/31 on 2 interfaces

[sfp]

No love with a /31 mask & default bcast addr.  Perhaps ARP reqs are failing to 
the subnet bcast address, didn't check.  To fix, specify an all ones broadcast.

Tested this against 6.0-Release & a Juniper box that was handy.

As always, YMMV.  Oliver has a point though, why not just use a /30 & save 
yourself the grief?

ifconfig vlan20 10.10.20.2 netmask 255.255.255.254 broadcast 255.255.255.255

[root at itestbox itest1]# ifconfig vlan20
vlan20: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        inet6 fe80::210:4bff:fe14:bc03%vlan20 prefixlen 64 scopeid 0x11 
        inet 10.10.20.2 netmask 0xfffffffe broadcast 255.255.255.255
        ether 00:02:b3:ab:66:3c
        media: Ethernet 100baseTX <full-duplex>
        status: active
        vlan: 20 parent interface: em1


set interfaces fxp1 unit 0 family inet address 10.10.20.3/31 broadcast 
255.255.255.255

jtest1 at jtestbox> show interfaces fxp1.0 extensive 
  Logical interface fxp1.0 (Index 66) (SNMP ifIndex 14) (Generation 3)
    Flags: SNMP-Traps Encapsulation: ENET2
    Protocol inet, MTU: 1500, Generation: 9, Route table: 0
      Flags: None
      Addresses, Flags: Is-Preferred Is-Primary
        Destination: 10.10.20.2/31, Local: 10.10.20.3, Broadcast: Unspecified, 
Generation: 23


[root at itestbox itest1]# ping -qnc2 10.10.20.3
PING 10.10.20.3 (10.10.20.3): 56 data bytes

--- 10.10.20.3 ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max/stddev = 0.369/0.439/0.508/0.070 ms


jtest1 at jtestbox> ping 10.10.20.2 count 2 
PING 10.10.20.2 (10.10.20.2): 56 data bytes
64 bytes from 10.10.20.2: icmp_seq=0 ttl=64 time=0.407 ms
64 bytes from 10.10.20.2: icmp_seq=1 ttl=64 time=0.412 ms

--- 10.10.20.2 ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max/stddev = 0.407/0.409/0.412/0.002 ms


jtest1 at jtestbox> monitor traffic interface fxp1 extensive 
Listening on fxp1, capture size 96 bytes

19:31:39.559772  In 0:2:b3:ab:66:3c > 0:2:b3:9:57:dd, ethertype IPv4 (0x0800), 
length 98: (tos 0x0, ttl  64, id 6478, offset 0, flags [none], proto: ICMP (1), 
length: 84) 10.10.20.2 > 10.10.20.3: ICMP echo request seq 0, length 64
19:31:39.559865 Out 0:2:b3:9:57:dd > 0:2:b3:ab:66:3c, ethertype IPv4 (0x0800), 
length 98: (tos 0x0, ttl  64, id 39612, offset 0, flags [none], proto: ICMP 
(1), length: 84) 10.10.20.3 > 10.10.20.2: ICMP echo reply seq 0, length 64
19:31:40.561499  In 0:2:b3:ab:66:3c > 0:2:b3:9:57:dd, ethertype IPv4 (0x0800), 
length 98: (tos 0x0, ttl  64, id 6482, offset 0, flags [none], proto: ICMP (1), 
length: 84) 10.10.20.2 > 10.10.20.3: ICMP echo request seq 1, length 64
19:31:40.561579 Out 0:2:b3:9:57:dd > 0:2:b3:ab:66:3c, ethertype IPv4 (0x0800), 
length 98: (tos 0x0, ttl  64, id 39615, offset 0, flags [none], proto: ICMP 
(1), length: 84) 10.10.20.3 > 10.10.20.2: ICMP echo reply seq 1, length 64
^C
4 packets received by filter
0 packets dropped by kernel



Quoting Unix-Solutions - Steven <steven at unix-solutions.be>:

> I'm need a point-to-point link between 2 servers
> with just a crossover cable between the servers.
> 
> ----- Original Message ----- 
> From: "fooler" <fooler at skyinet.net>
> To: "Kim Shrier" <kim at tinker.com>; "Unix-Solutions - Steven" 
> <steven at unix-solutions.be>
> Cc: <freebsd-net at freebsd.org>
> Sent: Wednesday, May 17, 2006 5:49 AM
> Subject: Re: /31 on 2 interfaces
> 
> 
> > ----- Original Message ----- 
> > From: "Kim Shrier" <kim at tinker.com>
> > To: "Unix-Solutions - Steven" <steven at unix-solutions.be>
> > Cc: <freebsd-net at freebsd.org>
> > Sent: Wednesday, May 17, 2006 9:20 AM
> > Subject: Re: /31 on 2 interfaces
> >
> >
> >>A /31 is not a point-to-point link.
> >
> > see RFC 3021
> >
> >> A /31 is a network with nothing but a loopback and broadcast address.
> >
> > not a loopback address but a network address (all zeros bit) and a 
> > broadcast address (all ones bit)
> >
> > fooler.
> >
> > _______________________________________________
> > freebsd-net at freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-net
> > To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
> > 
> 
> 
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"
>