Secure your connection to Elasticsearch

edit

Secure your connection to Elasticsearch

edit

The Logstash Elasticsearch output, input, and filter plugins, as well as monitoring and central management, support authentication and encryption over HTTPS.

Elasticsearch clusters are secured by default (starting in 8.0). You need to configure authentication credentials for Logstash in order to establish communication. Logstash throws an exception and the processing pipeline is halted if authentication fails.

In addition to configuring authentication credentials for Logstash, you need to grant authorized users permission to access the Logstash indices.

Security is enabled by default on the Elasticsearch cluster (starting in 8.0). You must enable TLS/SSL in the Elasticsearch output section of the Logstash configuration in order to allow Logstash to communicate with the Elasticsearch cluster.

Elasticsearch security on by default

edit

Elasticsearch generates its own default self-signed Secure Sockets Layer (SSL) certificates at startup.

Logstash must establish a Secure Sockets Layer (SSL) connection before it can transfer data to a secured Elasticsearch cluster. Logstash must have a copy of the certificate authority (CA) that signed the Elasticsearch cluster’s certificates. When a new Elasticsearch cluster is started up without dedicated certificates, it generates its own default self-signed Certificate Authority at startup. See Starting the Elastic Stack with security enabled for more info.

Elasticsearch Service uses certificates signed by standard publicly trusted certificate authorities, and therefore setting a cacert is not necessary.

Security to Elastic Cloud serverless

Elasticsearch on serverless simplifies safe, secure communication between Logstash and Elasticsearch.

Configure the Logstash Elasticsearch output plugin to use cloud_id and an api_key to establish safe, secure communication between Logstash and Elasticsearch on serverless. No additional SSL configuration steps are needed.

Configuration example:

  • output {elasticsearch { cloud_id => "<cloud id>" api_key => "<api key>" } }

For more details, check out Grant access using API keys.

Security to hosted Elasticsearch Service

Our hosted Elasticsearch Service on Elastic Cloud simplifies safe, secure communication between Logstash and Elasticsearch. When you configure the Logstash Elasticsearch output plugin to use cloud_id with either the cloud_auth option or the api_key option, no additional SSL configuration steps are needed. Our hosted Elasticsearch Service is available on AWS, GCP, and Azure, and you can try it for free.

Configuration example:

  • output {elasticsearch { cloud_id => "<cloud id>" cloud_auth => "<cloud auth>" } }
  • output {elasticsearch { cloud_id => "<cloud id>" api_key => "<api key>" } }

For more details, check out Grant access using API keys or Sending data to Elastic Cloud (hosted Elasticsearch Service).

Secure communication with an on-premise Elasticsearch cluster

edit

If you are running Elasticsearch on your own hardware and using the Elasticsearch cluster’s default self-signed certificates, you need to complete a few more steps to establish secure communication between Logstash and Elasticsearch.

You need to:

  • Copy the self-signed CA certificate from Elasticsearch and save it to Logstash.
  • Configure the elasticsearch-output plugin to use the certificate.

These steps are not necessary if your cluster is using public trusted certificates.

Copy and save the certificate

edit

By default an on-premise Elasticsearch cluster generates a self-signed CA and creates its own SSL certificates when it starts. Therefore Logstash needs its own copy of the self-signed CA from the Elasticsearch cluster in order for Logstash to validate the certificate presented by Elasticsearch.

Copy the self-signed CA certificate from the Elasticsearch config/certs directory.

Save it to a location that Logstash can access, such as config/certs on the Logstash instance.

Configure the elasticsearch output

edit

Use the elasticsearch output's cacert option to point to the certificate’s location.

Example

output {
  elasticsearch {
    hosts => ["https://...] 
    cacert => '/etc/logstash/config/certs/ca.crt' 
  }
}

Note that the hosts url must begin with https

Path to the Logstash copy of the Elasticsearch certificate

For more information about establishing secure communication with Elasticsearch, see security is on by default.

Configuring Logstash to use basic authentication

edit

Logstash needs to be able to manage index templates, create indices, and write and delete documents in the indices it creates.

To set up authentication credentials for Logstash:

  1. Use the the Management > Roles UI in Kibana or the role API to create a logstash_writer role. For cluster privileges, add manage_index_templates and monitor. For indices privileges, add write, create, and create_index.

    Add manage_ilm for cluster and manage and manage_ilm for indices if you plan to use index lifecycle management.

    POST _security/role/logstash_writer
    {
      "cluster": ["manage_index_templates", "monitor", "manage_ilm"], 
      "indices": [
        {
          "names": [ "logstash-*" ], 
          "privileges": ["write","create","create_index","manage","manage_ilm"]  
        }
      ]
    }

    The cluster needs the manage_ilm privilege if index lifecycle management is enabled.

    If you use a custom Logstash index pattern, specify your custom pattern instead of the default logstash-* pattern.

    If index lifecycle management is enabled, the role requires the manage and manage_ilm privileges to load index lifecycle policies, create rollover aliases, and create and manage rollover indices.

  2. Create a logstash_internal user and assign it the logstash_writer role. You can create users from the Management > Users UI in Kibana or through the user API:

    POST _security/user/logstash_internal
    {
      "password" : "x-pack-test-password",
      "roles" : [ "logstash_writer"],
      "full_name" : "Internal Logstash User"
    }
  3. Configure Logstash to authenticate as the logstash_internal user you just created. You configure credentials separately for each of the Elasticsearch plugins in your Logstash .conf file. For example:

    input {
      elasticsearch {
        ...
        user => logstash_internal
        password => x-pack-test-password
      }
    }
    filter {
      elasticsearch {
        ...
        user => logstash_internal
        password => x-pack-test-password
      }
    }
    output {
      elasticsearch {
        ...
        user => logstash_internal
        password => x-pack-test-password
      }
    }

Granting access to the Logstash indices

edit

To access the indices Logstash creates, users need the read and view_index_metadata privileges:

  1. Create a logstash_reader role that has the read and view_index_metadata privileges for the Logstash indices. You can create roles from the Management > Roles UI in Kibana or through the role API:

    POST _security/role/logstash_reader
    {
      "cluster": ["manage_logstash_pipelines"]
    }
  2. Assign your Logstash users the logstash_reader role. If the Logstash user will be using centralized pipeline management, also assign the logstash_admin role. You can create and manage users from the Management > Users UI in Kibana or through the user API:

    POST _security/user/logstash_user
    {
      "password" : "x-pack-test-password",
      "roles" : [ "logstash_reader", "logstash_admin"], 
      "full_name" : "Kibana User for Logstash"
    }

    logstash_admin is a built-in role that provides access to system indices for managing configurations.

Configuring Logstash to use TLS/SSL encryption

edit

If TLS encryption is enabled on an on premise Elasticsearch cluster, you need to configure the ssl and cacert options in your Logstash .conf file:

output {
  elasticsearch {
    ...
    ssl => true
    cacert => '/path/to/cert.pem' 
  }
}

The path to the local .pem file that contains the Certificate Authority’s certificate.

Hosted Elasticsearch Service simplifies security. This configuration step is not necessary for hosted Elasticsearch Service on Elastic Cloud. Our hosted Elasticsearch Service is available on AWS, GCP, and Azure, and you can try it for free.

Configuring the Elasticsearch output to use PKI authentication

edit

The elasticsearch output supports PKI authentication. To use an X.509 client-certificate for authentication, you configure the keystore and keystore_password options in your Logstash .conf file:

output {
  elasticsearch {
    ...
    keystore => /path/to/keystore.jks
    keystore_password => realpassword
    truststore =>  /path/to/truststore.jks 
    truststore_password =>  realpassword
  }
}

If you use a separate truststore, the truststore path and password are also required.

Configuring credentials for Logstash monitoring

edit

If you want to monitor your Logstash instance with Elastic Stack monitoring features, and store the monitoring data in a secured Elasticsearch cluster, you must configure Logstash with a username and password for a user with the appropriate permissions.

The security features come preconfigured with a logstash_system built-in user for this purpose. This user has the minimum permissions necessary for the monitoring function, and should not be used for any other purpose - it is specifically not intended for use within a Logstash pipeline.

By default, the logstash_system user does not have a password. The user will not be enabled until you set a password. See Setting built-in user passwords.

Then configure the user and password in the logstash.yml configuration file:

xpack.monitoring.elasticsearch.username: logstash_system
xpack.monitoring.elasticsearch.password: t0p.s3cr3t

If you initially installed an older version of X-Pack and then upgraded, the logstash_system user may have defaulted to disabled for security reasons. You can enable the user through the user API:

PUT _security/user/logstash_system/_enable

Configuring credentials for Centralized Pipeline Management

edit

If you plan to use Logstash centralized pipeline management, you need to configure the username and password that Logstash uses for managing configurations.

You configure the user and password in the logstash.yml configuration file:

xpack.management.elasticsearch.username: logstash_admin_user 
xpack.management.elasticsearch.password: t0p.s3cr3t

The user you specify here must have the built-in logstash_admin role as well as the logstash_writer role that you created earlier.

Grant access using API keys

edit

Instead of using usernames and passwords, you can use API keys to grant access to Elasticsearch resources. You can set API keys to expire at a certain time, and you can explicitly invalidate them. Any user with the manage_api_key or manage_own_api_key cluster privilege can create API keys.

Tips for creating API keys:

  • API keys are tied to the cluster they are created in. If you are sending output to different clusters, be sure to create the correct kind of API key.
  • Logstash can send both collected data and monitoring information to Elasticsearch. If you are sending both to the same cluster, you can use the same API key. For different clusters, you need an API key per cluster.
  • A single cluster can share a key for ingestion and monitoring purposes.
  • A production cluster and a monitoring cluster require separate keys.

For security reasons, we recommend using a unique API key per Logstash instance. You can create as many API keys per user as necessary.

Create an API key

edit

You can create API keys using either the Create API key API or the Kibana UI. This section walks you through creating an API key using the Create API key API. The privileges needed are the same for either approach.

Here is an example that shows how to create an API key for publishing to Elasticsearch using the Elasticsearch output plugin.

POST /_security/api_key
{
  "name": "logstash_host001", 
  "role_descriptors": {
    "logstash_writer": { 
      "cluster": ["monitor", "manage_ilm", "read_ilm"],
      "index": [
        {
          "names": ["logstash-*"],
          "privileges": ["view_index_metadata", "create_doc"]
        }
      ]
    }
  }
}

Name of the API key

Granted privileges

The return value should look similar to this:

{
  "id":"TiNAGG4BaaMdaH1tRfuU", 
  "name":"logstash_host001",
  "api_key":"KnR6yE41RrSowb0kQ0HWoA" 
}

Unique id for this API key

Generated API key

Create an API key for publishing
edit

You’re in luck! The example we used in the Create an API key section creates an API key for publishing to Elasticsearch using the Elasticsearch output plugin.

Here’s an example using the API key in your Elasticsearch output plugin configuration.

output {
  elasticsearch {
    api_key => "TiNAGG4BaaMdaH1tRfuU:KnR6yE41RrSowb0kQ0HWoA" 
  }
}

Format is id:api_key (as returned by Create API key)

Create an API key for reading
edit

Creating an API key to use for reading data from Elasticsearch is similar to creating an API key for publishing described earlier. You can use the example in the Create an API key section, granting the appropriate privileges.

Here’s an example using the API key in your Elasticsearch inputs plugin configuration.

input {
  elasticsearch {
    "api_key" => "TiNAGG4BaaMdaH1tRfuU:KnR6yE41RrSowb0kQ0HWoA" 
  }
}

Format is id:api_key (as returned by Create API key)s

Create an API key for filtering
edit

Creating an API key to use for processing data from Elasticsearch is similar to creating an API key for publishing described earlier. You can use the example in the Create an API key section, granting the appropriate privileges.

Here’s an example using the API key in your Elasticsearch filter plugin configuration.

filter {
  elasticsearch {
    api_key => "TiNAGG4BaaMdaH1tRfuU:KnR6yE41RrSowb0kQ0HWoA" 
  }
}

Format is id:api_key (as returned by Create API key)

Create an API key for monitoring
edit

To create an API key to use for sending monitoring data to Elasticsearch, use the Create API key API. For example:

POST /_security/api_key
{
  "name": "logstash_host001", 
  "role_descriptors": {
    "logstash_monitoring": { 
      "cluster": ["monitor"],
      "index": [
        {
          "names": [".monitoring-ls-*"],
          "privileges": ["create_index", "create"]
        }
      ]
    }
  }
}

Name of the API key

Granted privileges

The return value should look similar to this:

{
  "id":"TiNAGG4BaaMdaH1tRfuU", 
  "name":"logstash_host001",
  "api_key":"KnR6yE41RrSowb0kQ0HWoA" 
}

Unique id for this API key

Generated API key

Now you can use this API key in your logstash.yml configuration file:

xpack.monitoring.elasticsearch.api_key: TiNAGG4BaaMdaH1tRfuU:KnR6yE41RrSowb0kQ0HWoA 

Format is id:api_key (as returned by Create API key)

Create an API key for central management
edit

To create an API key to use for central management, use the Create API key API. For example:

POST /_security/api_key
{
  "name": "logstash_host001", 
  "role_descriptors": {
    "logstash_monitoring": { 
      "cluster": ["monitor"],
      "index": ["read"]
    }
  }
}

Name of the API key

Granted privileges

The return value should look similar to this:

{
  "id":"TiNAGG4BaaMdaH1tRfuU", 
  "name":"logstash_host001",
  "api_key":"KnR6yE41RrSowb0kQ0HWoA" 
}

Unique id for this API key

Generated API key

Now you can use this API key in your logstash.yml configuration file:

xpack.management.elasticsearch.api_key: TiNAGG4BaaMdaH1tRfuU:KnR6yE41RrSowb0kQ0HWoA 

Format is id:api_key (as returned by Create API key)

Learn more about API keys

edit

See the Elasticsearch API key documentation for more information:

See API Keys for info on managing API keys through Kibana.