<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Simon , Bob , Peter:<br>
Thank for sharing your wisdom on this, there is something I still
can not understand.<br>
Maybe a picture can help, please take a look bellow.<br>
On scenario 1 following should happend:<br>
Relay1 will set Gi-addr=1.1.0.1 so dhcpd will pick a lease from pool
1 and network 1.1.0.0/24 will use 1.1.1.1 as gateway.<br>
Relay2 will set Gi-addr=1.1.2.1 so dhcpd will pick a lease from pool
3 and network 1.1.2.0/24 will use 1.1.2.0 as gateway.<br>
This is ok;<br>
<br>
On Scenario 2<br>
After running out of /24 ips I will add remaining /24 networks
behind each relay.<br>
Add pool2 behind relay 1<br>
Add pool4 behind relay 2.<br>
How will it work ? <br>
My concern is that, if relay1 still using 1.1.0.1 for gi-addr ;
dhcpd can pick a lease from pool4 instead using pool2 since 1.1.0.1
falls into the 1.1.0.0/22 declared on shared network.<br>
If this happens the request will receive the option router = 1.1.3.1
witch ip is not set at any interface on relay1.<br>
And vice-versa, dhcp server can pick a lease from pool2 for a
request coming from relay2, so it will also receive an incorrect
router value.<br>
<br>
<img alt="relay" title="relay"
src="cid:part1.09060502.03070405@gmail.com" height="543"
width="830"><br>
<br>
<br>
bw: Thanks Simon for reading material sugestion, i will begin now.<br>
Thanks in advance.<br>
Leo.<br>
<br>
<br>
<br>
<br>
<div class="moz-cite-prefix">On 23/06/15 18:19, Simon Hobson wrote:<br>
</div>
<blockquote
cite="mid:51E21E2F-5E3C-4CF3-A785-35DC07E46A49@thehobsons.co.uk"
type="cite">
<pre wrap="">Leandro <a class="moz-txt-link-rfc2396E" href="mailto:ingrogger@gmail.com"><ingrogger@gmail.com></a> wrote:
</pre>
<blockquote type="cite">
<pre wrap="">After I run out of those ips, I can do two thinks:
a)change the network mask from /24 to /23.
b)Add a second /24 subnet behind the relay , for example 1.1.2.0/24 and set a second gateway ip 1.1.2.1/24.
option a) is not good since the broadcast domain at /23 could bring many collisions. (its just my opinion).
</pre>
</blockquote>
<pre wrap="">
I don't think it will make much difference. Don't forget that having two subnets in a shared network won't segregate the broadcast traffic. I think the only reduction would be from inter subnet traffic going via the router rather than using ARP to find the neighbour - but routing the traffic via the router rather than directly will more than outweigh any saving there.
</pre>
<blockquote type="cite">
<pre wrap="">option b) Could work but, how does relay agent knows witch ip to use for GI-Adrr ?
Can relay agent send both or more than one ips, on the GI-Addr field so dhcpd can figure out from witch range can serve the ip ?
</pre>
</blockquote>
<pre wrap="">
As already mentioned, as long as the GI-Addr value is within any of the subnets, then the server will work it out from the shared-network.
BTW - I'd suggest a read of "The DHCP Handbook" by Ralph Droms and Ted Lemon, it explains all this and a lot more, and is quite readable.
_______________________________________________
dhcp-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:dhcp-users@lists.isc.org">dhcp-users@lists.isc.org</a>
<a class="moz-txt-link-freetext" href="https://lists.isc.org/mailman/listinfo/dhcp-users">https://lists.isc.org/mailman/listinfo/dhcp-users</a>
</pre>
</blockquote>
<br>
</body>
</html>