This is a cache of https://www.elastic.co/guide/en/observability/current/profiling-get-started.html. It is a snapshot of the page at 2024-08-28T00:40:31.513+0000.
Get started with Universal Profiling | Elastic Observability [8.15] | Elastic

Get started with Universal Profiling

edit

Get started with Universal Profiling

edit

On this page, you’ll learn how to configure and use Universal Profiling. This page covers:

  • Prerequisites to setting up Universal Profiling
  • Setting up Universal Profiling in your Elastic Cloud deployment
  • Installing the Universal Profiling Agent
  • Installing the Universal Profiling Agent integration

We would appreciate feedback on your experience with this product and any other profiling pain points you may have. See the send feedback section of the troubleshooting documentation for more information.

Prerequisites

edit

Before setting up Universal Profiling, make sure you meet the following requirements:

  • An Elastic Stack deployment on Elastic Cloud at version 8.7.0 or higher. Universal Profiling is currently only available on Elastic Cloud.
  • The workloads you’re profiling must be running on Linux machines with x86_64 or ARM64 CPUs.
  • The minimum supported kernel version is either 4.19 for x86_64 or 5.5 for ARM64 machines.
  • The Integrations Server must be enabled on your Elastic Cloud deployment.
  • Credentials (username and password) for the superuser Elasticsearch role (typically, the elastic user).
Interpreters
edit

Universal Profiling is a system-wide profiling solution with additional support for PHP, Python, Java (or any JVM language), Go, Rust, C/C++, Node.js/V8, Ruby, .Net, and Perl.

The minimum supported versions of each interpreter are:

  • JVM/JDK: 7
  • Python: 3.6
  • V8: 8.1.0
  • Perl: 5.28
  • PHP: 7.3
  • Ruby: 2.5
  • .Net: 6
Deployment configuration example
edit

The following deployment configuration example was tested to support profiling data from a fleet of up to 500 hosts, each with 8 or 16 CPU cores, for a total of roughly 6000 cores:

Component

Size per zone (memory)

Zones

Elasticsearch

64 GB

2

Kibana

8 GB

1

Integrations Server

8 GB

1

Even if you’re profiling a smaller fleet, we recommend configuring at least two zones for Elasticsearch and 4 GB of memory each for the Integrations Server and Kibana.

Set up Universal Profiling on an Elastic Cloud deployment

edit

To set up Universal Profiling on your Elastic Cloud deployment, you need to configure data ingestion first.

Configure data ingestion
edit

After enabling Universal Profiling on your deployment for the first time, select any subheading under Universal Profiling in the navigation menu to open the following page:

profiling setup popup

Click Set up Universal Profiling to configure data ingestion.

To configure data ingestion, you need elevated privileges, typically the elastic user.

If you’re upgrading from a previous version with Universal Profiling enabled, see the upgrade guide.

Install the Universal Profiling Agent

edit

You have the following options when installing the Universal Profiling Agent:

Install the Universal Profiling Agent using the Elastic Agent
edit

To install the Universal Profiling Agent using the Elastic Agent and the Universal Profiling Agent integration, complete the following steps:

  1. Copy the secret token and Universal Profiling Collector url from the Elastic Agent Integration

    profiling elastic agent
  2. Click Manage Universal Profiling Agent in Fleet to complete the integration.
  3. On the Integrations page, click Add Universal Profiling Agent.
  4. In Universal Profiling Agent → Settings, add the information you copied from the Add profiling data page:

    1. Add the Universal Profiling collector URL to the Universal Profiling collector endpoint field.
    2. Add the secret token to the Authorization field.

      profililing elastic agent creds
  5. Click Save and continue.

Install the Universal Profiling Agent in standalone mode

edit

The Universal Profiling Agent profiles your fleet. You need to install and configure it on every machine that you want to profile. The Universal Profiling Agent needs root / CAP_SYS_ADMIN privileges to run.

After clicking Set up Universal Profiling in the previous step, you’ll see the instructions for installing the Universal Profiling Agent. You can also find these instructions by clicking the Add data button in the top-right corner of the page.

The following is an example of the provided instructions for Kubernetes:

profiling k8s hostagent
Universal Profiling Agent configuration notes
edit

Consider the following when configuring your Universal Profiling Agent:

  • The instructions in Kibana work well for testing environments. For production environments, we recommend setting an immutable version.
  • Before Elastic Stack version 8.9 the Universal Profiling Agent versioning scheme was not aligned with the Elastic Stack version scheme.
  • You can find a list of container image versions in the Elastic container library repository.
  • For Kubernetes deployments, the Helm chart version is already used to configure the same container image, unless overwritten with the version parameter in the Helm values file.
  • For Elastic Stack version 8.8 or higher, use v3 host agents. For version 8.7, use v2. v3 host agents are incompatible with 8.7 Elastic Stack versions.