Falcon LogScale 1.267.0 GA (2026-11-17)

Version?Type?Release Date?Availability?End of SupportSecurity UpdatesUpgrades From?Downgrades To?Config. Changes?
1.267.0GA2026-11-17

Cloud

Next LTSNo1.177.01.177.0No

Hide file download links

Show file download links

Bug fixes and updates

Advance Warning

The following items are due to change in a future release.

Removed

Items that have been removed as of this release.

GraphQL API

  • The GraphQL mutation deleteFeatureFlag() has been removed. This mutation was deprecated in version 1.254 and announced for removal in LTS version 1.258.

Deprecation

Items that have been deprecated and may be removed in a future release.

  • The GraphQL field meta has been deprecated and now requires authentication. It will be completely removed in LogScale 1.304. To achieve similar results, use the loginInfo and clusterConfig fields instead.

    To temporarily opt out of the authentication requirement, the feature flag UnauthenticatedMeta can be enabled.

  • The userId parameter for the updateDashboardToken GraphQL mutation has been deprecated and will be removed in version 1.273.

  • The configuration variable USING_EPHEMERAL_DISKS is now slated for deprecation and removal in LogScale version 1.282.

    Because this configuration variable can only be enabled on clusters with bucket storage and there is no use case for disabling it, the option will be deprecated. Instead, LogScale will adopt the correct behavior based on whether bucket storage is configured. When bucket storage is enabled, nodes will be ephemeral; when bucket storage is disabled, nodes will be non-ephemeral.

    Clusters that have configured bucket storage but have not set USING_EPHEMERAL_DISKS will be silently migrated to ephemeral disks.

    Clusters that have configured bucket storage and have explicitly set USING_EPHEMERAL_DISKS to false will crash on boot. Administrators should unset USING_EPHEMERAL_DISKS on such clusters.

    To boot a cluster with bucket storage and include USING_EPHEMERAL_DISKS with it set to false as a temporary measure, set ALLOW_USING_NON_EPHEMERAL_DISKS_WITH_BUCKET_STORAGE to true. This configuration option will also be removed in 1.282, and exists only as an escape hatch.

Behavior Changes

Scripts or environment which make use of these tools should be checked and updated for the new configuration:

  • GraphQL API

    • An issue has been fixed where GraphQL responses with a size of 1 GB or greater could crash a LogScale node with an OutOfMemoryError notification while the response was being written.

      GraphQL responses from the /graphql endpoint are now streamed to the client in chunks rather than rendered in full prior to sending. This changes HTTP responses in the following ways:

      • Responses use chunked transfer encoding (Transfer-Encoding: chunked) and no longer include a Content-Length header.

      • The request header Accept is now no longer used to negotiate the response content type. Instead, responses are now always application/json. Previously, a successful request whose Accept header did not allow application/json was answered with the error response 406 Not Acceptable; it now receives the JSON response. Error responses are already sent as application/json regardless of the Accept header, so they are unchanged.

      • If the connection is interrupted while a response is being sent, the client sees a truncated response body rather than an error status, because the status line has already been sent.

      To disable this change, set the feature flag StreamChunkedGraphqlResponse to false. This opt-out will be available until 2027-02-10, after which the feature flag will be removed.

Upgrades

Changes that may occur or be required during an upgrade.

  • Security

    • The FasterXML Jackson BOM dependencies (jackson-core, jackson-databind, jackson-dataformat-cbor, jackson-dataformat-csv, jackson-dataformat-yaml, jackson-datatype-jdk8, jackson-datatype-jsr310, and jackson-module-scala_2.13) have been updated to version 2.22.4 to address a CVE.

    • The dependencies for the org.ow2.asm framework (asm, asm-analysis, asm-commons, asm-tree, and asm-util) have been updated to version 9.11.1 to address a CVE.

  • Other

    • The dependency io.modelcontextprotocol.sdk:mcp-core has been updated to version 2.0.1 to address a CVE.

New features and improvements

  • GraphQL API

    • The GraphQL mutation setCapacitiesForProductType() has been added; Set the capacity for a product type of several organizations. Organizations not given are left untouched. If any part of the input is invalid, nothing is changed. Returns the number of organizations whose capacity changed.

    • The GraphQL mutation clearCapacitiesForProductType() has been added; Clear the capacity for a product type of several organizations. Organizations not given, and other product types, are left untouched. If any part of the input is invalid (i.e., a non-existent organization ID is given), nothing is changed. Returns the number of organizations whose capacity was cleared.

  • API

    • A new query export feature provides HTTP API endpoints that run non-aggregate searches and write matching events to bucket storage rather than returning them in the HTTP response. Use this feature to export result sets that are too large to stream through a single HTTP response.

      This feature is off by default. Enable it globally using the feature flag ExportQueries with enableFeature(). The API shape is experimental and not covered by any API stability guarantee.

      Four new endpoints are provided:

      • POST /api/v1/repositories/{view}/export-query — submits an export and returns the query ID.

      • GET /api/v1/repositories/{view}/export-query/{id} — polls export status and returns the list of exported objects.

      • GET /api/v1/repositories/{view}/export-query/download/{id} — streams the result through LogScale.

      • DELETE /api/v1/repositories/{view}/export-query/{id} — cancels an export.

      All endpoints require read access and must target a query-capable node; ingest-only nodes reject the request.

  • Configuration

    • The environment variable PDF_RENDER_SERVICE_SHARED_SECRET has been added. When set, LogScale sends its value as a bearer token with every request to the PDF render service. Set the same value in the render service's SHARED_SECRET environment variable to make the render service reject report requests that do not carry it.

  • Functions

    • The drop() function now supports glob patterns with the new mode parameter. Set mode to glob to drop multiple fields by pattern rather than listing each field explicitly.

      For example, to drop all fields whose names start with error_:

      logscale
      drop(fields=["error_*"], mode=glob)

      Multiple patterns are supported:

      logscale
      drop(fields=["tmp_*", "*_internal"], mode=glob)

      Glob patterns also work for arrays. To drop all entries of an array including non-contiguous indices:

      logscale
      drop(fields=["errors[*]"], mode=glob)

      To drop specific sub-fields across all entries of an object array:

      logscale
      drop(fields=["events[*].severity", "events[*].internal_id"], mode=glob)

      The original behavior is unchanged and remains the default (mode=exact).

      For more information, see drop().

Fixed in this release

  • User Interface

    • An issue has been fixed in the Fields panel where selecting Sample more could display the notification Showing fields from latest 0 events and disable percentages, even though the query had events. This occurred when some fields were counted more often than the total number of events. The panel now shows the actual event count and explains why percentages are unavailable.

  • Storage

    • An issue has been fixed where temporary files could be left on disk after a decryption failure when downloading data from bucket storage. The temporary file is now deleted on failure.

  • Dashboards and Widgets

    • An issue has been fixed in Schedule PDF Reports where widget time windows were resolved in UTC instead of the scheduled report's associated time zone. Time ranges that depend on the time zone such as calendar-based or anchored ranges could start and end at different times in the PDF than on the dashboard. Widget queries in scheduled reports now use the schedule time zone. The $startTime and $endTime values in note and text widgets also now use the schedule's time zone, so they may show different times than in earlier reports.

  • Functions

    • An issue has been fixed where the function sample() accepted the parameter value percentage=NaN as a valid value. Because comparisons with NaN always return false, this was equivalent to passing false and produced no events. percentage=NaN is now disallowed, the same as percentage=0.

    • An issue has been fixed with the fieldStats() function, where some fields could be counted twice in certain scenarios.

Known Issues

  • Storage

    • For clusters using secondary storage where the primary storage on some nodes in the cluster may be getting filled (that is, the storage usage on the primary disk is halfway between PRIMARY_STORAGE_PERCENTAGE and PRIMARY_STORAGE_MAX_FILL_PERCENTAGE), those nodes may fail to transfer segments from other nodes. The failure will be indicated by the error java.nio.file.AtomicMoveNotSupportedException with message "Invalid cross-device link".

      This does not corrupt data or cause data loss, but will prevent the cluster from being fully healthy, and could also prevent data from reaching adequate replication.

Improvement

  • Automation and Triggers

    • Slack action fields can now be reordered by dragging them into a different position in the Slack Action form.

  • GraphQL API

    • The GraphQL datatype ScheduledReport now exposes the field createdInfo, which contains structured creation metadata. Data points include author and timestamp, which is consistent with other asset types such as filter alerts, dashboards, and parsers.

      The existing fields createdBy and creationDate are now deprecated and will be removed in version 1.306. Migration to the createdInfo field is recommended, as it provides richer author information (user, token, and/or system) alongside the creation timestamp.

      Existing scheduled reports are automatically migrated, no action is required.

    • The GraphQL query searchDataDistribution() now accepts the new argument includeEmptyRepositories. By default, repositories that have no matching datasources with segments in the time range are now excluded from results. To include them, set the argument to true. Repositories the query can never target, such as those excluded by a #repo filter or a view connection filter, are always excluded. The default for the argument maxDatasources was also raised from 200 to 1000.

  • Configuration

    • The distribution mechanism for Indicator of Compromise (IOC) data within a LogScale cluster has been updated to reduce the load on the node responsible for fetching new data from the remote IOC data source.

      Two new dynamic configuration parameters control how often IOC data is updated:

      • IocRemoteUpdateInterval - determines how often the responsible node downloads new IOC data from the remote source. The default value is 45 minutes (m).

      • IocFollowerUpdateInterval - determines how often the other nodes fetch IOC data from the responsible node. The default value is 15 minutes. Fetches from follower nodes are spread out over this interval.

      Both values accept duration strings such as 45 minutes or 1 hour (h) and must be positive. The worst-case delay for new IOC data to appear in ioc:lookup is unchanged at one hour, as it is the sum of both intervals.

  • Queries

    • The raw throughput of function predicates in match() statements has been improved by approximately 5-8%. For example, queries of the form field match { in(values=[A,B,C]) => ... ; ...} benefit from this change.

    • Queries using the sort() and selectLast() functions now automatically skip segments that cannot affect the final result, improving performance and reducing total query cost. Previously, this behavior required users to set the parameter allowEventSkipping to true on query submission.

      To disable this functionality and ensure comprehensive event counts, set the parameters showQueryEventDistribution or doneCriteria.minimumMatchedEvents to true. To force event skipping even when showQueryEventDistribution or doneCriteria.minimumMatchedEvents is set, set allowEventSkipping to true.

  • Fleet Management

    • The Fleet Management Configuration Editor now displays the minimum required Log Collector version in the Publish dialog for a configuration.

  • Metrics and Monitoring

    • Two new fields have been added to log lines for slow queries and starved searches log lines, as well as to the query-ended log line for non-streaming queries:

      • segmentBytesReadFromDisk - logs how many bytes the query read from disk from segment files, counted before decompression.

      • hashFilterBytesReadFromDisk - logs how many bytes the query read from disk from hash filter files.

  • Functions

    • The function readPersistedAggregation() now reports a single warning notifying the user that the read is incomplete when the query memory limit has been exceeded. This replaces the error notifications Fewer events than expected and Bucket is incomplete.