Troubleshooting High Memory Utilization on customer Node

Overview

The customer is experiencing high memory utilization on their node, specifically on port 8963, which is causing frequent alerts. Despite having a cache memory purging script already scheduled on the node, the memory consumption remains above the threshold. The issue has been escalated to the Professional Services Team for further investigation.

Solution

To resolve the high memory utilization issue on the node, particularly on port 8963, follow these steps:

1. Review the output of the `top` command to identify any processes that are consuming an unusually high amount of memory. Investigate these processes to determine if they are necessary or if they can be optimized.
2. Check the output of the `df -h` command to ensure there is enough disk space available. Low disk space can sometimes lead to high memory usage if the system is attempting to use swap space excessively.
3. Examine the `free -m` command output to understand the memory allocation, including used and free memory, as well as swap usage. If swap usage is high, it may indicate that the physical memory is insufficient for the workload, leading to swapping and potential performance degradation.
4. 
5. Implement a cache clear script to manage high cache memory and alleviate some of the memory pressure. Even if a cache memory purging script is already scheduled on the node, ensure it is functioning correctly and efficiently. You can verify the script from crontab.
6. If the issue persists, consider disabling any unnecessary services that could be contributing to the high memory utilization.
7. Provide the SAR report, (min-max) Memory -service - Type of service table, and current/capacity traffic to the support team for further investigation.

Summary

High memory utilization on a node, particularly on port 8963, can be caused by a variety of factors. By investigating the processes, checking disk space, understanding memory allocation, examining the service running on the problematic port, implementing a cache clear script, disabling unnecessary services, and providing necessary reports to the support team, the issue can be resolved. If the problem persists, it should be escalated to the Professional Services Team.