IMPORTANT: No additional bug fixes or documentation updates will be released for this version. For the latest information, see the current release documentation.
« Troubleshoot common problems Release notes »
Elastic Docs ›Fleet and Elastic Agent Guide [8.4] ›Troubleshoot

Frequently asked questions

edit

Frequently asked questions

edit

We have collected the most frequently asked questions here. If your question isn’t answered here, contact us in the discuss forum. Your feedback is very valuable to us.

Also read Troubleshoot common problems.

Why doesn’t my enrolled agent show up in the Fleet app?

edit

If Elastic Agent was successfully enrolled, but doesn’t show up in the Agents list, it might not be started. Make sure the elastic-agent process is running on the host. If it’s not running, use the run command to start it. The most common way to deploy an Elastic Agent is by using the install command. This command starts the Elastic Agent for you.

Where does Elastic Agent store logs after startup?

edit

The location of Elastic Agent logs varies by platform. In general, Elastic Agent stores logs under the data directory where Elastic Agent was started. For example, on macOS, you’ll find logs for the running Elastic Agent under path similar to:

/Library/Elastic/Agent/data/elastic-agent-08e204/logs/elastic-agent-20220105.ndjson

You’ll find logs for the Beats shippers, such as Metricbeat, under paths like:

/Library/Elastic/Agent/data/elastic-agent-08e204/logs/default/metricbeat-20220105.ndjson

If the log path does not exist, Elastic Agent was unable to start Metricbeat, which is a higher level problem to triage. Usually you can see these logs in the Fleet UI, unless there are problems severe enough that the Elastic Agent or its related processes cannot send data to Elasticsearch.

See Installation layout to find out the exact paths for each platform.

What policy is the Elastic Agent running?

edit

To find the policy file, inspect the elastic-agent.yml file in the directory where Elastic Agent is running. Not sure where the agent is running? See Installation layout.

If the agent is running in Fleet mode, this file contains the following citation:

fleet:
  enabled: true

The state.yml file (located under data/elastic-agent-*) contains the entire, unencrypted policy.

  • To see the Elasticsearch location, look at the hosts setting under outputs. For example:

    outputs:
  default:
    api_key: Aq-mPpcBDA7TmnriKCSD:Np6NAleNQ1mMpgN_JPYazw
    hosts:
    - https://3m63533c175a4036b3d8bbe7bd462fa3.us-east-1.aws.found.io:443
    type: elasticsearch

    This file also shows the version of all packages used by the current policy.

  • To see the Elastic Agent version, run:

    elastic-agent version

Why can’t I see the data Elastic Agent is sending?

edit

If Elastic Agent is set up and running, but you don’t see data in Kibana:

  1. Go to Management > Dev Tools in Kibana, and in the Console, search your index for data. For example:

    GET metrics-*/_search

    Or if you prefer, go to the Discover app.

  2. Look at the data that Elastic Agent has sent and see if the name.host field contains your host machine name.

If you don’t see data for your host, it’s possible that the data is blocked in the network, or that a firewall or security problem is preventing the Elastic Agent from sending the data.

Although it’s redundant to install stand-alone Metricbeat, you might want to try installing it to see if it’s able to send data successfully to Elasticsearch. For more information, see Metricbeat quick start.

If Metricbeat is able to send data to Elasticsearch, there is possibly a bug or problem with Elastic Agent, and you should report it.

How do I restore an Elastic Agent that I deleted from Fleet?

edit

It’s okay, we’ve got your back! The data is still in Elasticsearch. To add Elastic Agent to Fleet again, Stop Elastic Agent, re-enroll it on the host, then run Elastic Agent.

How do I restart Elastic Agent after rebooting my host?

edit

Elastic Agent should restart automatically when you reboot your host. If it doesn’t, you can start it manually by running:

elastic-agent run

If the process is already running, you can restart it by running:

elastic-agent restart

Does Elastic Agent or Kibana download integration packages?

edit

Elastic Agent does not download integration packages. When you add an integration in Fleet, Kibana connects to the Elastic Package Registry at epr.elastic.co, downloads the integration package, and stores its assets in Elasticsearch. This means that you no longer have to run a manual setup command to load integrations as you did previously with Beats modules.

By default, Kibana requires an internet connection to download integration packages from the Elastic Package Registry. If network restrictions prevent Kibana from reaching the public Elastic Package Registry, you can use a proxy server or host your own Elastic Package Registry. To learn more, refer to Air-gapped environments.

Does Elastic Agent download anything from the Internet?

edit
  • In version 7.10 and later, a fully capable artifact can be installed with no connection to the Elastic download site. However, if it is in use, the Endpoint and Cloud Security process is instructed to attempt to download newer released versions of the integration-specific artifacts it uses. Some of those are, for example, the malware model, trusted applications artifact, exceptions list artifact, and others. Elastic Endpoint will continue to protect the host even if it’s unable to download updates. However, it won’t receive updates to protections until Elastic Agent is upgraded to a new version. For more information, refer to the Elastic Security documentation.
  • Elastic Agent requires internet access to download artifacts for binary upgrades. In air-gapped environments, you can host your own artifact registry. For more information, refer to Air-gapped environments.

Do I need to set up the Beats managed by Elastic Agent?

edit

You might have noticed that Elastic Agent runs Beats under the hood. But note that the Beats managed by Elastic Agent are set up and run differently from standalone Beats.

For example, standalone Beats use modules and require you to run a setup command on the host to load assets, such as ingest pipelines and dashboards. In contrast, Beats managed by Elastic Agent use integration packages that Kibana downloads from the Elastic Package Registry at epr.elastic.co. This means that Elastic Agent does not need extra privileges to set up assets because Fleet manages the assets.

What is the Endpoint and Cloud Security integration in Fleet?

edit

The Endpoint and Cloud Security integration provides protection on your Elastic Agent controlled host. The integration monitors your host for security-related events, allowing for investigation of security data through the Elastic Security app in Kibana. The Endpoint and Cloud Security integration is managed by Elastic Agent in the same way as other integrations. Try it out! For more information, refer to the Elastic Security documentation.

How are communications secured between Elastic Security and Elastic Agent?

edit

Elastic Security connects to the agent over loopback TLS on port 6788. Elastic Security validates that the agent has root (Linux and macOS) or SYSTEM (Windows) permissions.

How are secrets secured when entered into integration policies or agent policies in Kibana?

edit

Credentials that you provide for an agent or integration policy are stored in Elasticsearch. They can be read by any user who has read permissions to the .fleet-* and .kibana* indices in Elasticsearch. By default these are the superuser, fleet-server service account tokens, and the kibana_system user. These secrets are also included in agent policies and shared with agents via Fleet through TLS. When you use the Elastic Agent installer and enroll agents in Fleet, the policies are stored on the host file system and, by default, can only be read by root.

Which Elasticsearch and Kibana ports need to be accessible?

edit

The policy generated by Fleet already contains the correct Elasticsearch address and port for your setup. If you run everything locally, the address is 127.0.0.1:9200. If you use our hosted Elasticsearch Service on Elastic Cloud, you can copy the Elasticsearch endpoint URL from the overview page of your deployment. If you’re not running in Elastic Cloud, make sure the Kibana and Elasticsearch HTTPS ports are both accessible; by default these are 5601 and 9200 respectively.

If I delete an integration dashboard asset from Kibana, how do I get it back?

edit

To reinstall the assets for a specific integration, you can use the Fleet UI. For more information, see Reinstall integration assets.

Alternatively, you can use the Fleet API using the package name and version. This needs to be run against the Kibana API and not the Elasticsearch API to be successful. To reinstall package assets, execute the following call with the force parameter in the body:

POST api/fleet/epm/packages/[package name]/[package version]
{ "force": true }

So, for example, to reinstall the system v1.0.0 package, POST:

POST api/fleet/epm/packages/system/1.0.0
{ "force": true }

The package version is shown in the Integrations view in Kibana.

« Troubleshoot common problems Release notes »