Infrastructure Processes page: Inspect process performance

Use the Processes page in New Relic Infrastructure to get information about processes running on your infrastructure, and to set alerts on process metrics. To view the Processes page: Go to infrastructure.newrelic.com > Processes.

You cannot start, stop, or restart processes from the UI. Instead, follow standard procedures to start, stop, check, or restart the Infrastructure agent manually.

Infrastructure Processes UI page

infrastructure.newrelic.com > Processes: Example of Infrastructure Processes page that shows you which processes on which hosts are consuming the most CPU by percentage.

Inspect processes and set alerts

The Processes page shows data such as CPU percentage, I/O bytes, and memory usage for individual processes or groupings of processes. The page also includes an events heatmap, which provides a snapshot of the events occurring within the same time range as the displayed metrics. Use this information to:

  • Pinpoint processes that are causing performance issues.
  • Create alerts for conditions such as CPU percentage and memory usage. On the Processes page, CPU percent is scoped to individual processes, rather than hosts. Because of this, the CPU percent metric does not take into account the resources of the entire system. Instead, it shows how much of a single CPU core each process is taking.

Example

Here's an example of how to pinpoint an issue and set an alert:

  1. You notice on the Infrastructure Hosts page that a cluster has a CPU percentage much higher than normal.
  2. You go to the Processes page and filter down to that cluster's processes.
  3. You notice that several instances of the same process have excessive CPU percentage.
  4. After fixing the issue, you decide to create an alert condition that triggers when the CPU percentage for that process type exceeds a certain threshold.

Processes page features

To filter and group processes and to examine process details, use these functions:

If you want to... Do this...
Only show hosts matching certain criteria Use filter sets.
Aggregate results or group by host attributes

Use Group by to change how the processes are grouped in the chart and the list. The dropdown contains host-related attributes such as hostID and entityName. If you are monitoring AWS EC2 instances, EC2-related attributes such as awsRegion will be available.

Select a time range

Use either of these options:

  • Select any of the time picker options at the top of the page.
  • Click and drag on the chart to select a custom time range.

After you select a time range, it carries over from one Infrastructure page to another.

View Docker metrics To view process usage by Docker containers, see Docker instrumentation.
Change charts

Select the dropdown beside the chart's name to switch what metric the chart displays. Choices include:

  • Memory: Resident or virtual size
  • CPU percentage
  • Network I/O: Sent or received
  • Disk I/O: Read or write
Search and filter the list

Use either of these options:

  • Type in the Search processes search bar.
  • Select the list item or user name you want to filter for.

The selected filters appear beside the filter icon above the chart, where you can select and remove them as needed.

Set an alert condition for a metric Mouse over a chart. Select the [ellipses] icon and then Create alert (not available on APM charts).
Share data and view in New Relic Insights

Mouse over a chart. Select the [ellipses] icon and then View query (not available on APM charts).

View host's alert threshold violation

Select the host's Critical times-circle-o icon or Warning exclamation-triangle icon.

Chart data attributes

For a technical explanation of the attributes used to populate the Processes page chart, see ProcessSample attributes.

For more help

Recommendations for learning more: