Redistribution Question

amb1s1

If I have a 2 BR with three protocols, is the proper way to distribute is to do redistribute for the three protocols. To understand my question you can click this link
http://www.routeralley.com/ra/labs/redistribution_lab.pdf. Like you see in the middle Router 5 and 8 are BRs and they have 3 protocols to redistribute RIP, OSPF and EIGRP.

tech-airman

amb1s1 wrote:

If I have a 2 BR with three protocols, is the proper way to distribute is to do redistribute for the three protocols. To understand my question you can click this link
http://www.routeralley.com/ra/labs/redistribution_lab.pdf. Like you see in the middle Router 5 and 8 are BRs and they have 3 protocols to redistribute RIP, OSPF and EIGRP.

amb1s1,

What does task 5 on pdf page 2 of 6 say?

amb1s1

Tech, I'm not going by what the task ask, I just want to know the proper way.

gojericho0

If you want the entire network to talk you would have to distribute all 3 routing protocols. EIGRP in one direction, OSPF and RIP the other direction

tech-airman

amb1s1 wrote:

Tech, I'm not going by what the task ask, I just want to know the proper way.

amb1s1,

The proper way is the required way according to the design document. Any deviation from that design document is declared an improper way. For example, if the design document stated to redistribute RIP into EIGRP, but you decide instead to redistribute EIGRP into RIP, then by defintion, what you did was the improper way. To put it into perspective, if an exam question asked you to do redistribution a certain way, if you do it another way (because someone on techexams.net said this is the proper way) then you're going to get that question graded wrong which may or may not be a factor if you pass that exam or not. So the point I'm trying to make is that what the task states IS the proper way.

amb1s1

Hi, Tech I understand what you are saying, but if you an engineer and you are the one designing and you are the one to choose the way to do, what would you choose. I'm not doing that toppology for taking a test, but just to understand the right way to do it.

amb1s1

How can I accomplish task 7.
7. Ensure that neither route feedback nor routing loops occur after
redistribution is complete.

scheistermeister

amb1s1 wrote:

How can I accomplish task 7.
7. Ensure that neither route feedback nor routing loops occur after
redistribution is complete.

Use route-maps to change the redistributed administrative distance.

amb1s1

After I do a route-map, how do I know if I'm getting feedback. Is there a Sh command for that or a way tyo know. Thanks

amb1s1

scheistermeister wrote:

amb1s1 wrote:

How can I accomplish task 7.
7. Ensure that neither route feedback nor routing loops occur after
redistribution is complete.

Use route-maps to change the redistributed administrative distance.

You are saying to change the AD, but I thought the router knows which protocol is better to route. Like I have RIP and OSPF, I would think the router choose OSPF instead of RIP.

kpjungle

amb1s1 wrote:

scheistermeister wrote:

amb1s1 wrote:

How can I accomplish task 7.
7. Ensure that neither route feedback nor routing loops occur after
redistribution is complete.

Use route-maps to change the redistributed administrative distance.

You are saying to change the AD, but I thought the router knows which protocol is better to route. Like I have RIP and OSPF, I would think the router choose OSPF instead of RIP.

Yes, but thats not always the best choice. When multiple boundary routes are involved, they will learn from the routing protocol with the lowest AD, which might be a suboptimal path, if the local boundary router has learned a better path from a routing protocol with a higher AD, you might want to artificially set the administrative distance with route-maps, or the distance command with a proper ACL attached.

amb1s1

How do I confirm if I'm doing the route-map right?

kpjungle

amb1s1 wrote:

How do I confirm if I'm doing the route-map right?

If you are changing the administrative distance with route-maps, you can check if that new distance takes effect on your selected routes.

amb1s1

kpjungle wrote:

amb1s1 wrote:

How do I confirm if I'm doing the route-map right?

If you are changing the administrative distance with route-maps, you can check if that new distance takes effect on your selected routes.

, But How?

scheistermeister

amb1s1 wrote:

kpjungle wrote:

amb1s1 wrote:

How do I confirm if I'm doing the route-map right?

If you are changing the administrative distance with route-maps, you can check if that new distance takes effect on your selected routes.

, But How?

Show ip route will tell you.

amb1s1

Finally I got it. Thanks guys for help. Now I have another question on Question 5.
"5. IGP updates should not be forwarded out inappropriate interfaces. For
Example, Router 5 should neither send nor accept RIP or OSPF updates
out its Ethernet0 interface." I did a Passive-interface on all protocols under each BR. I just did a debug EIGRP on Router 5 with the passive interface on and I didn't eigrp hello send to any of the RIP or OSPF router, but I went a disable all the Passive interface and I still don't hello been send to RIP or OSPF. Can anyone explain me what is the purpose of the passive-interface in this topology? Thanks

tech-airman

amb1s1 wrote:

Finally I got it. Thanks guys for help. Now I have another question on Question 5.
"5. IGP updates should not be forwarded out inappropriate interfaces. For
Example, Router 5 should neither send nor accept RIP or OSPF updates
out its Ethernet0 interface." I did a Passive-interface on all protocols under each BR. I just did a debug EIGRP on Router 5 with the passive interface on and I didn't eigrp hello send to any of the RIP or OSPF router, but I went a disable all the Passive interface and I still don't hello been send to RIP or OSPF. Can anyone explain me what is the purpose of the passive-interface in this topology? Thanks

amb1s1,

On R5, which routing protocols are running on it?

nicklauscombs

im going to set this lab up this weekend hopefully

amb1s1

tech-airman wrote:

amb1s1 wrote:

Finally I got it. Thanks guys for help. Now I have another question on Question 5.
"5. IGP updates should not be forwarded out inappropriate interfaces. For
Example, Router 5 should neither send nor accept RIP or OSPF updates
out its Ethernet0 interface." I did a Passive-interface on all protocols under each BR. I just did a debug EIGRP on Router 5 with the passive interface on and I didn't eigrp hello send to any of the RIP or OSPF router, but I went a disable all the Passive interface and I still don't hello been send to RIP or OSPF. Can anyone explain me what is the purpose of the passive-interface in this topology? Thanks

amb1s1,

On R5, which routing protocols are running on it?

Is running Rip EIGRP and OSPF and Im doing redistribution to all the protocols.

gojericho0

amb1s1 wrote:

Finally I got it. Thanks guys for help. Now I have another question on Question 5.
"5. IGP updates should not be forwarded out inappropriate interfaces. For
Example, Router 5 should neither send nor accept RIP or OSPF updates
out its Ethernet0 interface." I did a Passive-interface on all protocols under each BR. I just did a debug EIGRP on Router 5 with the passive interface on and I didn't eigrp hello send to any of the RIP or OSPF router, but I went a disable all the Passive interface and I still don't hello been send to RIP or OSPF. Can anyone explain me what is the purpose of the passive-interface in this topology? Thanks

Copy and paste sh ip protocols on this guy

amb1s1

gojericho0 wrote:

amb1s1 wrote:

Finally I got it. Thanks guys for help. Now I have another question on Question 5.
"5. IGP updates should not be forwarded out inappropriate interfaces. For
Example, Router 5 should neither send nor accept RIP or OSPF updates
out its Ethernet0 interface." I did a Passive-interface on all protocols under each BR. I just did a debug EIGRP on Router 5 with the passive interface on and I didn't eigrp hello send to any of the RIP or OSPF router, but I went a disable all the Passive interface and I still don't hello been send to RIP or OSPF. Can anyone explain me what is the purpose of the passive-interface in this topology? Thanks

Copy and paste sh ip protocols on this guy

My lab is little be different because only used 10 router so instead of router 5 on my topology ther router is 4. The following is the ip route on router 5 (4 on my lab):

Router_4#sh ip route
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
       D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area 
       N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
       E1 - OSPF external type 1, E2 - OSPF external type 2
       i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
       ia - IS-IS inter area, * - candidate default, U - per-user static route
       o - ODR, P - periodic downloaded static route

Gateway of last resort is not set

O E2 192.168.89.0/24 [110/20] via 192.168.46.2, 00:04:54, Serial0/1
C    192.168.46.0/24 is directly connected, Serial0/1
R    192.168.12.0/24 [120/4] via 192.168.45.2, 00:00:03, Serial0/0
O E2 192.168.90.0/24 [110/20] via 192.168.46.2, 00:04:45, Serial0/1
C    192.168.45.0/24 is directly connected, Serial0/0
     4.0.0.0/24 is subnetted, 1 subnets
C       4.4.4.0 is directly connected, Loopback0
R    192.168.57.0/24 [70/1] via 192.168.45.2, 00:00:03, Serial0/0
O E2 192.168.78.0/24 [110/20] via 192.168.46.2, 00:05:49, Serial0/1
D    192.168.23.0/24 [90/2172416] via 192.168.34.1, 00:05:44, FastEthernet1/0
O    192.168.67.0/24 [110/128] via 192.168.46.2, 00:05:49, Serial0/1
C    192.168.34.0/24 is directly connected, FastEthernet1/0

gojericho0

ok passive interface normally blocks the sending of updates from an interface. The interface can normally still form neighbors and accept incoming routes.

The exception is EIGRP. If using a passive interface with this protocols neighbors will not form, and consequently you will not be able to send or receive updates.

If you want the same effect as passive interface on other protocols create a distribute list that blocks any network going out

amb1s1

gojericho0 wrote:

ok passive interface normally blocks the sending of updates from an interface. The interface can normally still form neighbors and accept incoming routes.

The exception is EIGRP. If using a passive interface with this protocols neighbors will not form, and consequently you will not be able to send or receive updates.

If you want the same effect as passive interface on other protocols create a distribute list that blocks any network going out

there is command to check if the update are going thru an interface. I would like to know the difference before and after I add the passive-interface.

tech-airman

gojericho0 wrote:

ok passive interface normally blocks the sending of updates from an interface. The interface can normally still form neighbors and accept incoming routes.

The exception is EIGRP. If using a passive interface with this protocols neighbors will not form, and consequently you will not be able to send or receive updates.

If you want the same effect as passive interface on other protocols create a distribute list that blocks any network going out

gojericho0,

Are you sure about the above?

scheistermeister

tech-airman wrote:

gojericho0 wrote:

ok passive interface normally blocks the sending of updates from an interface. The interface can normally still form neighbors and accept incoming routes.

The exception is EIGRP. If using a passive interface with this protocols neighbors will not form, and consequently you will not be able to send or receive updates.

If you want the same effect as passive interface on other protocols create a distribute list that blocks any network going out

gojericho0,

Are you sure about the above?

He is correct, you use the eigrp stub receive only [or the other parameters] to make an eigrp interface act like a passive-interface.

amb1s1

scheistermeister wrote:

tech-airman wrote:

gojericho0 wrote:

ok passive interface normally blocks the sending of updates from an interface. The interface can normally still form neighbors and accept incoming routes.

The exception is EIGRP. If using a passive interface with this protocols neighbors will not form, and consequently you will not be able to send or receive updates.

If you want the same effect as passive interface on other protocols create a distribute list that blocks any network going out

gojericho0,

Are you sure about the above?

He is correct, you use the eigrp stub receive only [or the other parameters] to make an eigrp interface act like a passive-interface.

Ok, I think you are right, but what command should I use to confirm that those update aren't passing to the other protocols.

scheistermeister

You can use the debugs commands to see what is being sent where and you can also check the routing tables on the routers that should not be receiving the routes to see if it has the routes it shouldn't in it's databases.

aaronbalchunas

Howdy,

I am the author of the lab in question. If I can be of any help, or answer any questions, please let me know. That's one of the more difficult labs I've inflicted on my students, though they seem to enjoy it. It takes them, on average, 6 to 8 hours to fully complete it and eliminate all loops/feedback.

The question was asked, how do you know if route feedback is occuring? Traceroute is your friend. The loops/suboptimal paths should be quite apparent. You should also see several unreachable routes.

Adjusting administrative distances will correct some, but not all of the loops. Using route-tags with route-maps, in conjunction with fiddling with AD's, is a more robust solution.

Ah, the joys of ridiculously overcomplicated labs.

- Aaron Balchunas
http://www.routeralley.com

gojericho0

tech-airman wrote:

gojericho0 wrote:

ok passive interface normally blocks the sending of updates from an interface. The interface can normally still form neighbors and accept incoming routes.

The exception is EIGRP. If using a passive interface with this protocols neighbors will not form, and consequently you will not be able to send or receive updates.

If you want the same effect as passive interface on other protocols create a distribute list that blocks any network going out

gojericho0,

Are you sure about the above?

Yup

http://www.cisco.com/en/US/tech/tk365/technologies_q_and_a_item09186a008012dac4.shtml#eleven

gojericho0

aaronbalchunas wrote:

Howdy,

I am the author of the lab in question. If I can be of any help, or answer any questions, please let me know. That's one of the more difficult labs I've inflicted on my students, though they seem to enjoy it. It takes them, on average, 6 to 8 hours to fully complete it and eliminate all loops/feedback.

The question was asked, how do you know if route feedback is occuring? Traceroute is your friend. The loops/suboptimal paths should be quite apparent. You should also see several unreachable routes.

Adjusting administrative distances will correct some, but not all of the loops. Using route-tags with route-maps, in conjunction with fiddling with AD's, is a more robust solution.

Ah, the joys of ridiculously overcomplicated labs.

- Aaron Balchunas
http://www.routeralley.com

Hey Aaron,

thanks for registering with our community. as you see we are having quite the discussion about one of your labs and learning a lot. it would be great if you drop by this site every now and then and help us with some of our questions and maybe even throw us a couple of curve balls

kpjungle

gojericho0 wrote:

aaronbalchunas wrote:

Howdy,

I am the author of the lab in question. If I can be of any help, or answer any questions, please let me know. That's one of the more difficult labs I've inflicted on my students, though they seem to enjoy it. It takes them, on average, 6 to 8 hours to fully complete it and eliminate all loops/feedback.

The question was asked, how do you know if route feedback is occuring? Traceroute is your friend. The loops/suboptimal paths should be quite apparent. You should also see several unreachable routes.

Adjusting administrative distances will correct some, but not all of the loops. Using route-tags with route-maps, in conjunction with fiddling with AD's, is a more robust solution.

Ah, the joys of ridiculously overcomplicated labs.

- Aaron Balchunas
http://www.routeralley.com

Hey Aaron,

thanks for registering with our community. as you see we are having quite the discussion about one of your labs and learning a lot. it would be great if you drop by this site every now and then and help us with some of our questions and maybe even throw us a couple of curve balls

But please limit the amount of routers, my PC doesnt have enough ram to run them all in dynamips