Mastering Cisco Routing: When to Use Multiple Protocols, Access Lists, and Redistribution

1. When to Use Multiple Routing Protocols

Multiple routing protocols are required when two different protocols must exchange routing information; route redistribution can also achieve this. Situations where multi‑protocol use is unnecessary include upgrading an IGP version, retaining an old protocol while adding a new one, terminating internal routing to avoid interference, or operating in a multi‑vendor environment.

2. Distance‑Vector vs. Link‑State Routing Protocols

Distance‑vector protocols (e.g., RIP, IPX RIP, AppleTalk RTMP, IGRP) are suited for small networks, use hop count as metric, and generate periodic full‑table updates (every 30 seconds). RIP limits hops to 15. Link‑state protocols (e.g., OSPF, IPX NLSP, IS‑IS) are better for large networks, consume more CPU, send hello packets every 10 seconds, and calculate metrics based on link bandwidth, allowing virtually unlimited cost values.

3. Access Lists

Access lists are administrator‑defined rule sets that control packet ingress and egress. They are processed sequentially; once a packet matches a rule, no further rules are evaluated. Each interface can have only one access list per protocol.

Supported Types

IP standard access‑list numbers: 1‑99

IP extended access‑list numbers: 100‑199

IPX standard access‑list numbers: 800‑899

IPX extended access‑list numbers: 1000‑1099

AppleTalk access‑list numbers: 600‑699

Creating an IP Standard Access List

access-list 10 permit 192.168.1.0 0.0.0.255

Key fields: access-list number: identifier (1‑99 for standard) permit | deny: action source: source IP address source‑mask: wildcard mask (bits set to 0 are ignored)

Example output from show access-lists:

Standard IP access list 1
 deny 204.59.144.0 0.0.0.255
 permit any

4. Route Redistribution

Redistribution is used on routers that learn routes from one autonomous system (AS) and need to advertise them into another AS. It is often automatic for IGRP/EIGRP.

Preparation Steps

Decide where the new protocol will be added.

Identify the autonomous‑system border router (ASBR).

Determine which protocol runs in the core and which at the edge.

Choose the redistribution direction (inbound or outbound).

Example Command (OSPF redistribution)

router(config-router)#redistribute protocol [process-id] [metric metric‑value] [metric-type type‑value] [subnets]

Parameters: protocol: source protocol (bgp, eigrp, isis, ospf, static, connected, rip, etc.) process-id: OSPF process identifier metric: optional metric for the redistributed routes (default 0)

5. Administrative Distance

Administrative distance (AD) ranks routing protocols by trustworthiness; lower AD means higher preference. When two protocols provide routes to the same destination, the router selects the route with the lower AD.

6. Importance of Neighbor Detection

In small networks, neighbor loss is less critical because other routers can converge quickly. In large networks, rapid neighbor detection (e.g., hello packets every 10 seconds for link‑state protocols) speeds convergence and reduces outage time.

7. BGP Basics

BGP (Border Gateway Protocol) exchanges routing information between autonomous systems. It supports two session types:

External BGP (eBGP): sessions between routers in different ASes.

Internal BGP (iBGP): sessions within the same AS, used to synchronize routing.

BGP advertises the full routing table initially, then only incremental updates. Keepalive messages maintain session health; notifications signal errors.

8. Metrics and Their Modification

Metrics quantify route cost. Common metric types include hop count, bandwidth, delay, reliability, load, MTU, etc. IGRP uses a composite metric consisting of bandwidth, delay, reliability, load, and MTU.

Metrics can be adjusted with offset-list commands. Example to add a delay offset of 10 on all outgoing routes: offset-list out 10 To apply the same offset to access‑list 121:

offset-list out 10 121

9. Information Every Router Needs to Find a Route

Destination address

Neighbor identification

Route discovery (knowledge of reachable networks)

Route selection based on metrics

Route maintenance (routing table storage)

10. Compatibility of Cisco Routing Protocols

Except for proprietary IGRP/EIGRP, Cisco routers implement standard routing protocols that are compatible with other vendors' implementations.

11. Cisco 3600 Series Hardware Support

The 3600 series supports WIC‑2T and WIC‑2A/S WAN interface cards on IOS 12.007XK or later, provided a Fast‑Ethernet hybrid network module is present. Supported modules include NM‑1FE2W, NM‑2FE2W, NM‑1FE1R2W, and NM‑2W; modules such as NM‑1E2W, NM‑2E2W, NM‑1E1R2W do not support these cards.

Maximum asynchronous and synchronous rates for supported NM and WIC cards are 115.2 kbps and 128 kbps, respectively.

12. Switch Stacking and EtherChannel

Catalyst 3550XL/2950XL stacking uses dedicated stacking cables (1 m or 0.5 m) and GigaStack GBIC cards. Two stacking methods are available: daisy‑chain (up to 9 switches, 1 Gbps) and point‑to‑point (up to 8 switches, 2 Gbps), both supporting redundancy.

EtherChannel aggregates 2 or 4 ports (10/100 Mbps or 1 Gbps) to increase bandwidth (up to 4 Gbps for 1 Gbps ports) and provides load balancing and link backup. Selection of the egress link is based on a hash of source/destination MAC address bits.

PAgP is an enhanced EtherChannel protocol that supports Spanning‑Tree, PortFast, and automatic channel configuration.

13. Catalyst 6000 Series Details

Backplane bandwidth: up to 32 Gbps; packet‑forwarding rate: up to 15 Mpps. Supervisor engines (e.g., WS‑X6K‑SUP1‑2GE, WS‑X6K‑SUP1A‑2GE) provide two Gigabit ports (GBIC required) and optional MSFC/PFC modules. Redundant engines must match the primary engine’s feature set.

The series supports routing protocols OSPF, IGRP, EIGRP, BGP4, IS‑IS, RIP, and RIP‑II, as well as non‑IP protocols NLSP, IPX RIP/SAP, IPX EIGRP, RTMP, AppleTalk EIGRP, and DECnet.

Three‑layer switching is achieved via an MSFC‑equipped engine; the MSFC cannot be ordered separately.

Differences between Catalyst 6000 and 6500: the 6000’s backplane is 32 Gbps versus the 6500’s up to 256 Gbps, making direct hardware upgrades impossible despite using the same line cards.