From: Ravi Singh (way2ccie@googlemail.com)
Date: Thu Mar 19 2009 - 03:22:18 ART
Hi Group,
I don't know what am I missing here but this thing is really making me
crazy. I am configuring PIM sparse mode on 3 routers connected
linearly i.e
R1(F0/0) <-------> (F0/0) R3 (F1/0) <-------> (F1/0) R5
(F0/0)---->joined 224.1.1.1
| | | |
(10.1.13.1) (10.1.13.3) (10.1.35.3) (10.1.35.5)
All the routers have loopback0 addresses configured on them as
1.1.1.1/8 , 3.3.3.3/8 and 5.5.5.5/8 . R5 is the RP . RP address is
configured as 5.5.5.5 in all the three routers statically.OSPF runs in
all the three routers on all the interfaces with them being in area 0.
I have done all the required config i.e enabled sparse-mode on the
interfaces , defined the static rp-address but when I ping the
multicast 224.1.1.1 from R1 , it times out. A debug shows me that the
ping is being sourced from loopback 0 of R1 (1.1.1.1) and R1 says it
as unroutable. I don't get anything if I debug ip pim or debug ip
mpacket detail. The only output I get is when I debug ip packet detail
which is this
R1#ping 224.1.1.1
Type escape sequence to abort.
Sending 1, 100-byte ICMP Echos to 224.1.1.1, timeout is 2 seconds:
IP: s=1.1.1.1 (local), d=224.1.1.1 (FastEthernet0/0), len 100, sending
broad/multicast
ICMP type=8, code=0
IP: s=1.1.1.1 (local), d=224.1.1.1 (Loopback0), len 100, sending broad/multicast
ICMP type=8, code=0
IP: s=1.1.1.1 (Loopback0), d=224.1.1.1, len 100, unroutable
ICMP type=8, code=0.
Also, here is the debug ip mpacket detail output on R3, if it helps. I
don't know why it fails an RPF check when the routing looks ok to me.
R3#
PIM(0): Check RP 5.5.5.5 into the (*, 224.1.1.1) entry
IP(0): MAC sa=cc00.0928.0000 (FastEthernet0/0), IP last-hop=10.1.13.1
IP(0): IP tos=0x0, len=100, id=32, ttl=254, prot=1
IP(0): s=1.1.1.1 (FastEthernet0/0) d=224.1.1.1 id=32, ttl=254, prot=1,
len=114(100), not RPF interface
R3#
Now, If I change the source interface or IP address in the ping
command to F0/0 , everything works. If I enable sparse-mode on
Loopback 0, everything works again.Why is R1 sourcing the packet from
Loopback 0 ? Is this expected behaviour ? Has anyone else encountered
this before, or is it just sheer stupidity that I am enjoying here ? I
am using dynamips by the way and and it's the same problem, in three
different machines where I run Dynamips.
Can anyone help me get out of this please .
TIA
Ravi
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Mon Apr 06 2009 - 06:44:05 ART