HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Hours:
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
BladeSystem Virtual Connect
cancel
Showing results for 
Search instead for 
Did you mean: 

Update Virtual Connect from v3.17 to v3.51 (with stacking)

 
chuckk281
Trusted Contributor

Update Virtual Connect from v3.17 to v3.51 (with stacking)

Question from Michael on updating VC firmware:

 

****************

 

i just wanted to ask if there are any people out there who already updated C-class Enclosures VC to 3.51 from 3.17 or similar versions. Reason is: Lately we experienced trouble with stacked enclosures as the firmware was not rolled out correctly among all modules in the domain. OA was 3.32 and I was not able to explain why this happened.

 

As we are facing the same setup on another customer next week we are not sure if we will run into the same issue again.

 

If somebody has done this before, could you share your way of doing this?

 

******************

 

Input from Jos:

 

****************

 

Check the known issues chapter of the VCSU 1.6.0 Release Notes on http://bizsupport2.austin.hp.com/bc/docs/support/SupportManual/c02923547/c02923547.pdf

 

There is a workaround for updating from VC 3.15 , VC 3.17 and VC 3.18.

 

***************

 

And good info from Lawrence:

 

*******************

 

There are a few things to be aware of and I just had to sort some of these things out for a colleagues, so it’s fresh in my head.

 

  1. Ideally your customer can give you a complete outage to your stacked enclosures so that you can also upgrade the blade servers to the right firmware version of BIOS, PMC, iLO, LOM firmware, etc using SPP.  If not, then your customer will have to manage the rolling outages required to get all the servers up to date.  SPP will have updates for firmware as well as OS updates.  Some updates like the Windows NCU update may require you to dissolve and reteam NICs.  Be careful that the customer is not on a very old PSP like ver. 8.30 and they want to just jump straight to SPP 2012.01.0 – the Windows OS upgrades may not work properly.
  2. Once all the blade servers are updated, the customer may want to upgrade the VC modules with the servers online – if that’s the case, you may want to let them know that FCoE connectivity may be interrupted if they have Mezz cards - http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c02950493 
  3.  Before upgrading the OA, check to make sure that the customer’s SIM server is at least version 6.3, and preferably patched with the latest hotfix patch Dec 2011. 
  4.  I would recommend upgrading the OA first before doing the VC firmware, and at minimum, I’d recommend OA 3.32 for VC 3.51.  Better still to be on OA 3.50 or 3.55 but please read this advisory to decide on which one to go for: http://sawpro.atlanta.hp.com/km/saw/view.do?docId=emr_na-c03264762 
  5. Before upgrading VC firmware from 3.17 to 3.51, you will need to know if the customer has VCEM.  If they have VCEM, then you need to make sure that it is updated to something recent that supports 3.51, e.g. VCEM 6.3.1.  If indeed you need to update VCEM be sure to observe this advisory:  http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c03232068

(You’ll need to set VCEM into maintenance mode before doing upgrades described in the rest of the steps below.  Later you’ll need to change the VCEM’s Domain Group firmware mode using the “upgrade firmware mode” feature. IE 7 is better than IE 8 for VCEM)

 

6.  Check to see if the VC modules are connected with copper DAC cables or fibre optic cabling.  If it is copper, make sure that they are no longer than 5 meters otherwise there could be connectivity issues after the VC firmware is applied.  This is because cabling rules are enforced in VC 3.51

 

7.  Use VCSU 1.60 (my personal preference over HPSUM) and if doing the upgrade from a HP COE machine, be sure to disable McAfee.   You’ll of course need the customer’s OA and VC logins and passwords.

 

8.  Backup the VC domain, perform a health check to ensure modules are in sync and if everything looks good, go for the update.

 

9. Make the necessary changes in VCEM to get out of maintenance mode, as mentioned above in Step 5.

 

Hope this helps clarify some of the potential minefields.  I did not go into a lot of detail into VCEM but feel free to ask. 

 

**************************

 

Reply from Michael:

 

***************

 

Thanks for this huge block of information.  Very helpful indeed.

 

As we are using OA 3.32 atm and no VCEM we are good to go. Blades are pretty much up to date. Only VC is missing.

 

 

What bothers me is the enclosure-stacking. Is there something in particular which has to be done here? Or do we simply use the VCM IP with corresponding OA IP / Credentials for the VCSU? Rest is done automatically? Stacking Cookbook wasn’t that helpful I have to say. Only says the update will run “automatically”.

 

****************

 

Reply from Lee:

 

***************

 

When enclosures are stacked with a single VCM instance, then all VC modules in the stacked enclosures will be updated at the same time.  As the document says, this will be handled “automatically” by VCSU or HPSUM.   The reset of the modules to activate the new firmware will be staggered so that a network or SAN outage is avoided if all the servers have network teaming or multipath configured properly.

 

If you use VCSU to run a health check, a list of the modules will be output and you should see that all modules in the enclosure stack are listed.

 

If you are using VCSU, you specify the OA IP address and VCSU will request the VCM IP address from the OA.  VCSU will prompt you for the OA credentials and then the VCM credentials.

 

*******************

 

Long replies but good info. Any additional questions?