Figuring out a headless server's zeroconf addr

[Robert Moskowitz]

On 07/21/2014 11:25 AM, Tim wrote:
> On Mon, 2014-07-21 at 10:55 -0400, Robert Moskowitz wrote:
>> I have a headless system that I cannot connect to.  So I was thinking
>> to put a direct connection to it and my notebook.  Both ethernets
>> would use the zeroconf (169.254.0.0/16) addresses.  I could then use
>> fping
>>   
>> fping -g 169.254.0.0/16
>>   
>> And SHOULD be able to get its address, and then SSH into the box.
> I was under the impression that zeroconf did some rudimentary name
> resolution, and you ought to be able to connect to hostname.local
> (replacing "hostname" with the actual hostname).
>
> It'd be a bit dopey if a zero-configuration scheme required you to
> configure things...

I am doing a little testing, and zeroconf does not seem to be working.  
I am seeing the link up light on my ethernet port.  I am seeing a 
local-scope v6 address, but no v4 address:

p6p1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
         inet6 fe80::ea9a:8fff:fe8d:7b56  prefixlen 64  scopeid 0x20<link>
         ether e8:9a:8f:8d:7b:56  txqueuelen 1000  (Ethernet)
         RX packets 7  bytes 2130 (2.0 KiB)
         RX errors 0  dropped 0  overruns 0  frame 0
         TX packets 46  bytes 4948 (4.8 KiB)
         TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

Note that it is receiving packets from the other system.   It also has a 
local-scope ipv6 addr, but no zeroconf addr (both systems are Fedora 20).

So how do I start zeroconf?  Of course on the other system, I can't do 
that...

So given ipv6 local-scope, how do I learn the other system's addr. 
Trying to figure out fping6.   How do I restrict it to the desired 
interface?

>
>> Any other thoughts?  I can't get to the box to recable it and reboot
>> it (as that is the only way I can figure out for it to readdress eth0)
>> until this evening.
> Only that:  Are you on the same network?  169.254 connections can't be
> expected to be reachable outside of their own net.

Crossover cable.  Is that enough of a 'same network'?  :)

And I have considerable routing and addressing knowledge.  Besides being 
one of the authors of rfc 1918, and worked on CIDR, here at IETF I 
contribute to ipv6ops and ipv6man.