Multilayer Campus Architecture and Design Principles 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 1
Enterprise-Class Availability Resilient Campus Communication Fabric Campus Systems Approach to High Availability Network-level redundancy System-level resiliency Enhanced management Human ear notices the difference in voice within 150 200 msec 10 consecutive G711 packet loss Video loss is even more noticeable 200 msec end-to end-campus convergence Ultimate Goal..100% Next-Generation Apps Video conf., Unified Messaging, Global Outsourcing, E-Business Business, Wireless Ubiquity Mission Critical Apps. Databases, Order-Entry, CRM, ERP Desktop Apps E-mail, File & Print APPLICATIONS DRIVE REQUIREMENTS FOR HIGH AVAILABILITY NETWORKING 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 2
Next Generation Campus Design Unified Communications Evolution VoIP is now a mainstream technology Ongoing evolution to the full spectrum of Unified Communications High-Definition Executive Communication Application requires stringent Service-Level Agreement (SLA) Reliable Service High Availability Infrastructure Application Service Management QoS 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 3
Agenda Multilayer Campus Design Principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 4
High-Availability Campus Design Structure, Modularity, and Hierarchy Access Distribution Core Distribution S Access WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 5
Hierarchical Campus Network Structure, Modularity and Hierarchy Not This!! Server Farm WAN Internet PSTN 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 6
Hierarchical Network Design Without a Rock Solid Foundation the Rest Doesn t Matter Access Distribution Core Offers hierarchy each layer has specific role Modular topology building blocks Easy to grow, understand, and troubleshoot Creates small fault domains clear demarcations and isolation Promotes load balancing and redundancy Promotes deterministic traffic patterns Distribution Incorporates balance of both Layer 2 and Layer 3 technology, leveraging the strength of both Utilizes Layer 3 routing for load Access balancing, fast convergence, scalability, and control Building Block 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 7
Access Layer Feature Rich Environment It s not just about connectivity Layer 2/Layer 3 feature rich environment; convergence, HA, security, QoS, IP multicast, etc. Intelligent network services: QoS, trust boundary, broadcast suppression, IGMP snooping Core Intelligent network services: PVST+, Rapid PVST+, EIGRP, OSPF, DTP, PAgP/LACP, UDLD, FlexLink, etc. Cisco Catalyst integrated security features IBNS (802.1x), (CISF): port security, DHCP snooping, DAI, IPSG, etc. Distribution Automatic phone discovery, conditional trust boundary, power over Ethernet, auxiliary VLAN, etc. Access Spanning tree toolkit: PortFast, UplinkFast, BackboneFast, LoopGuard, BPDU Guard, BPDU Filter, RootGuard, etc. 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 8
Distribution Layer Policy, Convergence, QoS, and High Availability Availability, load balancing, QoS and provisioning are the important considerations at this layer Aggregates wiring closets (access layer) and uplinks to core Protects core from high density peering and problems in access layer Route summarization, fast convergence, redundant path load sharing HSRP or GLBP to provide first hop redundancy Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 9
Core Layer Scalability, High Availability, and Fast Convergence Backbone for the network connects network building blocks Performance and stability vs. complexity less is more in the core Aggregation point for distribution layer Separate core layer helps in scalability during future growth Keep the design technology-independent Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 10
Do I Need a Core Layer? It s Really a Question of Scale, Complexity, and Convergence No Core Fully meshed distribution layers Physical cabling requirement Routing complexity Second Building Block 4 New Links 4th Building Block 12 New Links 24 Links Total 8 IGP Neighbors 3rd Building Block 8 New Links 12 Links Total 5 IGP Neighbors 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 11
Do I Need a Core Layer? It s Really a Question of Scale, Complexity, and Convergence Dedicated Core Switches Easier to add a module Fewer links in the core Easier bandwidth upgrade Routing protocol peering reduced Equal cost Layer 3 links for best convergence 2nd Building Block 8 New Links 4th Building Block 4 New Links 16 Links Total 3 IGP Neighbors 3rd Building Block 4 New Links 12 Links Total 3 IGP Neighbors 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 12
Design Alternatives Come Within a Building (or Distribution) Block Access Layer 2 Access Routed Access Virtual Switching System Distribution Core Distribution S Access WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 13
Layer 3 Distribution Interconnection Layer 2 Access No VLANs Span Access Layer Tune CEF load balancing Match CatOS/IOS EtherChannel settings and tune load balancing Summarize routes towards core Limit redundant IGP peering STP Root and HSRP primary tuning or GLBP to load balance Layer 3 on uplinks Point to Set trunk mode on/no-negotiate Point Link Disable EtherChannel unless needed Set port host on access layer ports: Disable Trunking Disable EtherChannel Enable PortFast RootGuard or BPDU-Guard Use security features VLAN 20 Data 10.1.20.0/24 VLAN 120 Voice 10.1.120.0/24 VLAN 40 Data 10.1.40.0/24 VLAN 140 Voice 10.1.140.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 14
Layer 2 Distribution Interconnection Layer 2 Access Some VLANs Span Access Layer Tune CEF load balancing Match CatOS/IOS EtherChannel settings and tune load balancing Summarize routes towards core Limit redundant IGP peering STP Root and HSRP primary or GLBP and STP port cost tuning to load balance on uplinks Set trunk mode on/no-negotiate Disable EtherChannel unless needed RootGuard on downlinks LoopGuard on uplinks Set port host on access Layer ports: Disable Trunking Disable EtherChannel Enable PortFast RootGuard or BPDU-Guard Use security features VLAN 20 Data 10.1.20.0/24 Layer 2 Trunk VLAN 120 Voice VLAN 140 Voice 10.1.120.0/24 10.1.140.0/24 VLAN 250 WLAN 10.1.250.0/24 VLAN 40 Data 10.1.40.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 15
Routed Access and Virtual Switching System Evolutions of and Improvements to Existing Designs Core Layer 3 VSS Link New Concept Distribution P-t-P Link VLAN 20 Data 10.1.20.0/24 VLAN 120 Voice 10.1.120.0/24 VLAN 40 Data 10.1.40.0/24 VLAN 140 Voice 10.1.140.0/24 VLAN 20 Data VLAN 10.1.20.0/24 40 Data VLAN 10.1.40.0/24 120 Voice VLAN 10.1.120.0/24 140 Voice VLAN 10.1.140.0/241 250 140 WLAN 0/24 10.1.250.0/24 Access See RST-3035 Advanced Enterprise Campus Design Alternatives: Routed Access and Virtual Switching System (VSS) 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 16
Virtual Switch System (VSS) Hub and Spoke VLANs can Span Access Layer Tune CEF load balancing Match CatOS/IOS EtherChannel settings and tune load balancing Summarize routes towards core Set trunk mode on/nonegotiate Use PaGP and Multi-Chassis EtherChannel RootGuard on downlink (MEC) LoopGuard on uplink (MEC) Set port host on access Layer ports: Disable trunking Disable EtherChannel Enable PortFast t RootGuard or BPDU-Guard on access ports Use security features VLAN 20 Data 10.1.20.0/24 VSS Link New Concept VLAN 40 Data 10.1.40.0/24 VLAN 120 Voice VLAN 140 Voice 10.1.120.0/241 120 0/24 10.1.140.0/241 140 0/24 VLAN 250 WLAN 10.1.250.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 17
Virtual Switching System 1440 Network System Virtualization Core/Distribution Data Center Access Features Network System Virtualization Benefits of VSS Increased Operational Efficiency via mplified Network Inter-Chassis Stateful Switch Over (SSO) Multi-Chassis EtherChannel (MEC) Boost Non-stop Communication Scale the System Bandwidth Capacity to 1.4 Tbps 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 18
Agenda Multilayer Campus Design Principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 19
Foundation Services Layer 1 physical things Layer 2 redundancy spanning tree Layer 3 routing protocols Trunking protocols (ISL/.1q) Unidirectional link detection Load balancing EtherChannel link aggregation CEF equal cost load balancing First hop redundancy protocols VRRP, HSRP, and GLBP Routing HSRP Spanning Tree 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 20
Best Practices Layer 1 Physical Things Use point-to-point to interconnections no L2 aggregation points between nodes Use fiber for best convergence (debounce timer) Tune carrier delay timer Layer 3 Equal Cost Links Layer 3 Equal Cost Links Use configuration on the physical interface not VLAN/SVI when possible WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 21
Redundancy and Protocol Interaction Link Neighbour Failure Detection ti Indirect link failures are harder to detect With no direct HW notification of link loss or topology change convergence times are dependent on SW notification Hellos Hub Indirect failure events in a bridged environment are detected by Spanning Tree Hellos In certain topologies the need for TCN updates or dummy multicast flooding (uplink fast) is necessary for convergence You should not be using hubs in a high availability design BPDUs Hub 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 22
Redundancy and Protocol Interaction Link Redundancy d and Failure Detection ti Direct point-to-point fiber provides for fast failure detection IEEE 802.3z and 802.3ae link negotiation define the use of Remote Fault Indicator and Link Fault gnaling mechanisms Bit D13 in the Fast Link Pulse (FLP) can be set to indicate a physical fault to the remote side Do not disable auto-negotiation on GigE and 10GigE interfaces 1 The default debounce timer on GigE and 10GigE fiber linecards is 10 msec The minimum debounce for copper is 300 msec Carrier-Delay 3560, 3750 and 4500 0 msec 6500 leave it set at default 3 Cisco IOS Throttling: Carrier Delay Timer 2 Linecard Throttling: Debounce Timer 1 Remote IEEE Fault Detection Mechanism 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 23
Redundancy and Protocol Interaction Layer 2 and 3 Why Use Routed Interfaces Configuring L3 routed interfaces provides for faster convergence than an L2 switch port with an associated L3 SVI L3 L2 1. Link Down 2. Interface Down 3. Routing Update ~ 8 msec ~ 150-200 loss msec loss 1. Link Down 2. Interface Down 3. Autostate 4. SVI Down 5. Routing Update 21:38:37.042 UTC: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet3/1, changed state to down 21:38:37.050 UTC: %LINK-3-UPDOWN: Interface GigabitEthernet3/1, changed state to down 21:38:37.050 UTC: IP-EIGRP(Default-IP-Routing- Table:100): Callback: route_adjust GigabitEthernet3/1 21:32:47.813 UTC: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet2/1, changed state to down 21:32:47.821 UTC: %LINK-3-UPDOWN: Interface GigabitEthernet2/1, changed state to down 21:32:48.069 UTC: %LINK-3-UPDOWN: Interface Vlan301, changed state to down 21:32:48.069 UTC: IP-EIGRP(Default-IP-Routing- Table:100): Callback: route, adjust Vlan301 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 24
Best Practices Spanning Tree Configuration Only span VLAN across multiple access layer switches when you have to! Use Rapid PVST+ for best convergence More common in the data center Required to protect against user side loops Required to protect against operational accidents (misconfiguration or hardware failure) Take advantage of the spanning tree toolkit Same VLAN Same VLAN Same VLAN Layer 3 Equal Cost Links WAN Layer 2 Loops Data Center Layer 3 Equal Cost Links Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 25
Multilayer Network Design Layer 2 Access with Layer 3 Distribution Vlan 10 Vlan 20 Vlan 30 Vlan 30 Vlan 30 Vlan 30 Each access switch has unique VLANs No layer 2 loops Layer 3 link between distribution No blocked links At least some VLANs span multiple access switches Layer 2 loops Layer 2 and 3 running over link between distribution Blocked links 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 26
Optimizing L2 Convergence PVST+, Rapid PVST+ or MST Rapid-PVST+ greatly improves the restoration times for any VLAN that requires a topology convergence due to link UP Rapid-PVST+ also greatly improves convergence time over backbone fast for any indirect link failures PVST+ (802.1d) Traditional spanning tree implementation Rapid PVST+ (802.1w) Scales to large size (~10,000 logical ports) Easy to implement, proven, scales MST (802.1s) Permits very large scale STP implementations (~30,000 logical ports) Not as flexible as Rapid PVST+ http://www.cisco.com/en/us/products/hw/switches/ps708/products_co nfiguration_example09186a00807b0670.shtml store Data a Flows (se ec) Time to Re 35 30 25 20 15 10 5 0 PVST+ Upstream Downstream Rapid PVST+ 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 27
Layer 2 Hardening Spanning Tree Should Behave the Way You Expect Place the root where you want it Root primary/secondary macro The root bridge should stay where you put it RootGuard LoopGuard UplinkFast UDLD Only end-station traffic should be seen on an edge port BPDU Guard RootGuard PortFast Port-security STP Root RootGuard LoopGuard UplinkFast LoopGuard BPDU Guard or RootGuard PortFast Port Security 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 28
Best Practices Layer 3 Routing Protocols Typically deployed in distribution to core, and core to core interconnections Used to quickly re-route around dfailed node/links while providing load balancing over redundant paths Build triangles not squares for deterministic convergence Only peer on links that you intend to use as transit Insure redundant L3 paths to avoid black holes Summarize distribution to core to limit EIGRP query diameter or OSPF LSA propagation Tune CEF L3/L4 load balancing hash to achieve maximum utilization of equal cost paths (CEF polarization) Layer 3 Equal Cost Links WAN Data Center Layer 3 Equal Cost Links Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 29
Best Practice Build Triangles Not Squares Deterministic vs. Non-Deterministic Triangles: Link/Box Failure Does NOT Squares: Link/Box Failure Requires Require Routing Protocol Convergence Routing Protocol Convergence Model A Model B Layer 3 redundant equal cost links support fast convergence Hardware based fast recovery to remaining path Convergence is extremely fast (dual equal-cost paths: no need for OSPF or EIGRP to recalculate a new path) 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 30
Best Practice Passive Interfaces for IGP Limit OSPF and EIGRP Peering Through the Access Layer Limit unnecessary peering using passive interface: Four VLANs per wiring i closet 12 adjacencies total Memory and CPU requirements increase with no real benefit Creates overhead for IGP Distribution Access Routing Updates OSPF Example: Router(config)#routerospf 1 Router(config-router)#passiveinterfaceVlan 99 Router(config)#routerospf 1 Router(config-router)#passiveinterface default Router(config-router)#no passiveinterface Vlan 99 EIGRP Example: Router(config)#routereigrp 1 Router(config-router)#passiveinterfaceVlan 99 Router(config)#routereigrp 1 Router(config-router)#passiveinterface default Router(config-router)#no passiveinterface Vlan 99 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 31
Why You Want to Summarize at the Distribution Limit EIGRP Queries and OSPF LSA Propagation It is important to force summarization at the distribution towards the core For return path traffic an OSPF or EIGRP re-route is required By limiting the number of peers an EIGRP router must query or the number of LSAs an OSPF peer must process we can optimize this re-route EIGRP example: No Summaries Queries Go Beyond the Core Rest of Network Core Distribution interface Port-channel1 description to Core#1 ip address 10.122.0.34 255.255.255.252 ip hello-interval eigrp 100 1 ip hold-time eigrp 100 3 ip summary-address eigrp 100 10.1.0.0 255.255.0.0 5 10.1.1.0/24 10.1.2.0/24 Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 32
Why You Want to Summarize at the Distribution Reduce the Complexity of IGP Convergence It is important to force summarization at the distribution towards the core For return path traffic an OSPF or EIGRP re-route is required By limiting the number of peers an EIGRP router must query or the number of LSAs an OSPF peer must process we can optimize his re-route For EIGRP if we summarize at the distribution we stop queries at the core boxes for an access layer flap For OSPF when we summarize at the distribution (area border or L1/L2 border) the flooding of LSAs is limited to the distribution switches; SPF now deals with one LSA not three Summaries Stop Queries at the Core Rest of Network 10.1.1.0/24 Summary: 10.1.0.0/16 10.1.2.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 33
Best Practice Summarize at the Distribution Gotcha Distribution-to-Distribution Link Required Best practice summarize at the distribution layer to limit EIGRP queries or OSPF LSA propagation p Gotcha: Upstream: HSRP on left distribution takes over when link fails Return path: old router still advertises summary to core Return traffic is dropped on right distribution switch Summarizing requires a link between the distribution switches Summary: 10.1.0.0/16 Alternative design: Use the access layer for transit 10.1.1.0/24 10.1.2.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 34
Provide Alternate Paths What happens if fails? No route to the core anymore? Allow the traffic to go through the access? Do you want to use your access switches as transit nodes? How do you design for scalability if the access used for transit traffic? Install a redundant link to the core Best practice: install redundant link to core and utilize L3 link between ee distribution Layer A B (summarization coming) ngle Path to Core Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 35
EIGRP Design Rules in the Campus Leverage the Tools Provided d The greatest advantages of EIGRP are gained when the network has a structured addressing plan that allows for use of summarization and stub routers when appropriate EIGRP provides the ability to implement multiple tiers of summarization and route filtering Minimize the number and time for query response to speed up convergence Summarize distribution block routes upstream to the core If routing in the access configure all access switches as EIGRP stub routers If routing in the access layer filter routes sent down to access switches 10.10.0.0/16 10.10.0.0/17 10.10.128.0/17 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 36
OSPF Design Rules in the Campus Where Are the Areas? Area design based on address summarization Area boundaries should define buffers between fault domains Summarize routes from the distribution block upstream into the core Minimize the number of LSAs and routes in the core Reduce the need for SPF calculations due to internal distribution block changes ABR for a regular area forwards Summary LSAs (Type 3) ASBR summary (Type 4) Specific externals (Type 5) Stub area ABR forwards Summary LSAs (Type 3) Summary default (0.0.0.0) A totally stubby area ABR forwards Summary default (0.0.0.0) Area 100 Area 110 Area 120 WAN Data Center Area 0 Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 37
Equal Cost Multi-Path Optimizing CEF Load-Sharing Depending on the traffic flow patterns and IP Addressing in use one algorithm may provide better load-sharing results than another Be careful not to introduce polarization in a multi- tier design by changing the default to the same thing in all tiers/layers of the network 30% of Flows 70% of Flows Catalyst t 4500 Load-Sharing Options Original Universal* Include Port Src IP + Dst IP Src IP + Dst IP + Unique ID Src IP + Dst IP + (Src or Dst Port) + Unique ID Load-Sharing mple Catalyst 6500 PFC3** Load-Sharing Options Default* Full Full Exclude Port mple Full mple Src IP + Dst IP + Unique ID Src IP + Dst IP + Src Port + Dst Port Src IP + Dst IP + (Src or Dst Port) Src IP + Dst IP Src IP + Dst IP + Src Port + Dst Port * = Default Load-Sharing Mode ** = PFC3 in Sup720 and Sup32 Supervisors Load-Sharing Full mple Load-Sharing mple 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 38
CEF Load Balancing Avoid Underutilizing Redundant Layer 3 Paths Distribution Default L3 Hash Core Default L3 Hash Distribution Default L3 Hash Redundant Paths Ignored L L R R CEF polarization: without some tuning CEF will select the same path left/left or right/right Imbalance/overload could occur Redundant paths are ignored/underutilized The default CEF hash input is L3 We can change the default to use L3 + L4 information as input to the hash derivation 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 39
CEF Load Balancing Avoid Underutilizing Redundant Layer 3 Paths Distribution L3/L4 Hash Core Default L3 Hash Distribution L3/L4 Hash L All Paths Used R L R L R The default will for Sup720/32 and latest hardware (unique ID added to default). However, depending on IP addressing, and flows imbalance could occur Alternating L3/L4 hash and L3 hash will give us the best load balancing results Use simple in the core and full simple in the distribution to add L4 information to the algorithm at the distribution and maintain differentiation tier-to-tier 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 40
Best Practices Trunk Configuration Typically deployed on interconnection between access and distribution layers Use VTP transparent mode to decrease potential for operational error Hard set trunk mode to on and encapsulation negotiate off for optimal convergence Change the native VLAN to something unused to avoid VLAN hopping Manually prune all VLANS except tthose needed d Disable on host ports: CatOS: set port host Cisco IOS: switchport host Layer 3 Equal Cost Links WAN 802.1q Trunks Data Center Layer 3 Equal Cost Links Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 41
VTP Virtual Trunk Protocol Centralized VLAN management VTP server switch propagates VLAN database to VTP client switches Runs only on trunks Four modes: Server: updates clients and servers Client: receive updates cannot make changes Transparent: let updates pass through Off: ignores VTP updates A Set VLAN 50 Server Trunk Ok, I Just Learnt VLAN 50! Client Trunk Drop VTP Updates Off Trunk Trunk F Transparent C Client Pass Through Update Ok, I Just Learnt VLAN 50! B 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 42
DTP Dynamic Trunk Protocol Automatic formation of trunked switch-to-switch interconnection On: always be a trunk Desirable: ask if the other side can/will Auto: if the other sides asks I will Off: don t become a trunk Negotiation of 802.1Q or ISL encapsulation ISL: try to use ISL trunk encapsulation 802.1q: try to use 802.1q encapsulation Negotiate: negotiate ISL or 802.1q encapsulation with peer Non-negotiate: always use encapsulation that is hard set On/On Trunk Auto/Desirable Trunk Off/Off NO Trunk Off/On, Auto, Desirable NO Trunk 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 43
Optimizing Convergence: Trunk Tuning Trunk Auto/Desirable Takes Some Time DTP negotiation tuning improves link up convergence time CatOS> (enable) set trunk <port> nonegotiate dot1q <vlan> IOS(config-if)# switchport mode trunk IOS(config-if)# switchport nonegotiate 25 2.5 Time to Co onverge in Seconds 2 1.5 1 0.5 Two Seconds of Delay/Loss Tuned daway Voice Data 0 Trunking Desirable Trunking Nonegotiate 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 44
Trunking/VTP/DTP Quick Summary VTP Transparent should be used; there is a trade off between administrative overhead and the temptation to span existing VLANS across multiple access layer switches Emerging technologies that do VLAN assignment by name (IBNS, NAC, etc.) require a unique VLAN database per access layer switch if the rule: A VLAN = A Subnet = AN access layer switch is going to be followed One can consider a configuration that uses DTP ON/ON and NO NEGOTIATE; there is a trade off between performance/ha impact and maintenance and operations implications An ON/ON and NO NEGOTIATE configuration is faster from a link up (restoration) perspective than a desirable/desirable alternative. However, in this configuration DTP is not actively monitoring the state of the trunk and a misconfigured trunk is not easily identified. It s really a balance between fast convergence and your ability to manage configuration and change control 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 45
Best Practices UDLD Configuration Typically deployed on any fiberoptic interconnection Use UDLD aggressive mode for best protection Turn on in global configuration to avoid operational error/ misses Config example Cisco IOS: udld aggressive Layer 3 Equal Cost Links Fiber Interconnections Layer 3 Equal Cost Links WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 46
Unidirectional Link Detection Protecting Against One Way Communication Highly-available networks require UDLD to protect against one-way communication or partially failed links and the effect that they could have on protocols like STP and RSTP Primarily used on fiberoptic links where patch panel errors could cause link up/up with mismatched transmit/receive pairs Each switch port configured for UDLD will send UDLD protocol packets (at L2) containing the port s own device/port ID, and the neighbor s device/port IDs seen by UDLD on that port Neighboring ports should see their own device/port ID (echo) in the packets received from the other side If the port does not see its own device/port ID in the incoming UDLD packets for a specific duration of time, the link is considered unidirectional and is shutdown Are You Echoing My Hellos? 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 47
UDLD Aggressive and UDLD Normal Timers are the same 15 second hellos by default Aggressive Mode after aging on a previously bi-directional link tries 8 times (once per second) to reestablish connection then err-disables port UDLD Normal Mode Only err-disable the end where UDLD detected other end just sees the link go down UDLD Aggressive err-disable BOTH ends of the connection due to err-disable when aging g and re-establishment of UDLD communication fails 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 48
Best Practices EtherChannel Configuration Typically deployed in distribution to core, and core to core interconnections Used to provide link redundancy while reducing peering complexity Tune L3/L4 load balancing hash to achieve maximum utilization of channel members Deploy in powers of 2 (2, 4, or 8) Match CatOS and Cisco IOS PAgP settings 802.3ad LACP for interop if you need it Disable unless needed CatOS: set port host Cisco IOS: switchport host Layer 3 Equal Cost Links WAN Data Center Layer 3 Equal Cost Links Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 49
Understanding EtherChannel Link Negotiation Options PAgP P and LACP Packet Aggregation Protocol Link Aggregation Protocol On/On Channel On/On Channel On/Off No Channel On/Off No Channel Auto/Desirable Channel Active/Passive Channel Off/On, Auto, Desirable No Channel Passive/Passive No Channel On: always be a channel/bundle member Desirable: ask if the other side can/will Auto: if the other side asks I will Off: don t become a member of a channel/bundle On: always be a channel/bundle member Active: ask if the other side can/will Passive: if the other side asks I will Off: don t become a member of a channel/bundle 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 50
PAgP Tuning PAgP Default Mismatches Matching EtherChannel Configuration on Both des Improves Link Restoration Convergence Times set port channel <mod/port> off 7 Time to Conver rge in Seconds 6 5 4 3 2 1 0 PAgP Mismatch As Much As Seven Seconds of Delay/Loss Tuned Away PAgP Off 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 51
EtherChannels or Equal Cost Multipath 10/100/1000 How Do You Aggregate It? Typical 4:1 Data Over- Subscription Core Distribution 10GE and 10GE channels Typical 20:1 Data Over- Subscription Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 52
EtherChannels or Equal Cost Multipath Reduce Complexity/Peer Relationships More links = more routing peer relationships and associated overhead EtherChannels allow you to reduce peers by creating single logical interface to peer over Layer 3 Equal Cost Links Layer 3 Equal Cost Links On single link failure in a bundle OSPF running on an IOS-based switch will reduce link cost and re-route route traffic OSPF running on a hybrid switch will not change link cost and may overload remaining links EIGRP may not change link cost and may overload remaining links WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 53
EtherChannels or Equal Cost Multipath Why 10-Gigabit Interfaces Layer 3 Equal Cost Links WAN More links = more routing peer relationships and associated overhead EtherChannels allow you to reduce peers by creating single logical interface to peer over However, a single link failure is Layer 3 Equal Cost Links not taken into consideration by routing protocols. Overload possible. Data Center Internet ngle 10-Gigabit links address both problems. Increased bandwidth without increasing complexity or compromising routing gprotocols ability to select best path. 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 54
EtherChannels Quick Summary For Layer-2 EtherChannels: Desirable/Desirable is the recommended configuration so that PAgP is running across all members of the bundle insuring that an individual link failure will not result in an STP failure For Layer-3 EtherChannels: One can consider a configuration that uses ON/ON. There is a trade-off between performance/ha impact and maintenance and operations implications. An ON/ON configuration is faster from a link-up p( (restoration)perspective p than a Desirable/Desirable alternative. However, in this configuration PAgP is not actively monitoring the state of the bundle members and a misconfigured bundle is not easily identified. Routing protocols may not have visibility into the state of an individual member of a bundle. LACP and the minimum links option can be used to bring the entire bundle down when the capacity is diminished. OSPF has visibility to member loss (best practices pending investigation). EIGRP does not When used to increase bandwidth no individual flow can go faster than the speed of an individual member of the link Best used to eliminate i single points of failure (i.e. link or port) dependencies d from a topology 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 55
Best Practices First Hop Redundancy Used to provide a resilient default gateway/first hop address to end-stations HSRP, VRRP, and GLBP alternatives VRRP, HSRP and GLBP provide millisecond timers and excellent convergence performance VRRP if you need multivendor interoperability GLBP facilitates uplink load balancing Preempt timers need to be tuned to avoid black-holed traffic Layer 3 Equal Cost Links WAN 1 st Hop Redundancy Data Center Layer 3 Equal Cost Links Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 56
First Hop Redundancy with VRRP IETF Standard RFC 2338 (April 1998) A group of routers function as one virtual router by sharing one virtual IP address and one virtual MAC address One (master) router performs packet forwarding for local hosts The rest of the routers act as back up in case the master router fails Backup routers stay idle as far as packet forwarding from the client side R1 Master, Forwarding Traffic; R2, Backup VRRP ACTIVE IP: 10.0.0.254 MAC: 0000.0c12.3456 vip: 10.0.0.10 vmac: 0000.5e00.0101 R1 Distribution-A VRRP Active Access-a VRRP BACKUP IP: 10.0.0.253 MAC: 0000.0C78.9abc vip: vmac: R2 Distribution-B VRRP Backup is concerned IP: 10.0.0.1 MAC: aaaa.aaaa.aa01 GW: 10.0.0.10 ARP: 0000.5e00.0101 IP: 10.0.0.2 MAC: aaaa.aaaa.aa02 GW: 10.0.0.10 ARP: 0000.5e00.0101 IP: 10.0.0.3 MAC: aaaa.aaaa.aa03 GW: 10.0.0.10 ARP: 0000.5e00.0101 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 57
First Hop Redundancy with HSRP RFC 2281 (March 1998) A group of routers function as one virtual router by sharing one virtual IP address and one virtual MAC address One (active) router performs packet forwarding for local hosts The rest of the routers provide hot standby in case the active router fails Standby routers stay idle as far as packet forwarding from the client side is R1 Active, Forwarding Traffic; R2 Hot Standby, Idle HSRP ACTIVE IP: 10.0.0.254 MAC: 0000.0c12.3456 vip: 10.0.0.10 vmac: 0000.0c07.ac00 R1 Distribution-A HSRP Active Access-a HSRP STANDBY IP: 10.0.0.253 MAC: 0000.0C78.9abc vip: vmac: R2 Distribution-B HSRP Backup concerned IP: 10.0.0.1 MAC: aaaa.aaaa.aa01 GW: 10.0.0.10 ARP: 0000.0c07.ac00 IP: 10.0.0.2 MAC: aaaa.aaaa.aa02 GW: 10.0.0.10 ARP: 0000.0c07.ac00 IP: 10.0.0.3 MAC: aaaa.aaaa.aa03 GW: 10.0.0.10 ARP: 0000.0c07.ac00 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 58
Why You Want HSRP Preemption Spanning Tree Root and HSRP Primary aligned When Spanning Tree Root is re-introduced, traffic will take a two-hop path to HSRP Active Core Spanning Tree Root HSRP HSRP Preempt Active HSRP Active Spanning Tree Root Distribution HSRP Preemption will allow HSRP to follow Spanning Tree topology Access Without Preempt Delay HSRP Can Go Active Before Box Completely Ready to Forward Traffic: L1 (Boards), L2 (STP), L3 (IGP Convergence) standby 1 preempt delay minimum 180 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 59
First Hop Redundancy with GLBP Cisco Designed, Load Sharing, Patent Pending All the benefits of HSRP plus load balancing of default gateway utilizes all available bandwidth A group of routers function as one virtual router by sharing one virtual IP address but using multiple virtual MAC addresses for traffic forwarding Allows traffic from a single common subnet to go through multiple redundant gateways using a single virtual IP address R1- AVG; R1, R2 Both Forward Traffic GLBP AVG/AVF, SVF IP: 10.0.0.254 MAC: 0000.0c12.3456 vip: 10.0.0.100 0 vmac: 0007.b400.0101 R1 Distribution-A GLBP AVG/ AVF, SVF Access-a GLBP AVF, SVF IP: 10.0.0.253 MAC: 0000.0C78.9abc vip: 10.0.0.10 vmac: 0007.b400.0102 Distribution-B GLPB AVF, SVF IP: 10.0.0.1 MAC: aaaa.aaaa.aa01 IP: 10.0.0.2 MAC: aaaa.aaaa.aa02 IP: 10.0.0.3 MAC: aaaa.aaaa.aa03 GW: 10.0.0.10 GW: 10.0.0.10 GW: 10.0.0.10 ARP: 0007.B400.0101 ARP: 0007.B400.0102 ARP: 0007.B400.0101 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 60
First Hop Redundancy with Load Balancing Cisco Gateway Load Balancing Protocol (GLBP) Each member of a GLBP redundancy group owns a unique virtual MAC address for a common IP address/default gateway When end-stations ARP for the common IP address/default gateway they are given a load balanced virtual MAC address Host A and host B send traffic to different GLBP peers but have the same default gateway GLBP 1 ip 10.88.1.10 vmac 0000.0000.0001 vip R1 10.88.1.10 R2.1 ARP.2 Reply GLBP 1 ip 10.88.1.10 vmac 0000.0000.0002 10.88.1.0/24.4.5 A B ARPs for 10.88.1.10 ARPs for 10.88.1.10 Gets MAC 0000.0000.0001 Gets MAC 0000.0000.0002 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 61
Optimizing Convergence: VRRP, HSRP, GLBP Mean, Max, and Min Are There Differences? VRRP not tested with sub-second second timers and all flows go through a common VRRP peer; mean, max, and min are equal HSRP has sub-second timers; however all flows go through same HSRP peer so there is no difference between mean, max, and min GLBP has sub-second timers and distributes the load amongst the GLBP peers; so 50% of the clients are not affected by an uplink failure Time in Seconds to Converge e 1.2 1 0.8 0.6 0.4 0.2 0 Distribution to Access Link Failure Access to Server Farm VRRP HSRP GLBP 50% of Flows Have ZERO Loss W/ GLBP Longest Shortest Average GLBP Is 50% Better 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 62
If You Span VLANS, Tuning Required By Default, Half the Traffic Will Take a Two-Hop L2 Path Both distribution switches act as default gateway Blocked uplink caused traffic to take less than optimal path Core Layer 3 Distribution Layer 2/3 Core Distribution-A Distribution-B GLBP Virtual GLBP Virtual MAC 1 MAC 2 Access Layer 2 F: Forwarding B: Blocking Access-a VLAN 2 Access-b VLAN 2 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 63
Agenda Multilayer Campus Design principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 64
Daisy Chaining Access Layer Switches Avoid Potential Black Holes Return Path Traffic Has a 50/50 Chance of Being Black Holed Core Layer 3 Distribution Layer 2/3 Distribution-A Layer 3 Link Distribution-B 50% Chance That Traffic Will Go Down Path with No Connectivity Access Layer 2 Access-a Access-n Access-c VLAN 2 VLAN 2 VLAN 2 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 65
Daisy Chaining Access Layer Switches New Technology Addresses Old Problems Stackwise/Stackwise-Plus technology eliminates the concern Loopback links not required No longer forced to have L2 link in distribution If you use modular (chassis-based) switches, these problems are not a concern Forwarding HSRP Active Layer 3 Catalyst 3750-E Forwarding HSRP Standby or Catalyst 2975 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 66
What Happens if You Don t Link the Distributions? STPs slow convergence can cause considerable periods of traffic loss STP could cause non-deterministic traffic flows/link load engineering STP convergence will cause Layer 3 convergence STP and Layer 3 timers are independent Unexpected Layer 3 convergence and re-convergence could occur Even if you do link the distribution switches dependence on STP and link state/connectivity can cause HSRP irregularities and unexpected state transitions STP Root and HSRP Active Traffic Dropped Until Transition to Forwarding; As much as 50 Seconds F 2 Access-a VLAN 2 Core Hellos STP Secondary Root and HSRP Standby B 2 Access-b VLAN 2 Traffic Dropped Until MaxAge Expires Then Listening and Learning 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 67
What if You Don t? Black Holes and Multiple Transitions Core Layer 3 Distribution Layer 2/3 STP Root and HSRP Active Core Hellos STP Secondary Root and HSRP Standby Aggressive HSRP timers limit black hole #1 Backbone fast limits time (30 seconds) HSRP Active to event #2 (Temporarily) Even with Rapid PVST+ at least one second before event #2 Access Layer 2 Access-a VLAN 2 F: Forwarding B: Blocking Access-b VLAN 2 MaxAge Seconds Before Failure Is Detected Then Listening and Learning Blocking link on access-b will take 50 seconds to move to forwarding traffic black hole until HSRP goes active on standby HSRP peer After MaxAge expires (or backbone fast or Rapid PVST+) converges HSRP preempt causes another transition Access-b used as transit for access-a s traffic 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 68
What if You Don t? Return Path Traffic Black Holed Core Layer 3 Distribution Layer 2/3 STP Root and HSRP Active Core Hellos STP Secondary Root and HSRP Standby 802.1d: up to 50 seconds PVST+: backbone fast 30 seconds Rapid PVST+: address by the protocol (one second) Access Layer 2 F: Forwarding B: Blocking Access-a VLAN 2 Access-b VLAN 2 Blocking link on access-b will take 50 seconds to move to forwarding return traffic black hole until then 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 69
Asymmetric Routing (Unicast Flooding) Affects redundant topologies with shared L2 access One path upstream and two paths downstream CAM table entry ages out on standby HSRP Without a CAM entry packet is flooded to all ports in the VLAN Asymmetric Equal Cost Return Path CAM Timer Has Aged out on Standby HSRP Downstream Packet Flooded Upstream Packet Unicast to Active HSRP VLAN 2 VLAN 2 VLAN 2 VLAN 2 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 70
Best Practices Prevent Unicast Flooding Assign one unique data and voice VLAN to each access switch Traffic is now only flooded down one trunk Access switch unicasts correctly; no flooding to all ports If you have to: Tune ARP and CAM aging timers; CAM timer exceeds ARP timer Bias routing metrics to remove equal cost routes Asymmetric Equal Cost Return Path Downstream Packet Flooded on ngle Port VLAN 3 VLAN 4 VLAN 5 Upstream Packet Unicast to Active HSRP VLAN 2 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 71
Agenda Multilayer Campus Design Principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 72
Building a Converged Campus Network Infrastructure Integration, QoS, and Availability Access layer Auto phone detection Inline power QoS: scheduling, trust boundary and classification Fast convergence Distribution ib ti layer Core High availability, redundancy, fast convergence Policy enforcement QoS: scheduling, trust boundary and classification High availability, redundancy, fast convergence QoS: scheduling, trust boundary Access Distribution Core Distribution Access Layer 3 Equal Cost Links WAN Data Center Layer 3 Equal Cost Links Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 73
Infrastructure Integration Extending the Network Edge Switch Detects IP Phone and Applies Power CDP Transaction Between Phone and Switch IP Phone Placed in Proper VLAN DHCP Request est and Call Manager Registration Phone contains a three-port switch that is configured in conjunction with the access switch and CallManager 1. Power negotiation 2. VLAN configuration 3. 802.1x interoperation 4. QoS configuration 5. DHCP and CallManager registration 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 74
Infrastructure Integration: First Step Power Requirement Negotiation Cisco pre-standard devices initially receive 6.3 watts and then optionally negotiate via CDP 802.3af devices initially receive 12.95 watts unless PSE able to detect specific PD power classification Class Usage Minimum Power Levels Output at the PSE Maximum Power Levels at the Powered Device 0 Default 15.4W 0.44 to 12.95W 1 Optional 4.0W 0.44 to 3.84W 2 Optional 7.0W 3.84 to 6.49W 3 Optional 15.4W 6.49 to 12.95W 4 Reserved for Future Treat as Class 0 Use Reserved for Future Use: a Class 4 gnature Cannot Be Provided by a Compliant Powered Device 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 75
Enhanced Power Negotiation 802.3af Plus Bi-Directional CDP (Cisco 7970) PSE Power Source Equipment Cisco 6500,4500, 3750, 3560 PD Plugged in Switch Detects IEEE PD 3750 3560 PD Is Classified Power Is Applied Phone Transmits a CDP Power Negotiation Packet Listing Its Power Mode Switch Sends a CDP Response with a Power Request PD Powered Device Cisco 7970 Based on Capabilities Exchanged Final Power Allocation Is Determined Using bi-directional CDP exchange exact power requirements are negotiated after initial power-on 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 76
Design Considerations for PoE Power Management Switch manages power by what is allocated not by what is currently used Device power consumption is not constant A 7960G requires 7W when the phone is ringing g at maximum volume and requires 5W on or off hook Understand the power behavior of your PoE devices Utilize static power configuration with caution Dynamic allocation: power inline auto max 7200 Static allocation: power inline static max 7200 Use power calculator to determine power requirements http://www.cisco.com/go/powercalculator Discover Cisco Enhanced PoE at the World of Solutions 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 77
Infrastructure Integration: Next Steps VLAN, QoS and 802.1x Configuration Phone VLAN = 110 (VVID) PC VLAN = 10 (PVID) 802.1Q encapsulation with 802.1p Layer 2 CoS Native VLAN (PVID) No Configuration Changes Needed on PC During initial CDP exchange phone is configured with a Voice VLAN ID (VVID) Phone also supplied with QoS configuration via CDP TLV fields Additionally switch port currently bypasses 802.1x authentication for VVID if detects Cisco phone 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 78
Agenda Multilayer Campus Design principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 79
Best Practices Quality of Service Must be deployed end-toend to be effective; all layers play different but equal roles End to End QoS Ensure that mission critical applications are not impacted by link or transmit queue congestion Aggregation and rate transition points must enforce QoS policies Layer 3 Equal Cost Links Layer 3 Equal Cost Links Multiple queues with configurable admission criteria and scheduling are required WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 80
Transmit Queue Congestion 10/100m Queued 128k Uplink WAN Router 100 Meg in 128 Kb/S out Packets Serialize in Faster than They Serialize out Packets Queued as They Wait to Serialize out Slower Link 1 Gig Link Queued 100 Meg Link Distribution Switch Access Switch 1 Gig In 100 Meg out Packets Serialize in Faster than They Serialize out Packets Queued as They Wait to Serialize out Slower Link 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 81
Auto QoS VoIP Making It Easy Configures QoS for VoIP on Campus Switches Access-Switch(config-if)#auto qos voip? cisco-phone Trust the QoS marking of Cisco IP Phone cisco-softphone Trust the QoS marking of Cisco IP SoftPhone trust Trust the DSCP/CoS marking Access-Switch(config-if)#autoqosvoipcisco-phone p Access-Switch(config-if)#exit! interface FastEthernet1/0/21 srr-queue bandwidth share 10 10 60 20 srr-queue bandwidth shape 10 0 0 0 Mls qos trust device cisco-phone Mls qos trust cos auto qosvoipcisco-phone end 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 82
Agenda Multilayer Campus Design principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 83
Best Practices Campus Security New stuff that we will cover! Catalyst Integrated Security Feature Set! Dynamic Port Security, DHCP Snooping, Dynamic ARP Inspection, IP Source Guard Things you already know we won t cover End-to-End Security Use SSH to access devices instead of Telnet Enable AAA and roles-based access control (RADIUS/TACACS+) for the CLI on all devices Enable SYSLOG to a server. Collect and archive logs When using SNMP use SNMPv3 Disable unused services: no service tcp-small-servers no service udp-small-servers Use FTP or SFTP (SSH FTP) to move images and configurations around avoid TFTP when possible Install VTY access-lists to limit which addresses can access management and CLI services Enable control plane protocol authentication where it is available (EIGRP, OSPF, BGP, HSRP, VTP, etc.) Apply basic protections offered by implementing RFC2827 filtering i on external edge inbound WAN Internet interfaces For More Details, See BRKSEC-2002 Session, Understanding and Preventing Layer 2 Attacks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 84
BPDU Guard Prevent Loops via WLAN (Windows XP Bridging) Problem: WLAN APs do not forward BPDUs Multiple Windows XP machines can create a loop in the wired VLAN via the WLAN Solution: BPDU Guard configured on all end-station switch ports will prevent loop from forming STP Loop Formed Win XP Bridging Enabled BPDU Discarded BPDU Guard Disables Port Win XP Bridging Enabled BPDU Generated 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 85
Problem: Prevalence of Rogue APs Example: 59 APs in Seven Miles in SJ Commute The majority of WLAN deployments are unauthorized by well intended employees (rogue APs) many are insecure A daily drive to work taken within the car at normal speeds with an IPAQ running a freeware application (mix of residences and enterprises) Insecure enterprise rogue AP s are a result of: Well intentioned staff install due to absence of sanctioned WLAN deployment An infrastructure that is not wireless ready to protect against rogue AP s Insecure APs 59 APs Found War Chalking 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 86
Basic 802.1x Access Control Controlling When and Where APs Are Connected Authorized User No 802.1x Here Who Are You? I Am Joe Cisco 802.1x Enabled on user Facing Ports ots Who Are You? Rogue AP Disabled D on Authorized WLAN AP Ports CatOS Configuration Example set dot1x system-auth-control enable set dot1x guest-vlan 250 set radius server 10.1.125.1 auth-port 1812 primary set radius key cisco123 set port dot1x 3/1-48 port-control auto Cisco IOS Configuration Example radius-server host 10.1.125.1 radius-server key cisco123 aaa new-model aaa authentication dot1x default group radius aaa authorization default group radius aaa authorization config-commands dot1x system-auth-control Cisco IOS Per-Port configuration Authorized dap int range fa3/1-48 dot1x port-control auto 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 87
Securing Layer 2 from Surveillance Attacks Cutting off MAC-Based Attacks 250,000 Bogus MACs per Second 00:0e:00:aa:aa:aa0 Only 3 MAC 00:0e:00:bb:bb:bb Addresses Allowed on the Port: Shutdown PROBLEM: Script Kiddie Hacking Tools Enable Attackers Flood Switch CAM Tables with Bogus Macs; Turning the VLAN into a Hub and Eliminating Privacy Switch CAM Table Limit Is Finite Number of Mac Addresses SOLUTION: Port Security Limits MAC Flooding Attack and Locks down Port and Sends an SNMP Trap switchport port-security switchport port-security maximum 10 switchport port-security violation restrict switchport port-security aging time 2 switchport port-security aging type inactivity 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 88
DHCP Snooping Protection Against Rogue/Malicious DHCP Server 1 1000s of DHCP Requests to Overrun the DHCP Server 2 DHCP Server DHCP requests (discover) and responses (offer) tracked Rate-limit it requests on trusted t interfaces; limits it DoS attacks on DHCP server Deny responses (offers) on non trusted interfaces; stop malicious or errant DHCP server 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 89
Securing Layer 2 from Surveillance Attacks Protection Against ARP Poisoning Dynamic ARP inspection protects against ARP poisoning (ettercap, dsnif, arpspoof) Uses the DHCP snooping binding table Tracks MAC to IP from DHCP transactions Rate-limits ARP requests from client ports; stop port scanning Gateway = 10.1.1.1 MAC=A Gratuitous ARP 10.1.1.50=MAC_B Gratuitous ARP 10.1.1.1=MAC_B Drop BOGUS gratuitous ARPs; stop ARP poisoning/mim i attacks Attacker = 10.1.1.251 1 Victim = 10.1.1.501 1 MAC=B MAC=C 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 90
IP Source Guard Protection Against Spoofed IP Addresses IP source guard protects against spoofed IP addresses Uses the DHCP snooping binding table Tracks IP address to port associations Gateway = 10.1.1.1 MAC=A Dynamically programs port ACL to drop traffic not originating from IP address assigned via DHCP Attacker = 10.1.1.251 1 Victim = 10.1.1.501 1 Hey, I m 10.1.1.50! 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 91
Catalyst Integrated Security Features Summary Cisco IOS IP Source Guard Dynamic ARP Inspection DHCP Snooping Port Security Port security prevents MAC flooding attacks DHCP snooping prevents client attack on the switch and server Dynamic ARP Inspection adds security to ARP using DHCP snooping table IP source guard adds security to IP source address using DHCP snooping table ipdhcp snooping ipdhcp snooping vlan 2-10 iparp inspection vlan 2-10! interface fa3/1 switchport port-security switchport port-security max 3 switchport port-security violation restrict switchport port-security aging time 2 switchport port-security aging type inactivity iparp inspection limit rate 100 ipdhcp snooping limit rate 100 ip verify source vlandhcp-snooping! Interface gigabit1/1 ipdhcp snooping trust iparp inspection trust 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 92
Agenda Multilayer Campus Design principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 93
Hierarchical Campus Access Distribution Core Distribution WAN Data Center Internet Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 94
Layer 3 Distribution Interconnection Layer 2 Access No VLANs Span Access Layer Tune CEF load balancing Match CatOS/IOS EtherChannel settings and tune load balancing Summarize routes towards core Limit redundant IGP peering STP Root and HSRP primary tuning or GLBP to load balance on uplinks Set trunk mode on/nonegotiate Disable EtherChannel unless needed Set port host on access layer ports: Disable Trunking Disable EtherChannel Enable PortFast RootGuard or BPDU-Guard Use security features VLAN 20 Data 10.1.20.0/24 VLAN 120 Voice 10.1.120.0/24 Layer 3 Point to Point Link VLAN 40 Data 10.1.40.0/24 VLAN 140 Voice 10.1.140.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 95
Layer 2 Distribution Interconnection Layer 2 Access Some VLANs Span Access Layer Tune CEF load balancing Match CatOS/IOS EtherChannel settings and tune load balancing Summarize routes towards core Limit redundant IGP peering STP Root and HSRP primary or GLBP and STP port cost tuning to load balance on uplinks Set trunk mode on/nonegotiate Disable EtherChannel unless needed RootGuard on downlinks LoopGuard on uplinks Set port host on access Layer ports: Disable Trunking Disable EtherChannel Enable PortFast RootGuard or BPDU-Guard Use security features VLAN 20 Data 10.1.20.0/24 Layer 2 Trunk VLAN 120 Voice VLAN 140 Voice 10.1.120.0/24 10.1.140.0/24 VLAN 250 WLAN 10.1.250.0/24 VLAN 40 Data 10.1.40.0/24 Core Distribution Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 96
Routed Access and Virtual Switching System Evolutions of and Improvements to Existing Designs Core Layer 3 VSS Link New Concept Distribution P-t-P Link VLAN 20 Data 10.1.20.0/24 VLAN 120 Voice 10.1.120.0/24 VLAN 40 Data 10.1.40.0/24 VLAN 140 Voice 10.1.140.0/24 VLAN 20 Data VLAN 10.1.20.0/24 40 Data VLAN 10.1.40.0/24 120 Voice VLAN 10.1.120.0/24 140 Voice VLAN 10.1.140.0/241 250 140 WLAN 0/24 10.1.250.0/24 Access See RST-3035 Advanced Enterprise Campus Design Alternatives: Routed Access and Virtual Switch System (VSS) 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 97
High Availability Campus Design mplified with VSS Access Distribution Core Distribution Access WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 98
SmartPorts Predefined Configurations Access-Switch#show parser macro brief default global : cisco-global default interface: cisco-desktop default interface: cisco-phone default interface: cisco-switch default interface: cisco-router default interface: cisco-wireless Access-Switch(config-if)#$ macro apply cisco-phone $access_vlan 100 $voice_vlan 10 Access-Switch#show run int fa1/0/19! interface FastEthernet1/0/19 switchport access vlan 100 switchport mode access switchport voice vlan 10 switchport port-security maximum 2 switchport port-security switchport port-security aging time 2 switchport port-security violation restrict switchport port-security aging type inactivity srr-queue bandwidth share 10 10 60 20 srr-queue bandwidth shape 10 0 0 0 mls qos trust device cisco-phone mls qos trust cos macro description cisco-phone auto qosvoipcisco-phone spanning-tree portfast spanning-tree bpduguard enable end 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 99
Agenda Multilayer Campus Design principles Data Center Services Block Foundation Services Campus Design Best Practices IP Telephony Considerations QoS Considerations Security Considerations Putting It All Together Summary Distribution Blocks 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 100
Summary Offers hierarchy each h hlayer has specific role Modular topology building blocks Access Easy to grow, understand, and troubleshoot Creates small fault domains Clear demarcations and isolation Distribution Promotes load balancing and redundancy Promotes deterministic traffic patterns Layer 3 Equal Cost Links Layer 3 Equal Cost Links Core Incorporates balance of both Layer 2 and Layer 3 technology, leveraging the strength of both Distribution Utilizes Layer 3 Routing for load balancing, fast convergence, scalability, and control WAN Data Center Internet Access 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 101
Q and A 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 102
Hierarchical Network Design Without a Rock Solid Foundation the Rest Doesn t Matter Access Distribution Core Distribution HSRP Access Routing Spanning Tree Building Block 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 103
Reference Materials http://www.cisco.com/go/srnd High Availability Campus Design Guide High Availability Campus Convergence Analysis High Availability Campus Design Guide Routed Access EIGRP and OSPF 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 104
2008 Cisco Systems, Inc. All rights reserved. Cisco Public 105
Optimal Redundancy When Is More Less? Core and distribution engineered with redundant nodes and links to provide maximum redundancy and optimal convergence Network bandwidth and capacity engineered to withstand t node or link failure Access Distribution Core Redundan t Nodes 120 200ms to converge around most events Distribution Access WAN Data Center Internet 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 106
ngle Points of Termination SSO/NSF Avoiding Total Network Outage Access L2 = SSO L3 = SSO/NSF Distribution Core The access layer is candidate for supervisor redundancy L2 access layer SSO L3 access layer SSO and NSF Network outage until physical replacement or reload vs. one to three seconds 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 107
Supervisor Processor Redundancy Stateful Switch Over (SSO) SP RP PFC Active Supervisor SP RP PFC Standby Supervisor Active/standby supervisors run in synchronized mode Redundant supervisor is in hotstandby mode Switch processors synchronize L2 port state information, (e.g., STP, 802.1x, 802.1q) PFCs synchronize L2/L3 FIB, NetFlow and ACL tables DFCs are populated with L2/L3 FIB, NetFlow and ACL tables Line Card DFC Line ecad Card DFCC Line Card DFC 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 108
Non-Stop Forwarding (NSF) NSF Recovery DFC enabled line cards continue to forward based on existing FIB entries Following SSO recovery and activation of standby Sup synchronized PFC continues to forward traffic based on existing FIB entries Hot-Standby MSFC RIB is detached from the FIB isolating FIB from RP changes Hot-Standby MSFC activates routing processes in NSF recovery mode MSFC re-establishes adjacency indicating this is an NSF restart Peer updates restarting MSFC with it s routing information Restarting MSFC sends routing updates to the peer RIB reattaches to FIB and PFC and DFCs updated with new FIB entries No Route Flaps During Recovery 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 109
Non-Stop Forwarding (NSF) NSF Capable vs. NSF Awareness Two roles in NSF neighbor graceful restart NSF Capable NSF Aware An NSF-Capable router is capable of continuous forwarding while undergoing a switchover An NSF-Aware router is able to assist NSF-Capable routers by: Not resetting adjacency Supplying routing information for verification after switchover NSF capable and NSF aware peers cooperate using Graceful Restart extensions to BGP, OSPF, ISIS and EIGRP protocols NSF-Aware NSF-Capable 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 110
Design Considerations for NSF/SSO NSF and Hello Timer Tuning? NSF is intended to provide availability through route convergence avoidance Fast IGP timers are intended to provide availability through fast route convergence In an NSF environment dead timer must be greater than SSO Recovery + RP restart + time to send first hello Switches running Native IOS OSPF 2/8 seconds for hello/dead EIGRP 1/4 seconds for hello/hold Switches running Hybrid OSPF 3/12 seconds for hello/dead EIGRP 2/8 seconds for hello/hold Neighbor Loss, No Graceful Restart 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 111
Design Considerations for NSF/SSO Where Does It Make Sense? Redundant topologies with equal cost paths provide sub-second convergence NSF/SSO provides superior availability in environments with non-redundant paths? 6 Second ds of Los st Voice 5 RP Convergence Is Dependent 4 on IGP and Tuning 3 2 1 0 Link Node Failure Failure NSF/SSO OSPF Convergence 2008 Cisco Systems, Inc. All rights reserved. Cisco Public 112