servers are showing higher-than-expected CPU utilization

High CPU utilization issues on Linux servers are usually easier to diagnose by identifying which processes are consuming the most CPU resources.

These commands may help:

ps aux --sort=-%cpu | head

top

htop

In many cases, background services, runaway scripts, container workloads, or stuck processes can silently consume CPU for long periods.

I would also recommend checking:

iostat

vmstat 1

sar

to determine whether the issue is actually CPU saturation, memory pressure, or I/O bottlenecks.

This guide may also help understand Linux process monitoring and process analysis:

https://www.linuxteck.com/ps-command-in-linux-with-examples/

Hello @Abhishek,

Thanks for using Q and A forum.

 First: Identify Where the CPU Is High

Start in the Azure Portal:

Go to Microsoft Azure → Virtual Machines → Affected VM

Open Metrics

CPU Percentage

Time range (last 1h / 24h / 7d)

Aggregation = Average + Max

Key question: Is CPU constantly high, or only spiking at specific times?

This distinction determines the next steps.

 

Check If This Is Expected Workload Load

High CPU is not always a problem.

Validate:

Has traffic increased?

Are there batch jobs, backups, cron jobs, or reports running?

Any recent:

Application deployment

Plugin/module update

Configuration change

If CPU increase aligns with business activity, scaling may be the correct solution rather than troubleshooting.

 

Inspect the VM from Inside (Critical Step)

SSH (Linux) or RDP (Windows) into the affected VM.

Linux:

top

htop

uptime

Look for:

Processes consuming high %CPU

High load average vs CPU core count

Long-running or stuck processes

Windows:

Task Manager → Processes

Resource Monitor → CPU tab

This usually reveals the culprit within minutes.

 

Common Root Causes We See in Azure

Application-Level

Inefficient code or loops

Memory pressure causing CPU thrashing

PHP/Java/.NET worker saturation

Excessive logging

OS-Level

Too many background services

Antivirus or monitoring agents

Unpatched kernel issues

Azure-Specific

VM size too small for workload

Burstable VM (B-series) CPU credits exhausted

Disk I/O waits causing CPU spikes

Network packet processing overhead

If the VM is B-series, check CPU Credit Balance immediately — this is a very common cause.

 

Review Azure Monitoring & Logs

Enable / review:

Azure Monitor

Log Analytics Workspace

VM Insights

Look for:

CPU vs Memory correlation

Disk queue length

Network throughput

Repeated restarts or failures

---

If the Answer is helpful, please click Accept Answer and Up-Vote 👍, so that this can be beneficial to other community members.

 

Hello @Abhishek ,

The following factors are common in a high-CPU situation:

• A recent code change or deployment that is mostly applicable to apps such as Internet Information Services (IIS), Microsoft SharePoint, Microsoft SQL Server, or third-party applications.
• A recent update that might be related to an OS-level update, or to application-level cumulative updates and fixes.
• A query change or outdated indexes. SQL Server and Oracle data tier applications also have a query plan optimization as another factor. Data changes or a lack of appropriate indexes can cause several queries to become more compute intensive.
• Azure VM-specific. There are certain processes such as RDAgent, and extension-specific processes such as Monitoring Agent, MMA agent, or Security client, that can cause high-CPU consumption. These processes must be viewed from either a configuration or known issues perspective.

Let's scope the issue first-

Is there a pattern to the issue? For example, does the high-CPU issue occur at a certain time every day, week, or month? If so, can you correlate this issue to a job, report, or user login?

Did the high-CPU issue start after a recent code change? Did you apply an update in Windows or in an application?

Did the high-CPU issue start after a change in workload, such as an increase in the number of users, a higher data influx, or a larger number of reports?

For Azure, did the high-CPU issue start in any of the following conditions?

• After a recent redeployment or restart
• When an SKU or VM type changed
• When a new extension was added
  • After load balancer changes were made

Run and analyze reports using PerfInsights through the Azure portal. Please refer to the following link, which provides guidance on identifying and diagnosing unexpectedly high CPU utilization- https://learn.microsoft.com/en-us/troubleshoot/azure/virtual-machines/windows/troubleshoot-high-cpu-issues-azure-windows-vm#perfinsights

If any additional questions come up, please feel free to reach out. I’m always here to support.