Fixed issues

The following customer-reported issues have been fixed in this release:

  • (HUB-45342). Fixed an issue that caused the ignored status flag in version_bom_risk_warning to become inconsistent with the remediation status due to residual data in version_bom_risk_status. The issue was traced to leftover data in version_bom_risk_status from migration scripts, which caused ignored flags to revert incorrectly. The fix involved rerunning migration scripts and deleting specific records where the status was REMEDIATED but ignored was false, ensuring proper synchronization between ignored flags and remediation status.

  • (HUB-45481). Resolved an issue where the upgrade to version 2025.4.1 took an excessive amount of time due to an unoptimized upgrade script. The script has been optimized to significantly improve upgrade performance.

  • (HUB-45490). Addressed an issue where the KB update job occasionally failed to update certain project versions with newly identified vulnerabilities. During the investigation, it was identified that vulnerability counts for component versions were out of sync across various project versions. As part of this update, the BOM for affected component versions has been refreshed to improve accuracy. However, new components that appear after the BOM refresh are not automatically updated at this time. Further improvements are being actively developed to provide a more comprehensive solution in future releases.

  • (HUB-45609). Fixed a SQL error caused by incorrect argument ordering in the query for the endpoint /api/vulnerabilities/{vulnerability_id}/affected-bom-components. The issue resulted in a BadSqlGrammarException and a PostgreSQL error indicating a mismatch between the ~~* operator and argument types (text vs. UUID). The fix corrected the query's argument order to ensure proper execution across all namespaces.

  • (HUB-45699). Improved the performance of the components-in-use API by optimizing the query plan to significantly reduce execution time and resource usage. The previous implementation caused excessive data processing and disk spill during sorting. By streamlining the query logic, the fix greatly reduces system overhead and improves response times, making the API more efficient and usable.