cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3975
Views
0
Helpful
5
Replies

Catalyst 6509 VSS IOS upgrade

jackson.ku
Level 3
Level 3

Hi,

We have a Catalyst 6509 VSS system, each chassis have 2 supervisor engines. The IOS version is 12.2(33)SXI4a. We should upgrade to 12.2(33)SXI12.

The following document mention 2 upgrade method : FSU & eFSU

http://www.cisco.com/en/US/docs/switches/lan/catalyst6500/ios/12.2SX/configuration/guide/vss.html#wp1170391

We can not use eFSU due to the images with release dates more than 18 months apart, so we can use FSU only. And there is some note for FSU :

Note VSS mode supports only one supervisor engine in each chassis. If another supervisor engine resides in the chassis it will act as the DFC

It make me some confuse.... What is the correct procedule to upgrade the Caytalyst 6509 VSS IOS ( each chassis with 2 supervisor engine )?

Best Regards,

1 Accepted Solution

Accepted Solutions

Hello Jackson,

SXI4a to SXI 12 cannot undergo ISSU directly. If you really want to do only an ISSU then there will be more than one iteration and involving multiple upgrades.

The more straightforward method and involving less reboot is the FSU method. It requires

downtime and maximum 20 to 40 min is what we expect if all goes well.

Haihua

View solution in original post

5 Replies 5

Maria Yurkina
Level 1
Level 1

Hello.

I can say, we upgraded VSS IOS last year. SXI5 to SXJ4. It's more than 18 month b/w releases (see compatiability matrix). We used eFSU. Traffic disruption was 1 ping on the commitversion.

BUT. It's better if u can test.

Can't tell about the second sup. May be later

Haihua Rong
Level 1
Level 1

Hello Jackson,

Please take a on the next post which may answered your questions:

https://supportforums.cisco.com/thread/2188244

Intrachassis Availability

The initial release of the Cisco Virtual Switching System supports only a single supervisor per chassis. If a second, or redundant, supervisor is installed in an individual chassis then the redundant supervisor will not fully boot. The redundant supervisor will stop the boot process at the ROMMON stage.

In this configuration any device connected to the chassis in a single-homed, or single-attach, manner must rely on the availability of the single supervisor. Therefore the recommendation for connecting to the VSS is to always dual-attach devices.

As a result of the single supervisor per chassis support the recovery period for replacing a failed supervisor module is undeterministic in that the recover process requires manual intervention in order to install and initialize a new supervisor in the chassis.

Beginning in the 12.2(33)SXI4 software release, Quad-Sup Uplink Forwarding is supported which allows for a redundant supervisor to fully boot Cisco IOS Software, thereby providing a deterministic recovery option for redundant supervisors in a VSS chassis.

Refer:

http://www.cisco.com/en/US/prod/collateral/switches/ps5718/ps9336/white_paper_c11_429338.pdf 

The link that you mentioned describe how to configure a VSS from release 12.2(33)SXH1.

Below is a step by step explanation of the upgrade process and the downtimes associated with each step:
 
http://www.cisco.com/en/US/prod/collateral/switches/ps5718/ps708/white_paper_c11-729039.html

ISSU restrictions and guidelines.
 
http://www.cisco.com/en/US/docs/switches/lan/catalyst6500/ios/12.2SX/configu
ration/guide/vss.html 
 
Video:
 
https://supportforums.cisco.com/videos/2650
 

Best regards,

Haihua

Hi,

Thanks for your reply. I checked the SX/Y eFSU (ISSU Phase 1) Compatibility Matrix, the status for upgrade from 12.2(33)SXI4a to 12.2(33)SXI12 is "P (Pruned) : The two images are not tested for an eFSU. eFSU upgrade or downgrade is only fully supported between images released within 18-month window"

Can I still use eFSU to upgrade my VSS?

Best Regards,

Hello Jackson,

SXI4a to SXI 12 cannot undergo ISSU directly. If you really want to do only an ISSU then there will be more than one iteration and involving multiple upgrades.

The more straightforward method and involving less reboot is the FSU method. It requires

downtime and maximum 20 to 40 min is what we expect if all goes well.

Haihua

Leo Laohoo
Hall of Fame
Hall of Fame
It make me some confuse.... What is the correct procedule to upgrade the Cay#talyst 6509 VSS IOS ( each chassis with 2 supervisor engine )?

Doing an IOS upgrade of a chassis with dual supervisor card is easy.  Doing this on a VSS with dual cards (per chassis) can get mighty confusing.

Before you start doing the below steps it is VERY IMPORTANT you compare the MD5 has value of the IOS found in your TFTP server against the MD5 hash value found in the Cisco website.  If they do not match, then you need to download the IOS again and compare.  They must match.  If they don't match and you put them into your chassis, the supervisor cards will go into ROMmon.

Here's how it's done.

1.  Copy the IOS from your TFTP server to your sup-bootdisk: (primary card, primary chassis);

2.  Then issue this command:  tftp flash sup-bootdisk:IOS.bin

3.  Next, copy the IOS from your sup-bootdisk to the redundant supervisor card.  I don't remember what the command is but it should be something like this:  copy sup-bootdisk:IOS.bin backup-sup-bootdisk:

4.  Once that is done, copy the IOS from your sup-bootdisk: to your secondary chassis using the command (again, I am recalling):  copy sup-bootdisk:IOS.bin slavesup-bootdisk:

5.  Do the same to copy the IOS from the sup-bootdisk: to the secondary chassis' secondary card.

And finally

6.  Once you've done this, change the bootstatement.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: