ospf broken on trunk interfaces?

[Paul B. Henson]

I'm trying to setup ospf on a trunk interface. I've had it configured
and working fine on a regular interface for quite some time, and now am
trying to add another neighbor on a trunk interface, and it just shows
the interface as down:

# ospfctl show i
Interface   Address            State  HelloTimer Linkstate  Uptime    nc ac
trunk0      10.128.0.9/30      DOWN   -          active     00:00:00   0 0
lo1         10.128.0.4/32      LOOP   -          unknown    17w5d04h   0 0
re0         10.128.0.1/30      BCKUP  00:00:05   active     17w5d08h   1 1

The trunk is definitely up:

# ifconfig trunk0
trunk0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        lladdr 00:30:18:a8:7c:cc
        priority: 0
        trunk: trunkproto lacp
        trunk id: [(8000,00:30:18:a8:7c:cc,4094,0000,0000),
                 (8000,f0:25:72:53:82:00,0001,0000,0000)]
                trunkport re1 active,collecting,distributing
        groups: trunk
        media: Ethernet autoselect
        status: active
        inet 10.128.0.9 netmask 0xfffffffc broadcast 10.128.0.11
        inet6 fe80::230:18ff:fea8:7ccc%trunk0 prefixlen 64 scopeid 0x12

I currently only have one physical port in the trunk (planning to add a second
later once it's all working):

# ifconfig re1
re1: flags=8b43<UP,BROADCAST,RUNNING,PROMISC,ALLMULTI,SIMPLEX,MULTICAST> mtu 
1500
        lladdr 00:30:18:a8:7c:cc
        priority: 0
        trunk: trunkdev trunk0
        media: Ethernet autoselect (1000baseT full-duplex)
        status: active
        inet6 fe80::230:18ff:feac:b83a%re1 prefixlen 64 scopeid 0x2

Traffic is definitely passing, I can ping the other side (which is a cisco 
layer3
switch):

# ping 10.128.0.10
PING 10.128.0.10 (10.128.0.10): 56 data bytes
64 bytes from 10.128.0.10: icmp_seq=0 ttl=255 time=4.427 ms

tcpdump on trunk0 shows ospf hello packets from the cisco gear:

22:26:09.862595 cisco-bart.pbhware.com > ospf-all.mcast.net: OSPFv2-hello  
44[80]: rtrid cisco.nms.pbhware.com backbone dr cisco-bart.pbhware.com [tos 
0xc0] [ttl 1]

I found this mailing list posting with exactly the same problem it would seem:

http://old.nabble.com/trunk-and-ospf-on-openbsd-4.8-td31833059.html

But there were no responses. I'm running OpenBSD 5.0, and in case I'm doing
something stupid here's the ospf config:

router-id 10.128.0.4
redistribute default
redistribute connected

area 0.0.0.0 {

        interface lo1:10.128.0.4 { passive }

        interface re0 {
                auth-type crypt
                auth-md 1 "XXXXX"
                auth-md-keyid 1
        }
        interface trunk0 {
                auth-type crypt
                auth-md 1 "XXXXX"
                auth-md-keyid 1
        }
}


Any suggestions? Is ospf not supported on trunk interfaces as surmised by the
other mailing list posting?

Thanks much for any assistance...