A systems administrator manages a critical legacy application server that exhibits progressive performance degradation over several days. Monitoring data confirms that RAM utilization consistently increases until the system becomes unresponsive. A manual server reboot resolves the problem, but the symptoms reappear after approximately one week. The application vendor has identified this as a known memory leak and confirmed a patch will not be available for several months. Which of the following is the BEST course of action for the administrator to take to maintain service availability in the interim?
Increase the size of the server's page file to accommodate the increased memory usage.
Create a script to automatically restart the application service whenever RAM utilization exceeds 90%.
Configure a weekly scheduled reboot during a low-traffic maintenance window.
Immediately upgrade the server's physical RAM to double its current capacity.
The correct answer is to configure a weekly scheduled reboot. The scenario describes a classic memory leak, where an application fails to release memory, eventually consuming all available RAM and causing the system to fail. Since a patch is not immediately available and the issue is predictable (occurring weekly), a scheduled reboot is the most appropriate proactive measure. Scheduling the reboot during a low-traffic maintenance window minimizes disruption while ensuring the server's resources are regularly cleared, preventing the inevitable crash and maintaining service availability. Increasing the page file would lead to severe performance issues from disk thrashing, not solve the problem. Adding more RAM is an expensive, temporary measure that only delays the failure without addressing the root software cause. Restarting only the application process might not be sufficient if the leak affects other system components or if the process doesn't fully release all memory upon restart; a full system reboot is a more reliable method to reset the system's state completely.