EIGRP: Active QUERY and distribution-list filters

Answered Question
Jan 18th, 2008

Hi,

let say we have 3 routers

(many networks)-R1---R2---R3

all have EIGRP, R1 anounce all networks to R2, R2 has "distribute-list out" configured and announce only some networks to R3.

If the link between R1 and R2 goes down then R2 will go Active and will send QUREY message to R3.

Question is, for which netwoks will be send the QUERY message? for all networks or only for subset, which defined by "distribute-list out"? Has "distribute-list out" any impact or influence on the QUERY's?

Correct Answer by Jon Marshall about 9 years 1 month ago

Hi Konstantin

Already had EIGRP setup in lab for something else so did a quick test

R1 -> R2 -> R3

R1 advertises to R2 the following subnets

192.168.22.0

172.16.31.0

172.16.32.0

R2 has the following distribute list on the interface to R3

router eigrp 1

distribute-list 1 out FastEthernet0/0

access-list 1 deny 172.16.32.0

access-list 1 deny 172.16.31.0

access-list 1 permit any

Comfirmed that R3 only received the 192.168.22.0/24 subnet.

Then i shutdown the link between R1 & R2 and ran "debug eigrp fsm" on R3

*Jan 18 11:17:29.530: DUAL: rcvquery: 192.168.22.0/28 via 192.168.6.1 metric 4294967295/4294967295, RD is 33280

*Jan 18 11:17:29.530: DUAL: Find FS for dest 192.168.22.0/28. FD is 33280, RD is 33280

*Jan 18 11:17:29.530: DUAL: 192.168.6.1 metric 4294967295/4294967295 not found Dmin is 4294967295

*Jan 18 11:17:29.530: DUAL: Peer total/stub 1/0 template/full-stub 1/0

*Jan 18 11:17:29.530: DUAL: Dest 192.168.22.0/28 (Split Horizon) not entering active state.

*Jan 18 11:17:29.530: DUAL: send REPLY(r1/n1) about 192.168.22.0/28 to 192.168.6.1

*Jan 18 11:17:29.530: DUAL: dest(172.16.32.0/24) not active

*Jan 18 11:17:29.530: DUAL: rcvquery: 172.16.32.0/24 via 192.168.6.1 metric 4294967295/4294967295, RD is 4294967295

*Jan 18 11:17:29.530: DUAL: send REPLY(r1/n1) about 172.16.32.0/24 to 192.168.6.1

*Jan 18 11:17:29.530: DUAL: dest(172.16.31.0/24) not active

*Jan 18 11:17:29.530: DUAL: rcvquery: 172.16.31.0/24 via 192.168.6.1 metric 4294967295/4294967295, RD is 4294967295

*Jan 18 11:17:29.530: DUAL: send REPLY(r1/n1) about 172.16.31.0/24 to 192.168.6.1

*Jan 18 11:17:29.554: DUAL: Removing dest 192.168.22.0/28, nexthop 192.168.6.1

*Jan 18 11:17:29.554: DUAL: No routes. Flushing dest 192.168.22.0/28

*Jan 18 11:17:29.554: DUAL: Removing dest 172.16.32.0/24, nexthop 192.168.6.1

*Jan 18 11:17:29.554: DUAL: No routes. Flushing dest 172.16.32.0/24

*Jan 18 11:17:29.554: DUAL: Removing dest 172.16.31.0/24, nexthop 192.168.6.1

*Jan 18 11:17:29.554: DUAL: No routes. Flushing dest 172.16.31.0/24

As you can see R2 still sends queries for 172.16.31.0/24 and 172.16.32.0/24 even though it is filtering these from R1.

HTH

Jon

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (1 ratings)
Loading.
Correct Answer
Jon Marshall Fri, 01/18/2008 - 04:47

Hi Konstantin

Already had EIGRP setup in lab for something else so did a quick test

R1 -> R2 -> R3

R1 advertises to R2 the following subnets

192.168.22.0

172.16.31.0

172.16.32.0

R2 has the following distribute list on the interface to R3

router eigrp 1

distribute-list 1 out FastEthernet0/0

access-list 1 deny 172.16.32.0

access-list 1 deny 172.16.31.0

access-list 1 permit any

Comfirmed that R3 only received the 192.168.22.0/24 subnet.

Then i shutdown the link between R1 & R2 and ran "debug eigrp fsm" on R3

*Jan 18 11:17:29.530: DUAL: rcvquery: 192.168.22.0/28 via 192.168.6.1 metric 4294967295/4294967295, RD is 33280

*Jan 18 11:17:29.530: DUAL: Find FS for dest 192.168.22.0/28. FD is 33280, RD is 33280

*Jan 18 11:17:29.530: DUAL: 192.168.6.1 metric 4294967295/4294967295 not found Dmin is 4294967295

*Jan 18 11:17:29.530: DUAL: Peer total/stub 1/0 template/full-stub 1/0

*Jan 18 11:17:29.530: DUAL: Dest 192.168.22.0/28 (Split Horizon) not entering active state.

*Jan 18 11:17:29.530: DUAL: send REPLY(r1/n1) about 192.168.22.0/28 to 192.168.6.1

*Jan 18 11:17:29.530: DUAL: dest(172.16.32.0/24) not active

*Jan 18 11:17:29.530: DUAL: rcvquery: 172.16.32.0/24 via 192.168.6.1 metric 4294967295/4294967295, RD is 4294967295

*Jan 18 11:17:29.530: DUAL: send REPLY(r1/n1) about 172.16.32.0/24 to 192.168.6.1

*Jan 18 11:17:29.530: DUAL: dest(172.16.31.0/24) not active

*Jan 18 11:17:29.530: DUAL: rcvquery: 172.16.31.0/24 via 192.168.6.1 metric 4294967295/4294967295, RD is 4294967295

*Jan 18 11:17:29.530: DUAL: send REPLY(r1/n1) about 172.16.31.0/24 to 192.168.6.1

*Jan 18 11:17:29.554: DUAL: Removing dest 192.168.22.0/28, nexthop 192.168.6.1

*Jan 18 11:17:29.554: DUAL: No routes. Flushing dest 192.168.22.0/28

*Jan 18 11:17:29.554: DUAL: Removing dest 172.16.32.0/24, nexthop 192.168.6.1

*Jan 18 11:17:29.554: DUAL: No routes. Flushing dest 172.16.32.0/24

*Jan 18 11:17:29.554: DUAL: Removing dest 172.16.31.0/24, nexthop 192.168.6.1

*Jan 18 11:17:29.554: DUAL: No routes. Flushing dest 172.16.31.0/24

As you can see R2 still sends queries for 172.16.31.0/24 and 172.16.32.0/24 even though it is filtering these from R1.

HTH

Jon

Konstantin Dunaev Fri, 01/18/2008 - 04:54

hi,

thank you :)

*Jan 18 11:17:29.530: DUAL: Dest 192.168.22.0/28 (Split Horizon) not entering active state.

*Jan 18 11:17:29.530: DUAL: dest(172.16.32.0/24) not active

*Jan 18 11:17:29.530: DUAL: dest(172.16.31.0/24) not active

but do you know why the status is different, for 192.168.22.0/28 is "not entering the active mode" and for other is simply "not active"?

Jon Marshall Fri, 01/18/2008 - 04:57

Hi

R3 only has R2 as an EIGRP neighbor. I suspect that the entry for 192.168.22.0/28 is different because R3 already had a route for that from R2 so it cannot query back to R2 for that route.

R3 did not have routes for the other 2 as they were being filtered from R2.

Jon

Actions

This Discussion