[Xorp-users] Problem with XORP RIP packet loss

[Bogdan Krivokapic]

I am trying to test the RIP convergence time using the following network
configuration:

*host1*---net1---*router1*---net2---*router2*---net3---*host2*
                        |                      |
                        |                      |
                      net4                 net5
                        |                      |
                        |-----*router3*---- -|

net1: (10.10.10.50)---(10.10.10.1)
net2: (192.168.10.1)----(192.168.10.2)
net3: (172.16.10.1)-------(172.16.10.50)
net4: (10.0.0.1)-----(10.0.0.2)
net5: (10.0.1.2)-----(10.0.1.1)

Every router has a simple configuration file. For example, the
configuration file on router1 is:

protocols{
rip {
interface eth0 {
vif eth0 {
address 10.10.10.1 {
metric: 1
horizon: "split-horizon-poison-reverse"
disable: false
passive: false
accept-non-rip-request: true
accept-default-route: true
advertise-default-route: true
route-timeout: 180
deletion-delay: 120
triggered-delay: 120
triggered-jitter: 66
update-interval: 30
update-jitter: 16
request-interval: 30
interpacket-delay: 50
}
}
}
interface eth1 {
vif eth1 {
address 192.168.10.1 {
metric: 1
horizon: "split-horizon-poison-reverse"
disable: false
passive: false
accept-non-rip-request: true
accept-default-route: true
advertise-default-route: true
route-timeout: 180
deletion-delay: 120
triggered-delay: 120
triggered-jitter: 66
update-interval: 30
update-jitter: 16
request-interval: 30
interpacket-delay: 50
}
}
}
interface eth2 {
vif eth2 {
address 10.0.0.1 {
metric: 1
horizon: "split-horizon-poison-reverse"
disable: false
passive: false
accept-non-rip-request: true
accept-default-route: true
advertise-default-route: true
route-timeout: 180
deletion-delay: 120
triggered-delay: 120
triggered-jitter: 66
update-interval: 30
update-jitter: 16
request-interval: 30
interpacket-delay: 50
}
}
}
export: "connected"
}
}

policy {
policy-statement connected {
term export {
from {
protocol: "connected"
}
}
}
}
fea {
unicast-forwarding4 {
disable: false
}
}
interfaces {
interface eth0 {
disable: false
discard: false
vif eth0 {
disable: false
address: 10.10.10.1 {
prefix-length: 24
broadcast: 10.10.10.255
disable: false
}
}
}
interface eth1 {
disable: false
discard: false
vif eth1 {
disable: false
address: 192.168.10.1 {
prefix-length: 24
broadcast: 192.168.10.255
disable: false
}
}
}
interface eth2 {
disable: false
discard: false
vif eth2 {
disable: false
address: 10.0.0.1 {
prefix-length: 24
broadcast: 10.0.0.255
disable: false
}
}
}
}
I used the following procedure: First i started ping command from host1 to
host2 on 172.16.10.50 and then disabled router1 interface 192.168.10.1.
After all RIP timers expired router1 started routing through router3. But
then problems appeared.
Again i started ping from host1 to host2. After bringing the interface
192.168.10.1 back in action host1 showed packet loss. Router1 immediately
learned route to 172.16.10.0 through router2. Router1 routing table was
showing that packets are routed through router2. At the same time router2
routing table was showing that router2 reaches network 10.10.10.0 by
routing through router3, but packets from host1 were somehow lost. My
question is why? This packet loss continues until router2 learnes a shorter
route to 10.10.10.0 through periodic update.
  Another strange thing is that when i ping from router1 to host2 at the
same time there is no packet loss, also when i ping from router2 to host1
there is no packet loss. And at the end when i ping from host1 to interface
10.0.1.1 which belongs to router2 there is no packet loss.
I also tried quagga and i had the same problem.

Why is there packet loss when i ping from host1 to host2 in the moment when
i enable interface on router1 until router2 stops routing through router3?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20130621/b31989e2/attachment.html