How long has it been since you upgraded your NonStop’s operating system? One year? Five years? Fifty years? You have no idea? Don’t feel embarrassed if you answered with anything other than one year. You’re not alone. Many companies that operate within the HPE NonStop environment have chosen not to upgrade their OS for a variety of reasons. Even though NonStop was designed to withstand the stress of running continuously for years, it still needs to undergo routine OS upgrades just like any complex system.
Common Excuses for Delaying or Avoiding NonStop OS Upgrades
I’ve been around the NonStop block a time or two so let’s begin by addressing the most common excuses I’ve heard over the years.
• Staff resources are working on other tasks. It should come as no surprise this is the No. 1 excuse I’ve heard from clients across multiple industries. I get it. Upgrading the NonStop is a time-consuming process that takes weeks or even months to complete and involves multiple user groups depending on the complexity of the environment. If the upgrade is performed correctly, the majority of time will be spent on researching and examining the current system and any changes that have been made over the years. Like any proactive decision, it’s important for business leaders to understand the long-term value of allocating resources—whether internally or third-party—to upgrade one of their most critical business components.
• Fear of creating a service outage during upgrade. This is a legitimate concern, but it’s also one that can be minimized or avoided altogether. If you’re relying on a single system for business function, then upgrading will require that system to be turned off. If this scenario applies to your organization then you should plan your upgrade to occur when your volume is low, such as Saturday night or Sunday morning, depending on your business. On the flip side, upgrading is a great time to test your company’s disaster recovery plan, which will not only set you up better for the future but also help you minimize any disruptions that may occur during the upgrade.
• Lack of tested procedures for upgrading and cold loading. When it comes to upgrading and cold loading, documentation is king. Some companies have not upgraded or cold loaded their systems in several years, and the employees who were involved last time may no longer be there. Chances are high they did not leave behind proper documentation. So now the next crew has to start from scratch without any notes. No wonder they don’t want to do it! Ideally, you should be testing the environment every year and documenting everything.
• IT team is not experienced in upgrades and cold loading. This issue goes hand in hand with lack of testing procedures. When it comes to managing your NonStop, there’s only so much talent from which to choose. The sad reality is that the NonStop talent pool is getting smaller as older users retire and younger IT professionals are not interested in taking their places.
• Why fix something that isn’t “broken”? This is just a bad mantra to follow. Anyone in IT should know that an outdated system can expose a business to so many potential threats and disasters. As NonStop becomes more standardized, it’s also becoming more vulnerable to cybersecurity attacks. This is why new releases are focusing more on security. In addition, you run the risk of your organization falling out of compliance and relying on an operating system that is no longer covered through technical support.
How to Safely, Securely and Successfully Upgrade Your NonStop OS
After your CIO approves an upgrade, how do you successfully complete it while avoiding accidental interruptions and other potential problems? Here are a few tips and solutions to keep in mind:
• Review procedures to restart OS, utilities, and applications by comparing startup scripts and configuration files to current environment. If someone were to ask me what nightmares are made of, I would say losing years and years of configurations after restarting a NonStop system. Oh, the horror! That’s exactly why this is a critical first step. Think back to the number of times you have reconfigured your NonStop environment within the last year—either at the OS level, utilities, applications or database. Now think back to all the changes that have been made since your last cold load—five, seven, ten years ago or longer. Because NonStop allows you to add components and configurations without restarting the system, you will lose all those changes after you perform a cold load. Add this step to the list of reasons why documentation is king.
• Review release notes, HPE Hotstuff updates, and support documents in an OS release. I always place this step high up in the preparation process. Before restarting your NonStop system, you need to be aware of all the HPE’s latest updates and software product releases so you can identify potential problems and product dependencies.
• Upgrade test system to identify and correct any issues before upgrading production environments. In order to accurately upgrade your testing system, you should first create a checklist of everything that needs to be tested. For example, you may have several functions or applications that will need to be verified before you can upgrade your OS. Automated testing scripts and formal regression testing should be included in your overall testing plans. I’ve also learned over the years that it’s important to schedule tests far apart—typically more than a week—to allow enough time for issues to present themselves.
• Avoid OS and product versions that are too recent. This may come across as a no-brainer, but it’s always a good reminder. You never want to be the guinea pig when it comes to software updates. Allow at least six months to pass before installing the update. That should give the developer enough time to fix any bugs.
• Develop a detailed, task-based upgrade plan and schedule including tasks before, during, and after cold load. Creating a highly detailed checklist will not only serve as the blueprint for your current upgrade but all future upgrades as well. I am constantly tweaking and adding to my checklist. Your plan should be as detailed as possible. In fact, it’s probably impossible to have too many details. Some topics I’ve included are roles and responsibilities, questions and open issues, pre-outage tasks and step-by-step cold load process, such as how to stop and restart applications and shut down the environment and OS utilities. I also recommend assigning someone to record detailed notes of the entire process that you can incorporate into future checklists. Documentation is king!
• Partner with a managed IT services provider. If you lack the necessary resources to perform an upgrade, I recommend partnering with a managed IT services provider like Odyssey Information Services. With 20 years of NonStop experience, Odyssey’s team will work closely with your IT department to learn your NonStop environment inside and out and provide the tools and subject matter experts, in addition to 24/7 coverage, needed to help your business thoroughly and successfully upgrade its NonStop OS.
About the Author