This documentation contains work-in-progress information for future Elastic Stack and Cloud releases. Use the version selector to view supported release docs. It also contains some Elastic Cloud serverless information. Check out our serverless docs for more details.

« Monitoring settings in Kibana Search sessions settings in Kibana »

› › ›

Reporting settings in Kibana

edit

Reporting settings in Kibana

edit

You can configure xpack.reporting settings in your kibana.yml to:

Enable reporting

edit

xpack.reporting.enabled: When true, enables the reporting features. Set this to false to disable reporting features entirely. The default is true.

Disabling the reporting features is discouraged. If you need to turn off the ability to generate reports, configure the roles and spaces in the Kibana application privileges.

If needed, you can also prevent a Kibana instance from claiming reporting work by setting xpack.reporting.queue.pollEnabled: false.

Encryption key setting

edit

By default, an encryption key is generated for the reporting features each time you start Kibana. If a static encryption key is not persisted in the Kibana configuration, any pending reports fail when you restart Kibana.

If you are load balancing across multiple Kibana instances, each instance needs to have the same reporting encryption key. Otherwise, report generation fails if a report is queued through one instance, and another instance picks up the job from the report queue. The instance that picks up the job is unable to decrypt the reporting job metadata.

xpack.reporting.encryptionKey: The static encryption key for reporting. Use an alphanumeric text string that is at least 32 characters. By default, Kibana generates a random key when it starts, which causes pending reports to fail after restart. Configure xpack.reporting.encryptionKey to preserve the same key across multiple restarts and multiple Kibana instances.

xpack.reporting.encryptionKey: "something_secret"

Security settings

edit

Reporting privileges are configured with Kibana application privileges. You can control the spaces and applications where users are allowed to generate reports.

Background job settings

edit

Reporting generates reports on the Kibana server as background tasks, and jobs are coordinated using documents in Elasticsearch. Depending on how often you generate reports and the overall number of reports, you might need to change the following settings.

xpack.reporting.capture.maxAttempts: If capturing a report fails for any reason, Kibana will re-queue the report job for retry, as many times as this setting. Defaults to 3.
xpack.reporting.queue.indexInterval: [8.15.0] Deprecated in 8.15.0. This setting has no effect. How often Reporting creates a new index to store report jobs and file contents. Valid values are year, month, week, day, and hour. Defaults to week. NOTE: This setting exists for backwards compatibility, but is unused. Use the built-in ILM policy provided for the reporting plugin to customize the rollover of Reporting data.
xpack.reporting.queue.pollEnabled: When true, enables the Kibana instance to poll Elasticsearch for pending jobs and claim them for execution. When false, allows the Kibana instance to only add new jobs to the reporting queue, list jobs, and provide the downloads to completed reports through the UI. This requires a deployment where at least one other Kibana instance in the Elastic cluster has this setting to true. The default is true.

Running multiple instances of Kibana in a cluster for load balancing of reporting requires identical values for xpack.reporting.encryptionKey and, if security is enabled, xpack.security.encryptionKey.

xpack.reporting.queue.pollInterval: Specifies the time that the reporting poller waits between polling the index for any pending Reporting jobs. Can be specified as a number of milliseconds. Defaults to 3s.
xpack.reporting.queue.timeout: How long each worker has to produce a report. If your machine is slow or under heavy load, you might need to increase this timeout. If a Reporting job execution goes over this time limit, the job is marked as a failure and no download will be available. Can be specified as a number of milliseconds. Defaults to 4m.

PNG/PDF settings

edit

We recommend using PNG/PDF reports to export moderate amounts of data only. The feature enables a high-level export capability, but it’s not intended for bulk export. If you need to export several pages of image data, consider using multiple report jobs to export a small number of pages at a time. If the screenshot of exported dashboard contains a large number of pixels, consider splitting the large dashboard into smaller artifacts to use less memory and CPU resources.

For the most reliable configuration of PDF/PNG reporting features, consider installing Kibana using Docker or using Elastic Cloud.

To generate PDF and PNG files, Reporting uses an internal "screenshotting" plugin which manages a headless browser that captures screenshots from Kibana.

The following settings control the capturing process.

xpack.screenshotting.capture.timeouts.openUrl: Specify the time to allow the Reporting browser to wait for the "Loading…" screen to dismiss and find the initial data for the page. If the time is exceeded, a screenshot is captured showing the current page, and the download link shows a warning message. Can be specified as number of milliseconds. Defaults to 1m.
xpack.screenshotting.capture.timeouts.waitForElements: Specify the time to allow the Reporting browser to wait for all visualization panels to load on the page. If the time is exceeded, a screenshot is captured showing the current page, and the download link shows a warning message. Can be specified as number of milliseconds. Defaults to 1m.
xpack.screenshotting.capture.timeouts.renderComplete: Specify the time to allow the Reporting browser to wait for all visualizations to fetch and render the data. If the time is exceeded, a screenshot is captured showing the current page, and the download link shows a warning message. Can be specified as number of milliseconds. Defaults to 2m.

If any timeouts from xpack.screenshotting.capture.timeouts.* settings occur when running a report job, Reporting will log the error and try to continue capturing the page with a screenshot. As a result, a download will be available, but there will likely be errors in the visualizations in the report.

xpack.screenshotting.capture.loadDelay: [8.0.0] Deprecated in 8.0.0. This setting has no effect. Specify the amount of time before taking a screenshot when visualizations are not evented. All visualizations that ship with Kibana are evented, so this setting should not have much effect. If you are seeing empty images instead of visualizations, try increasing this value. NOTE: This setting exists for backwards compatibility, but is unused and therefore does not have an affect on reporting performance.

Chromium headless browser settings

edit

For PDF and PNG reports, Reporting spawns a headless Chromium browser process on the server to load and capture a screenshot of the Kibana app. When installing Kibana on Linux and Windows platforms, the Chromium binary comes bundled with the Kibana download. For Mac platforms, the Chromium binary is downloaded the first time Kibana is started.

xpack.screenshotting.browser.chromium.disableSandbox: It is recommended that you research the feasibility of enabling unprivileged user namespaces. An exception is if you are running Kibana in Docker because the container runs in a user namespace with the built-in seccomp/bpf filters. For more information, refer to Chromium sandbox. Defaults to false for all operating systems except CentOS, Debian, and Red Hat Linux, which use true.
xpack.screenshotting.browser.chromium.proxy.enabled: Enables the proxy for Chromium to use. When set to true, you must also specify the xpack.screenshotting.browser.chromium.proxy.server setting. Defaults to false.
xpack.screenshotting.browser.chromium.proxy.server: The uri for the proxy server. Providing the username and password for the proxy server via the uri is not supported.
xpack.screenshotting.browser.chromium.proxy.bypass: An array of hosts that should not go through the proxy server and should use a direct connection instead. Examples of valid entries are "elastic.co", "*.elastic.co", ".elastic.co", ".elastic.co:5601".

Kibana server settings for headless browser connection

edit

To generate screenshots for PNG and PDF reports, Reporting opens the Kibana web interface using a local connection to the server. In most cases, using a local connection to the Kibana server presents no issue. If you prefer the headless browser to connect to Kibana using a specific hostname, there are a number of settings that allow the headless browser to connect to Kibana through a proxy, rather than directly.

The xpack.reporting.kibanaServer settings are optional. Take caution when editing these settings. Adding these settings can cause the PDF/PNG reporting features to fail. If reports fail, inspect the server logs and pay attention to errors regarding the headless browser being unable to connect to the server. The full Kibana URL that Reporting is attempting to open is logged during report execution.

xpack.reporting.kibanaServer.port: The port for accessing Kibana.
xpack.reporting.kibanaServer.protocol: The protocol for accessing Kibana, typically http or https.
xpack.reporting.kibanaServer.hostname: The hostname for accessing Kibana.

Network policy settings for headless Chromium restrictions

edit

To generate PDF reports, Reporting uses a headless Chromium browser to fully load the Kibana page on the server. This potentially involves sending requests to external hosts. For example, a request might go to an external image server to show a field formatted as an image, or to show an image in a Markdown visualization.

If the headless Chromium browser is asked to send a request that violates the network policy, it will stop processing the page before the request goes out, and the report is marked as a failure. Additional information about the event is in the Kibana server logs.

Kibana installations are not designed to be publicly accessible over the internet. The Reporting network policy and other capabilities of the Elastic Stack security features do not change this condition.

xpack.screenshotting.networkPolicy: Capturing a screenshot from a Kibana page involves sending out requests for all the linked web assets. For example, a Markdown visualization can show an image from a remote server.
xpack.screenshotting.networkPolicy.enabled: When false, disables the headless browser network policy. Defaults to true.
xpack.screenshotting.networkPolicy.rules: A policy is specified as an array of objects that describe what to allow or deny based on a host or protocol. If a host or protocol is not specified, the rule matches any host or protocol.

The rule objects are evaluated sequentially from the beginning to the end of the array, and continue until there is a matching rule. If no rules allow a request, the request is denied.

# Only allow requests to placeholder.com
xpack.screenshotting.networkPolicy:
  rules: [ { allow: true, host: "placeholder.com" } ]

# Only allow requests to https://placeholder.com
xpack.screenshotting.networkPolicy:
  rules: [ { allow: true, host: "placeholder.com", protocol: "https:" } ]

A final allow rule with no host or protocol allows all requests that are not explicitly denied:

# Denies requests from http://placeholder.com, but anything else is allowed.
xpack.screenshotting.networkPolicy:
  rules: [{ allow: false, host: "placeholder.com", protocol: "http:" }, { allow: true }];

A network policy can be composed of multiple rules:

# Allow any request to http://placeholder.com but for any other host, https is required
xpack.screenshotting.networkPolicy
  rules: [
    { allow: true, host: "placeholder.com", protocol: "http:" },
    { allow: true, protocol: "https:" },
  ]

The file: protocol is always denied, even if no network policy is configured.

CSV settings

edit

We recommend using CSV reports to export moderate amounts of data only. The feature enables analysis of data in external tools, but it is not intended for bulk export or to backup Elasticsearch data. Report timeout and incomplete data issues are likely if you are exporting data where:

More than 250 MB of data is being exported
Data is stored on slow storage tiers
Any shard needed for the search is unavailable
Network latency between nodes is high
Cross-cluster search is used
ES|QL is used and result row count exceeds the limits of ES|QL queries

To work around the limitations, use filters to create multiple smaller reports, or extract the data you need directly with the Elasticsearch APIs.

For more information on using Elasticsearch APIs directly, see Scroll API, Point in time API, ES|QL or SQL with CSV response data format. We recommend that you use an official Elastic language client: details for each programming language library that Elastic provides are in the Elasticsearch Client documentation.

Reporting parameters can be adjusted to overcome some of these limiting scenarios. Results are dependent on data size, availability, and latency factors and are not guaranteed.

xpack.reporting.csv.maxSizeBytes: The maximum byte size of a CSV file before being truncated. This setting exists to prevent large exports from causing performance and storage issues. Can be specified as a number of bytes. Defaults to 250mb.
xpack.reporting.csv.scroll.size: Number of documents retrieved from Elasticsearch for each scroll iteration during a CSV export. Defaults to 500.

You may need to lower this setting if the default number of documents creates a strain on network resources.

xpack.reporting.csv.scroll.duration: Amount of time allowed before Kibana cleans the scroll context during a CSV export. Valid option is either auto or time, Defaults to 30s.

If search latency in Elasticsearch is sufficiently high, such as if you are using cross-cluster search, you may either need to increase the time setting or set this config value to auto. When the config value is set to auto the scroll context will be preserved for as long as possible, before the report task is terminated due to the limits of xpack.reporting.queue.timeout.

xpack.reporting.csv.scroll.strategy: Choose the API method used to page through data during CSV export. Valid options are scroll and pit. Defaults to pit.

Each method has its own unique limitations which are important to understand.

Scroll API: Search is limited to 500 shards at the very most. In cases where data shards are unavailable or time out, the export may return partial data.
PIT API: Permissions to read data aliases alone will not work: the permissions are needed on the underlying indices or datastreams. In cases where data shards are unavailable or time out, the export will be empty rather than returning partial data.

xpack.reporting.csv.checkForFormulas: Enables a check that warns you when there’s a potential formula included in the output (=, -, +, and @ chars). See OWASP: https://www.owasp.org/index.php/CSV_Injection. Defaults to true.
xpack.reporting.csv.escapeFormulaValues: Escape formula values in cells with a '. See OWASP: https://www.owasp.org/index.php/CSV_Injection. Defaults to true.
xpack.reporting.csv.useByteOrderMarkEncoding: Adds a byte order mark (\ufeff) at the beginning of the CSV file. Defaults to false.

« Monitoring settings in Kibana Search sessions settings in Kibana »