I've had my Qube crash 3 times now with this info:

db> bt
80224304+214 (ff00,f9,c6,2000) ra 800d2c0c sz 0
800d25fc+610 (ff00,f9,c6,2000) ra 0 sz 0
User-level: curlwp NULL
db> sync
syncing disks... trap: TLB miss (load or instr. fetch) in kernel mode
status=0xff02, cause=0xbc08, epc=0x8022736c, vaddr=0x0
curlwp == NULL ksp=0xcb5cf590
Stopped at      netbsd:pmap_clear_modify+0x194: lw      v1,16(v0)
db>

I don't know what's doing it, and I don't know what time of day it's
happening in particular. I don't think it's happening at the same
time.

Here's the kernel:

# uname -a
NetBSD taz 3.1 NetBSD 3.1 (TAZ3) #0: Mon Dec 11 07:20:22 MST 2006  root@foghorn:
/usr/obj-cobalt-3.1/sys/arch/cobalt/compile/TAZ3 cobalt

This thing had been stable for months, and I even built the entire
-current release on it which took over 2 days. So I don't know why
suddenly it's starting to crash.

I'll put 4.0 on it when it hits.

Andy