The differences between the two that you’ll most likely encounter during the upgrade are in the way the two operating systems are installed through Cisco’s appliance image (ISO). UCOS 12.5 is based on CentOS Linux rather than the RedHat Linux so many of us are familiar with. While 2 hours per server is a fair amount of time, it will mostly take that amount of time and doesn’t include a lot of spare time for downloading / uploading files. Know how the business operates and propose time frames that are likely to be the least impactful to the business.įinally, have all the necessary upgrade files and images already staged in the environment before the upgrade. The client in my recent upgrade is a 24/7/365 medical facility, so finding the perfect 18 hour window was very challenging keep this in mind when crafting your business statement for the upgrade expectation. Ultimately, I was able to get it done in just under 14 hours, and that is with a few road bumps encountered. In the 12.5 upgrade I completed recently, there were 8 UC servers involved so I gave the client an 18 hour total expectation (I always add 2 extra outage hours to every actual outage estimate). Personally, I like to estimate a 2 hour maintenance window per UC server for a refresh upgrade. You must consider that while this won’t render a UC cluster of servers completely unusable, it will created degraded experiences and, UC applications without a cluster or HA server will be completely unusable during the upgrade. The most substantial difference is that the server is powered down during the upgrade rather than being able to remain up while the upgrade is installed into the inactive partition.
Understand that an upgrade to version 12.5 is a refresh upgrade which is different than non refresh upgrades. Its generally a bad idea, it is not supported and if it isn’t done with precision, can cause way more problems than it was ever meant to solve. Some UC ninjas have tried snapshots while the VM is powered on, but I’m not going to discuss or endorse that. However, DRS is the only way to backup with the server powered on, and be officially supported. That said, if you have the option, allowance and ability powering down each server and copying the entirety of the server’s VM files to a separate location will do just fine as well. This is, ultimately, your only officially and completely supported fail-safe to recover from a total loss server failure. Cisco Unified Communications Manager v10.5Īs a matter of good practice, ALWAYS have a complete and recent DRS backup set from each Cisco UC server application before attempting any upgrade.In this blog, I’ll be covering the upgrade experience from the following applications I won’t go into procedural detail about Cisco UCOS upgrades generally, as its beyond the scope of this blog, but I will share some upgrade basics and then some of the unique experiences in the 12.5 upgrade that are different from other version upgrades.
I recently completed an v10.5 to v12.5 upgrade in a large environment and it was completely successful ( flawless victory for the 90s Midway / Acclaim Entertainment fans.), but there were several things I learned. I’ve decided a blog detailing the upgrade experience in Cisco’s UCOS 12.5 UCCX 11.6 release would be a good enough read for a blog, as there are some some items to look out for.