Showing results for 
Search instead for 
Did you mean: 
Cisco Employee

Top challenges with BE6000 upgrades and migrations

How many of you are looking at migrating to Collaboration System Release 10.6 or 11.0?

What are you migrating from?   BE6000 8.x/9.x, BE5000 or an older-version UCM environment on MCS 7800?

What are your biggest obstacles getting to 10.x / 11.0?

Very interested in getting a "fresh pulse" from the community.  We've seen mass moves first to virtualization and then to packaged collaboration solutions like BE6000 solving many of the issues from the past, but what is in your way today?


Joe Cooke

Additional app support for PCD-based "migrate" is a BIG ask from the field.  Works great for CUCM - need to see additional apps support it as there is a significant installbase currently on MCS.


“Works great for CUCM” – please say that again into my tape recorder… ☺

For additional apps, how would you prioritize / tier the apps ? (I put the list below so you don’t have to remember) In what priority order would you put the remaining apps in the suite?

Already PCD-Migrate-able: UCM (including SME) and IM&Presence

Providing alternative to PCD Migration: Unity Connection (use COBRAS to get to 10x/11x), Contact Center Express (their own process to get to 10x/11x), after that both are PCD-Upgrade-able.

The rest of the apps suite we preload on BE6000 (and BE7000):

Prime Collaboration Provisioning

Paging Server

Emergency Responder


TelePresence Server

Video Communication Server


TelePresence Management Suite

TelePresence Content Server


Trad Tobaishat

On BE6000 and BE7000, Can we use the PCI and LOM for NIC Teaming? Is this suppported? are we free to use the networking part or are there any restriction on NICs?

The BE6000 Document just talk about the teaming vnic0 and vnic1 but nothing about the rest of the NICs. I get this question alot but I couldnt find answers for it.

Also, What is the recomended process for migration that has a domain change? Can PCD change the Domain on version 11.0?

The factory preloads are setup a default way for easy out of the box and making CTOP and preconfigured image VMs work right.

But for those who know how there isn’t a support restriction on using the other ethernet ports … as long as you’re doing something allowed by UCS C.

E.g. teaming a LOM port and a NIC port is a common physical link redundancy tactic. Note we only put NICs in the BE6000H as the BE6000M VM density isn’t high enough to need lots of pnics, and we tend to see less VLAN trunking and other situations in BE6000M-sized deals.

General network best practices for virtualized collab here (some not relevant to BE6000):


One of the biggest issue I see is to fit the Prime collaboration provisioning on the Brown field environment. Is there any recommendation for it?


Are you talking add of PCP to existing BE6000 or virtualized deployment, or migration of what you have to a BE6000 with PCP ?

Can you provide more details on what your brownfield looks like? Servers (# and which), which apps ( & #app nodes or VMs), etc.


Hi Trad,

Prime Provisioning will allow you to import an existing configuration for brown field deployments. There is no requirement to start from scratch, if that’s what you are asking.


Hi Andrew,

Could you please point to proper guide that describe this part?



Hi Trad,

You should find what you need in this section of the PCP Admin Guide.



The past 3 BE6K's that I've installed came pre-installed and pre-configured with IP addresses on some bogus IP subnet, with configured route partitions, calling search spaces, dns, ntp, etc. It is much easier to blow away that VM, redeploy a new .ova, and re-install the application (CUCM, CUC, CUPS, etc) than it is to go through everything that needs to be changed.

Please, tell us how to order without this pre-built configuration that takes more time to un-do than it is supposed to save.

Previous distributions of BE6K server came with pre-deployed .ova's, with the .iso simply loaded in the dvd.

Recall we have a few ways to perform install and first-time-setup with a BE6000 appliance, to provide some efficiency and cost reduction options vs. differences in how various partner practices work.

1. Traditional way (find OVA download file, find ISO image file, deploy OVA, install app from ISO, manually config via CLI/GUI, etc.) … many of these have moved to a “skip install” where the OVA is deployed and the app is partially installed but not configured.

2. Cisco Config to Order Portal (order server from distributor, fill out web page, they do the rest and ship to you with the config settings you entered)

3. Preconfigured Image VMs (predeployed OVA, fully preinstalled app, preconfigured with “canned / one size fits most” common configuration, if this works for customer use in lieu of #1 or #2)

There are different files in the preload datastore for #1 vs. #3.

E.g. #1 is the “OVA containing blank VMs” like you’d pull from Software Download, and ISO installer file you’d pull from PUT or be eDelivered / shipped on a DVD for new purchase.

Vs. #3 has a bunch of “OVAs containing preinstalled and preconfigured apps”.

At time of install you would pick the appropriate files for #1 or #3 based on which way you wanted to go.

See the Release Notes, Preload Summaries and Software load docs here for more details:

The BE6K Install Guide is also changing to call out these options.



When we introduced the preconfigured images, we would have liked to continue also having the predeployed “skip” images too… Unfortunately, there just wasn’t enough disk space to accommodate both options.

We do still ship the BE7000 with the skip install VMs though.

We’re evolving the CTOP offer, such that we’ll probably return to predeploying the skip VMs again in the not too distant future.

It’s worth a quick poll though…. Of those of you reading this thread, who is using the preconfigured images and would like to keep them? Who would prefer a return to the skip VMs?


Recognize Your Peers
Content for Community-Ad