gce/High Vm Memory Utilization

Diagnoses high memory utilization issues in a Compute Engine VM.

Product: Compute Engine
Step Type: AUTOMATED STEP

Description

This step evaluates memory usage through available monitoring data or, as a fallback, scans serial logs for Out of Memory (OOM) indicators. It distinguishes between VMs which has exported metrics and those without, and employs a different strategy for ‘e2’ machine types to accurately assess memory utilization.

Failure Reason

Memory utilization on this VM has reached levels that may compromise its VM application performance.

Failure Remediation

Elevated memory usage can result in slow or unresponsive or termimated applications. Consider enhancing the VM’s memory capacity by changing to a machine type with more memory. Guidance on stopping and changing the machine type can be found here:

Changing machine type: https://cloud.google.com/compute/docs/instances/changing-machine-type-of-stopped-instance#gcloud For deeper analysis of memory issues:

Additionally, use the GCE observability metrics for an in-depth analysis to pinpoint high-usage processes: https://cloud.google.com/compute/docs/instances/observe-monitor-vms#memory_utilization

Or connect via the Serial Console if SSH is not available to mitigate the issue. https://cloud.google.com/compute/docs/troubleshooting/troubleshooting-using-serial-console

Success Reason

Memory utilization on this VM is within optimal range.