Transition from Servers to New Relic Infrastructure

This transition guide helps you understand the return on investment you gain by switching to New Relic Infrastructure, and explains how to move your current configuration from New Relic Servers to Infrastructure. This guide also provides tips about how your Dev/Ops teams can make the most out of Infrastructure's enhanced features.

Comparison

New Relic Servers was originally created as a built-in accompaniment to New Relic APM, where the focus is more on app performance than a business's supporting ecosystem. Infrastructure is specifically designed for complex environments that need flexible, dynamic server monitoring, from a physical datacenter to thousands of Amazon Elastic Compute Cloud (Amazon EC2) instances.

Pricing depends on your New Relic Infrastructure subscription. Take advantage of the free trial period!

Standard features New Relic Servers New Relic Infrastructure
Operating systems support Legacy Linux and Windows OS only

Modern OS, including:

  • Amazon
  • Linux
  • Windows
Dashboard views Individual servers only All your hosts unified in a single view
Data retention in New Relic Insights

3 or 13 months

(depends on subscription)

Configuration change tracking [check icon]
Real-time inventory [check icon]
Dynamic targeting with filters and alerts [check icon]
AWS native support [check icon]
SaaS, including AWS and other integrations

[check icon]

Process alerting [check icon]
Alerting for host not reporting [check icon]
On-host, custom integrations [check icon]

Switch to Infrastructure

With real-time data collection and a UI that scales from a handful of hosts to thousands, Infrastructure is designed for modern Operations teams with fast-changing systems. To truly understand the power of the events and inventory data that Infrastructure collects, New Relic recommends installing the Infrastructure agent on at least 25 hosts.

As you make the transition, the Infrastructure agent can run seamlessly alongside any legacy Servers agents you may have installed on the same host, without any interference.

1. Install the Infrastructure agent.

To install New Relic Infrastructure on individual hosts:

  1. Review the compatibility and requirements checklists.
  2. Follow installation procedures as applicable for your operating system:

To simplify rollout with your automation tools, use New Relic's existing modules to automate installation on multiple hosts:

2. Configure the Infrastructure agent.

Installation is quick, requiring minimal additional configuration. You only need to configure your New Relic license key and any other settings as needed; for example, additional configuration settings or custom attributes.

3. Configure your Amazon EC2 integrations.

If you have an Amazon account, configure your Amazon EC2 and other integrations with Infrastructure. The EC2 integration only requires a Read-only role, and it automatically imports your EC2 tags. It also automatically imports metadata about all the AWS services on your account you select to monitor.

Partnership accounts: Follow separate instructions to install Infrastructure for AWS Elastic Beanstalk.

4. Set up your alert configuration.

You can configure new, more flexible alert conditions directly from the hosts you are monitoring in Infrastructure. If you previously used legacy alerts or New Relic Alerts with New Relic Servers, you will need to set up new alert configurations.

5. Optional: Explore your Infrastructure data in New Relic Insights.

Replacing the New Relic REST API queries for New Relic Servers, you can now query and share comprehensive Infrastructure data via New Relic Insights. To make the transition from Servers to Infrastructure, recast any REST API queries in terms of the corresponding NRQL requests, and use the Insights API.

6. Optional: Uninstall New Relic Servers.

You can continue to run New Relic Servers, although the data collected by New Relic Infrastructure is much more comprehensive. Historical data will continue to appear in the Servers dashboards until you remove your servers or until you reach the data retention limit for your subscription level.

Whenever you are ready:

  1. Follow standard procedures to uninstall New Relic Servers for Linux or Windows.
  2. Follow standard procedures to delete servers from the New Relic Servers UI, or use the REST API to delete servers.
Operating system Uninstall New Relic Servers
Linux

Packages:

Windows Windows Server

Use the Infrastructure UI

If you are used to New Relic Servers, here are some tips to get the most out of New Relic Infrastructure UI.

  • Use Infrastructure's custom attributes to assign classes of servers. This will help identify specific hosts or instances important to your business and reduce long lists of servers in the UI.
  • Use unique hostnames. You can filter on complete or partial names.

Here are some additional tips for exploring your detailed data in Infrastructure's UI.

Servers index = Infrastructure Inventory and Events

New Relic Servers provided a long index of individual or tagged groups of servers.

Use the Infrastructure Inventory page to view detailed configuration information per host for your entire inventory at a glance, including system modules, configuration files, metadata, packages, services, user sessions, etc.

In addition, take advantage of the Infrastructure Events page as a live feed for all your host metrics and log analytics events, to understand correlations between them and your environment. Search and filter through your events to decrease the mean time to repair and the mean time to detect issues within your infrastructure.

Servers Overview = Infrastructure Compute

The New Relic Servers Overview page provided summary and detail information about a specific server.

Use the Infrastructure Compute page to better understand your CPU, load, and memory for one or more hosts or your filter sets.

Servers Processes = Infrastructure Processes and Storage

The New Relic Servers Processes page provided memory or by CPU usage for an individual server.

Use the Infrastructure Processes page to examine CPU percentage, I/O bytes, and memory usage for individual processes or groupings of processes, then use this information to:

  • Pinpoint processes that are causing performance issues.
  • Create alerts directly from this page for conditions such as CPU percentage and memory usage.

In addition, use the Infrastructure Storage page to monitor your resources' capacity and efficiency. This can help you:

  • Examine unexpected patterns; for example, a cluster of mirrored machines that do not process I/O tasks uniformly.
  • Monitor usage levels before disk space completely runs out.
  • Set alert conditions to notify you about problems with one or more hosts; for example, long processing times for read or write operations, disk availability or utilization based on percentage or bytes, etc.
  • Make informed decisions about where to redistribute hosts with higher or lower than normal processing requests.
  • Use data to help plan for additions or changes to your infrastructure budget before an emergency arises.
Servers Network = Infrastructure Network

The New Relic Servers Network page provided information about bandwidth, packets per second, and errors.

Use the Infrastructure Network page to measure and compare capacity. For example:

  • Examine saturation levels across your network or filter sets.
  • Compare load balances between specific resources.
  • Identify unexpected differences in capacity patterns between similar hosts.
  • Evaluate the top five network errors that New Relic Infrastructure automatically presents for the selected time period.

Troubleshoot your transition

The transition from Servers to Infrastructure is designed to have minimal disruption to your work environment. If you have problems, refer to New Relic Infrastructure's troubleshooting procedures.

Here are some additional tips as you make the transition from Servers to Infrastructure.

Concerns about lost data

New Relic Servers and New Relic Infrastructure are completely independent products. You do not need to be concerned about losing data from New Relic Servers, assuming you do not uninstall or remove non-reporting servers.

However, if you used any REST API integrations with New Relic Servers, that data will not be collected via Infrastructure. Instead, you will need to record the data you want to collect via NRQL and the New Relic Insights API.

Other concerns

If you have other concerns about making the transition from Servers to New Relic Infrastructure:

For more help

Discuss New Relic Servers in the New Relic Online Technical Community! Troubleshoot and ask questions, or discuss Servers for Linux or Servers for Windows in detail.

If you need additional help, get support at support.newrelic.com.