VXLAN Deployment - Use Cases and Best Practices
VXLAN Deployment - Use Cases and Best Practices
Practices
BRKDCT-1301
Azeem Suleman
Contributions
Thanks to the team:
Abhishek Saxena
Mehak Mahajan
Lilian Quan
Bradley Wong
Mike Herbert
Sandeep Subramaniam
BRKDCT-1301
Cisco Public
Abstract
With growing adoption of virtualization in customer environment and large
number of workload mobility requirements in data center; overlays are
becoming key technology. VXLAN is one the overlay technology.
This session will discuss about VXLAN design, deployment and best
practices in Data Center environment. It will also cover:
BRKDCT-1301
Cisco Public
Agenda
Overlays
Introduction to VXLAN
VXLAN Design
Deployment Steps
Troubleshooting
Key Takeaways
References
BRKDCT-1301
Cisco Public
Overlays
Why Overlays?
Robust Underlay/Fabric
BRKDCT-1301
Cisco Public
Overlay Taxonomy
Overlay Control Plane
Encapsulation
Edge Device
Edge Devices
Hosts
(end-points)
Underlay Network
Underlay Control Plane
BRKDCT-1301
Cisco Public
Layer 2 Overlays
Layer 3 Overlays
BRKDCT-1301
Cisco Public
Host Overlays
Integrated Overlays
Fabric DB
V
M
O
S
Physical
Physical
V
M
O
S
Virtual
V
M
O
S
V
M
O
S
A
p
A
p
p
O
S
p
O
S
Virtual
Virtual
Physical
Router/switch end-points
Resiliency + Scale
Traditional VPNs
x-organizations/federation
Open Standards
Tunnel End-points
BRKDCT-1301
Cisco Public
Introduction to VXLAN
VXLAN Overview
Challenges VXLAN addresses:
VM mobility restricted within a VLAN VXLAN encapsulates L2 frame in IPUDP header allowing L2 adjacency across router boundaries
MAC-in-UDP encapsulation
Leverage ECMP to achieve optimal path usage over the transport network
BRKDCT-1301
Cisco Public
VXLAN Overview
VXLAN can be implemented on both Hypervisor-based Virtual Switches to allow for scalable VM deployments,
as well as on Physical switches, which provides the ability to bridge VXLAN segments back into VLAN
segments. In these cases, the Physical Switch instantiates a VTEP, and function as a VXLAN Gateway
VNI 1000
VLAN 10
vSwitch
VTEP
Switch
VTEP
vSwitch
VTEP
VLAN 20
VNI 2000
BRKDCT-1301
Cisco Public
Each of these segments represents a unique Layer 2 broadcast domain, and can be administered
in such a way that it can uniquely identify a given tenants address space or subnet.
Outer
Ethernet
Outer
IP
Outer
UDP
VXLAN
Ethernet
Header
Payload
FCS
Inner
Ethernet
Payload
New
FCS
8 Bytes
Flags
Reserved
1 Byte
Rsvd
BRKDCT-1301
Rsvd
Instance ID
Reserved
Outer UDP Destination Port = VXLAN (originally 8472, recently updated to 4789)
Outer UDP Source Port = Hash of Inner Frame Headers (optional)
2014 Cisco and/or its affiliates. All rights reserved.
Cisco Public
Transport VLAN
Outer
MAC
DA
Outer
MAC
SA
Outer
802.1Q
Outer
IP DA
A Layer-2 Gateway
bridges traffic to VLAN
based on MAC DA
A Layer-3 Gateway
routes traffic to VLAN
based on IP DA
Outer
IP SA
Outer
UDP
VXLAN ID
(24 bits)
Inner Inner
MAC MAC
SA
DA
Identifies packet as a
VXLAN packet
BRKDCT-1301
CRC
Outer
IP Header
BRKDCT-1301
VNID
Reserved
32
Reserved
32
VXLAN
RRRR1RRR
16
Checksum
0x0000
Outer
Dst. IP
UDP
Length
Outer
Src. IP
72
8 Bytes
VXLAN
Port
Header
Checksum
16
FCS
8 Bytes
Protocol
0x11
16
Original
FCS L2 Frame
20 Bytes
IP Header
Misc Data
16
Ether
Type
0x0800
VLAN ID
Tag
Src.
MAC
Addr.
48
VLAN Type
0x8100
Dst.
MAC Addr.
14 Bytes
(4 bytes optional)
48
VXLAN
Header
UDP Header
UDP
Src. Port
Outer
Mac Header
16
16
16
16
24
24
Cisco Public
BUM
BRKDCT-1301
Cisco Public
VXLAN VTEP
VXLAN terminates its tunnels on VTEPs (Virtual Tunnel End Point).
Each VTEP has two interfaces - one to provide bridging function for local hosts, the
other has an IP identification in the core network for VxLAN encapsulation/deencapsulation.
Transport IP Network
VTEP
BRKDCT-1301
VTEP
IP Interface
IP Interface
End System
End System
End System
2014 Cisco and/or its affiliates. All rights reserved.
End System
Cisco Public
End
System
VTEP-3
VTEP 3
IP-3
Mcast Group
VTEP-1
VTEP-2
IP Network
End System A
MAC-A
IP-A
BRKDCT-1301
VTEP 1
IP-1
VTEP 2
IP-2
End System B
MAC-B
IP-B
Cisco Public
VXLAN Overview
VTEP Discovery & Address Learning
End System
End System
3
VTEP3
VTEP 3
IP-3
S-MAC: MAC-1
D-MAC:
00:01:5E:01:01:01
VxLAN ID
Remote VTEP
10
IP-1
UDP
VXLAN VNID: 10
7
ARP Response from IP
B
Src MAC: MAC-B
Dst MAC: MAC-A
End System A
MAC-A
IP-A
ARP Request for IP B
Src MAC: MAC-A
Dst MAC: FF:FF:FF:FF:FF:FF
BRKDCT-1301
VTEP 1
IP-1
MAC-1
MAC
6
VxLAN ID
Remote VTEP
10
IP-2
Address
MAC-B
VTEP 2
IP-2
MAC-2
Mcast Group
2 239.1.1.1
VTEP-1
VTEP-2
S-MAC: MAC-2
D-MAC: MAC-1
Outer S-IP: IP-2
Outer D-IP: IP-1
UDP
VXLAN VNID: 10
ARP Response
from IP B
Src MAC: MAC-B
Dst MAC: MAC 2014 Cisco and/or its affiliates. All rights reserved.
A
End System B
MAC-B
IP-B
VxLAN ID
Remote VTEP
10
IP-1
MAC-A
Cisco Public
VXLAN Overview
Unicast Forwarding Packet Flow
Outer S-MAC: MAC-3
Outer D-MAC: MAC-4
Routed Based on
Outer IP header
Router-1
S-MAC: MAC-A
D-MAC: MAC-B
VTEP-1
S-MAC: MAC-A
D-MAC: MAC-B
S-IP: IP-A
D-IP: IP-B
UDP
VXLAN VNID: 10
IP Network
VXLAN VNID: 10
S-IP: IP-A
D-IP: IP-B
MAC-2
IP-2:
165.123.1.2
S-MAC: MAC-A
D-MAC: MAC-B
Router-2
MAC-3
IP-3:
140.123.1.2
MAC-1
IP-1:
165.123.1.1
S-IP: IP-A
D-IP: IP-B
4
MAC-4
IP-4:
140.123.1.1
S-IP: IP-A
D-IP: IP-B
Host-A
MAC-A
IP-A:
10.1.1.100
BRKDCT-1301
VTEP-2
S-MAC: MAC-A
D-MAC: MAC-B
Host-B
MAC-B
IP-B:
10.1.1.101
Cisco Public
Supported Platforms
Platform
NX-OS version
Nexus 9500
Nexus 9300
Nexus 3100 (3132/3172)
Nexus 6000
Nexus 7000 with F3
Minimum
Recommended
6.1.2I3.1.x
6.1.2I2.1.x
6.x
7.0(0)N1x)
7.0.x
6.1.2.I3.1.x
6.1.2.I2.1.x
Q2 CY14
Q3 CY14
Q4 CY14
Cisco Public
VXLAN ID
10
1010
20
1020
VxLAN VTEP
VLAN 10
VLAN 20
(VxLAN Gateway)
L3
Network
VxLAN
VTEP
VxLAN
VTEP
VNI
1010
BRKDCT-1301
VNI
1020
Cisco Public
VTEP
interface et4/13
switchport
switchport access vlan 10
no shut
interface nve1
no shutdown
source-interface loopback0
overlay-encapsulation vxlan
member vni 1010 mcast-group 230.1.1.1
VxLAN ID
Remote
VTEP
AA:AA:AA:AA:AA:AA
1010
10.1.1.2
BB:BB:BB:BB:BB:BB
1020
10.1.1.3
vlan 10
vn-segment 1010
switch# show nve vni
Interface
VNI
Multicast-group
VNI State
--------------------------------------------nve1
1010
230.1.1.1
up
switch# show nve peers
Interface
Peer-IP
VNI
Up Time
----------------------------------------nve1
10.1.1.2
1010
00:52:24
switch#
BRKDCT-1301
VXLAN Encap
VXLAN ID
10
1010
20
1020
Ethernet/ 802.1Q
Cisco Public
VXLAN ID
10
1010
20
1020
VxLAN VTEP
VLAN 10
VLAN 20
(VxLAN Bridging)
L3
Network
VxLAN
VTEP
VxLAN
VTEP
VNI
1010
BRKDCT-1301
VNI
1020
Cisco Public
VXLAN
VLAN
ALE
NFE
T2-96
Encap/Decap
ASIC
VLAN
BRKDCT-1301
VXLAN
Cisco Public
Bridging and
Gateway
mode
support
Any to Any
Behavior
Fabric 1
NFE
NFE
Fabric 2
NFE
NFE
Fabric 3
NFE
NFE
Fabric 4
NFE
NFE
Fabric 5
NFE
NFE
Fabric 6
NFE
NFE
ALE
ALE
ALE
ALE
ALE
ALE
NFE
ASIC
T2
Decap
NFE
ASIC
Encap
T2
NFE
ASIC
Decap
T2
BRKDCT-1301
Cisco Public
VLAN 20
VLAN 10
L3
Network
VxLAN
VTEP
VxLAN
VTEP
VNI
1010
BRKDCT-1301
VNI
1020
Cisco Public
ALE
Insieme
Recirculate
Insieme
Recirculate
NFE
Route
Packet
VLAN
Subnet 10.20.20.0/24
BRKDCT-1301
NFE
Route
Packet
Encap/
Decap
VXLAN
Subnet 10.10.10.0/24
VLAN
Subnet 10.20.20.0/24
Cisco Public
Encap/
Decap
VXLAN
Subnet 10.10.10.0/24
Fabric 3
T2
T2
Fabric 2
NFE
NFE
LPM IP Route
ALE
IP
ASIC
Route
VLAN
BRKDCT-1301
ALE
NFE
IP
Route
ALE
VXLAN
Route
Encap
Decap
VXLAN
Fabric 6
Fabric 5
NFE
NFE LPM
T2 IP Route T2
Fabric 4
NFE
NFE
VXLAN
ALE
IP
Route
NFE
VLAN
VXLAN
Route
Decap
VLAN
Cisco Public
If required LPM
route lookup on
Fabric
/32 route
lookup on
ingress T2
VXLAN Design
VXLAN Designs
Explore a variety of VXLAN designs and evaluate how they meet key design
criteria
Assumption is the choice to go Overlay/VXLAN has already been made
Objective is not to debate "Layer 2 vs. Layer 3" or "why overlay in the Data Center?"
BRKDCT-1301
Cisco Public
VXLAN Designs
High-level design options considered in this presentation are in the following areas:
Routed Access + IP Mobility
L2 extension across Pod / Multi-tenancy
Datacenter Interconnect (DCI)
BRKDCT-1301
Cisco Public
Classic Design
Simplest design option
Extension of traditional aggregation/access designs in
spine / leaf
L3
SVIs
SVIs
Immediate benefits:
L2/L3 boundary
Simplified configuration
Extension of VLANS
Traffic distribution using ECMP
IP Network
IP Mobility
BRKDCT-1301
Cisco Public
VTE
P
VTEP
VXLAN
Enabled
Hypervisor
VXLAN
Enabled
Hypervisor
BRKDCT-1301
Cisco Public
VXLAN Forwarding
vPC
VXLAN
AnyCast
VTEP
AnyCast
VTEP
VLAN
Cisco Public
VXLAN Forwarding
Design Considerations
VXLAN
VTEP
VLAN
Cisco Public
L3 Core
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VXLAN Overlay
(VLAN Extension)
Pod 2
Pod 1
IP GW
IP GW
Layer-2 VLAN Domain
BRKDCT-1301
Cisco Public
Nexus
VXLAN Overlay
Routing Between VXLAN
VLAN Subnets
(Bridging)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
BRKDCT-1301
Cisco Public
Spine
VXLAN Overlay
(Routing)
(Bridging)
Leaf
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
BRKDCT-1301
Cisco Public
Nexus
VXLAN Overlay
Routing Between VXLAN
Extended VLANs
(Bridging)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Bridging
+
Routing)
VTEP
(Layer-2 only)
VLAN BLUE
VLAN GREEN
VLAN GREEN
VLAN BLUE
BRKDCT-1301
Cisco Public
Nexus
VXLAN Overlay
Routing Between
VLAN A & B Subnets
(Bridging)
VTEP
(Layer-2 only for VLAN A & B))
VTEP
(IP GW for VLAN A
Learn other VLAN/VXLAN subnets)
VTEP
(IP GW for VLAN B
Learn other VLAN/VXLAN subnets)
BRKDCT-1301
Cisco Public
Spine
VXLAN Overlay
(Routing)
(Bridging)
Leaf
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
VTEP
(Layer-2 only)
BRKDCT-1301
Cisco Public
Spine
VXLAN Overlay
(Routing)
(Bridging)
Leaf
VTEP
(IP GW for VLAN A
L2 for Other VLANs)
VTEP
(IP GW for VLAN B
L2 for Other VLANs)
VTEP
(IP GW for VLAN C
L2 for Other VLANs)
VTEP
(IP GW for VLAN D
L2 for Other VLANs)
BRKDCT-1301
Cisco Public
Nexus
Routing to Outside
VXLAN Overlay
Routing Between
VXLAN VLAN Subnets
(Bridging)
VTEP
(IP GW for All VXLAN VLANs)
VTEP
(IP GW for All VXLAN VLANs)
VTEP
(IP GW for All VXLAN VLANs)
BRKDCT-1301
Cisco Public
Spine
VXLAN Overlay
Leaf
(Bridging)
Routing to Outside
Routing Between
VXLAN VLAN Subnets
VTEPs
(IP GW for All VXLAN VLANs)
BRKDCT-1301
Cisco Public
IP
Transport
N9000
VTEP
VTEP
VTEP
HOST
HOST
BRKDCT-1301
Cisco Public
N9000
VTEP
N9000
VTEP
FEX
FE
X
Hypervisor
Hypervisor
Hypervisor
VTEP
Hypervisor
BRKDCT-1301
N9000
VTEP
N9000
VTEP
FEX
FE
X
Hypervisor
VTEP
Cisco Public
Hypervisor
VTEP
VXLAN Configuration
Enabling VXLAN feature
switch(config)# [no] feature nv overlay
Enable Multicast
switch(config)# [no] feature pim
Make sure source interface and next hop L3 interface should have multicast enabled
switch(config)# interface loopback 0
switch(config-if)# ip pim sparse-mode
BRKDCT-1301
Cisco Public
VXLAN Configuration
Default UDP Port is 4789
Configure VXLAN UDP Port
switch(config)# [no] vxlan udp port <number>
BRKDCT-1301
Cisco Public
BRKDCT-1301
Cisco Public
Lpbk0:
100.100.100.1/32
.1
e2/1
.2
20.1.1.0/30
30.1.1.0/30
.1
e2/1
e1/1
e1/1
Host A
(VLAN 10)
BRKDCT-1301
Host B
(VLAN 10)
Cisco Public
Lpbk0:
100.100.100.2/32
BRKDCT-1301
Cisco Public
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
RP: 10.1.1.1
Host B
(VLAN 10)
BRKDCT-1301
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
RP: 10.1.1.1
Host A
(VLAN 10)
Cisco Public
Host B
(VLAN 10)
BRKDCT-1301
Cisco Public
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
RP: 10.1.1.1
Host B
(VLAN 10)
BRKDCT-1301
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
RP: 10.1.1.1
Host A
(VLAN 10)
Cisco Public
Host B
(VLAN 10)
.2
.6
20.1.1.4/30
e1/1
vPC
Loop back 0:
200.200.200.2/32
100.100.100.1/32 (Secondary)
Host A
(VLAN 10)
BRKDCT-1301
.2
.5
30.1.1.0/30
.1
e2/1
Cisco Public
Loop backk0:
100.100.100.2/32
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
vPC
Host A
(VLAN 10)
Cisco Public
Host B
(VLAN 10)
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
vPC
Host A
(VLAN 10)
BRKDCT-1301
Cisco Public
Host B
(VLAN 10)
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
vPC
Host A
(VLAN 10)
Cisco Public
Host B
(VLAN 10)
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
BRKDCT-1301
vPC
Host A
(VLAN 10)
Cisco Public
Host B
(VLAN 10)
BRKDCT-1301
L3 Transport Network
(OSPF and IP PIM)
Multicast group:
230.1.1.1
vPC
Host A
(VLAN 10)
Cisco Public
Host B
(VLAN 10)
VXLAN Troubleshooting
Troubleshooting
NVE Interface State
Restrictions
Only 1 NVE local interface is supported
VXLAN gateway shouldnt be connected to multiple core routers via a layer 2 switch
Support for downstream multicast receivers
BRKDCT-1301
Cisco Public
Troubleshooting
Verify Overlay Peer
switch# show nve peers
Interface
Peer-IP
---------------------------nve1
10.1.1.2
VNI
-------10000
Up Time
------00:52:24
VNI
-------10000
Up Time
------00:52:24
VNI
-------10000
Up Time
------00:52:24
BRKDCT-1301
Multicast-group
--------------230.1.1.1
VNI State
--------up
Cisco Public
Troubleshooting (Cont.)
Verify MAC address is learnt
Cisco Public
Troubleshooting (Cont.)
Few Debug commands
switch# debug nve errors
switch# debug nve events
switch# debug nve all
switch# debug nve pim-library
BRKDCT-1301
Cisco Public
Alternatives:
BGP Control Plane
LISP Control Plane
API
BRKDCT-1301
Cisco Public
VXLAN is efficient
Proper utilization of ECMP
Optimal path between any two nodes
VXLAN is scalable
Can extend a bridged domain without extending the risks generally associated with
Layer 2 and beyond 4K VLAN limit
BRKDCT-1301
Cisco Public
Provides multiple design options to help you build a network that meets your
requirements
BRKDCT-1301
Cisco Public
References
VXLAN Overview: Cisco Nexus 9000 Series Switches
http://www.cisco.com/en/US/prod/collateral/switches/ps9441/ps13386/whitepaper-c11-729383_ns1261_Networking_Solutions_White_Paper.html
* Cisco Nexus 9000 Series NX-OS VXLAN Configuration Guide
VXLAN: A Framework for Overlaying Virtualized Layer 2 Networks over Layer 3
Networks
http://www.ietf.org/id/draft-mahalingam-dutt-dcops-vxlan-07.txt
BRKDCT-1301
Cisco Public
Call to Action
Visit the World of Solutions: Cisco Campus
Walk-in Labs
Technical Solutions Clinics
Meet the Engineer
Cisco Public
74
Presentation_ID
Cisco Public
75