce03950c,ce03953c,c03ee6b0,6,c06e0675)
 at netbsd:nfs_receive+0x627
nfs_reply(cdc11a80,cdfb6340,c349e300,cdc11a80,cdfb6340,0,4,0,c349b500,20) at 
netbsd:nfs_reply+0x95
nfs_request(cdc5e560,c349b500,1,cdfb6340,cc74af00,ce0396c4,ce0396c0,ce0396c8,0,cd0de006)
 at netbsd:nfs_request+0x7ef
nfs_getattr(ce039724,3c,a,0,c07336c0,cdc5d458,ce039760,cc74af00,3c,0) at netbsd:
nfs_getattr+0x4be
VOP_GETATTR(cdc5d458,ce039760,cc74af00,c03f78d4,ce0398b0,cdc81df4,88,0,0,0) at 
netbsd:VOP_GETATTR+0x62
nfs_flushstalebuf(cdc5d458,cc74af00,cdfb6340,1,6,6000000,ce03983c,c03377dc,ce039806,6)
 at netbsd:nfs_flushstalebuf+0x11e
nfs_bioread(cdc5d458,ce039984,0,cc74af00,0,c0733c80,0,1,ce039940,cdc5d458) at ne
tbsd:nfs_bioread+0x1a9
nfs_read(ce039940,1,ce03995c,c0733740,cdc5d458,ce039984,0,cc74af00,ce039998,0) 
at netbsd:nfs_read+0x5d
VOP_READ(cdc5d458,ce039984,0,cc74af00,0,0,ce0399ac,c0648d79,ce0399a0,1) at 
netbsd:VOP_READ+0x68
vn_rdwr(0,cdc5d458,cdd3e800,404,0,0,1,0,cc74af00,ce039a0c) at netbsd:vn_rdwr+0xc
5          
check_exec(cdfb6340,ce039afc,400,ce039c58,20,0,ce039a7c,c025a13c,1008,ce039af8) 
at netbsd:check_exec+0x1b1
execve1(cdfb6340,bb9120b0,bfbfecf4,bb91d080,c032ea9d,0,ce039cbc,c0550bcb,cdfb63 
0,ce039ce4) at netbsd:execve1+0x28a
sys_execve(cdfb6340,ce039ce4,ce039d0c,0,c0537976,cdf1b00c,ce039cbc,c0550796,cdfb6340,cdfb6340)
 at netbsd:sys_execve+0x36
sy_call(c074b3a4,cdfb6340,ce039ce4,ce039d0c,c0482842,c1f9e94c,0,c1f9e140,bb9120b0,bfbfecf4)
 at 0xc0550bcb
syscall(ce039d48,b3,bfbf00ab,bfbf001f,bb91001f,23,bfbfff62,bfbfecb8,0,bfbfecf4) 
at netbsd:syscall+0x1ef

might be where it is stuck?

Thoughts?

Patrick