6500 upgrade to 12.2.33SXI

Unanswered Question
May 5th, 2009


We are in the process of upgrading a network of 12 6509's from 12.2.18SXF5 to 12.2.33SXI1. So far we have upgraded 8 nodes, and everything has been running smooth.

But - yesterday I upgraded another node and got lots of problems. For example:

- I got HSRP flapping on physical interfaces and subinterfaces (not SVI's)

- The CPU util increased dramatically. Sometimes i couldn't even save the configuration file.

It seem like it has "calmed down" a bit now, but I am evaluating to downgrade back to 12.2.18SXF5.

1. Is it any issues with downgrading from 12.2.33 SXI to 12.2.18SXF5?

2. If we install a modular image - is the downgrade process more complex then?

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (1 ratings)
stephen.sanyaol... Wed, 05/06/2009 - 00:50

Hello Jorgen Sorqvist,

Good to know you have already upgraded 8 of the switches successfully. I am about doing this in couple of days. Can I have the step by step process you took for the upgrade. I will equally appreciate any useful documentation for this.


stephen.sanyaol... Wed, 05/06/2009 - 02:36

The current image on the 6509E is System image file is "sup-bootdisk:s72033-ipservicesk9_wan-mz.122-33.SXH3.bin". We are looking to upgrade to 12.2(33)SXI1. I need the step by step process to upgrade this including what to avoid in the process. Any useful documentation will be good for me. Just to leverage on your experience that you have upgraded 8 of the switches already


jorgensor Tue, 05/12/2009 - 05:51

Shouldn't be any problem to upgrade frpm 12.2.33SXH to SXI1.

Of course you have to check rel.notes to make sure (it depends on the services you use in the network).

From the image file name I can see that you are running standard (not modular) sw.

mz - not modular

vz - modular

jorgensor Tue, 05/12/2009 - 05:48

It depends on if you shall install modular sw or not.

1: Ordinary SW (not modular)

- download correct image to the node (f.ex sup-bootdisk:).

- backup the config (just to make sure)

- leave the old image if you need to rollback

- change bootvar:

conf t

no boot system

boot system sup-bootdisk:


- Reload

See also http://www.cisco.com/en/US/products/hw/switches/ps708/products_configuration_example09186a0080116ff0.shtml

2. Modular image:

All steps under 1: and in addition:

- install file sup-bootdisk: sup-bootdisk:/sys

- change bootvar:

conf t

no boot system

install bind sup-bootdisk:/sys



- Reload

See also http://www.cisco.com/en/US/docs/ios/swmod/configuration/guide/12_2SX/sw_mod_12_2SX_book.html

stephen.sanyaol... Wed, 06/10/2009 - 06:28

The following is the procedure that was used. This may useful for other professionals.

1. Take a text copy on a laptop

2. copy start tftp ---- to back up the configuration

3. copy tftp sup-bootdisk: slavesup-bootdisk: ---- to download the image to the Supervisor Engine bootflashes of both active and standby supervisors

4. no boot system sup-bootdisk:oldiamge.bin ---- This removes the current boot statement

5a. boot system sup-bootdisk:newimage.bin ---- This modifies the boot statement to boot from the new image

5b. boot system sup-bootdisk:oldimage.bin

6. copy run start ---- This also triggers a manual synchronization of startup configuration on the standby Supervisor Engine.

7. sh bootvar ---- This verifies the modified boot statements.

8. hw-module module 6 reset ---- This resets the standby Supervisor Engine in slot 6. Before you reset the standby Supervisor Engine, make sure you wait long enough to ensure that all configuration synchronization changes have completed.

9a. show module ---- Verify the redundancy status

9b. show redundancy states ---- Verify the redundancy status

10a. redundancy force-switchover ---- This reloads the active unit and forces switchover to standby. The Supervisor Engine in slot 5 now becomes the standby. The old active Supervisor Engine in slot 5 reboots with the new image and becomes the standby Supervisor Engine.

NB: Establish a console connection to the Supervisor Engine in slot 6 simultaneously when the switchover is initiated from Supervisor Engine in slot 5.

10b Pause for the firewall to synchronize before doing step 11

11. redundancy force-switchover ---- This reloads the active unit and forces switchover to standby. The active Supervisor Engine in slot 6 reboots with the new image and becomes the standby Supervisor Engine forcing Slot 5 to become the Master


This Discussion