Frequently our customers experience downtime in their HPE Application Lifecycle Management (ALM)/Performance Center (PC) environment for various reasons. How many do you think ever consider maintaining their HPE ALM/PC environment?
Not many.
You should think of your HPE ALM/PC environment as if it were your vehicle or your home. Like your vehicle or home, you must perform regular maintenance checks to keep things running smoothly and looking great. If you ignore your routine maintenance checks, you end up with a costly repair, and the time it takes to make those repairs may be significant. You need to apply this same concept to your HPE ALM/PC environment to keep it running smoothly, because failure to preform maintenance could cost the business time and money.
To help you with developing a maintenance plan for your HPE ALM/PC environment, we have listed some suggested tasks a schedule below. This list of tasks should be the minimum required to keep your HPE ALM/PC properly maintained. If you don’t do everything on the list when you are supposed to, it isn’t a deal breaker; however, you should try to make a similar schedule a routine part of the HPE ALM/PC environment administrator role. If you consistently miss these maintenance tasks, eventually you will have some issues that cause an inopportune down time at a critical point in a project’s testing— that’s just Murphy’s law.
Each category of tasks is cumulative. In other words, if you are planning to do the monthly tasks, you should also perform the daily and weekly tasks before starting.
Daily:
- Review any hosts that have gone non-operational.
- If a Check Host comes up good, return the host to operational status and put the host on the watch list.
- If a machine is regularly going offline, consider turning on debug logging to get more information about why. Once you have more details, open a case with Orasi support to work through the problem.
- Run the System Health Check in Lab Management
- Review all the load tests in Test Runs in Lab Management. Any that have a failed to run, try to determine why.
- Incremental backup of all HPE ALM databases and the HPE ALM repositories. (Note: Only perform backups when no users in the system and no tests running.)
Weekly:
- Reboot HPE ALM server(s).
- Reboot HPE PC server(s).
- Full back up of all HPE ALM databases and HPE ALM repositories. (Note: Only perform backups when no users are in the system and no tests are running.)
- Check the amount of free disk space on the HPE ALM databases and HPE ALM repositories. Ensure you maintain at least 50 GB of free disk space each.
Monthly:
- Install the latest operating system patches.
- Review HPE ALM access reports and logs.
- Review any event logs (hardware, application, and system) and identify and resolve any new hardware-\software-based issues that have come up since the last check.
- Reboot all hosts.
- Reboot all LGs.
- Reboot all data processors.
Quarterly:
- Disk clean up on all servers and components (temp folders). Do not delete any of the HPE software files. If you are not sure, DO NOT DELETE.
- Verify all test runs have been collated.
- Intelligently delete old test run data (PC hosts and SA LGs).
- Clear out all old test run data (PC hosts that have been controllers).
- Clear out all old logs.
- Clear Windows event log.
- Defragment hard disk(s).
- In HPE ALM Site Admin, check that all projects are active and in use. If a project is not going to be used for more than a few months, deactivate the project until access is
needed. - Install latest HPE PC patches.
- HPE ALM patch must be supported by the latest HPE PC server patch.
- HPE PC host patch must be supported by latest HPE PC server patch.
- Stand-alone apps (VuGen, Analysis, and Load Generator) must be supported by latest PCS patch.
- Review each project directory structure for compliance to standards.
- Are users putting scripts in the correct folders?
- Are new releases being maintained properly in the project?
- Are there training opportunities to help end users get better use of the tools?
Semi-Annually:
- Verify that only authorized and active users are in HPE ALM and Lab Management.
- Clear out any old users from each project.
- Prepare an HPE ALM usage report to verify that the current HPE ALM hardware is adequate for concurrent use.
- Prepare an HPE PC load test report to verify that the current HPE PC server hardware is adequate for concurrent use.
- Review any newly released versions (not patches) of HPE PC that have been released.
- If there has been a new release:
- Are there new features or enhancement requests that would be beneficial to your company?
- Are there resolved defects in the new version that are currently causing your users to use a workaround?
- Are there new protocols available only in the new version?
- Has it been 12-18 months since the current version was released (not to be confused with when your implementation was installed)?
If the answer to any of the above questions is yes, you should strongly consider planning an effort to upgrade to the new version.
A routine maintenance schedule, when followed, for your HPE ALM/PC environment can be crucial to project timelines. The suggested schedule of tasks in this post can be a starting place for you as an HPE administrator to build a plan for maintenance. Use your plan to keep an environment where testing can always happen and downtime is rarely experienced.
Special thanks to HPE for the substance contained in this document.