Upgrading code on 6500 w/ Redundant Supes

Unanswered Question
Oct 23rd, 2009

I am trying to see if there is a way to upgrade the code on a 6513 with redundant supervisors in a way that has no downtime.

It seems that all is well when I have SSO redundancy but then once I upgrade on one supe, SSO redundancy changes to RPR redundancy which means a code upgrade requires the supe to reload when it takes over.

Right now at best I am seeing about 250 seconds of downtime when I upgrade.

Let me know if anyone has a way around this.



  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
jfraasch Fri, 10/23/2009 - 04:36

Thanks for the link. I will check it out and report back. I have the 6513s in a lab so I can see if it works.


Reza Sharifi Fri, 10/23/2009 - 05:08

Hi James,

I used ISSU in VSS mode using the same document Giuslar provided. It worked fine with no down time at all.

Let us know how it goes with 1 box and 2 sups.



jfraasch Fri, 10/23/2009 - 07:35

Not good so far.

It worked the same as the old way but there is an error somewhere that says that my modules in the Show ISSU OUTAGE SLOT ALL command are all MDR State Invalid with error number 3.

So I still had 3 minutes of downtime when I tried.

I'll update once I hear back from TAC.


jfraasch Mon, 10/26/2009 - 06:49

So here is where it stands at the moment.

As per previous replies, the eFSU (enhanced Fast Software Upgrade- or even sometimes called ISSU) is in effect if you have Sup720 or above and you have WS-X67xx modules installed on your chassis, and you have two identical supes in your chassis.

Unfortunately for me, I have an Ace module and a FWSM and some 10/100 blades for some old servers

Since I have so many modules that are not compliant with the WS-x67xx part of the requirement, I will always have a reload as part of my upgrade process.

It seems that ALL the modules need to be compliant in order to be able to do the eFSU without chassis downtime.

If anyone know anything different let me know.




This Discussion