some configuration questions and the SIMS

sleemie

The only source for testing the sim questions I have is the exsim program and it raises some questions on some of the configuration and i'm wondering if the real test is as sketchy on the info they give you as the exsims.

1. when configuring some of the access lists it shows you to apply it to the interface before defining it, and in other cases it applies it to the interface after defining the list. I thought you were always supposed to apply it to the interface after it's defined.

2. when entering a line for blocking or granting access via telnet, instead of eq 23 at the end, it uses eq telnet. I thought it was always supposed to be eq 23, i've never seen any other place teach you to use the word telnet.

3. it gave me a question that told me to configure a routing protocol on routerA for the subnetwork on the ethernet interface of routerB, but you couldn't access routerB to check the IP address. maybe i'm missing the piece of knowledge of how you determine that from routerA. I tried show cdp neigbhors and with the detail, but neither gave me the info. when I did the "show me" in the sim to show me how it's done, they came up with an ip address for the interface on routerb...not sure how I was supposed to find it.

4. another scenario where it told you to configure eigrp between routera and routerb, and it tells you the ip addresses are configured. I did a show ip interface brief on both router a and b and what I did was on router a entered the command "nework (network addresses on router b)" and visa-versa...but when i told it to show me how it's done, on routera it entered the network addresses of the routera interfaces, and the same with router b. again, is that something I'm not understanding correctly? I thought you entered the network addresses of connected router, not the network addresses of the interfaces on the same router.

Humper

1. I would define the ACL first, then write the command to apply it to the interface, in case you made an error in the ACL and you could block all traffic on the interface.

2.You can use telnet or port 23..Word/port it doesnt matter.


3. The question should give you the subnet for Router B's e0 interface.

4. I think you are getting static routes and dynamic routes confused. When type a static route you are specifying the destination network you want your router to forward/send traffic too. Example:

ip route <destination network> <subnet mask> <exit interface|next hop ip address>

ip route 10.0.0.0 255.0.0.0 10.0.0.1
or
ip route 0.0.0.0 0.0.0.0 f0/0 (this is often referred as a defaulte-route)

Now with EIGRP,RIP,OSPF,IGRP..etc you are advertising your directly connected routes to the OTHER routers that are connected to you.

Example:
Your router has a serial port and an ethernet port.
E0: 192.168.1.0/24
S0:192.168.2.0/24

With EIGRP you are advertising these routes to neighbours:

router eigrp 1
network 192.168.1.0 0.0.0.255 (this is a wildcard mask)
network 192.168.2.0 0.0.0.255

Do you see the difference? I can try and make myself more clear if you need me to be.

sleemie

okay, I see...but with RIP, aren't you supposed to use the network address of the directly connected router in the "network..." command?

and you don't use a wildcard mask with IGRP, do you? I thought you only use the wildcard mask with OSPF, but not when configuring the other routing protocols??

Humper

sleemie wrote:

okay, I see...but with RIP, aren't you supposed to use the network address of the directly connected router in the "network..." command?

and you don't use a wildcard mask with IGRP, do you? I thought you only use the wildcard mask with OSPF, but not when configuring the other routing protocols??

No with RIP it is the same as the rest of the dynamic routing protocols, but with RIP,RIP2, and IGRP you are specifying the subnet at classful boundaries, as opposed to OSPF & EIGRP where its specified at classless boundaries.

lwwarner

1. As ModemH said, you typically define the acl first, then triple check it, then check it again, then apply it. Hoever, IOS doesn't care what order you do it in.

2. Here are more protocol names you can use:

R1(config)#access-list 101 permit tcp any any eq ?
  <0-65535>    Port number
  bgp          Border Gateway Protocol (179)
  chargen      Character generator (19)
  cmd          Remote commands (rcmd, 514)
  daytime      Daytime (13)
  discard      Discard (9)
  domain       Domain Name Service (53)
  echo         Echo (7)
  exec         Exec (rsh, 512)
  finger       Finger (79)
  ftp          File Transfer Protocol (21)
  ftp-data     FTP data connections (20)
  gopher       Gopher (70)
  hostname     NIC hostname server (101)
  ident        Ident Protocol (113)
  irc          Internet Relay Chat (194)
  klogin       Kerberos login (543)
  kshell       Kerberos shell (544)
  login        Login (rlogin, 513)
  lpd          Printer service (515)
  nntp         Network News Transport Protocol (119)
  pim-auto-rp  PIM Auto-RP (496)
  pop2         Post Office Protocol v2 (109)
  pop3         Post Office Protocol v3 (110)
  smtp         Simple Mail Transport Protocol (25)
  sunrpc       Sun Remote Procedure Call (111)
  syslog       Syslog (514)
  tacacs       TAC Access Control System (49)
  talk         Talk (517)
  telnet       Telnet (23)
  time         Time (37)
  uucp         Unix-to-Unix Copy Program (540)
  whois        Nicname (43)
  www          World Wide Web (HTTP, 80)

3. You don't need the IP address of the routerB ethernet interface. You just need the subnet & mask.

4. You seem to have a fundamental misunderstanding of what routing protocols do & how to configure them. You use config-router network statements to specify the networks that you want your router to advertise to it's neighbors. I suggest you go back and carefully reread the appropriate chapters in whatever books you are using...

sleemie

maybe I am missing something.... What I said was that when using the network statement that you should enter the network number of the connected networks, but by me making that statement you said I must not understand how the routing protocols work. Or was it another statement that made you say that. But, in all of my Boson labs it tells you to enter the network address of the the connected devices, and in my book here's a statement about RIP..

"Note the fact the you need to type in every directly connected network you want RIP to advertise."

Here's one about EIGRP

"You then enter the network numbers connected to the router using the network command followed by the network number."

What am I missing? What you're doing is telling the routers to advertise the networks that you enter using the network command, and the networks you enter using that command are the networks of the directly connected networks.

And all of this started because some of my exsim simulation questions didn't tell me what the IP address of the of the interfaces of the connected devices were so I had no way of knowing which network to enter under the network statement because I didn't know the connected networks.

What about the wildcard masks? humper used an EIGRP example with a wildcard mask, but my book tells me you only need the wildcard mask for OSPF, not the other protocols. what's the story on that?

lwwarner

sleemie wrote:

What am I missing? What you're doing is telling the routers to advertise the networks that you enter using the network command, and the networks you enter using that command are the networks of the directly connected networks.

Yes. That is right, but then you say:

And all of this started because some of my exsim simulation questions didn't tell me what the IP address of the of the interfaces of the connected devices were so I had no way of knowing which network to enter under the network statement because I didn't know the connected networks.

This is the problem. You did have all the info you needed. When you configure a routing protocol on router X, you do not need any information from any other routers. Everything you need to know is already known in the router X config. In a way, this is the whole point of using a dynamic routing protocol. Initially, each router only knows it's own configuration, which it shares (advertises) with it's neighbors, and vice-versa.

I may be misreading, or reading too much into your questions, but it just seems to me that if you really understood the underlying concepts here (the how & why of the routing protocols themselves) that you wouldn't be having this problem understanding the configuration issues.

What about the wildcard masks? humper used an EIGRP example with a wildcard mask, but my book tells me you only need the wildcard mask for OSPF, not the other protocols. what's the story on that?

This is the kind of factual question that is easily answered with just a little digging. Try this:
http://www.google.com/search?q=cisco+eigrp+network+command+mask

Basically, the mask is optional. If you don't specify a mask eigrp will assume the default class mask. That point is important to remmember if you move beyond very simple configs. The mask can be entered using either subnet notation or wildcard notation. If you enter the mask using subnet notation, IOS will convert it to wildcard notation for you. For example, if you enter:

Router(config-router)#net 172.17.1.0 255.255.255.252

This will be converted to:

network 172.17.1.0 0.0.0.3

The interesting part is when do you want or need to use a mask with your eigrp network command? Try this senario[*]. Show the eigrp config for a router with the following interface configs. Do not enable eigrp on int s0. Anyone?

s0: 192.168.1.249/30
s1: 192.168.1.253/30
e0: 192.168.1.1/25
e1: 192.168.1.129/27
e2: 192.168.1.161/27
e3: 192.168.2.1/23

[*] This is probably pushing beyond the CCNA envelope a bit, but this is where things start to get interesting!

sleemie

i admit i am a little sketchy on some of it...this is what I understand. The purpose of the routing protocols is to determine the best path to the various networks to which a router is directly or indirectly connected. This is done by building and maintaining a routing table, which contains the best path to each network.
Each protocol uses a different method of determining what it considers the best path - a metric. The distance vector ones only consider the hop count, EIGRP uses bandwidth and delay as the default but are capable of using 4 and I can't remember the other 2...I think it's load and MTU....and not sure what OSPF uses off the top of my head. The routing protocols share their routing table information with the other routers..each one doing it differently..some sending the entire table, some only sending the updates to the table and OSPF using a designated router for that purpose so that there's only one source for keeping the tables up to date.

That's the basics of it, of course there's more such as each protocol's administrative distance and the router choosing the one with the lowest value, with EIGRP having the lowest(90) and RIP having the highest(120)...unless you count the EIGRP's external routes as having the highest(170). And of course the interface is the lowest at O and then a static the next lowest at 1. OSPF is 110 and IGRP is 100.

So..what would you say I'm missing? Just in terms of understanding how the protocols work, not so much the details of each one.

This is what I know off the top of my head without looking at the book or anything.

i was thinking about the whole thing with thinking I needed to know the ip address and mask of the connected router in order to set up the protocol, but can't you simply use that information from the router your setting up because the network has to be the same for that router and the one that it's directly connect to, right?

lwwarner

Actually, it sounds like your getting a pretty good handle on things. I'll just make a couple of comments.

sleemie wrote:

Each protocol uses a different method of determining what it considers the best path - a metric. The distance vector ones only consider the hop count,

Not necessarily. A DV metric can be just about anything. RIP happens to use hop count, but IGRP uses the same composite metric as EIGRP.

EIGRP uses bandwidth and delay as the default but are capable of using 4 and I can't remember the other 2...I think it's load and MTU....and not sure what OSPF uses off the top of my head.

The (E)IGRP composite metric is based on bandwidth, delay, reliability & load. For Cisco, the default OSPF metric is 10^8/bandwidth. Note that the choice of metric is not defined in the OSPF spec, and other vendors use different metrics.

i was thinking about the whole thing with thinking I needed to know the ip address and mask of the connected router in order to set up the protocol, but can't you simply use that information from the router your setting up...

Yes! That is the key to your original config questions.