Service Pack 12

Release date: August 24, 2026

Service Pack availability

Service Pack 12 is available for the following release family:

  • v.8: Release v8.0.2
Important: v8.0.2 Data Agent supported version

A new data agent version (16.1.2) is required for the v8.0.2 release to incorporate the fixes introduced in this service pack.

Recommendation

It is strongly recommended to upgrade to the latest available service pack within your release family to ensure optimal stability, security, and continued support coverage.

Fixed issues

Bug IdentifierSupport TicketDescriptionApplicable Releases
XT-5447The data agent auto-upgrade could fail due to insufficient memory because the data agent’s version-check step inherited the same large JVM memory allocated to the main processes. The version-check step no longer inherits the large memory flags, allowing it to run without requesting excessive memory.
The fix is available in Data Agent version 16.1.2.
v8.0.2
INC-103064Spark SQL views built on top of Analyzer views weren't invalidating their cache when the underlying schema was updated.v8.0.2
INC-102615
INC-102445
69922
68992
The Join, Fuzzy Join, and Unpivot recipes could fail with a generic error: Could not get the schema, due to unhandled cases. The fix includes the following: Improved and descriptive error messages Join and Fuzzy Join recipes: Data Studio now automatically renames columns that share the same name with different cases. Unpivot recipe: The Header Column Name and Value Column Name fields are now mandatory before saving the recipe.v8.0.2
INC-10318770857The calculation of the expected time to execute a load job did not work correctly, causing the Load job taking longer than expected notification not to be sent when the Loader Service was down.v8.0.2
INC-10305270525Extra horizontal lines appeared on dashboards when row gridlines were enabled in tables.v8.0.2
INC-102794
INC-103100
INC-102779
INC-102715
70013
70165
70310
69825
70010
70139
70567
A large number of concurrent short-lived connections could exhaust memory and cause SQLi to become unresponsive. The fix has improved connection handling and memory management under heavy loads and improved SQLi resilience in throttling mode by:
  ●  Introducing a new CMC option, Max number of concurrent connections to SQLi, to limit the allowed concurrent connections to the SQLi service. Connections beyond the limit are rejected instead of queuing up. The default is 300; increasing this limit requires increasing the SQLi allocated memory.
  ●  Decreasing the default Timeout to allocate memory before SQLi rejects connection to 10 seconds.
v8.0.2
INC-10315770829Pivot and Aggregate tables with Merge Columns enabled threw an error: Cannot read properties of undefined (reading 'style').v8.0.2
INC-10270970112A schema could incorrectly show Finished With Errors status (with a blank error page) although everything actually completed successfully, while the non-mandatory COMPACTION step was still running at status check.v8.0.2
INC-102243
INC-103192
68999
70299
Long-running queries in the Grouping and Aggregation phase were not interrupted after exceeding the configured timeout, causing subsequent dashboard requests to fail.v8.0.2
INC-10321270854Large table extractions from Databricks would start normally, make some progress, and then get stuck until manually cancelled or timed out.v8.0.2
INC-103355
INC-98959
70992
63318
64017
64069
Rich Text insights displayed is not a constructor in dashboards after upgrading from 2024.7.6, 2025.7.4, or earlier releases to a later release.v8.0.2
INC-10321170925Dashboard paths were displayed in black after upgrading to v8.0.0 - 2026.8 when custom CSS was configured.v8.0.2
INC-10332470388
71108
During On-Premises upgrades to v8.0.0 - 2026.8, customers could encounter some intermittent environment-related issues.v8.0.2
INC-10343371304SSO login would fail after the upgrade if Distributed User Session was enabled for the cluster.v8.0.2