JEX 11.a C7 HighAvailability

Junos Enterprise Switching

Chapter 7: High Availability

© 2011 Juniper Networks, Inc. All rights reserved. | www.juniper.net | Worldwide Education Services

Chapter Objectives  After successfully completing this chapter, you will be

able to: • List and describe features that promote high availability • Configure and monitor high availability features • Describe key concepts and components of a Virtual Chassis • Explain the operational details of a Virtual Chassis • Implement a Virtual Chassis and verify its operations

© 2011 Juniper Networks, Inc. All r ights reserved.

Worldwide Education Services

www.junip er.net | 7-2

Agenda: High Availability Overview of High Availability Networks  Link Aggregation Groups  Redundant Trunk Groups  Virtual Chassis

• Overview of Virtual Chassis • Configuring and Monitoring a Virtual Chassis




What Is High Availability?  The ability to ensure a high degree of operational

continuity to a resource for a user community • Typically measured over a specific period of time




High Availability Networks  High availability networks include provisions that allow

for the maximum amount of uptime and accessibility 1. Is the base functionality of the network’s hardware, software, protocols, and overall operations protected? 2. When a failure occurs, how will the overall system or network react and what level of recovery can be expected?




High Availability Features  EX Series switches support a number of features to

help increase uptime and availability including: • Link aggregation groups (LAGs) • Redundant Trunk Groups (RTG) • Virtual Chassis




Agenda: High Availability  Overview of High Availability Networks Link Aggregation Groups  Redundant Trunk Groups  Virtual Chassis





What Is It?  Link aggregation combines multiple Ethernet

interfaces in to a single link layer interface, also known as a link aggregation group (LAG) or bundle • Defined in the 802.3ad standard Participating interfaces are known as member links

Commonly used on trunk port connections Switch-1

ge-0/0/12

Switch-2

ge-0/0/13




Benefits of 802.3ad Link Aggregation  Benefits of 802.3ad link aggregation include:

• Increases bandwidth • Provides link efficiency • Creates physical layer redundancy

Switch-1

ge-0/0/12

Switch-2

ge-0/0/13




Link Requirements and Considerations  Interface requirements and considerations include:

• Duplex and speed must match • Up to eight member links per LAG • Member links do not need to be contiguous ports nor must

they be on the same switch when part of a Virtual Chassis Switch-2

Switch-1 ge-0/0/1 … … … ge-0/0/5 ge-0/0/10 … … ge-0/0/13




Traffic Processing and Forwarding  Traffic processing and forwarding considerations:

• RE generated traffic is always sent on the lowest member link • IP traffic hashing uses Layer 2, Layer 3, and Layer 4 details • Non-IP traffic hashing uses source and destination MAC addresses RE traffic

Switch-2

Switch-1 ge-0/0/1

IP traffic Non-IP traffic ge-0/0/13




Link Aggregation Control Protocol  LACP performs link monitoring and controls the

member links that form a single logical channel  You can set the LACP mode as active or passive:

• Active mode initiates transmission of LACP packets • Passive mode responds to LACP packets Switch-1 ae1

Switch-2 ae1

ae0

ae0

Note: At least one device must be configured for active mode for the link to become operational. © 2011 Juniper Networks, Inc. All r ights reserved.



Implementing LAGs (1 of 2)  Create an aggregated Ethernet interface: {master:0}[edit chassis] user@Switch-1# run show interfaces terse | match ae0 {master:0}[edit chassis] user@Switch-1# set aggregated-devices ethernet device-count 1 {master:0}[edit chassis] user@Switch-1# commit configuration check succeedscommit complete {master:0}[edit chassis] user@Switch-1# run show interfaces terse | match ae0 ae0 up down

Link state remains down until operational member links are added to LAG

Switch-2

Switch-1 ge-0/0/12 ae0

ae0 ge-0/0/13




Implementing LAGs (2 of 2)  Configure the aggregated Ethernet interface and

associate desired member links with the LAG: {master:0}[edit interfaces] user@Switch-1# set ae0 unit 0 f amily ethernet-swit ching {master:0}[edit interfaces] user@Switch-1# set ae0 aggregated-ether-options lacp active {master:0}[edit interfaces] user@Switch-1# set ge-0/0/12 ether-options 802.3ad ae0 {master:0}[edit interfaces] user@Switch-1# set ge-0/0/13 ether-options 802.3ad ae0 {master:0}[edit interfaces] user@Switch-1# commit configuration check succeedscommit complete {master:0}[edit interfaces] user@Switch-1# run show interfaces ge-0/0/12.0 up up ge-0/0/13.0 up up ae0 up up ae0.0 up up


terse | match ae0 aenet --> ae0.0 aenet --> ae0.0

Switch-2

Switch-1 ge-0/0/12 ae0

ae0 eth-switch


ge-0/0/13


Monitoring LAGs  Use the

show interfaces output to determine

state information for aggregated interfaces: {master:0} user@Switch-1> show interfaces terse | match ae0 ge-0/0/12.0 up up aenet --> ae0.0 ge-0/0/13.0 up up aenet --> ae0.0 ae0 up up ae0.0

up

up

eth-switch

{master:0} user@Switch-1> show interfaces extensive ae0.0 | LACP Statistics: LACP Rx LACP Tx ge-0/0/12.0 26 516 ge-0/0/13.0 25 519 Marker Statistics: Marker Rx Resp Tx ge-0/0/12.0 0 0 ge-0/0/13.0 0 0 Protocol eth-switch, Generation: 195, Route Flags: None

find "LACP Statistics:" Unknown Rx Illegal Rx 0 0 0 0 Unknown Rx Illegal Rx 0 0 0 0 table: 0

• Note that other commands exist that show similar details




Test Your Knowledge  Given the sample outputs, can you guess what might

be causing ae0.0 to remain in the down state? {master:0} lab@Switch-1> show interfaces terse | match ae0 ge-0/0/12.0 up up aenet --> ae0.0 ge-0/0/13.0 up up aenet --> ae0.0 ae0 up down ae0.0

up

down eth-switch

{master:0} lab@Switch-1> show interfaces extensive ae0.0 | find "LACP Statistics:" LACP Statistics: LACP Rx LACP Tx Unknown Rx Illegal Rx ge-0/0/12.0 0 224 0 0 ge-0/0/13.0 0 223 0 0 Marker Statistics: Marker Rx Resp Tx Unknown Rx Illegal Rx ge-0/0/12.0 0 0 0 0 ge-0/0/13.0 0 0 0 0 Protocol eth-switch, Generation: 195, Route table: 0 Flags: None

• The remote device does not have LACP enabled




Agenda: High Availability  Overview of High Availability Networks  Link Aggregation Groups Redundant Trunk Groups  Virtual Chassis





What Is It?  A redundant trunk group provides a quick and simple

failover mechanism for redundant Layer 2 links • Can be used as a replacement for STP on access switches that are connected to two aggregation switches Switch-1

ge-0/0/1

Switch-2

RSTP ge-0/0/2

RTG Switch-3


Switch-4


Switch-5


How Does It Work?  Only active link is used to forward traffic, other links

serve as backup links and do not forward traffic • When active link fails, a backup link becomes active

Switch-1

ge-0/0/1

Switch-2

ge-0/0/2

Switch-3




Configuration Considerations  RTG is typically only configured on access switches

• RTG and STP are mutually exclusive on a given port • STP BPDUs received on RTG links are discarded • STP is configured on aggregation switches Switch-1 RSTP

ge-0/0/1

Switch-2

ge-0/0/2

ge-0/0/8

ge-0/0/10

RTG Switch-3

Switch-4

Switch-5

Note: A maximum of 16 redundant trunk groups per switch. © 2011 Juniper Networks, Inc. All r ights reserved.



Case Study: Topology and Objectives  Objectives:

• Implement RTG on Switch-3 to ensure that only a single path is available toward the aggregation switches • Ensure that Switch-3 forwards user traffic out ae0.0 whenever it is operational Switch-1

Switch-2

Switch-3




Case Study: Configuring RTG  You configure RTG under the [edit ethernet-

switching-options] hierarchy: {master:0}[edit ethernet-switching-options] user@Switch-3# set redundant-trunk-group group rtg-1 interface ae0.0 primary {master:0}[edit ethernet-switching-options] user@Switch-3# set redundant-trunk-group group rtg-1 interface ge-0/0/10.0 {master:0}[edit ethernet-switching-options] user@Switch-3# show redundant-trunk-group { group rtg-1 { interface ge-0/0/10.0; interface ae0.0 { primary; } } }

Interface marked as primary is always active when operational

Switch-1

Switch-2

Switch-3

Note: If primary knob is omitted, highest-numbered interface initially becomes active link but does not preempt lower-numbered interfaces functioning as the active link in failure and recovery scenarios © 2011 Juniper Networks, Inc. All r ights reserved.



Case Study: Monitoring RTG  Use

show redundant-trunk-group to monitor

RTG operations: {master:0} user@Switch-3> show redundant-tr unk-group Group Interface State Time of last flap name rtg-1

ae0.0 Up/Pri/Act ge-0/0/10.0 Up

Flap count

Never Never

0 0

(Pri) = Primary interface with preemption enabled (Act) = Active interface currently forwarding traffic Switch-1

Switch-2

ae0.0 is selected as the primary interface for the rtg-1 RTG group

Switch-3 © 2011 Juniper Networks, Inc. All r ights reserved.



Lab 6: Implementing LAGs and RTG  Configure and monitor LAGs and RTG.




Agenda: High Availability  Overview of High Availability Networks  Link Aggregation Groups  Redundant Trunk Groups Virtual Chassis Overview

of Virtual Chassis

• Configuring and Monitoring a Virtual Chassis




What Is It?  Two or more interconnected EX Series switches

operating as a single Virtual Chassis system  Valid chassis combinations

• Two or more (up to 10) EX4200s • Two EX4500s • Two EX4500s with up to eight EX4200s • Two EX8200s including an external routing engine

Up to 10 switches can be interconnected © 2011 Juniper Networks, Inc. All r ights reserved.



Benefits of Virtual Chassis  Virtual Chassis benefits include:

• High availability: Redundant REs • Enables the use of NSR

• Simplified network design: Single network entity to manage,

configure, monitor;switches potential elimination of STP; flexible placementand of member




Virtual Chassis Components  Virtual Chassis components include:

• EX Series switches • Participating switches serve as REs or line cards (LCs)

• Virtual Chassis ports (can be dedicated VCPs or uplink ports

converted to extended VCPs through configuration)

• VCPs interconnect PFEs from one switch to another to form a single backplane

Location 1


Location 2



Virtual Chassis Cabling Options (1 of 3)  Dedicated Virtual Chassis daisy-chained ring method

• Longest cable spans the entire Virtual Chassis; maximum length between end systems is 5 meters

Note: The EX4200 Series switches come with a .5 meter cable. You can acquire longer cables in lengths of 1, 3, and 5 meters. © 2011 Juniper Networks, Inc. All r ights reserved.



Virtual Chassis Cabling Options (2 of 3)  Dedicated Virtual Chassis braided ring method

• Maximum length between end devices in a Virtual Chassis consisting of 10 EX Series switch is 22.5 meters




Virtual Chassis Cabling Options (3 of 3)  Extended Virtual Chassis ring method

• Maximum circumference of ring using is 100 km; uses 1 GbE or 10 GbE uplinks to extend the distance of the Virtual Chassis

This option is required when the circumference of a Virtual Chassis exceeds 22.5 meters. This option is often used to interconnect wiring closets or data center racks or rows. © 2011 Juniper Networks, Inc. All r ights reserved.



Recommended RE Placement (1 of 2)  RE placement recommendations:

• Ensure spatial redundancy and avoid split Virtual Chassis issues and other failure conditions by evenly spacing the master and backup REs (by hop) and by separating REs from uplinks Uplink Uplink

Uplink

Uplink

Dedicated Virtual Chassis daisy-chained ring


Dedicated Virtual Chassis braided ring



Recommended RE Placement (2 of 2)  RE placement recommendations (contd.):

• When implementing an extended Virtual Chassis, you should place the master and backup REs in different locations

Location 2

Location 1

Extended Virtual Chassis ring




Master and Backup Election  Mastership determination: 1. Member with the highest user-configured priority • Priority range is 1–255, factory-default value is 128

2. Member previously functioning as master prior to reboot 3. Member with the longest standing uptime • Difference must be greater than 1 minute

4. Member with the lowest MAC address

• Used as tie breaker if all is equal through the first three determination steps

5. Second member in election decision tree becomes backup switch; all other members are line cards Note: If a master or backup fails, one of the line card switches is elected the new backup switch using the same criteria.




Member ID and Interface Names  All member switches are assigned a member ID (0 –9)

• Member ID is assigned manually through configuration or dynamically from the master switch (usually member ID 0) • Member ID is preserved through reboots • Member ID serves as slot number for interface naming




Replacing a Member Switch  When a member switch is removed, its member ID is

not automatically released and made available nor will the replacement switch automatically inherit the configuration associated with the previous switch • Recommended replacement steps: 1.

Recycle the member ID of the switch being replaced so it becomes the next-lowest available unused member ID

2.

Add replacement switch which should automatically be assigned the recycled member ID and inherit the required configuration

{master:0} user@Switch-1> request virtual-chassis recycle member-id

Note: You recycle member IDs on the master switch. © 2011 Juniper Networks, Inc. All r ights reserved.



Think About It!  Given the member ID assignments, what are the

interface names for the highlighted interfaces? • Hint: All member switches are 48 port models and all interfaces are Gigabit Ethernet interfaces Location #2




Management Connectivity (1 of 2)  Single management interface and IP address

• Individual management Ethernet ports (me0) on member switches are tied to a special management VLAN associated with a Layer 3 virtual management Ethernet (VME) interface • The Virtual Chassis system is managed as a single network element; therefore, it has only one management IP address {master:0}[edit] user@Switch-1# show interfaces vme unit 0 { family inet { address 10.210.14.148/27; } } {master:0}[edit] user@Switch-1# run show interfaces terse vme Interface Admin Link Proto Local vme up up vme.0 up up inet 10.210.14.148/27



Remote


Management Connectivity (2 of 2)  Single virtual console

• Connection to a console on any member switch in a Virtual Chassis system is redirected to the master switch by the virtual console software running on all member switches




Software Upgrades  You can perform software upgrades for a single

Virtual Chassis member or for all members from the master switch {master:0} user@Switch-1> request system software add member ? Possible completions: Install package on VC Member (0..9)

 Master switch performs a software compatibility

check for each member added to the Virtual Chassis • If a version mismatch exists, a syslog message is generated and the newly added switch is placed in an inactive state • Upgrade incompatible switches from the master switch manually or enable the automatic software upgrade option




Topology Discovery (1 of 3)  Virtual Chassis members use VCCP to create a loop-

free topology • LSA-based discovery messages are exchanged between all PFEs and build the member switch and PFE topology maps • Each switch runs the shortest-path first (SPF) algorithm for each PFE which creates PFE map tables between all PFEs • Each PFE builds source ID egress filter tables used to prevent broadcast and multicast packets from looping Switch-1

Switch-2

Switch-3

VCCP LSA-based messages Virtual Chassis cables and backplane




Topology Discovery (2 of 3)  Topology discovery example:




Topology Discovery (3 of 3)  Topology discovery example (contd.):




Inter-Chassis Packet Flow  Packets always take the shortest path through a

Virtual Chassis • Shortest path is determined by hop count and bandwidth • Inter-chassis packet flow examples: 1.

Packets going from ge-0/0/10 to ge-3/0/14 pass through member 4 because it is a shorter path

2.

Packets going from ge-0/0/28 to ge-2/0/47 pass through member 1 because it is a shorter path




Agenda: High Availability  Overview of High Availability Networks  Link Aggregation Groups  Redundant Trunk Groups Virtual Chassis

• Overview of Virtual Chassis Configuring

and Monitoring a Virtual Chassis




Virtual Chassis Configuration  You enable Virtual Chassis configuration options under

the [edit virtual-chassis] hierarchy: {master:0}[edit virtual-chassis] user@Switch-1# set ? Possible completions: + apply-groups Groups from which to inherit configuration data + apply-groups-except Don't inherit configuration data from these groups > Auto > auto-sw-update fast-failover Fast software failover update mechanism id Virtual Chassis identifier, of type ISO system-id > mac-persistence-timer How long to retain MAC address when member leaves Virtual Chassis > member Member of Virtual Chassis configuration no-split-detection Disable split detection. Only recommended in a 2 member setup preprovisioned Only accept preprovisioned members > traceoptions Global tracing options for Virtual Chassis

• To minimize traffic interruption during an RE failover scenario, enable graceful Routing Engine switchover: {master:0}[edit chassis] user@Switch-1# set redundanc y graceful-swit chover? Possible completions: > graceful-switchover Enable graceful switchover on supported hardware © 2011 Juniper Networks, Inc. All r ights reserved.



Dynamic Configuration Process (1 of 2)  Dynamic configuration steps:

1. Install desired master switch: •

Power up desired master switch, switch becomes master and obtains member ID 0, assign mastership priority 255

Master (Active RE) ON

Backup (Backup RE) ON

Linecard

2. Add desired backup switch: •

Connect to master switch using Virtual Chassis cable, power up desired backup switch, switch is elected as backup and dynamically assigned member ID 1, assign mastership priority 255

Linecard

Linecard

{master:0}[edit virtual-chassis] user@Switch-1# set member mastership-pr iority © 2011 Juniper Networks, Inc. All r ights reserved.



Dynamic Configuration Process (2 of 2)  Dynamic configuration steps (contd.): Master (Active RE)

3. Add line card switch: •

ON

Connect to switch above with VCB cable, power up third switch —switch becomes line card and is dynamically

Backup (Backup RE)

assigned member ID 2, assign desired mastership priority

Linecard

ON

ON

4. Repeat Step 3 to add subsequent line card switches: •

Last line card switch completes loop by connecting with master

Linecard ON

Linecard ON

{master:0}[edit virtual-chassis] user@Switch-1# set member mastership-pr iority © 2011 Juniper Networks, Inc. All r ights reserved.



Preprovisioned Configuration Example {master:0}[edit virtual-chassis] user@Switch-1# show preprovisioned; member 0 { role routing-engine; serial-number BM0208105168; } member 1 { role line-card; serial-number BM0208124111;

Note when preprovisioned option is used, you do not specify a mastership priority but rather only assign the role to a given device.

} member 2 { role routing-engine; serial-number BM0208124231; } member 3 { role line-card; serial-number BM0208124333; }

Note: You should power on the switch designated as RE0 first, create and activate the desired preprovisioned configuration, and then add the remaining switches. © 2011 Juniper Networks, Inc. All r ights reserved.



Monitoring Virtual Chassis Operations  Use the

show virtual-chassis commands to

monitor Virtual Chassis operations: {master:0} user@Switch-1> show virtual-chassis ? Possible completions: <[Enter]> Execute this command active-topology Virtual Chassis active topology device-topology PFE device topology fast-failover Fast failover status login protocol Show Virtual Chassis protocol information status Virtual Chassis information vc-path Show virtual-chassis packet path vc-port Virtual Chassis port information | Pipe through a command




Verifying Virtual Chassis Port State  Use

show virtual-chassis vc-port to view

VCP state and details: {master:0} user@Switch-1> show virtual-chassis vc-port fpc0: -------------------------------------------------------------------------Interface Type Trunk Status Speed Neighbor or ID (mbps) ID Interface PIC / Port vcp-0 Dedicated 2 Up 32000 1 vcp-0 vcp-1 Dedicated 1 Up 32000 1 vcp-1 fpc1: -------------------------------------------------------------------------Interface Type Trunk Status Speed Neighbor or ID (mbps) ID Interface PIC / Port vcp-0 Dedicated 2 Up 32000 0 vcp-0 vcp-1 Dedicated 1 Up 32000 0 vcp-1




Enabling and Disabling Virtual Chassis Ports  Built-in VCPs (vcp-0/vcp-1) are enabled by default

• You can disable or re-enable VCPs in operational mode: {master:0} VC user@Switch-1> request virtual-chassis vc-port set interface vcp-0 disable {master:0} user@Switch-1> show virtual-chassis vc-port fpc0: -------------------------------------------------------------------------Interface Type Trunk Status Speed Neighbor or ID (mbps) ID Interface PIC / Port vcp-0 Dedicated 2 Disabled 32000 vcp-1 Dedicated 1 Up 32000 1 vcp-1 ... {master:0} user@Switch-1> request virtual-chassis vc-port set interface vcp-0 {master:0} user@Switch-1> show virtual-chassis vc-port fpc0: -------------------------------------------------------------------------Interface Type Trunk Status Speed Neighbor or ID (mbps) ID Interface PIC / Port vcp-0 Dedicated 2 Down 32000 vcp-1 Dedicated 1 Up 32000 1 vcp-1 ... © 2011 Juniper Networks, Inc. All r ights reserved.



Verifying Status Information  Use

show virtual-chassis status to view

Virtual Chassis status information: {master:0} user@Switch-1> show configuration virtual-chassis preprovisioned; member 0 { role routing-engine; serial-number BM0208105168; } member 1 { role line-card; serial-number BM0208124231; } {master:0} user@Switch-1> show virtual-chassis status Preprovisioned Virtual Chassis Virtual Chassis ID: 8d5c.a77f.8de8 Member ID 0 (FPC 0)

Status Prsnt

Mastership Serial No Model priority BM0208105168 ex4200-24t 129

Role Master*

1 (FPC 1)

Prsnt

BM0208124231 ex4200-24t

Linecard


0


Neighbor List ID Interface 1 vcp-0 1 vcp-1 0 vcp-0 0 vcp-1 www.junip er.net | 7-59

Summary  In this chapter, we:

• Listed and described features that promote high availability • Configured and monitored high availability features • Described key concepts and components of a Virtual

Chassis • Explained the operational details of a Virtual Chassis • Implemented a Virtual Chassis and verified its operations




Review Questions 1. What is the purpose of LACP? 2. In what situation would you likely find RTG? 3. List some benefits of implementing a Virtual Chassis.




Lab 7: Virtual Chassis  Implement and verify operations of a Virtual Chassis.





JEX 11.a C7 HighAvailability

Recommend Documents