Switch VSS Guide
Switch VSS Guide
LTRCRS-2004
Cisco Catalyst Virtual Switching System - Instant
Access (VSS - IA) Hands On Lab
LAB GUIDE
Lab Overview
Cisco Catalyst Virtual Switching System (VSS) combined with Instant Access (IA) solution brings
operational simplicity and high availability to the Campus Catalyst Ethernet switching network. It creates a
single Virtual Switch across the distribution and access layers enabling a single network touch point;
ultimately driving simplified operations, distribution layer (Catalyst 6500/6800) features at the access
layer, consistent CLI and lower total cost of ownership (TCO).
The Virtual Switching System (VSS) is an innovative technology that allows the bonding of two VSS-
enabled Catalyst switches chassis such that it is represented as a single logical entity to the rest of the
network from a control plane perspective as well as from a configuration and management standpoint.
The VSS technology provides increased uptime with less configuration complexity by embracing new
innovations in EtherChanneling, NSF/SSO (Non-stop forwarding; stateful switchover) and EFSU
(Enhanced Fast Software Upgrade), whilst replacing the existing reliance on STP, HSRP and Fast
Routing hellos.
Cisco Catalyst Instant Access (IA) brings “fabric extension” (also known as FEX) technology, with high
availability and operational simplicity, to the Catalyst Campus Ethernet Switching line. This technology
creates a single configuration & management environment, across both Distribution and Access-layer
switches and has been specifically tailored to meet the needs of Campus / Enterprise network
deployments.
This demonstration lab was developed to give you hands-on experience with the fundamental simplicity of
provisioning, configuring and managing the VSS and IA solutions.
Product Overview
Benefits of Catalyst Virtual Switching Systems:
The solution simplifies enterprise campus networks by providing provisioning and operational simplicity.
The image below depicts an example of a single touch-point for a 21 Access switch Distribution block
(1008 ports).
Starting with version 15.2SY scalability has increased to up to 2,016 ports with specific platforms. For
more specific information please refer to the release notes at www.cisco.com.
Lab Topology
POD Assignment
Each POD includes:
NOTE: Every 2 students will share 1 POD. Each student can work on one 6500/6800 chassis and one IA
stack. You will be assigned a POD number by the proctor.
Lab Topology
Refer to the topology diagram below, for lab use.
Copyrighted Cisco® Systems Inc. December 2015 4
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Getting Started
This hand on lab consists on two separate sections. In the first section we will provision the VSS system
while in the second section we will configure the Instant Access solution. Please note that VSS needs to
be configured in order to enable Instant Access.
For each of the sections, we will provide a list of exercises together with the information on how to access
the assigned POD
Exercise Description
The conversion of the two 6500 switches into a single VSS is relatively straightforward and can be
summarized into the following steps:
1. Configure the Virtual Switch Domain on both devices and designate each switch as either Switch
1 or Switch 2
2. Configure the Virtual Switch Links
3. Execute the Conversion command after which the switches will reboot
4. Re-configure Standby Switch’s VSL on the Active Switch and save the configuration
The Virtual Switch Domain is defined as the grouping of the 2 members of the Virtual Switching System
as is assigned an ID; this ID is configured in global configuration mode and is recommended to be unique
for the Layer 2 network it will be part of. Virtual Switch Domain IDs can be in the range of 1-255, and must
match on both members of the Virtual Switch Domain. Refer to the figure below for more information on
this relationship.
Exercise Objective
The objective of this lab is to allow the student experience the simplicity of the VSS conversion as well as
to become familiar with the VSS infrastructure and terminology.
NOTE: For the purposes of this lab, use your Pod number as the Virtual Switch Domain ID.
Step 2 We are going to configure the Virtual Switch Domain on both devices using the POD number as
the domain id and we will designate each switch as either Switch 1 or Switch 2. Connect to the
console of the first switch to enter the commands below.
Router#conf t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)#hostname podX-VSS-SW1 ß
configure
hostname
for
easier
identification;
X=POD#
podX-VSS-SW1(config)#switch virtual domain X ß
configure
the
virtual
switch
domain
with
your
POD
NUMBER
Domain ID X config will take effect only after the exec command 'switch convert mode
virtual' is issued
Copyrighted Cisco® Systems Inc. December 2015 6
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Router#conf t
Enter configuration commands, one per line. End with CNTL/Z.
Router(config)#hostname podX-VSS-SW2 ß
configure
hostname
for
easier
identification;
X=POD#
Enter configuration commands, one per line. End with CNTL/Z.
podX-VSS-SW2(config)#switch virtual domain X ß
configure
the
virtual
switch
domain
with
your
POD
NUMBER
Domain ID X config will take effect only after the exec command 'switch convert mode
virtual' is issued
podX-VSS-SW2(config-vs-domain)#switch 2 ß
we
identify
this
switch
as
switch
#2
podX-VSS-SW2(config-vs-domain)# exit
The Virtual Switch Link is a special link, containing one or more physical ports that will allow configuration
and stateful (SSO) information to be exchanged between the two physical switches. The VSL appends a
special header (Virtual Switch Header) onto each frame that is sent across the wire. Refer the figure
below for more details:
Only specific interfaces can be used to form the VSL. For the purposes of this lab, the 10GE uplink ports
on the Supervisor 2T will be used to form the VSL. For complete lists of the interfaces that can be used
for the VSL, consult the VSS Configuration chapters of the 12.2SX, 15..0(1)SY and 15.1(1)SY
Configuration Guides for the Catalyst 6500/6800.
Step 4 In the next two steps we will be configuring the interfaces for VSL and bundle them into a port
channel. Enter the commands below on Switch 1
podX-VSS-SW1(config)#interface port-channel 1
podX-VSS-SW1(config)#no shut
podX-VSS-SW1(config-if)#switch virtual link 1
podX-VSS-SW1(config-if)#exit
podX-VSS-SW1(config)#interface range tenGigabitEthernet 1/4 - 5
podX-VSS-SW1(config-if-range)#no shut
podX-VSS-SW1(config-if-range)#channel-group 1 mode on
podX-VSS-SW1(config-if-range)#end
NOTE: The Supervisor 2T comes with 3 x 1G (1/1,1/2 and 1/3) and 2 x 10Gig uplink interfaces (Ten1/4
and Ten1/5). For the VSL configuration we are using the 2 x 10G uplinks.
Step 5 We will now be creating the VSL interfaces in the second switch. Enter the commands below on
Switch 2
To proceed with the conversion process, execute the “switch convert mode virtual” command in
enable mode. By doing so, it will convert the interface names to a 3-numbered notation (switch-
number/slot/port), save the configuration and reload the switch. The reload is necessary for a few
reasons:
Step 6 In the next two steps we will be converting the two standalone chassis into VSS mode. That is,
we will first save the configuration and we will then enter the commands for the VSS conversion:
podX-VSS-SW1#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-VSS-SW1(config)#
boot sys bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin ß
set
the
boot
variable
for
the
current
IOS
version)
Copyrighted Cisco® Systems Inc. December 2015 8
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
podX-VSS-SW1(config)#end
podX-VSS-SW1#wr mem
Destination filename [startup-config]? ß
Press
<ENTER>
to
accept
the
destination
filename
Building configuration...
[OK]
podX-VSS-SW1#
podX-VSS-SW1#switch convert mode virtual ß
note
that
we
are
in
enable
mode
This command will convert all interface names to naming convention "interface-type
switch-number/slot/port", save the running config to startup-config and
reload the switch.
Do you want to proceed? [yes/no]: yes ß
accept
the
reload
Converting interface names
Building configuration...
<..snip..>
Destination filename [startup-config.converted_vs-20140325-123157]? Enter ß
accept
Step 7 We will now proceed with the conversion of switch 2. First we will save the configuration and
then we will proceed with conversion of switch 2:
podX-VSS-SW2#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-VSS-SW2(config)#
boot sys bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.binß
set
the
boot
variable
for
the
current
IOS
version)
podX-VSS-SW2(config)#end
podX-VSS-SW2#wr mem
Destination filename [startup-config]?
Building configuration...
[OK]
podX-VSS-SW2#
podX-VSS-SW2#switch convert mode virtual ß
note
that
we
are
in
enable
mode
This command will convert all interface names to naming convention "interface-type
switch-number/slot/port",save the running config to startup-config and reload the
switch.
Do you want to proceed? [yes/no]: yes ß
accept
the
reload
Converting interface names
Building configuration...
<..snip..>
Destination filename [startup-config.converted_vs-20140325-123157]? Enter ß
accept
Step 8 At this stage, you will observe the reboot messages scrolling across the screen. On switch 1:
Initializing ATA monitor library...
<…snip…>
<…snip…>
Step 9 Observe the reboot messages on switch 2. As you can see in the output below, each switch
reboots independently, but during the boot process, the configuration is pre-parsed to detect
which interfaces form the VSL. The switch will then negotiate Active and Standby roles using
VSL Protocol (VSLP) and will continue the boot up process.
<…snip…>
<…snip…>
*Oct 11 13:31:45.555: %VSL_BRINGUP-6-MODULE_UP: VSL module in slot 1 switch 2 brought
up
*Oct 11 13:32:19.323: %VSLP-5-RRP_ROLE_RESOLVED: Role resolved as STANDBY by VSLP
Step 10 Upon the completion of the bootup process, you will notice that only the Active switch (Switch 1
in this case) CLI is accessible. Switch 2’s console is no longer available, since all management
and troubleshooting can be accomplished on the Active Switch’s console. Also note that the
prompt of Switch 2 has changed:
podX-VSS-SW1-sdby>
Standby console disabled
podX-VSS-SW1-sdby>
Standby console disabled
podX-VSS-SW1-sdby>
Standby console disabled
podX-VSS-SW1-sdby>
Step 11 CONGRATULATIONS!! At this stage, both 6500 switches have been converted into a single
Virtual Switching System, so you might want to rename your newly-configured VSS device and
proceed to save the configuration:
podX-VSS-SW1>en
Password:
podX-VSS-SW1#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-VSS-SW1(config)#hostname podX-vss ß X is POD number, eg. POD #1 would be pod1-vss
podX-vss(config)#exit
podX-vss#
00:13:40: %SYS-5-CONFIG_I: Configured from console by console
podX-vss#
podX-vss#wr mem
Building configuration...
[OK]
podX-vss#
NOTE: Configuration changes cannot be made till the two switches and Line cards have completely
come online and the two supervisors are in complete SSO sync them.
podX-VSS-SW1#conf t
Config mode cannot be entered during Standby initialization
After the conversion process, you can now take some time to get familiar with the Virtual Switch System
environment. First, you can verify the modules you now have control of from the Active Virtual Switch:
podX-vss#show module switch 1 ß “switch” keyword has been added to select either Switch 1 or Switch 2 of the VSS
Switch Number: 1 Role: Virtual Switch Active
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 5 Supervisor Engine 2T 10GE w/ CTS (Acti VS-SUP2T-10G SAL16052XQX
2 4 CEF720 4 port 10-Gigabit Ethernet WS-X6704-10GE SAL1019MC05
4 20 DCEF2T 4 port 40GE / 16 port 10GE C6800-16P10G-XL SAL1834ZAKB
podX-vss#sh module switch 2 ß new “switch” keyword has been added to select either Switch 1 or Switch 2 of the VSS
Switch Number: 2 Role: Virtual Switch Standby
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 5 Supervisor Engine 2T 10GE w/ CTS (Hot) VS-SUP2T-10G SAL16127TU2
2 4 CEF720 4 port 10-Gigabit Ethernet WS-X6704-10GE SAL08435CUF
4 20 DCEF2T 4 port 40GE / 16 port 10GE C6800-16P10G SAL1834ZAA7
NOTE: If you run these commands before the initialization process is completed, some modules with
show up with “Unknown”
Step 12 The same “switch” keyword can be used for others commands like “show running-configuration”.
In this step we will be checking the configuration for SW1 and SW2 (and we will do this from the
VSS active):
Flags : V - Valid
Step 14 Configuring the VSS for connectivity: Now that you have your Pod converted to VSS mode, we
should complete the remainder of the configuration for connectivity to the rest of the network.
The following diagram depicts the layout of your Pod, where X represents your Pod number.
Step 15 The next steps will take place in Virtual Switching System: first we set up a loopback address for
routing purposes
NOTE: Replace X with your pod’s ID. Pod’s ID can be captured from Host name (“ podX-vss” ) of
devices your logged-in
podX-vss#config t
podX-vss(config)#interface loopback 0
podX-vss(config-if)#ip address 1X.1X.1X.1X 255.255.255.255 ß
X
is
POD
number
podX-vss(config-if)#exit
Step 16 Also in the Virtual Switching System, we will set up the Port Channel interface and its
associated members that connect to the Core switch
Step 17 Next, we configure OSPF as the routing protocol to peer to the Core switch, and enable NSF:
podX-vss(config)#router ospf 10
podX-vss(config-router)#network 1X.1X.1X.1X 0.0.0.0 area 0 ß X is POD number
podX-vss(config-router)#network 10.252.0.0 0.0.255.255 area 0
podX-vss(config-router)#exit
Step 18 Still in the Virtual Switching System configure connectivity to the 4948-10G as a Layer 3
Etherchannel:
podX-vss(config)#interface range tenGigabitEthernet 1/2/2,tenGigabitEthernet 2/2/2
podX-vss(config-if-range)#no shut
podX-vss(config-if-range)#no switchport
podX-vss(config-if-range)#channel-group 3 mode on
podX-vss(config-if-range)#exit
podX-vss(config)#int port-channel 3
Creating a port-channel interface Port-channel 3
podX-vss(config-if)#no shut
podX-vss(config-if)#no switchport
podX-vss(config-if)#ip address 10.252.5X.1 255.255.255.0 ß X is POD number
podX-vss(config-if)#end
podX-vss#wr mem
Step 19 Now we are going to proceed to configure the 4948-10G to setup the connectivity to the VSS in
your POD as Layer 3 Etherchannel. Connect to the console of the 4948-10G switch and enter
the following commands:
Switch>en
Password: lab-cert
Switch#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-4948-sw(config)#interface port-channel 3
podX-4948-sw(config-if)#no shut
podX-4948-sw(config-if)#no switchport
podX-4948-sw(config-if)#ip address 10.252.5X.2 255.255.255.0 ß
X
is
POD
number
podX-4948-sw(config-if)#exit
Step 20 You can now verify connectivity of the 4948-10G by pinging its default gateway on the VSS and
by pinging the core:
podX-4948-sw#ping 100.100.100.100
Step 21 We will proceed to enable the IP address for management in the VSS pair so that we can use
telnet to access the VSS distribution in the next section:
Step 23 You can confirm the correct configuration of the management IP address with the commands
below (you can also try to access the VSS pair directly via telnet using the MGMT IP address in
the table provided in “Access to the Lab” section”):
podX-vss#show run int gig 1/1/1
Building configuration...
Step 24 The next exercise in this section, Exercise #2, is optional. If you are planning on doing
Exercise #2 please skip this step (it will be taken care of when doing the exercise).
Otherwise, please configure NSF (Non Stop Forwarding) for the OSPF process. NSF must be
explicitly configured:
podX-vss#conf t
podX-vss(config)#router ospf 10
podX-vss(config-router)#nsf
podX-vss(config-router)#end
podX-vss#show run partition router ospf 10
<..snip..>
router ospf 10
nsf
network 10.252.0.0 0.0.255.255 area 0
network 1X.1X.1X.1X 0.0.0.0 area 0 ß
X
is
the
POD#
!
!
end
podX-vss#write mem ß
Save
your
configuration
Exercise Description
Now that you have your Pod all up and running, it’s time to break it and see how it behaves. We will
examine 3 different failure scenarios and verify the effect on traffic as it passes through the VSS domain.
We will be performing the following tests:
• Switch Failure
Exercise Objective
The objective of this section is to verify how well a VSS environment handles anomalies in the network.
NOTE: You will be using extended ping on the 4948 as your testing mechanism for this exercise
Step 2 The first test will be the Access Switch Uplink Failure. In this test, we will test the uplink failure
that connects the 4948-10G to the VSS.
Have your 4948 configured as follows, but do not accept “Sweep range of sizes [n]:” until
you are ready to shut down your first interface of the test. We will generate traffic via this
ping sweep and let it run until it finishes.
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß DO NOT accept right now. Leave it ready for next step
Step 3 On the VSS issue the following command to determine which interfaces to shut down:
podX-vss#show etherchannel 3 summary | beg Number ß portchannel to the 4948 access switch
Number of channel-groups in use: 4
Number of aggregators: 4
Step 4 You can now proceed to shut/no shut each of the above member ports and verify the traffic loss
experienced when this action is issued. You should notice the traffic loss to be minimal – at
most 1 or 2 ping losses. We are also going to start the <ping> from step above.
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß
accept
the
ping
in
the
4948
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#interface tengigabitEthernet 1/2/2
podX-vss(config-if)#shut ß
shut
one
interface
connecting
to
the
access
switch
podX-vss(config-if)#no shut ß
no
shut
one
interface
connecting
to
the
access
switch
podX-vss(config-if)#exit
podX-vss(config)#interface tengigabitEthernet 2/2/2
podX-vss(config-if)#shut ß
shut
the
other
interface
connecting
to
the
access
switch
podX-vss(config-if)#no shut ß
no
shut
the
other
interface
connecting
to
the
access
switch
podX-vss(config-if)#end
Step 5 Check the output of the ping command in the 4948 and you will see that packet loss in minimal
<..snip..>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!
Success rate is 99 percent (39999/40000), round-trip min/avg/max = 1/1/168 ms podX-
4948-sw#
Step 6 The second test will be the core uplink failure. We will perform a similar test as above, but this
time, we will test the re-convergence between the VSS and the Core Switch, by shut/no shut the
member interfaces of the EtherChannel connecting the 2 devices.
Have your 4948 configured as follows, but do not accept “Sweep range of sizes [n]:” until
you are ready to shut down your first interface of the test:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Copyrighted Cisco® Systems Inc. December 2015 19
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Step 7 To determine which member interfaces make up this bundle, you can execute the following
command, replacing the ‘X’ in the port channel number with your own Pod assignment.
We will also accept the ping from the step above. Let it run until it finishes. Once again, you
should notice that by shut/no-shutting these member interfaces, the loss on the 4948 is also
minimal.
podX-vss#sh etherchannel X0 sum | begin Group ß
portchannel
to
the
core
switch
/
X
is
the
POD#
Group Port-channel Protocol Ports
------+-------------+-----------+--------------------------------------------
10 Pox(RU) PAgP Te1/2/1(P) Te2/2/1(P) ß
X
is
the
POD#
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß
accept
the
ping
in
the
4948
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#interface tengigabitEthernet 1/2/1
podX-vss(config-if)#shut ß
shut
one
interface
connecting
to
the
core
switch
podX-vss(config-if)#no shut ß
no
shut
one
interface
connecting
to
the
core
switch
podX-vss(config-if)#exit
podX-vss(config)#interface tengigabitEthernet 2/2/1
podX-vss(config-if)#shut ß
shut
the
other
interface
connecting
to
the
core
switch
podX-vss(config-if)#no shut ß
no
shut
the
other
interface
connecting
to
the
core
switch
podX-vss(config-if)#end
Step 8 Check the output of the ping command in the 4948 and you will see that packet loss in minimal
<..snip..>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!
Success rate is 99 percent (39999/40000), round-trip min/avg/max = 1/1/1412 ms
podX-4948-sw#
Step 9 The third test will be the switch failure. This exercise will involve the complete switchover of the
active Supervisor (switch) by reloading the box. First, we can identify which particular switch is
the active switch and the code level that is in operation, by executing the following commands:
<..snip..>
In dual-active recovery mode: No
Step 11 Have your 4948 configured as follows, but do not accept “Sweep range of sizes [n]:” until
you are ready to accept the reload:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß DO NOT accept right now. Leave it ready for next step
Step 12 We can now proceed to reload the Active Virtual Switch by executing the command below. We
will also be starting the <ping> from the step above:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß
accept
the
ping
in
the
4948
Step 13 Once again, if you are observing the ping results on the 4948, you should notice only 1 or 2
packets lost as a result of this operation.
If you lost more than 1 or 2 packets, check to see how many packets you lost by looking at the
“Success rate” information at the end of the extended ping. Note how many packets you
lost. Why do you think this happened?
Here is a clue: check your OSPF configuration with the “show run partition router
ospf 10” command. Is anything missing that could have prevented these packet drops?
We’ll fix this in a minute.
Step 14 Now that the console of Switch 2 is active, you are able to execute the following commands:
<..snip..>
Step 15 Shortly after this, Switch 1 will complete the bootup process and assume a Standby role after it
has completed being brought up:
<..snip..>
Switch Switch Status Priority Role Local Remote
Number Oper(Conf) SID SID
--------------------------------------------------------------------
LOCAL 2 UP 100(100) ACTIVE 0 0
REMOTE 1 UP 100(100) STANDBY 2038 8419
Step 16 Now, let’s fix the packet drop issue. Going back to the “show run partition router ospf
10” command, your output should look something like this:
Step 17 What is missing is NSF (Non Stop Forwarding). Remember that NSF must be explicitly
configured, but we forgot to do this when we did the initial ospf configuration in Exercise 1.
Therefore we will do it in this step. You should then see nsf in the ospf configuration.
podX-vss#conf t
podX-vss(config)#router ospf 10
podX-vss(config-router)#nsf
podX-vss(config-router)#end
podX-vss#show run partition router ospf 10
<..snip..>
router ospf 10
nsf
network 10.252.0.0 0.0.255.255 area 0
network 1X.1X.1X.1X 0.0.0.0 area 0 ß
X
is
the
POD#
!
!
end
podX-vss#write mem ß
Save
your
configuration
Step 18 Check that your VSS system is in SSO and that the second switch is ready for the switchover
CONFIG_FILE =
BOOTLDR =
Configuration register = 0x2102
Fabric State = ACTIVE
Control Plane State = ACTIVE
Step 19 Have your 4948 configured as follows, but do not accept “Sweep range of sizes [n]:” until you
are ready to accept the reload:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß DO NOT accept right now. Leave it ready for next step
Step 20 We can now proceed to reload the Active Virtual Switch by executing the command below. We
will also start the <ping> from step above:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 100.100.100.100
Repeat count [5]: 40000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß
accept
the
ping
in
the
4948
Step 21 How many pings were dropped this time? With NSF enabled, the hardware forwarding table is
maintained while the software forwarding table is rebuilt. Once the software table is complete,
only incremental changes are made to the hardware forwarding changes. Since none of our
forwarding entries changed, the only packet loss is associated with the switchover of traffic in
the MEC, if that needs to happen. Since that is a sub-second process, there should only be
1 or 2 pings lost at most.
In this lab exercise, each student will automatically provision their IA Client in 2 simple steps, using 2 new
CLI commands, directly on the C6K VSS.
1. Create the Port-channel to connect the fabric links to the IA client…
1.1. Set the Port-channel to operate in FEX mode
1.2. Associate a unique FEX ID to the Port-channel
2. Associating the physical fabric links to the Port-channel created in Step 1
Throughout this section we will be configuring and using the topology below:
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX 101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask intructors IP addresses and
credentials).
Exercise Objective
The objective of this exercise is to demonstrate the overall simplicity of the IA solution, where the access
devices get provisioned automatically, without any direct console access or manual configuration at
access layer.
This exercise will also demonstrate that the only “work” required is a physical (L1) connection between
the IA Parent & IA Client switches, and 2 simple configuration steps on the IA Parent.
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask instructors for IP addresses and
credentials).
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#interface port-channel 10Z ß
Create
a
PortChannel
(Z=1
for
student
1
and
slno
Z=2
for
student
2)
podX-vss(config-if)#switchport
podX-vss(config-if)#switchport mode fex-fabric ß
Enables
Fex
fabric
All extraneous configs removed from interface Port-channel101!
podX-vss(config-if)#fex associate 10Z ß
Associate
FEX-‐ID
to
Port
Channel
(Z=1
for
student
1
and
Z=2
for
student
2)
podX-vss(config-if)#load-interval 30 ß
To
verify
traffic
stats
in
30s
interval
podX-vss(config-if)#no shut
podX-vss(config-if)#end
Step 2 Associate the Port-channel interface with its Physical member interface
podX-vss#conf t
podX-vss(config)#interface TenGigabitEthernet 1/4/4+Z ß
port
connected
to
FEX
;
Z=1
for
student
1
and
Z=2
for
student
2)
podX-vss(config-if)#switchport
podX-vss(config-if)#switchport mode fex-fabric
All extraneous configs removed from interface TenGigabitEthernet1/4/4+Z
podX-vss(config-if)#channel-group 10Z mode on ß
Associate
Physical
port
to
Port
Channel;
Z=1
for
student
1
and
Z=2
for
student
2)
podX-vss(config-if)#no shut
podX-vss(config-if)#end
NOTE: We are associating only one TenGig interface to the Port-channel for this exercise. Only one
physical link is required for IA to operate. We will add the other TenGig interface in a later exercise (for 2
x 10G fabric links), to demonstrate the simplicity of provisioning additional fabric link bandwidth.
Step 3 Verify if the IA Client is provisioned over the TenGigabitEthernet link and bound to the IA Parent
podX-vss#show fex 10Z detail ß
Z=1
for
student
1
and
Z=2
for
student
2
FEX: 10Z Description: FEX010Z state: online ß
Indicates
the
IA
status
(*)
FEX version: 15.0(2)EX5
Extender Model: C6800IA-48TD, Extender Serial: FOC1737W0PS
FCP ready: yes
(*) NOTE: You might see other states (e.g. “state: connected”) before you see “state: online”
(*) NOTE: The output above is for FEX101. Student 2 will have a slightly different output
Step 4 Get familiar with new “fex” commands in the VSS pair
podX-vss#show fex
FEX FEX FEX FEX
Number Description State Model Serial
---------------------------------------------------------------------------
101 FEX0101 online C6800IA-48TD FOC1737W0PS
102 FEX0102 online C6800IA-48TD FOC1737S245
Step 1 (Optional) IA also allows you to pre-provision a new FEX even before physically connecting it to
the parent C6K VSS. Below is an example of pre-provisioning a new FEX client, with 2 stack
members. Use the following command to pre-provision at the enable prompt (not config prompt)
podX-vss# module provision create fex 15Z type C6800IA-48TD slot 1 ß
Z=1
for
student
1
and
Z=2
for
student
2
FEX 15Z slot 1 module provisioning entry added. ß
Pre-‐Provisions
FEX
15Z
in
as
stack
member
1
podX-vss#module provision create fex 15Z type C6800IA-48TD slot 2 ß
Z=1
for
student
1
and
Z=2
for
student
2
FEX 15Z slot 2 module provisioning entry added. ß
Pre-‐Provisions
FEX
15Z
in
as
stack
member
2
podX-vss#show run fex 15Z ß
Z=1
for
student
1
and
Z=2
for
student
2.
Show
run
fex
confirms
the
pre-‐provisioning,
and
now
all
IA
hosts
ports
can
be
configured
at
this
point.
Building configuration...
podX-vss#show run fex 15Z mod 2 ß
Z=1
for
student
1
and
Z=2
for
student
2.
Show
interfaces
in
the
second
member
of
the
FEX
stack
Building configuration...
Exercise Objective
The purpose of the exercise is to demonstrate the overall simplicity of managing the IA solution, which
provides the following benefits:
• Single point of management for all configuration
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask instructors for IP addresses and
credentials).
podX-vss#show run fex 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
Building configuration...
podX-vss#show module fex 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
Switch Number: 10Z Role: FEX
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 48 C6800IA 48GE C6800IA-48TD FOC1737W0PS
Step 3 Verify the current Port-Channel & Fabric Link Status between IA Client & Parent
podX-vss#show etherchannel 10Z summary ß
Z=1
for
student
1
and
Z=2
for
student
2
Flags: D - down P - bundled in port-channel
I - stand-alone s - suspended
H - Hot-standby (LACP only)
R - Layer3 S - Layer2
U - in use N - not in use, no aggregation
f - failed to allocate aggregator
w - waiting to be aggregated
Number of channel-groups in use: 5
Number of aggregators: 5
NOTE: Remember that we have only associated one TenGig interface to the Port-channel for this
exercise. Only one physical link is required for IA to operate. We will add the other TenGig interface in a
later exercise (for 2 x 10G), to demonstrate the simplicity of provisioning additional fabric link bandwidth.
podX-vss#show environment status fex 10Z ß
Enables
env
check
of
remote
IA
line
card
from
Cat6k.
Z=1
for
student
1
and
Z=2
for
student
2
Fex 10Z Fan 1
FEX 10Z Fan 1 type: Built-in
FEX 10Z Fan 1 mode: Auto
Fex 10Z fan-tray 1 fan-fail: OK
Copyrighted Cisco® Systems Inc. December 2015 32
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
podX-vss#show power fex 10Z ß
Check
the
PSU
of
IA
Client.
Z=1
for
student
1
and
Z=2
for
student
2
FEX Switch Number: 10Z
system power redundancy mode for module:1 = non-redundant
system power total for module:1 = 135.00 Watts (11.25 Amps @ 12V)
system power used for module:1 = 41.04 Watts ( 3.42 Amps @ 12V)
system power available for module:1 = 93.96 Watts ( 7.83 Amps @ 12V)
Power-Capacity PS-Fan Output Oper
PS Type Watts A @12V Status Status State
---- ------------------ ------- ------ ------ ------ -----
1 Built-in 135.00 11.25 -- -- on
Pwr-Allocated Oper
Fan Type Watts A @12V State
---- ------------------ ------- ------ -----
1 Built-in 5.04 0.42 on
Pwr-Requested Pwr-Allocated Admin Oper
Slot Card-Type Watts A @12V Watts A @12V State State
---- ------------------ ------- ------ ------- ------ ----- -----
1 C6800IA-48TD 36.00 3.00 36.00 3.00 on on
Step 6 In situations where it is required to issue commands on the IA Client, like console access, the
user can open a remote session using the “attach” CLI command. Note that there is also a
traditional RJ45 Console port on the IA Client
podX-vss#attach fex-id 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2.
Equivalent
of
doing
a
session
to
a
C6k
line
card
Attach FEX:10Z ip:192.1.1.101
Trying 192.1.1.101 ... Open
????????
d - default port
<..snip..>
FEX-10Z#show int teng 1/0/1 ß Verify the FEX “Uplink” interface towards C6k which are
not visible from C6k CLI – Note: use the interface number
from the output above. In this example is Te1/0/1
TenGigabitEthernet1/0/1 is up, line protocol is up (connected)
Hardware is Ten Gigabit Ethernet, address is c025.5ca1.feb3 (bia c025.5ca1.feb3)
MTU 9198 bytes, BW 10000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive not set
Full-duplex, 10Gb/s, link type is auto, media type is SFP-10GBase-CX1
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input never, output never, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 3000 bits/sec, 1 packets/sec
5 minute output rate 3000 bits/sec, 1 packets/sec
1470 packets input, 928937 bytes, 0 no buffer
Received 1378 broadcasts (1377 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 1377 multicast, 0 pause input
0 input packets with dribble condition detected
1677 packets output, 915917 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out
FEX-10Z#exit
Step 8 Verify the FEX system usage across all FEX- ID’s
podX-vss#show fex system platform usage
FEX id usage details
Fex-ids inuse: 101, 102
Fex-ids online: 101, 102
Total Used Free
----- ---- ----
42 2 40 ß
Highlights
the
FEX-‐ID
(MAX,
USED
&
FREE)
usage
&
how
many
are
available
NOTE: The command above shows scalability with the IOS version currently running in the system. For
scalability information on specific platforms and IOS versions, please refer to the release notes
Exercise Objective
The purpose of the exercise is to give users the hands on experience of truly plug and play simplicity of
provisioning additional stack members.
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask instructors for IP addresses and
credentials).
podX-vss#show fex 10Z detail ß
Z=1
for
student
1
and
Z=2
for
student
2
FEX: 10Z Description: FEX010Z state: online ß
FEX
10Z
online
FEX version: 15.0(2)EX2 ß
You
might
see
a
different
software
version
Extender Model: C6800IA-48TD, Extender Serial: FOC1741S1F9
FCP ready: yes
Image Version Check: enforced
Fabric Portchannel Ports: 1
Fabric port for control traffic: Te1/2/(4+Z)
Fabric interface state:
Po10Z - Interface Up.
Te1/4/(4+Z) - Interface Up. state: bound
Step 2 Verify FEX 10Z shows up as line card module. We will use the command “show module fex
10Z” to verify how many stack members (modules) are part of FEX 10Z
podX-vss#show module fex 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
Switch Number: 10Z Role: FEX
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 48 C6800IA 48GE C6800IA-48TD FOC1741S1F9 ß
Shows one module of C6800IA-48TD
Step 3 The purpose of this exercise is to demonstrate the simplicity of adding stack members to an
existing IA stack. The stack members are already pre-cabled using flex-stack cable to your FEX
10Z, however they are not booted up. In this step we will simulate the stack member addition by
booting up the stack member manually.
To manually boot the stack member, connect to the console of the second FEX in the stack.
switch: ß This is the ROMMON prompt of IA Client to allow us to do manual booting.
Switch: boot ß
Type
“boot”
to
manually
boot
the
Stack
member.
Note
in
real
world
IA
deployments,
all
IA
Clients
automatically
boot
up
and
would
not
require
any
manual
boot
step.
Loading "flash:/c6800ia-universalk9-mz.150-2.EX2/c6800ia-universalk9-mz.150-
2.EX2.bin"...Verifying image flash:/c6800ia-universalk9-mz.150-2.EX2/c6800ia-
universalk9-mz.150-
2.EX2.bin.............................................................................
......................................................................................
.........................................................................Image passed
digital signature verification
@@@@@@@@@@@@@@@...
Step 5 The bootup process of the stack member will take approx. 5 minutes. After 5 minutes the IA
client stack members will automatically associate and show up as part of show module output
on C6k.
podX-vss#show module fex 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
Switch Number: 10Z Role: FEX
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 48 C6800IA 48GE C6800IA-48TD FOC1741S1F9
2 48 C6800IA 48GE C6800IA-48TD FOC1741S1FF ß
Module
2
automatically
shows
up
as
Stack
member
2
boots
up
Exercise Objective
The purpose of the exercise is to give users the hands on experience of Catalyst 6500/6800 feature
richness with the Instant Access solution.
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask instructors for IP addresses and
credentials).
podX-vss#show run int gig 10Z/1/0/1 ß
Z=1
for
student
1
and
Z=2
for
student
2
Building configuration...
Step 2 Add VLAN101 (for student 1) and VLAN102 (for student 2) for host access and assign an IP
address
podX-vss#conf t
podX-vss(config)#vlan 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-vlan)#exit
% Applying VLAN changes may take few minutes. Please wait...
podX-vss(config)#int vlan 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#ip add 10X.10Z.1.100 255.255.255.0 ß
X
is
the
POD#;
ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#no shut
podX-vss (config-if)#end
podX-vss#show run int vlan 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
Building configuration...
Step 3 Configure the first IA Client host port (interface Gig10Z/1/0/1) like any regular Access switch
host port. We are going to use access VLAN 10Z (Z = 1 for student 1 and Z = 2 for student 2)
podX-vss#conf t
podX-vss(config)#int gig 10Z/1/0/1 ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#switchport mode access
podX-vss(config-if)#switchport access vlan 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#load-interval 30 ß
To
verify
traffic
stats
in
30s
interval
podX-vss(config-if)#logging event link
podX-vss(config-if)#no shutdown
podX-vss(config-if)#end
podX-vss#write mem ß
Save
the
configuration
Step 4 We are going to add the network from the vlan into the OSPF process:
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#router ospf 10
podX-vss(config-router)#network 10X.10Z.0.0 0.0.255.255 area 0 ß
X=POD
number;
ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-router)#end
Step 5 Configure the 4948 switch as the host for Instant Access.
podX-4948-sw#conf t
podX-4948-sw(config)#int gig 1/Z ←
Z=1
for
student
1
and
Z=2
for
student
2
podX-4948-sw(config-if)#no switchport
podX-4948-sw(config-if)#ip add 10X.10Z.1.2 255.255.255.0
podX-4948-sw(config-if)#no shut
podX-4948-sw(config)#ip route 10X.10Z.10.1 255.255.255.255 10X.10Z.1.100
ß
X=POD
number;
Z=1
for
student
1
and
Z=2
for
student
2
podX-4948-sw(config)#end
podX-4948-sw#write mem ß
Save
the
configuration
Step 6 We are now going to generate traffic from the host (4948 switch) to the core switch but first we
are going to test basic connectivity:
podX-4948-sw#ping 10X.10Z.10.1 ß X= POD number / Z=1 for student 1 and Z=2 for student 2
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 101.101.10.1, timeout is 2 seconds:
.!!!!
Success rate is 80 percent (4/5), round-trip min/avg/max = 1/1/1 ms
Step 7 Verify the interface stats on IA Client host port directly from C6k console CLI
podX-vss#sho int gig 10Z/1/0/1 ßZ=1
for
student
1
and
Z=2
for
student
2
-‐
Verify
the
IA
Host
port
counters
on
C6k
console
GigabitEthernet10Z/1/0/1 is up, line protocol is up (connected)
Hardware is C6k 1000Mb 802.3, address is ece1.a959.6581 (bia ece1.a959.6581)
MTU 9216 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 2/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseT
input flow-control is off, output flow-control is unsupported
Clock mode is auto
ARP type: ARPA, ARP Timeout 04:00:00
Last input never, output never, output hang never
Last clearing of "show interface" counters 6d02h
Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
30 second input rate 106000 bits/sec, 12 packets/sec ß
Ingress
traffic
from
4948
Step 8 Verify the traffic is traversing from all the way up to C6k via Interface port-Channel (the port-
channel interface is the one we are using for the ports connected to the IA client)
podX-vss#show interface port-channel 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
Port-channel101 is up, line protocol is up (connected)
Hardware is EtherChannel, address is 0019.5507.4170 (bia 0019.5507.4170)
MTU 1500 bytes, BW 40000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 10Gb/s, media type is unknown
input flow-control is on, output flow-control is off
Members in this channel: Te1/2/15 Te2/2/15
ARP type: ARPA, ARP Timeout 04:00:00
Last input never, output never, output hang never
Last clearing of "show interface" counters 6d03h
Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
30 second input rate 1851000 bits/sec, 166 packets/sec <- Packet switching @ C6K
30 second output rate 1851000 bits/sec, 166 packets/sec
517281346 packets input, 534068995937 bytes, 0 no buffer
Received 495849496 broadcasts (495847353 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 0 multicast, 0 pause input
0 input packets with dribble condition detected
22710165 packets output, 21876977237 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 PAUSE output
0 output buffer failures, 0 output buffers swapped out
NOTE: The following steps simply demonstrate the capability to configure any of the advanced C6K
routing & switching features (e.g. L3 (IPv4/IPv6), VRF, MPLS, etc.) on IA Client host ports.
Step 9 (Optional) Configure L3 interface enabling routed access at IA client host port.
podX-vss#conf t
podX-vss(config)#int gig 10Z/1/0/15 ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#no switchport ß
Routed
Access
Interfaces
podX-vss(config-if)#ip add 10Z.15.1.1 255.255.255.0 ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#no shut
podX-vss(config-if)#end
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss (config)#vrf definition RED-Z ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss (config-vrf)#vnet tag 10Z ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss (config-vrf)#address-family ipv4
podX-vss (config-vrf-af)#end
podX-vss#conf t
podX-vss(config)#int gig 10Z/1/0/25 ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#no switchport
podX-vss(config-if)#vrf forwarding RED-Z ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#ip address 10Z.25.1.1 255.255.255.0 ß
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#mpls ip
podX-vss(config-if)#no shut
podX-vss(config-if)#end
Exercise Objective
The purpose of the exercise is to demonstrate the simplicity of provisioning additional bandwidth using a
single CLI command on C6K. This allows customers to deploy additional bandwidth directly from C6k
without any changes and configuration at access layer.
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask the instructors for IP addresses and
credentials).
podX-vss#show fex 10Z detail ß
Z=1
for
student
1
and
Z=2
for
student
2
FEX: 10Z Description: FEX010Z state: online
FEX version: 15.0(2)EX2
Extender Model: C6800IA-48TD, Extender Serial: FOC1709Z2V9
FCP ready: yes
Image Version Check: enforced
Fabric Portchannel Ports: 1
Fabric port for control traffic: Te1/2/(4+Z)
Fabric interface state:
Po10Z - Interface Up. ß
Z=1
for
student
1
and
Z=2
for
student
2
Te1/4/4+Z - Interface Up. state: bound ß
Z=1
for
student
1
and
Z=2
for
student
2.
We
have
one
10G
uplink
deployed
Step 3 Verify the additional Uplinks are provisioned to IA Fabric link between IA client and IA Parent
podX-vss#show fex 10Z detail ß Z=1
for
student
1
and
Z=2
for
student
2
FEX: 10Z Description: FEX010Z state: online
FEX version: 15.0(2)EX2
Extender Model: C6800IA-48TD, Extender Serial: FOC1709Z2V9
FCP ready: yes
Image Version Check: enforced
Fabric Portchannel Ports: 2
Fabric port for control traffic: Te1/2/(4+Z)
Fabric interface state:
Po10Z - Interface Up.
Te1/4/4+Z - Interface Up. state: bound ß Z=1
for
student
1
and
Z=2
for
student
2
Te2/4/4+Z - Interface Up. state: bound ß Two 10G Fabric Links
Exercise Objective
Compact switches can be deployed in standalone mode like traditional switches and the can be deployed
in FEX mode with Instant Access. The student will have hands-on experience on how to convert compact
switches to instant access mode. For this lab we will be using compact switch WS-C3560CX-8XPD-S.
We will also leverage IOS commands that allow for auto-provisioinng of FEX clients.
podX-vss#conf t
podX-vss(config)#fex auto-config ß enables
automatic
provisioning
of
FEX
clients
for
the
whole
system
podX-vss(config)#end
podX-vss#wr mem ß saves
configuration
NOTE: Both students will be working simultaneously in this lab since there’s one IA compact switch per student.
We are going to use 2 x FEX ID’s. Since this exercise uses the “auto-config” command, the IA parent will
automatically assign a FEX ID. Since we don’t know which FEX ID’s each student will be assigned, we will use
FEX XYZ to refer to the compact switch prosivioned as a FEX.
We will still use the concept of Z=1 for student 1 and Z=2 for student 2 for uplinks
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask instructors for IP addresses and
credentials). Each student will access the console of the assigned compact switch.
Step 2 Each student will connect to one compact switch via the console. Once connected, we will
check the fex mode for the compact switch. Compact switches can be deployed in standalone
mode like traditional switches and the can be deployed in FEX mode with Instant Access. The
compact switches are currently in standalone mode so we will convert them into FEX mode.
NOTE:
Student 1 should connect to Compact Switch 1 and use parameter Z=1 below.
Student 2 should connect to Compact Switch 2 and use parameter Z=2 below.
Copyrighted Cisco® Systems Inc. December 2015 46
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Step 3 We are now going to enable the interfaces connected to the Compact Switch Instant Access
client
podX-vss#conf t
podX-vss(config)#int rang teng 1/4/Z+2, teng 2/4/Z+2 ß
Z=1
for
student
1
and
Z=2
for
student
2)
podX-vss(config-if-range)#no shut
podX-vss(config)#end
podX-vss#wr mem ß saves
configuration
Step 4 We are now convert the standalone compact switch into FEX mode
Switch#fex-mode enable ß
Enter
this
commmand
from
the
console
of
the
compact
switch
System will reload after mode conversion.
Do you want to continue? [no]: yes ß Enter
“Yes”
and
<ENTER>
Step 5 We will now wait for the switch to reload which will take a few minutes. Once the FEX reloads,
we will check the FEX configuration for the IA parent / CAT6K-VSS
podX-vss#sh fex
FEX FEX FEX FEX
Number Description State Model Serial
---------------------------------------------------------------------------
101 FEX0101 online C6800IA-48TD FOC1737W0Q9
102 FEX0102 online C6800IA-48TD FOC1736W19L
XYZ FEX0XYZ online WS-C3560CX-8XPD-S FHH1840P00B ß
Locate
FEX
id
assigned
by
auto-‐fex
podX-vss#reload fex XYZ standalone dissociate ß
Use
FEX
id
assigned
by
auto-‐fex
ß
We
are
going
to
reload
the
compact
switch
in
standalone
mode
Proceed with reload of fex module and Convert to Standalone mode?[confirm] ß[ENTER]
podX-vss#sh fex
Copyrighted Cisco® Systems Inc. December 2015 47
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#no fex XYZ ß
Use
FEX
id
assigned
by
auto-‐fex;
we
are
now
removing
the
FEX
podX-vss(config)#end
podX-vss#sh fex
FEX FEX FEX FEX
Number Description State Model Serial
---------------------------------------------------------------------------
101 FEX0101 online C6800IA-48TD FOC1737W0Q9
102 FEX0102 online C6800IA-48TD FOC1736W19L
Step 7 Connect to the console of the compact switch and confirm that the switch is once again in
standalone mode. Please note that the reload has to finish in order to be able to log into the
switch. You can check on the boot up process via the console.
Exercise Objective
In this exercise, the goal is to demonstrate high availability across multiple fabric links when one of the
fabric link fails and data plane continues to forward traffic across the VSS pair with minimal loss.
NOTE: Both students will be working simultaneously in this lab since there’s one IA stack per student.
We are going to use 2 x FEX ID’s = FEX101 (student 1) and FEX 102 (student 2).
During the steps we will be referring to FEX 10Z; Z=1 for student 1 and Z=2 for student 2
The output of some of the “show” commands may vary based on the student number
Each student will now use TELNET access to the VSS and to the 4948 (ask instructors for IP addresses and
credentials).
Step 1 Telnet to the 4948 switch and generate traffic to the core switch (for telnet access information
refer to the table in section “Access your POD”)
Step 2 Now from the CAT6K VSS verify the traffic across 2 fabric links
podX-vss#sh int ten 1/4/4+Z | inc rate ß
Z=1
for
student
1
and
Z=2
for
student
2
Queueing strategy: fifo
30 second input rate 4200000 bits/sec, 380 packets/sec ß
Traffic
on
one
10G
link
30 second output rate 4200000 bits/sec, 380 packets/sec ß
podX-vss#sh int ten 2/4/4+Z| inc rate ß
Z=1
for
student
1
and
Z=2
for
student
2
Queueing strategy: fifo
30 second input rate 0 bits/sec, 0 packets/sec ß
Copyrighted Cisco® Systems Inc. December 2015 49
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#int ten 1/4/4+Z ß
Chose
the
interface
to
shutdown
from
the
step
above;
Z=1
for
student
1
and
Z=2
for
student
2
podX-vss(config-if)#shutdown
podX-vss(config-if)#end
podX-vss#sh int ten 1/4/4+Z | inc rate ß
Z=1
for
student
1
and
Z=2
for
student
2
Queueing strategy: fifo
30 second input rate 0 bits/sec, 0 packets/sec
30 second output rate 0 bits/sec, 0 packets/sec
podX-vss#sh int ten 2/4/4+Z | inc rate ß
Z=1
for
student
1
and
Z=2
for
student
2
Queueing strategy: fifo
30 second input rate 1851000 bits/sec, 166 packets/sec
30 second output rate 1851000 bits/sec, 166 packets/sec ß
Traffic
switched
to
other
10G
link
Note: The traffic loss, in this case 1 packet loss during the switchover across the fabric links
Step 5 Now restore the shutdown 10G fabric link back up.
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#int ten 1/4/4+Z ß
Chose
the
interface
to
shutdown
from
the
step
2
podX-vss(config-if)#no shutdown
ß
Restore
fabric
link
podX-vss(config-if)#end
podX-vss#
Step 6 In this step we will verify which 10G link is carrying the control information (heartbeat and control
packets) between C6k and C6800IA client.
Step 7 Shutdown the 10G link that is carrying the control information (heartbeat and control packets)
between C6k and C6800IA client and verify that IA Client maintains steady state of control plane
and data plane between C6k and C6800IA client. Note the interface number we will be shutting
down so that we can restore it later.
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#int ten 2/4/4+Z ß
Chose
the
interface
to
shutdown
from
the
step
above
podX-vss(config-if)#shutdown ß
Shut
the
control
traffic
10G
fabric
link
podX-vss(config-if)#end
Step 8 Now restore the shutdown 10G fabric link back up.
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#int ten 2/4/4+Z ß
Chose
the
interface
to
shutdown
from
the
step
2
podX-vss(config-if)#no shutdown
ß
Restore
fabric
link
podX-vss(config-if)#end
podX-vss#write mem ß
Save
the
configuration
podX-vss#conf t
Enter configuration commands, one per line. End with CNTL/Z.
podX-vss(config)#port-channel load-balance src-mac ß
change
Load
hash
at
C6k
podX-vss(config)#end
podX-vss#
Step 10 Verify the changes in Load balancing hash for both C6k and IA Client
podX-vss#show etherchannel load-balance
EtherChannel Load-Balancing Configuration:
src-mac ß
Load
sharing
hash
on
interface
port-‐channel
on
C6k
mpls label-ip
Copyrighted Cisco® Systems Inc. December 2015 52
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Exercise Description
The Cisco Catalyst 6500E and 6800X Series support enhanced Fast Software Upgrade (eFSU) in order
to increase network availability by reducing the downtime caused by software upgrades across two
supervisors in a VSS pair. It brings the active and standby supervisors into synchronous Stateful
Switchover (SSO) mode across two supervisor running two different software versions. It maintains an
active data plane on both switches in the VSS pair, providing increased network availability during the
upgrade process.
Enhanced Fast Software Upgrade is a software upgrade feature that has been supported since
12.2(33)SXI software release onwards.
Using EFSU, traffic loss during the upgrade process is minimized such that customers can perform
software upgrades without requiring scheduled maintenance windows. Using FSU with VSS (pre-
12.2(33)SXI), it is expected to lose 100% of the bandwidth available on VSS for at least 3-4 minutes
during RPR switchover. However, EFSU always allows VSS to maintain at least 50% of bandwidth during
a software upgrade provided the devices are dual-homed to the VSS pair (for single homed devices, VSS
Quad-SUP is recommended). EFSU allows different images on Active and Standby supervisors to coexist
while they are operating in SSO Redundancy mode.
EFSU capability is extended to support Instant Access client upgrades similar to how a line card is
upgraded. The client software image is bundled with the Catalyst 6500/6800 software image. A new CLI
is introduced, enabling the upgrade of the Instant Access client stack (FEX-IDs), which in turn enables an
upgrade of the Instant Access client's software version before the “issu commitversion (step 4)” of
the EFSU process.
EFSU is a process that requires the user to go through 4 steps to perform software upgrade on VSS. The
process is depicted in the picture below:
Exercise Objective
The objective of this lab is to allow users experience the Enhanced Fast Software Upgrade (eFSU)
procedure and to test the High Availability (HA) capability during software upgrade of both the distribution
switches of VSS pair and all the access layer Instant access clients from single point of management
while having minimal impact of traffic.
We will also be able to demonstrate the simplicity of upgrading the Instant Access clients.
NOTE: For this lab, we are going to upgrade from 15.2(1)SY1 to 15.2(1)SY1a
Both students will be working together in this software upgrade using just the console for the Active VSS
Switch. We recommend students have 3 windows opened: console to active VSS switch; console to
standby VSS switch and telnet to 4948 switch
Connect to the VSS active console using the information provided in section “Access your POD”
(please read the note in the blue box above). Both students will be working together during
the upgrade process. Issue the “show version” command in the VSS system to check the
IOS version used.
NOTE: Mark below, which is the version of software you are using (we will call it OLD_CODE) and also note
the version you will be upgrading to (we will call it NEW_CODE). You are going to use this information throughout
this lab:
Step 2 Ensure that the NEW_CODE is present in the disk of both the Active Virtual Switch and the
Standby Virtual Switch:
podX-vss#dir bootdisk:
Directory of bootdisk:/
podX-vss#dir sw2-slot1-bootdisk:
Directory of slavebootdisk:/
Step 3 We will confirm that FEX 101 and FEX 102 are online and both links connected. We will also
check that both compact switches have been removed from the Instant Access domain
podX-vss#show fex
FEX FEX FEX FEX
Number Description State Model Serial
---------------------------------------------------------------------------
101 FEX0101 online C6800IA-48TD FOC1737W0Q9
102 FEX0102 online C6800IA-48TD FOC1736W19L
Step 4 In Instant Access solution, the software image of C6800IA client is bundled with C6k image. To
verify what version of Image is running on IA Client. Please note that the image version of the
client will differ from the image version of the parent. However, all clients should be running the
same image. Issue the following command:
<..snip..>
podX-vss#show version fex 102
==> Showing version for FEX 102:
Cisco IOS Software, C6800IA Software (C6800IA-UNIVERSALK9-M), Version 15.2(3m)E1, RELEASE
SOFTWARE (fc2)
Technical Support: http://www.cisco.com/techsupport
Copyright (c) 1986-2015 by Cisco Systems, Inc.
Compiled Tue 28-Apr-15 22:53 by prod_rel_team
<..snip..>
Step 5 The software version that is running in each stack member can be verified by the following
Step 6 The Software version running on each of C6800IA client will be same as the software version
bundled with C6k image. To verify that you can issue the following command on C6k.
Copyrighted Cisco® Systems Inc. December 2015 57
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Step 7 There are whole set of new commands introduced with EFSU upgrade process. EFSU uses the
same infrastructure as ISSU (In Service Software Upgrade) so we will be leveraging ISSU-
based commands. Throughout this labs, we will be using the ISSU commands shown in the
output below to perform each of the steps to complete a software upgrade using EFSU:
podX-vss#issu ?
ISSU commands
abortversion abort ISSU process
acceptversion accept new IOS version on new Active
commitversion commit IOS version on Standby
loadversion load new IOS version on Standby
runversion run new IOS version on Standby and make it Active
Step 8 We will now get ready to generate traffic during our upgrade process. Open a window
connecting to the 4948 and have it configured as follows, but do not accept “Sweep range of
sizes [n]:” until you are ready to perform the “issu loadversion” command in the steps
below:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 10X.101.10.1
Repeat count [5]: 50000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß
WAIT!
Don’t
accept
the
“Sweep
range”
until
mid
next
step
Step 9 The first step of the EFSU process is “issu loadversion” and the purpose of this step is to
load the new software image onto the standby supervisor of the VSS pair. During this step, the
Standby switch will reload and boot up with the new software image. Please refer to the
“Exercise Description” section for the image that shows this action.
Now we are going to perform an “ISSU loadversion” typing the command from the output
below (use the image name of the NEW_IMAGE).
Note that you can use tab/autocomplete to make sure you type the correct image name.
podX-vss#
*Mar 14 14:24:51.626: %ISSU_PROCESS-SW1-3-LOADVERSION: Loadversion sequence will begin
in 60 seconds. Enter 'issu abortversion' to cancel.
While Switch-1 is booting with the new image, the VSS will run at 50% capacity (Standby Switch
2 reloading). We are going to generate traffic. So accept “Sweep Range of sizes [n]:”
on the 4948 now.
podX-4948-sw#
Sweep range of sizes [n]: ß
Hit
Enter
for
the
ping
sweep
in
4948
when
issuing
issu
loadversion
Type escape sequence to abort.
Sending 50000, 100-byte ICMP Echos to 10X.101.10.1, timeout is 2 seconds:
Step 10 Pay attention to the telnet window opened for VSS Standby (Switch 2) Console. You will
observe that the VSS Standby (Switch 2) is booting with the new-image. You will also notice that
even though the standby is running a different image when compared to VSS Active (Switch 1),
redundancy mode negotiated between them is SSO.
podX-vss-sdby>
*Mar 14 14:25:21.630: %ISSU_PROCESS-SW2_STBY-6-SELF_RELOAD: slot 33 countdown to self-
reload started, 30 second delay
<..snip..>
<..snip..>
<…snip…>
<..snip..>
Step 11 After successful boot up of the Switch 2 (Standby VSS) to the NEW_IMAGE, verify that the
redundancy mode between the two Supervisors is SSO state (Hot Standby). At this stage, the
VSS will be operating at 100 % capacity. Verify the redundancy state using the “show switch
virtual redundancy” command:
Step 12 In the console of the VSS active switch you will receive an ISSU message when the loadversion
stage is completed:
podX-vss#
*Mar 25 17:20:51.006: %ISSU_PROCESS-SW1-3-LOADVERSION: Loadversion has completed.
Please issue the 'issu runversion' command after all modules come online. ß
Note,
We
need
to
wait
for
all
the
line
cards
to
be
up
before
we
proceed
here.
Verify that all the Line cards are up in Standby Supervisor before issuing “issu runversion”. Also we
are going to check that the new code has been loaded. To verify this, run the command below:
podX-vss#show module sw 2 ß
Verify
if
the
Line
cards
on
Standby
chassis
are
up.
Switch Number: 2 Role: Virtual Switch Standby
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 5 Supervisor Engine 2T 10GE w/ CTS (Hot) VS-SUP2T-10G SAL16127TU2
2 4 CEF720 4 port 10-Gigabit Ethernet WS-X6704-10GE SAL08435CUF
4 20 DCEF2T 4 port 40GE / 16 port 10GE C6800-16P10G SAL1834ZAA7
We are also going to check, that the active switch still has the old code:
podX-vss#show module sw 1
Switch Number: 1 Role: Virtual Switch Active
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 5 Supervisor Engine 2T 10GE w/ CTS (Acti VS-SUP2T-10G SAL16052XQX
2 4 CEF720 4 port 10-Gigabit Ethernet WS-X6704-10GE SAL1019MC05
4 20 DCEF2T 4 port 40GE / 16 port 10GE C6800-16P10G-XL SAL1834ZAKB
Step 13 Since the TenGig intefaces on module 4 (int tengig 2/4/5, int tengig 2/4/6) on SW2 are
connected to FEX 101 and FEX 102 respectively,they will be in link state “down” till the line card
comes up. We will wait till then before proceeding:
Step 14 Eventually when running “show fex 101/102 detail” you should get to the output below
which indicates that you are ready for the next step in the issu process
Step 15 Check the results of the ping’s in the 4948 during the “issu loadversion”:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!
Success rate is 100 percent (50000/50000), round-trip min/avg/max = 1/1/36 ms
podX-4948-sw#
Step 16 We will now check the output of the command “show issu state”. Notice that there are two
boot variables for the Standby RP. This is because there is a timer running. If the conversion is
not complete by the time the timer expires, then the system will fall back to the older code,
which is listed second in the list.
Slot = 1/1
RP State = Active
ISSU State = Load Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12; ß
“OLD
IMAGE”
Slot = 2/1
RP State = Standby
ISSU State = Load Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1a.bin,12;bootdisk:s2t54-
adventerprisek9-mz.SPA.152-1.SY1.bin,12 ß
“NEW
IMAGE”
followed
by
“OLD
IMAGE”
102 FEX_UPGRADE_INIT ß
FEX
will
be
considered
as
a
part
of
ISSU
upgrade
101 FEX_UPGRADE_INIT ß
FEX
will
be
considered
as
a
part
of
ISSU
upgrade
Step 17 In this step we will run the “issu runversion” command, which causes the supervisor or
chassis switchover such that Switch 2 (previous Standby VSS) now assumes the Active role
whilst Switch 1 (previous Active VSS) is reloaded. At this time, Switch 2 transitions from an SSO
Hot Standby state to the Active state. During switchover ~200msec of traffic loss is expected.
Please refer to the “Exercise Description” section for the image that shows this action.
Verify traffic loss experienced with this step on the 4948, you should notice loss of 1 or less ping
packets. If tested with traffic generator, usually 200 msec of traffic loss is seen.
Have your 4948 configured as follows, but do not accept “Sweep range of sizes [n]:” until you
are ready to proceed with the “issu runversion” command in the step below:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 10X.101.10.1
Repeat count [5]: 50000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]:
ß
WAIT!
Don’t
accept
the
“Sweep
range”
until
mid
next
step
Now run the “issu runversion” command in the VSS system and hit <ENTER> in the 4948
to launch the ping sweep just before running “issu runversion”. You can check the output
taken from Switch 1 below:
podX-4948-sw#
Sweep range of sizes [n]: ß
Hit
Enter
for
the
ping
sweep
in
4948
just
prior
to
issuing
issu
runversion
Type escape sequence to abort.
Sending 50000, 100-byte ICMP Echos to 10X.101.10.1, timeout is 2 seconds:
podX-vss#issu runversion
System configuration has been modified. Save? [yes/no]: yes ß
Save
if
prompted
This command will reload the Active unit. Proceed ? [confirm] ß
Enter
to
confirm
the
runversion
%issu runversion initiated successfully
<..snip..>
<..snip..>
Step 18 Switch 2 immediately takes over as the Active Supervisor while previously active supervisor
(SW1) reloads. During this transition there is minimal packet loss (1 Packet). We can check this
in the output of the ping sweep in the 4948 switch:
<Snip>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!!!!!!!!!! ß1
Packet
loss
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Step 19 Shortly afterwards, Switch 1 will also return online with the old version of software. Verify the
redundancy state of the VSS using the “show switch virtual redundancy” command
(please note that we will have to enter this command in the console of the new active switch; the
console of the new standby switch is now disabled)
podX-vss#show switch virtual redundancy ß
new
active
switch
My Switch Id = 2
Peer Switch Id = 1
Last switchover reason = user forced
Configured Redundancy Mode = sso
Operating Redundancy Mode = sso
BOOT = bootdisk:s2t54-adventerprisek9-mz.SPA.152-
1.SY1.bin,12;
CONFIG_FILE =
BOOTLDR =
Configuration register = 0x2102
Fabric State = ACTIVE
Control Plane State = STANDBY
Step 20 Check that all linecards in switch 1 are up. Also Note that the data path from FEX 101 is UP for
both Active Switch (SW2) running the new code and the HotStandby Switch (SW1) running the
old code. You can optionally look also at FEX 102 and at switch 2.
podX-vss#show mod sw 1
Switch Number: 1 Role: Virtual Switch Standby
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 5 Supervisor Engine 2T 10GE w/ CTS (Hot) VS-SUP2T-10G SAL16052XQX
2 4 CEF720 4 port 10-Gigabit Ethernet WS-X6704-10GE SAL1019MC05
4 20 DCEF2T 4 port 40GE / 16 port 10GE C6800-16P10G-XL SAL1834ZAKB
podX-vss#show mod sw 2
Switch Number: 2 Role: Virtual Switch Active
---------------------- -----------------------------
Mod Ports Card Type Model Serial No.
--- ----- -------------------------------------- ------------------ -----------
1 5 Supervisor Engine 2T 10GE w/ CTS (Acti VS-SUP2T-10G SAL16127TU2
2 4 CEF720 4 port 10-Gigabit Ethernet WS-X6704-10GE SAL08435CUF
4 20 DCEF2T 4 port 40GE / 16 port 10GE C6800-16P10G SAL1834ZAA7
Step 21 Verify current ISSU state using the “show issu state” command. Notice that Switch 1 still
has OLD_CODE as the only option in its Boot Variable, but Switch 2 has both OLD_CODE and
NEW_CODE.
Slot = 2/1
RP State = Active
ISSU State = Run Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-
1.SY1a.bin,12;bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12
Slot = 1/1
RP State = Standby
ISSU State = Run Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12;
101 FEX_UPGRADE_INIT ß
FEX
will
be
considered
as
a
part
of
ISSU
upgrade
102 FEX_UPGRADE_INIT ß
FEX
will
be
considered
as
a
part
of
ISSU
upgrade
Step 22 “Issu Acceptversion” provides a window of time to try out the new software version
running on switch 2 and check if things are working as expected. If the “issu
acceptversion” command is not applied within the rollback timer period, the EFSU software
upgrade process will be terminated and the system will try to rollback to the old-image. In this
step we will verify the rollback timer using the “show issu rollback-timer” command
podX-vss#show issu rollback-timer ß
In
this
step
we
only
verify
that
the
rollback-‐timer
is
runnung
Rollback Process State = In progress
Configured Rollback Time = 00:45:00
Automatic Rollback Time = 00:31:02
podX-vss#issu acceptversion
% Rollback timer stopped. Please issue the 'issu commitversion' command.
podX-vss#show issu rollback-timer ß
In
this
step
we
only
verify
that
the
rollback-‐timer
is
runnung
Rollback Process State = Not in progress
Configured Rollback Time = 00:45:00
Step 24 Since our network also has Instant Access clients, we are going to proceed and upgrade those
clients in this step using the added command “issu runversion fex”. We have different
options to proceed with this upgrade: all the clients at once of in stages. We will look at the
options available running the command below:
Step 25 We are now going to upgrade all the Instance access clients at once running the command
“issu runversion fex all”
NOTE: Instead of doing “issu runversion fex all”. The FEX clients can be upgraded one at a time by issuing “issu
runversion fex <fex-id>. Or It can have rolling upgrade across multiple FEX’s by issuing “issu run version fex <fex-
id1>, <fex-id2>…
podX-vss#issu runversion fex all
Step 26 We can now monitor the status of the upgrade using the “show issu state” command.
Slot = 2/1
RP State = Active
ISSU State = Run Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-
1.SY1a.bin,12;bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12
Slot = 1/1
RP State = Standby
ISSU State = Run Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12;
101 FEX_UPGRADE_IN_PROGRESS ß
Image
download
initiated
for
FEX.
FEX
will
start
image
download
and
extraction
of
downloaded
tar
image
and
rebooting
with
new
image.
102 FEX_UPGRADE_IN_PROGRESS
Step 27 You will see the following output from VSS active console during the FEX upgrade:
podX-vss#
*Dec 7 14:28:50.423: %FEXMGR-SW2-6-IMAGE_DNLD_STATUS: (FEX 101) ISSU Image Download :
In progress
*Dec 7 14:28:50.439: %FEXMGR-SW2-6-IMAGE_DNLD_STATUS: (FEX 102) ISSU Image Download :
In progress
<..snip..>
<..snip..>
Step 28 We can now monitor the completion of the upgrade using the “show issu state” command.
This step takes a few minutes to you will have to wait to get to the
“FEX_UPGRADE_COMPLETE” state.
podX-vss#show issu state
The system is configured to be upgraded in staggered mode.
2 supervisor nodes are found to be online.
Summary: an in-tandem upgrade is in progress.
Slot = 2/1
RP State = Active
ISSU State = Run Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-
1.SY1a.bin,12;bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12
Slot = 1/1
RP State = Standby
ISSU State = Run Version
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12;
Step 29 The final step of the EFSU process is “issu commitversion”. Once the new image is ready
to be rolled out, perform the “issu commitversion” command. It will cause a reload of
Switch 1 (VSS Standby) so that it can load with the new software image. We will verify and we
will verify traffic loss experienced with this step on the 4948. You should notice a maximum of 1
ping packet being lost. If tested with traffic generator, usually 200 msec of traffic loss is seen.
Have your 4948 configured as follows, but do not accept “Sweep range of sizes [n]:” until you
are ready to run with the “issu commitversion” command in the step below:
podX-4948-sw#ping
Protocol [ip]:
Target IP address: 10X.101.10.1
Repeat count [5]: 50000
Datagram size [100]:
Timeout in seconds [2]:
Extended commands [n]:
Sweep range of sizes [n]: ß
WAIT!
Don’t
accept
the
“Sweep
range”
until
mid
next
step
Perform the “issu commitversion” in the VSS system and you can compare with the output
below taken from Switch 2, current active switch:
podX-4948-sw#
Sweep range of sizes [n]: ß
Hit
Enter
for
the
ping
sweep
in
4948
just
prior
to
issuing
issu
commitversion
Type escape sequence to abort.
Sending 50000, 100-byte ICMP Echos to 106.101.10.1, timeout is 2 seconds:
podX-vss#issu commitversion
%issu commitversion initiated successfully, upgrade sequence will continue shortly
podX-vss#
*Dec 7 15:39:45.699: %ISSU_PROCESS-SW2-3-COMMITVERSION: issu commitversion;
Commitversion sequence will begin in 60 seconds. Enter 'issu abortversion' to cancel.
*Dec 7 15:40:03.483: %SYS-SW2-5-CONFIG_I: Configured from console by console
*Dec 7 15:40:15.699: %ISSU_PROCESS-SW2-6-COMMITVERSION_INFO: Resetting Standby
shortly
<..snip..>
*Dec 7 15:45:42.439: %OIR-SW2-6-INSREM: Switch 1 Physical Slot 4 - Module Type
LINE_CARD inserted
Copyrighted Cisco® Systems Inc. December 2015 72
Cisco Catalyst Virtual Switching System and Instant Access Lab Guide
Step 30 Verify traffic loss experienced with this step on the 4948. You should notice a maximum of 1
ping packet being lost. If tested with traffic generator, usually ~200 msec traffic loss is seen. At
this point, the VSS will forward traffic with 50% bandwidth capacity until Switch 1 completely
boots up with the NEW_IMAGE
Step 31 Once Switch 1 has come online, execute “show issue state” and notice the process is
completed and returned to “Init” state:
Slot = 2/1
RP State = Active
ISSU State = Init
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-
1.SY1a.bin,12;bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12
Slot = 1/1
RP State = Standby
ISSU State = Init
Boot Variable = bootdisk:s2t54-adventerprisek9-mz.SPA.152-
1.SY1a.bin,12;bootdisk:s2t54-adventerprisek9-mz.SPA.152-1.SY1.bin,12
101 FEX_INIT ß
FEX
is
online
sate
(exists
only
when
ISSU
is
not
in
progress)
102 FEX_INIT
Step 32 Once again we now need to manually boot the second switch in the stack for FEX 101 and FEX
102.
Loading "flash:/c6800ia-universalk9-mz.152-3m.E2/c6800ia-universalk9-mz.152-
3m.E2.bin"...Verifying image flash:/c6800ia-universalk9-mz.152-3m.E2/c6800ia-
universalk9-mz.152-
3m.E2.bin.............................................................................
......................................................................................
......................................................................................
..........................Image passed digital signature
verification..........................................................................
......................................................................................
............................................................................Image
passed digital signature verification
@@@@@@@@@@@@@@@...
Loading "flash:/c6800ia-universalk9-mz.152-3m.E2/c6800ia-universalk9-mz.152-
3m.E2.bin"...Verifying image flash:/c6800ia-universalk9-mz.152-3m.E2/c6800ia-
universalk9-mz.152-
3m.E2.bin...........................................................
Step 33 Once Switch 1 has come online, execute “show module switch 1” and notice that the
version of code in the “Sw” column is NEW_IMAGE running on C6k. Execute “show module
switch 2”. Is the same information there for Standby C6k switch?
Step 34 Verify the software image in C6800IA clients are also upgraded. Issue the following command:
podX-vss#show version fex 101
==> Showing version for FEX 101:
Cisco IOS Software, C6800IA Software (C6800IA-UNIVERSALK9-M), Version 15.2(3m)E2,
RELEASE SOFTWARE (fc5)
Technical Support: http://www.cisco.com/techsupport
Copyright (c) 1986-2015 by Cisco Systems, Inc.
Compiled Fri 18-Sep-15 09:59 by prod_rel_team
<..snip..>