1 |
On Thu, Jan 19, 2012 at 4:55 PM, Grant Edwards |
2 |
<grant.b.edwards@×××××.com> wrote: |
3 |
> On 2012-01-19, Michael Mol <mikemol@×××××.com> wrote: |
4 |
> |
5 |
>> Indeed. Other reasons to avoid using LL addresses unless necessary: |
6 |
>> What if the MAC address on the server changes? |
7 |
> |
8 |
> It won't. It's an embedded device with a hard-wired MAC that the user |
9 |
> can't change. |
10 |
|
11 |
It was more a philosophical question, not one of the specific use |
12 |
case. In most systems, hardware NICs fail and may be replaced. (Well, |
13 |
virtualization is making that a bit odd, but still.) I have ideas |
14 |
about your use case, but I can't and won't judge because I don't know |
15 |
enough specifics. Your product, not mine. :) |
16 |
|
17 |
> |
18 |
>> What if your network grows to have hundreds of clients? |
19 |
> |
20 |
> Then people probably won't be using L-L addresses. However, for a |
21 |
> network that consists of 6 small devices all living inside a cabinet |
22 |
> with no router, DHCP server, or connection to the outside workd, L-L |
23 |
> is great. |
24 |
|
25 |
Sure, so long as various applications get fixed to understand LL |
26 |
addresses and are corrected to direct traffic to the appropriate |
27 |
interfaces, which is something I'd definitely like to see. |
28 |
|
29 |
>> Do you really want that much broadcast and wide multicast (think |
30 |
>> DNS-SD and NTP in multicast mode) traffic on the same Ethernet |
31 |
>> segment? |
32 |
> |
33 |
> That bit I don't understand. It's no worse that ARP, and we seem to |
34 |
> live with that quite easily. |
35 |
|
36 |
Not just arp, but actual broadcast/multicast data. If you've ever run |
37 |
PulseAudio and enabled network sources and sinks on a couple boxes, |
38 |
you might have accidentally discovered an easy way to bring a wireless |
39 |
network to its knees. And that's just something I've had personal |
40 |
experience with. Come to think of it, that's a good reason I should |
41 |
continue to keep my home wired and wireless networks on separate |
42 |
subnets, and not simply bridged as I'd done at the time. |
43 |
|
44 |
One anecdote a friend of mine gave me...there was a network he was |
45 |
brought in to manage where he discovered that a huge campus of over a |
46 |
thousand hosts was configured as one large ethernet segment with |
47 |
various-speed links bridging smaller islands. The slower links were |
48 |
absolutely flooded with arp and netbios broadcasts, and the network |
49 |
moved along at a crawl. Chopping that up into a few routed subnets |
50 |
gave the entire network a massive performance boost. |
51 |
|
52 |
> |
53 |
>> LL addresses are very useful for diagnostic and investigation |
54 |
>> purposes, of course. |
55 |
> |
56 |
> Indeed, and that's what I'm doing. |
57 |
|
58 |
-- |
59 |
:wq |