From: Jonathan V Hays (jhays@jtan.com)
Date: Thu Jul 31 2003 - 09:20:09 GMT-3
Let me verify your math.
access-list 33 permit 228.0.0.0 10.255.255.255
10 00001010 mask
228 11100100 prefix to be matched
1110x1x0
^ ^
This mask forces a match on every bit in the prefix
except those denoted by 'x' above.
1110x1x0
229 11100101 don't match
230 11100110 match
236 11101100 match
238 11101110 match
You are right - it looks like the 229 prefix should be filtered.
Perhaps as you say, this type of access list won't work in
this situation.
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
jfaure@sztele.com
Sent: Thursday, July 31, 2003 7:15 AM
To: ccielab@groupstudy.com
Subject: Distributing multicast groups to different RPs with AutoRP
Hi all:
As you know, you can use AutoRP in such a way that permits that certain
C-RPs were elected as RPs only for a "range" of multicasts groups. You
use
in the config of the C-RP some like this:
ip pim send-rp-announce Loopback0 scope 15 group-list 33
Where "33" is an ACL, that in my practices is:
access-list 33 permit 228.0.0.0 10.255.255.255
Whit this config, i though that this router will be the RP only for the
multicast groups that have a first octec like:
228
230
236
238
This is, i though the acl forces the exact match in the bits that you
have
put as "0" in the wilcard mask.
However, i'm seing that a multicast group like 229.6.6.6 is accepted too
by
this c-rp, and in deed the router is elected as RP for this group.
I've read the command reference guide for 12.1 (i'm using 12.1.19 IOS)
and
isn't very clear. Anyone can explain the reason of this issue? It seems
that the ACL is treated as a "range".
Any help would be highly apreciated
Juan Faure Ferrer
email: jfaure@sztele.com
Lmnea de Negocio de Telematica y CC
Ingeniero de Integracisn de Redes y Sistemas
------------------------------------------------------------------------
----SOLUZIONA TELECOMUNICACIONES Servicios Profesionales de UNION FENOSA Jerez, 3 28016 MADRID tel 91 579 30 00 fax 91 350 72 83 ------------------------------------------------------------------------ ---
This archive was generated by hypermail 2.1.4 : Wed Aug 06 2003 - 06:52:59 GMT-3