Brkaci 2001

Download as pdf or txt
Download as pdf or txt
You are on page 1of 96

Integration and Interoperation of

Existing Nexus Networks into an


ACI Architecture
Michael Herbert Principal Engineer INSBU
BRKACI-2001
Cisco Spark
Questions? 

Use Cisco Spark to communicate 

with the speaker after the session

How
1. Find this session in the Cisco Live Mobile App
2. Click “Join the Discussion”
3. Install Spark or go directly to the space
4. Enter messages/questions in the space

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 3
Agenda
▪ Introduction to the Intent/Application Based Data Centre
• What is ACI?
• What do we mean by Intent
• What do we mean by Applications Based Infrastructure
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 4
Next Gen Forwarding & Networking

STP
VPC FabricPath
VXLAN

Data Plane Based Endpoint Discovery

Control Plane Based Endpoint Location Tracking


VXLAN VXLAN
/EVPN /ACI APIC

MAN/WAN MAN/WAN

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 5
Identity Based Networking
 VTEP IP Group VNID Tenant Packet
Policy

Shared Application
Servers
Devices and users are Services

authenticated and
authorised into end-point Ingress and
ACI Fabric
groups (aka EPG’s or
Egress ➔
Enforcement
SGT’s) APIC

End Point Group Tags Enterprise


Backbone
(EPG’s, SGT’s) are
encoded in a VXLAN
header
Campus Switch Campus Switch
Policies between scalable
groups are established
following the provider/ Employee Tag

consumer model Supplier Tag


Non-Compliant Employee Voice Voice Employee Non-Compliant Tag

VLAN A VLAN B

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 6
Identity Based Networking
 VTEP IP Group VNID Tenant Packet
Policy

Shared Application
Servers
Devices and users are Services

authenticated and
authorised into end-point Ingress and
ACI Fabric
groups (aka EPG’s or
Egress ➔
Enforcement
SGT’s) APIC

End Point Group Tags Enterprise


Backbone
(EPG’s, SGT’s) are
encoded in a VXLAN
header
Campus Switch Campus Switch
Policies between scalable
groups are established
following the provider/ Employee Tag

consumer model Supplier Tag


Non-Compliant Employee Voice Voice Employee Supplier Non-Compliant Non-Compliant Tag

VLAN A VLAN B

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 6
Directory Enabled Networking

Data Base Defined Networking APIC

Common Operational Properties - AD, LDAP, …


System Management, Change Management, System Integrity, Correlation

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 7
Application Centric Infrastructure (ACI)

Identity and Intent Based Infrastructure

Embedded L4 Security
Next Gen Stateful L4-7
Embedded Sensors Visibility and Control
Policy Discovery, Identity and Policy
Monitoring and Control Federation

Multi-Tier Sensor
Data Gathering

MACSEC and INS-SEC Firewall at Each


Encryption Leaf switch
Servers (Physical, Virtual, Containers, Micro
Services)

Branch Web1 App1 DB Identity & Policy


QoS QoS QoS

Filter Service Filter


Driven Security
Architecture

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 8
Application Centric Infrastructure (ACI)

Identity and Intent Based Infrastructure

Embedded L4 Security
Next Gen Stateful L4-7
Embedded Sensors Visibility and Control
Policy Discovery, Identity and Policy
Monitoring and Control Federation

Multi-Tier Sensor
Data Gathering

MACSEC and INS-SEC Firewall at Each


Encryption Leaf switch
Servers (Physical, Virtual, Containers, Micro
Services)

Branch Web1 App1 DB Identity & Policy


QoS QoS QoS

Filter Service Filter


Driven Security
Architecture

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 8
Agenda
▪ Introduction to the Intent/Application Based Data Centre
• What is ACI?
• What do we mean by Intent
• What do we mean by Applications Based Infrastructure
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 9
Gartner’s View

Intent Based Infrastructure

http://blogs.gartner.com/andrew-lerner/2017/02/07/intent-based-networking/
BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 10
Treating Infrastructure Like a System

Intent Based Control
Define Business outcomes
(Actual Requirements - Policies)

Assurance Functions Logical Description ‘Write to’


(System Correctness and SLA Business Policies
& Outcomes
of Requirements Operations
Monitoring)

Centralised Data
Monitoring, Data Insights & Controllers Model (Database)
Discovery, Learning Analytics & Orchestrators

‘Read from’
Operations Streaming Telemetry
and Systems Coordinated Updates to
Programmable
Feedback Infrastructure Infrastructure Components

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 11
Data Centre Vision

Intent Based Infrastructure

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 12
Inter-dependent
Data Centre Vision
 feedback loops
Intent Based Infrastructure
ACI, UCS (Intent
1. Deployment and Based Automation)
Infrastructure
Provisioning Automation
Security

Cisco
Guarantees Network
CloudCentre Application
(Common Deployment Compliance Assurance
Consistency
Consumption across Engine
Hybrid IT)
(Formal
Methodologies)

ADM

Tetration Analytics Security 2. Operations and
Forensics
(Machine Learning Based Operations Management
and Security)

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 12
Agenda
▪ Introduction to the Intent/Application Based Data Centre
• What is ACI?
• What do we mean by Intent
• What do we mean by Applications Based Infrastructure
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 13
What Do We Mean by “Application”?

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 14
What Do We Mean by “Application Centric”?

VISIBILITY
Cisco CloudCentre

Application Profile

Application
Service Storage
Network
Profiles Profiles SECURITY
Profiles

Cisco ACI

PERFORMANCE

Cisco Workload
Optimisation Manager

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 15
Cisco Tetration
What Do We Mean by “Application Centric”? Analytics™

Correlation of Enforcement and Telemetry

• Cisco Tetration and ACI are designed to provide


complementary visibility, security and
operations
• Tetration platform provides network
performance monitoring functionalities in Cisco
ACI™ mode
• Following Cisco Nexus® 9000 series hardware
is required:
• Cisco Nexus 9300-FX based leaf switches
• Cisco Nexus 9500 series spine switches with N9K-
X9736C-FX line cards
• These functionalities require Cisco ACI release
3.1 or later

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 16
Cisco Tetration
What Do We Mean by “Application Centric”? Analytics™

Correlation of Enforcement and Telemetry

• Cisco Tetration and ACI are designed to provide


complementary visibility, security and
operations
• Tetration platform provides network
performance monitoring functionalities in Cisco
ACI™ mode
• Following Cisco Nexus® 9000 series hardware
is required:
• Cisco Nexus 9300-FX based leaf switches
• Cisco Nexus 9500 series spine switches with N9K-
X9736C-FX line cards
• These functionalities require Cisco ACI release
3.1 or later

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 16
Cisco Tetration
What Do We Mean by “Application Centric”? Analytics™

Correlation of Enforcement and Telemetry

• Cisco Tetration and ACI are designed to provide


complementary visibility, security and
operations
• Tetration platform provides network
performance monitoring functionalities in Cisco
ACI™ mode
• Following Cisco Nexus® 9000 series hardware
is required:
• Cisco Nexus 9300-FX based leaf switches
• Cisco Nexus 9500 series spine switches with N9K-
X9736C-FX line cards
• These functionalities require Cisco ACI release
3.1 or later

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 16
Cisco Tetration
What Do We Mean by “Application Centric”? Analytics™
Correlation with the view from the Server

Flow Inventory

Flow details

Process
Inventory

Process
details

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 17
What Do We Mean by “Application Centric”?
Correlating the Application ‘Transaction’ View with the Infrastructure View

APIC APIC

Exchanges Information on
Application Tiers, Nodes, Services, Get the context related to
Service Endpoints, End host, affected EP’s
Business Transactions, Health Create Troubleshooting
Status, Faults session

Provide an ‘Application Allow Application Admin


Transaction’ View of the to launch ACI
Network Troubleshooting Tools

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 18
Objective: View of Application as Related to Infrastructure

Multi-Domain View

Consistent Governance
Cisco Tetration
Analytics™
Application
Owner,
Administrator, …

Public Cloud
vPod
Infrastructure
Service VM Service VM Service VM
Hypervisor Hypervisor Hypervisor Administration

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 19
Agenda
▪ Introduction to the Intent/Application Based Data Centre
▪ ACI Foundations
▪ Forwarding and Network Availability
▪ ACI Constructs

▪ Integrating Existing Nexus Layer 2 and Layer 3


▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 20
ACI Fabric – An IP network with an Integrated Overlay

APIC

VTEP GPO VXLAN MACIP Payload IP Transport

VTEP VTEP VTEP VTEP VTEP VTEP

VTEP
vSwitch VTEP
vSwitch

• Cisco’s ACI solution leverages an integrated VXLAN based overlay


• IP Network for Transport
• VXLAN based tunnel end points (VTEP)
• VTEP discovery via infrastructure routing
• Directory (Mapping) service for EID (host MAC and IP address) to VTEP lookup

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 21
Host Level Forwarding Granularity

IP Forwarding: MAC Forwarding:

Forwarded using DIPi Forwarded using DMAC


address, HW learning of IP address, HW learning of
address MAC address

10.1.3.11 10.1.3.35 10.6.3.2 10.6.3.17

• Forward based on destination IP Address for intra and inter subnet (Default Mode)
• Bridge semantics are preserved for intra subnet traffic (no TTL decrement, no MAC
header rewrite, etc.)
• Non-IP packets will be forwarded using MAC address. Fabric will learn MAC’s for non-IP
packets, IP address learning for all other packets
• Route if MAC is router-mac, otherwise bridge (standard L2/L3 behaviour)
BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 22
Removing the Classic L2/L3 Boundaries
Layer 2 and Layer 3 integrated forwarding

10.1.1.10 10.1.3.11 10.6.3.2 10.1.3.35 10.1.3.11 10.1.3.35


10.1.1.10 10.6.3.2

Distributed Default Gateway Directed ARP Forwarding and/or Proxy-


ARP based default GW

• ACI Fabric supports full layer 2 and layer 3 forwarding semantics, no changes required to applications or end point IP
stacks
• ACI Fabric provides optimal forwarding for layer 2 and layer 3
• Fabric provides a pervasive SVI which allows for a distributed default gateway

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 23
ACI Multi-Pod
 BiDir PIM Multicast Requirement in Backbone
(HER planned for future release)

Inter-Pod Network
Pod ‘N’
Pod ‘A’

Layer 2/3 Host Mobility

Single APIC Cluster

▪ Multiple ACI Pods connected by an IP Inter-Pod L3 ▪ Forwarding control plane (IS-IS, COOP) fault isolation
network, each Pod consists of leaf and spine nodes (BGP between pods)
▪ Managed by a single APIC Cluster (single database ▪ Data Plane VXLAN encapsulation between Pods (any
domain) that can be geographically distributed (10 msec layer 2 VLAN and layer 3 subnet can be extended
RTT, will increase to 50 msec) across pods)

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 24
ACI Multi-Site

Network and Identity Extended between Fabrics
Identity information carried across Fabrics Network information carried across Fabrics
(Availability Zones) (Availability Zones)
VTEP IP Class-ID VNID Tenant Packet

No Multicast Requirement in
Backbone (HER for any BUM
IP Network traffic)

MP-BGP - EVPN

Multi-Site Orchestrator
BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 25
ACI Remote Physical Leaf
Available from ACI 3.1 Release All local to remote traffic is forwarded
via ACI VXLAN Data-Plane

VXLAN
Data-Plane
IP WAN/IPN
(No Multicast Required)

Any traffic that requires use of the Spine vSwitch


Hypervisor
Proxy will be forwarded to the primary site

* Spine services will be required in scenario’s like copy


All local traffic is switched directly between
service, BD Proxy Services, PBR endpoints, both virtual and bare metal

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 26
ACI Remote Virtual Leaf (Virtual Pod)
Scheduled for 2HCY18 DME/PE COOP BGP
Oracle RR

BGP EVPN Carries


Host and Network
Routing End to End

AVE AVE AVE


Hypervisor Hypervisor Hypervisor

Web App DB

IP Network

AVE vSwitch vSwitch DME/PE COOP BGP


Oracle RR
Hypervisor Hypervisor Hypervisor GPO VXLAN Provides
Network and Identity
Based Segmentation
Web App DB End to End
AVE AVE AVE
Hypervisor Hypervisor Hypervisor

Web App DB

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 27
ACI Anywhere
Scheduled for 2HCY18

Reachability - SDWAN
(Viptela), Direct
Connect, ExpressRoute

Web App DB

IP Network

AVE vSwitch vSwitch COOP BGP


DME/PE Oracle
Hypervisor Hypervisor Hypervisor GPO VXLAN Provides RR
Network and Identity
Based Segmentation
Web App DB End to End
AVE AVE AVE
Hypervisor Hypervisor Hypervisor

Web App DB

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 28
ACI Fault and Policy Domains

DC Wide Operational Domain – Single Domain for security groups (EPG’s), connectivity, …, with scoping for changes

Common WAN Services

DC Core

MP-BGP - EVPN

Pod ‘A’ Pod ‘B’ Pod ‘C’ Pod ‘D’


Switch
Maintenance
Zones

APIC Cluster APIC Cluster

Network Configuration Network Configuration Network Configuration Network Configuration


and Fault Domain and Fault Domain and Fault Domain and Fault Domain

Availability Zone Availability Zone

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 29
Agenda
▪ Introduction to the Intent/Application Based Data Centre
▪ ACI Foundations
▪ Forwarding and Network Availability
▪ ACI Constructs

▪ Integrating Existing Nexus Layer 2 and Layer 3


▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 30
Some new (or not so new) terms: Tenants, VRF (Context), Bridge Domains, Application Network

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 31
What is an EPG?

A Logical Group of Endpoints Attached to the Network

BD1 BD2 BD2


BD3

EPG1 EPG2 EPG3 EPG4 EPG5 EPG6 EPG7 EPG8 EPG9

• All of the endpoints (things attached to the network) in the same EPG are treated to the same rules (policy)
• A security group using the same access lists (similar to an SGT in TrustSec)
• A services group using the same QoS rules, same L4-7 services, …
• It could be as simple as all the servers on the same VLAN or subnet

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 32
What are Contracts and Filters?

The Network Rules Tenant

VRF
• Contracts are semantics to Bridge Domains Bridge Domains
specify EPG to EPG
communication in ACI EPGs EPGs EPGs EPGs

• Communication policy includes C C C

filters (ACLs), QoS, Route


C
Leaking, L4-7 Service Graphs
• Contract filters are similar to
Access Control Lists (e.g. match
this TCP port)
• Contracts can be defined VRF

between EPGs or between L3out Bridge Domains


C
External EPGs and regular EPGs EPGs C EPGs

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 33
What is an Application Profile?

A Logical Group of EPG’s and Associated Contracts

Health scores, statistics, logs


and audit data automatically
correlated and rolled up at
Application Profile level

EPG, uEPG, domain


associations, contract relations
and L4-7 Configuration

Developer: My application template NOC: What I monitor and troubleshoot

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 34
What is a Tenant?

A Virtual Private Cloud
Outside View: A Tenant is a group that owns a virtual Tenant
private cloud instance

Inside View: The portion of the database that your


login has access to

Context Context

BD BD BD

Subnet Subnet Subnet Subnet Subnet


A B C B C

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 35
Okay but Where the *&$# Did My Network Go? 

Bridge Domains and Contexts
• Context == Private Network
• Unique Layer 3 (L3) forwarding domain
• It’s a VRF

• Bridge Domain (BD)


• Unique Layer 2 (L2) forwarding domain
• By default it defines the Broadcast Domain
• Is associated with a VRF

• Subnet
• Is a subnet
• It is associated with a Bridge Domain 10.10.20.0/24
10.10.10.0/24
• You can have multiple subnets associated with one
192.168.4.0/24
BD (think secondary IP’s)

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 36
ACI Networking Foundations 

BDs and EPGs

Layer 2

Layer 2

Layer 2

Extend L2 domain beyond ACI fabric - 2 options


1. Manually assign a port to a VLAN which in turn mapped to an EPG. This extend EPG beyond ACI fabric
(EPG == VLAN)
2. Create a L2 connection to outside network. Extend bridge domain beyond ACI fabric. Allow contract
between EPG inside ACI and EPG outside of ACI

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 37
ACI Networking Foundations 

BDs and EPGs

Layer 2

Layer 2

Layer 2

Lets Look at the


Links

Extend L2 domain beyond ACI fabric - 2 options


1. Manually assign a port to a VLAN which in turn mapped to an EPG. This extend EPG beyond ACI fabric
(EPG == VLAN)
2. Create a L2 connection to outside network. Extend bridge domain beyond ACI fabric. Allow contract
between EPG inside ACI and EPG outside of ACI

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 37
VLAN, EPG and BD

VLAN == EPG

Layer 2
VLAN 30

VLAN 20

BD
EPG
Existing
App 100.1.1.5
100.1.1.3 100.1.1.99 100.1.1.7 100.1.1.3

• VLANs are localised to the leaf ports


• The same subnet, bridge domain, EPG can be configured as a ‘different’ VLAN on each leaf
switch

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 38
VLAN, EPG and BD

VLAN == EPG

VLAN 10

VLAN 30
VLAN 10 VLAN 10 VLAN 10 Layer 2
VLAN 20

BD
EPG Existing
App
100.1.1.3 100.1.1.99 100.1.1.7 100.1.1.5 100.1.1.3

• Single Policy Group (one extended EPG)


• Leverage vPC for interconnect (diagram shows a single port-channel which is an option)
• BPDU should be enabled on the interconnect ports on the ‘vPC’ domain
BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 39
VLAN, EPG and BD

Sub-divide the ’Broadcast Domain’

VLAN 10

Layer 2

VLAN 30
VLAN 10 VLAN 10 VLAN 10
VLAN 20

EPG EPG BD
Outside C Inside Existing
100.1.1.3 100.1.1.99 100.1.1.7 100.1.1.5 App
100.1.1.3

• External EPG (policy between the L2 outside EPG and internal EPG)
• Leverage vPC for interconnect (diagram shows a single port-channel which is an option)
• BPDU should be enabled on the interconnect ports on the ‘vPC’ domain
• L2 outside forces the same external VLAN << fewer operational errors
BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 40
Where did HSRP go?

Anycast Default Gateway

WAN

Anycast Default Gateway

L2 Trunk
10.10.10.3 10.20.20.50

10.10.10.5
10.20.20.7
10.10.10.20
100.1.1.3 100.1.1.99 10.20.20.20
100.1.1.7

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 41
Agenda
▪ Introduction to the Intent/Application Based Data Centre
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 42
Attach the existing legacy network to the ACI fabric via L2
double-sided vPC
Spines

Leafs

North-to-South

East-to-West

Existing Network Infrastructure

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 43
Move the Default Gateway to ACI or...
ACI
EPG1 The VRF and Bridge Domain on ACI becomes
the default gateway for servers
New Servers Attach to ACI
EPG Legacy

Legacy Network
Server A Server B

MAC A MAC B

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 44
Move the Firewall to ACI and Keep Default Gateway on the
Firewall
ACI
EPG1 EPG Firewall MPLS
New Servers Attach to ACI
EPG Legacy

North-to-South Legacy Network

Server A Server B
East-to-West

MAC A MAC B

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 45
Eventually Move the Server NICs to ACI Leafs


Spines
Firewalls

Leafs

Servers

MPLS
Virtual Machines
Legacy Network

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 46
Inside a VRF, Policy Enforcement is a Binary Decision
• Policy Enforce: no communication without contracts
• Policy Unenforced: all communication allowed

VRF – MyVRF VRF – MyVRF


L3Out L3Out

EPG-A EPG-B EPG-C EPG-A EPG-B EPG-C


External External
EPG EPG

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 47
Tools to simplify Contract (ACL) Management During
Migration - vzAny
Tenant ONE Tenant Shared Services

EPG1 VRF1 VRF Services


EPG Shared
Services

vzAny
EPG2

EPG3

Leverage an allow ‘any’ rule - vzAny

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 48
Start off with Contract Preferred Groups

All EPG’s within a preferred group are trusted
VRF – MyVRF

Preferred Group
No need for contracts L3Out
VLAN10 VLAN20 VLAN30
or to understand External
application behaviour EPG
because EPGs are
configured in Preferred
Group

Migrate Network Centric Model – EPG = VLAN

VLAN10 VLAN20

VLAN30

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 49
Simplify Transition from Network Centric (VLAN == EPG) to
Application Centric
VRF – MyVRF

Preferred Group
L3Out
VLAN10 VLAN20 VLAN30
External
EPG

When you identify


endpoints that require Tomcat MySQL
isolation or contracts you
configure more EPG/
uEPG APP-01 VLAN30

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 50
Considerations when Migrating an Existing L2 Network to an
ACI fabric

▪ Make sure how BD forwards broadcasts and BPDU traffic


▪ Understand built-in Loop prevention mechanisms in ACI
▪ Make sure you understand how Spanning-Tree interacts with ACI
▪ Limit the interactions with TCN BPDUs
▪ Harden the Fabric with MCP and with EP move detection
▪ Be careful with ARP (or proxy ARP)

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 51
ACI Fabric Loop Detection
▪ ACI prevents loops from being introduced in the
fabric as follows: APIC

▪ With LLDP, by detecting if a leaf port is connected


to another leaf of the same fabric
▪ With the Miscabling Protocol (MCP)
▪ By detecting MAC moves
▪ With BPDU-guard on physical ports
▪ With BPDU-guard on virtual ports (in the VMM LLDP BPDU
domain) MCP

▪ By forwarding Spanning-Tree BPDUs (but ACI LLDP Loop


doesn't run STP per se) within an EPG Detection MCP Loop
Detection
(supported with STP Loop
11.1 release) Detection

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 52
Enabling MCP Globally


BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 53
Creating an MCP Interface Policy


BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 54
Global MCP Policy Properties
Initial Delay (sec): The delay time before MCP starts taking action based on the value of
Loop Protection Action value configured by users. From the system bootup until the initial
delay timer timeout, the MCP will only create syslog entry if a loop is detected. 180 seconds
by default.

Loop Detect Multiplication Factor: The multiplication factor which MCP uses to determine
when a loop is formed. It denotes the number of continuous packets a port has to receive
before claiming a loop is formed. 3 by default.

Loop Protection Action: This determines how MCP will take action when a loop is
detected. MCP would error-disable the port or send syslog only based on this value.

Transmission Frequency (sec): How often we will send MCP PDUs. 2 seconds by default.

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 55
By default Flooding is BD-wide, not just Restricted to an
EPG

BD1

EPG1 EPG2 leaf1 BD1 leaf3 BD1 BD1 leaf4

EPG1 EPG2 EPG1 EPG2


leaf2
1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8
VLAN 5, 6, 7, 8 VLAN 9,10,11,12
VLAN 5, VLAN 6 VLAN 9, VLAN 10

VLAN 5, VLAN 7

Broadcast VLAN 9, VLAN 11

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 56
BPDU Forwarding Within the EPG is Based on the VLAN

leaf4
leaf1 BD1 leaf3 BD1
BD1

EPG1 leaf2 EPG2 EPG1 EPG2

1/1 1/2 1/3 1/4 1/5 1/6 1/7 1/8

VLAN 5, VLAN 6 VLAN 9, VLAN 10

BPDUs
VLAN 5, VLAN 7

BPDUs
VLAN 9, VLAN 11

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 57
TCNs are Useful in this Topology should the Active Path
Change from switch1 to switch2

leaf4
leaf1 BD1 leaf3 BD1 BD1

EPG1 leaf2 EPG2 EPG1 EPG2

switch1 switch2
STP Root Switch

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 58
But is there a need for TCNs in ACI with this topology? No

Limit STP impact on the fabric with vPC and Portfast/Trunkfast

There is only one L2 path from ACI
Spines to the L2 network outside

Leafs

Servers

Virtual Machines

Existing Network Infrastructure

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 59
With Topologies that Require TCNs, Limit the Impact of
TCNs
▪ When ACI receives a TCN on a VLAN it flushes the endpoints of the BD that are
associated with that VLAN
▪ If you have an EPG that has local endpoints connected
▪ And
▪ Is also connected to the outside via L2
▪ Use a different VLAN for the locally attached endpoints
▪ And
▪ For the L2 extended network

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 60
For legacy networks connected to ACI consider BD set to unknown
unicast flooding First frame is flooded

Spines

MAC A

MAC A

Server 1
MAC 1

Virtual Machines

Existing Network Infrastructure

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 61
Checklist for Migration / Coexistence of Legacy Infrastructure
with ACI
▪ With clustered servers check how do they use ARP:
▪ Check the type of servers that you need to • E.g. if you have MNLB servers you need to use dedicated
connect to ACI: L2 BDs
• If one node is master at any given time is the only one
▪ Which teaming do they use? answering ARP request while all other servers forward
• If they use Transmit Load Balancing change the traffic with the same source IP you may need a dedicated
teaming configuration L2 BD
• If you are using virtualised servers make sure they ▪ Make sure BPDUs are passed through the fabric only if
use mac pinning equivalents or port-channelign necessary
▪ Are servers doing routing? If so you need to ▪ Limit BPDU TCN impact on the fabric to the EPGs/
connected them to a L3out VLANs that really require the TCN

▪ Are servers aggregating multiple clients like a ▪ If there is only one L2 path outside of the ACI fabric,
firewall doing source NAT? enable spanning-tree portfast / trunkfast on the outside
L2 switches
• If yes make sure that the number of source IP per
MAC is within the limits ▪ Set BDs that have outside L2 connectivity to unknown
• Make sure individual IP addresses are aged out unicast flooding unless you can remove BPDU TCNs
independently and you force all migrated servers to be discovered by
ACI

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 62
Where to Go for More Best Practices Information
✓ ACI Design Guide
https://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/application-centric-infrastructure/white-
paper-c11-737909.pdf

✓ About endpoint learning and BD settings


https://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/application-centric-
infrastructure/white-paper-c11-739989.html
✓ About Service Graphs
https://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/application-centric-
infrastructure/white-paper-c11-734298.html
✓ About the use of PBR in a Service Graph
https://www.cisco.com/c/en/us/solutions/data-center-virtualization/application-centric-infrastructure/
white-paper-c11-739971.html

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 63
Agenda
▪ Introduction to the Intent/Application Based Data Centre
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 64
Integrating an Existing DC with OTV


Fabric ‘A’
BD extend via OTV/
VPLS with multiple
L2Out

OTV
EVPN

10.10.10.8 10.20.20.32 10.10.10.9


10.10.10.5 10.10.10.34 Subnet 10.10.10.0/24
DG = 10.10.10.1 DG = 10.10.10.1

• ACI fabrics can connect to any existing layer 2 DCI system, OTV, VPLS, VXLAN with EVPN, …
• Same basic L2 connection as connecting to any other Nexus or Catalyst switch

http://www.cisco.com/c/en/us/solutions/data-center-virtualization/application-centric-infrastructure/
white-paper-c11-737077.html

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 65
Integrating new ACI Fabrics with Existing OTV


Fabric ‘A’ Fabric ‘B’


BD extend via OTV/
VPLS with multiple
L2Out

OTV
EVPN

• ACI fabrics can connect to any existing layer 2 DCI system, OTV, VPLS, VXLAN
with EVPN, …
• Same basic L2 connection as connecting to any other Nexus or Catalyst switch
• Improvements with 11.2 ACI release provide an automatic common default
gateway across both connected fabrics
• Prior to 11.2 manual configuration is required for default GW and ARP config

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 66
Support for the same Pervasive GW IP & MAC on
independent fabrics
Fabric ‘A’ Fabric ‘B’
BD extend via OTV/
VPLS with multiple
L2Out

OTV
EVPN

10.10.10.5 10.10.10.34 Subnet 10.10.10.0/24 10.10.10.57 10.10.10.39


DG = 10.10.10.1 DG = 10.10.10.1 DG = 10.10.10.1 DG = 10.10.10.1

• Support for the same default gateway address on independent fabrics with BD extended
(same pervasive default gateway)
• Same subnet exists in independent fabrics with layer 2 transport (OTV, VPLS, VXLAN
EVPN)
• Common gateway to allow attachment of device with hard coded DG at either site
• Replicates the current vPC HSRP localisation design
BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 67
Removing the Classic L2/L3 Boundaries
DCI is different now

Site 1 DP-ETEP B Site 2


DP-ETEP A
IP Backbone
S1 S2 S3 S4 S5 S6 S7 S8

= VXLAN Encap/Decap
EP1 EP2

Communication between endpoints in separate sites (Layer 2 and/or Layer 3) is enabled


simply by creating and pushing a contract between the endpoints’ EPGs

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 68
ACI & SDWAN Integration

Target 2HCY18
▪ Purpose is to provide end to end path visibility/control to perform application based routing from
the DC to the branch WAN edge
▪ APIC and vManage (DNAC in the future) will exchange group, path and policy requirements
▪ Application traffic will receive appropriate WAN service level

ACI Control/forwarding Segment SDWAN Control/forwarding Segment

1 Internet
Path

Path 2
App A
MPLS

4G LTE
Path
3

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 69
Where to Go for More Information

✓ ACI Stretched Fabric White Paper


http://www.cisco.com/c/en/us/td/docs/switches/datacenter/aci/apic/sw/kb/b_kb-aci-stretched-
fabric.html#concept_524263C54D8749F2AD248FAEBA7DAD78
✓ ACI Multi-Pod White Paper
http://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/application-
centric-infrastructure/white-paper-c11-737855.html?cachemode=refresh
✓ ACI Dual Fabric Design Guide
http://www.cisco.com/c/en/us/solutions/data-center-virtualization/application-centric-
infrastructure/white-paper-c11-737077.html?cachemode=refresh
✓ ACI and GOLF High Level Integration Paper
http://www.cisco.com/c/en/us/solutions/collateral/data-center-virtualization/
application-centric-infrastructure/white-paper-c11-736899.html

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 70
Agenda
▪ Introduction to the Intent/Application Based Data Centre
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 71
Identity and Access Control
 VTEP IP Group
Policy
VNID Tenant Packet

(TrustSec)
Shared Application
Servers
▪ Devices and users are Services

authenticated and
Ingress and
authorised into end-point ACI Fabric
groups (aka EPG’s or
Egress ➔
Enforcement
SGT’s) APIC

▪ End Point Group Tags Enterprise


Backbone
(EPG’s, SGT’s) are
encoded in a VXLAN
header
Campus Switch Campus Switch
▪ Policies between scalable
groups are established
Employee Tag
following the provider/
consumer model Supplier Tag
Non-Compliant Employee Voice Voice Employee Non-Compliant Tag

VLAN A VLAN B

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 72
Identity and Access Control
 VTEP IP Group
Policy
VNID Tenant Packet

(TrustSec)
Shared Application
Servers
▪ Devices and users are Services

authenticated and
Ingress and
authorised into end-point ACI Fabric
groups (aka EPG’s or
Egress ➔
Enforcement
SGT’s) APIC

▪ End Point Group Tags Enterprise


Backbone
(EPG’s, SGT’s) are
encoded in a VXLAN
header
Campus Switch Campus Switch
▪ Policies between scalable
groups are established
Employee Tag
following the provider/
consumer model Supplier Tag
Non-Compliant Employee Voice Voice Employee Supplier Non-Compliant Non-Compliant Tag

VLAN A VLAN B

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 72
ACI Policy and Security Fundamentals 
 Contract
EPGs, Identity and Access Control Filters
Web Tier App Tier
▪ Each EPG is identified by a specific Group Policy End Points End Points
Filter
‘class-id’
Filter
▪ All traffic sourced from an endpoint is identified by the
class-id corresponding to it’s EPG membership Filter

▪ When one endpoint communicates with another


endpoint the Fabric checks that the class-id of the
source is permitted to communicate to the class-id of Provider identified by
Consumer identified by
the destination using the specific ports as defined by the D_Class
the S_Class (Source
the filters in the contract EPG Class)
(Destination EPG
Class)
▪ Communication policy is ‘directional’ (policy checks
both that both Web to App and App to Web are
allowed)
▪ Note: An EPG has a unique class-id, ‘source class’
and ‘destination class’ only refer to relative policy
Independent Policy for
enforcement (which direction is being enforced) reverse direction

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 73
ACI Policy and Security Fundamentals 
 Contract
EPGs, Identity and Access Control Filters
Web Tier App Tier
▪ Each EPG is identified by a specific Group Policy End Points End Points
Filter
‘class-id’
Filter
▪ All traffic sourced from an endpoint is identified by the
class-id corresponding to it’s EPG membership Filter

▪ When one endpoint communicates with another


endpoint the Fabric checks that the class-id of the
source is permitted to communicate to the class-id of Provider identified by
Consumer identified by
the destination using the specific ports as defined by the D_Class
the S_Class (Source
the filters in the contract EPG Class)
(Destination EPG
Class)
▪ Communication policy is ‘directional’ (policy checks
both that both Web to App and App to Web are
allowed)
▪ Note: An EPG has a unique class-id, ‘source class’
and ‘destination class’ only refer to relative policy
Independent Policy for
enforcement (which direction is being enforced) reverse direction

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 73
ACI Multi-Site

Network and Identity Extended between Fabrics
Network information carried across Identity information carried across
Fabrics (Availability Zones) Fabrics (Availability Zones)

VTEP IP VNID Class-ID Tenant Packet No Multicast Requirement in


Backbone, Head-End
Replication (HER) for any
IP Network Layer 2 BUM traffic)

MP-BGP - EVPN

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 74
Federating Identity between Domains
TrustSec-ACI Integration
▪ Sharing Groups between TrustSec and ACI domains began with ISE 2.1
▪ Allow TrustSec (SDA) security groups to be used in ACI policies
▪ Allow ACI EndPoint Groups to be used in policies in TrustSec domain

SDA Policy Domain ACI Policy Domain

Campus / Branch / Non-ACI DC


ISE 2.1
APIC
 Data Centre
TrustSec Policy Domain DC
APIC Policy Domain

Voice Employee Supplier BYOD


ACI Fabric Web App DB
Voice Data
VLAN VLAN

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 75
SDA and ACI

Border Leaf Leveraging IP Based EPG
ISE & APIC
SDA ISE Policy Domain Exchange Groups ACI Policy Domain
and Member
Security Groups information End Point Groups

Cisco APIC-DC

Cisco ISE ISE notifies APIC


about IP endpoint
APIC-EM IP, SGT mappings to EPG mappings IP-ClassId, VNI bindings

SDA & iWAN

L3Out

External EPG
User Switch Router* Classified and ACI Nexus9000 ACI Fabric Server
implements policy Border Leaf
Classification
LISP,SGT & VXLAN

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 76
Extended Access Control with Cisco Firepower
TrustSec & ACI Policy Groups in Access Rule

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 77
Extended Access Control with Cisco Firepower
TrustSec & ACI Policy Groups in Access Rule

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 77
Agenda
▪ Introduction to the Intent/Application Based Data Centre
▪ ACI Foundations
▪ Integrating Existing Nexus Layer 2 and Layer 3
▪ Integrating DCI and WAN
▪ Integrating with Other Security Domains (TrustSec, SDA, …)
▪ What about Public Cloud?

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 78
ACI Anywhere

ACI Network and Policy Extension Multisite Orchestrator

IP Network

vSpine +
vLeaf
AVE AVE AVE AVE
AVE
Hypervisor Hypervisor Hypervisor Hypervisor
Remote Physical Leaf AVE with non-N9K
(N9K)

Site 1 Site 2

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 79
ACI Anywhere
Multi-Cloud Policy and Management Extension

Multisite Orchestrator

IP Network

Site 1 Site 2

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 80
Why use Cloud Constructs?

Policy Mapping - AWS
User Account Tenant
Virtual Private Network VRF

VPC subnet BD Subnet

Security Group EPG

Contracts

Security Group Rule EPG Contracts

Outbound rule Consumed contracts


Source/Destination: Subnet or IP or Any or ‘Internet’
Protocol
Port

Inbound rule Provided contracts


EC2 Instance

Network Adapter

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 81
Why use Cloud Constructs? 

Policy Mapping - Azure
Resource Group Tenant

Virtual Network VRF

Subnet BD Subnet

Application Security Group EPG


(ASG)

Network Security Group


(NSG) EPG Contracts

Outbound rule Consumed contracts


Source/Destination: ASG or Subnet or IP or Any or ‘Internet’
Protocol
Port

Inbound rule Provided contracts


Virtual Machine

Network Adapter

BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 82
Connect the User Domain to the Application
End to End Identity and Intent Based Infrastructure
APIC & ACI Multi-Site

Tetration Analytics
CloudCentre Platform

vPod
Service Service Service
Hypervisor Hypervisor Hypervisor
VM VM VM

DB App Web

Application User Group

Group Based Policy (Intent) 83


BRKACI-2001 © 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public
Q&A

84
Complete Your Online 

Session Evaluation
• Give us your feedback and receive
a Cisco Live 2018 Cap by
completing the overall event
evaluation and 5 session
evaluations.
• All evaluations can be completed
via the Cisco Live Mobile App.

Don’t forget: Cisco Live sessions will be


available for viewing on demand after the
event at www.CiscoLive.com/Global.

© 2018 Cisco and/or its affiliates. All rights reserved. Cisco Public 85
Thank you

86

You might also like