The CMS service is no longer disabled on the Additional Polling Engine backup.

Jobs are no longer executed in the wrong order in the JobEngine recurring queue.

Active Diagnostics are no longer failing on CRL checks due to an invalid URL.

The default health checks for High Availability (HA) have been updated to fix an issue that could cause an HA Pool Member to be marked down incorrectly.

The WMI/WinRM cached connection no longer fails to perform polling if the target machine is restarted.

The All Active Alerts page keeps the selected Group By value after refreshing and no longer resets to Severity.

When the SolarWinds Observability Network Collector license expires, the Network Collector host machine no longer enters in to inconsistent state.

Recent search results in the Global Search feature in the top right corner of all pages are no longer missing.

Inefficient NotificationItemsGrouped view no longer causes the SolarWinds Platform Web Console to hang and overload the SQL.

The PerfStack chart correctly shows the parent node in Reports and Views.

When setting filters in Modern Dashboards, the following error is no longer displayed:

An error has occured. RunQuery failed, check fault information. No viable alternative at input 'WITH' in From clause.

In-process workers no longer spawn new SNMP jobs before the previous one finishes.

RabbitMQ message queues no longer increase significantly, causing delayed alerts and data sync issues.

NCM table views that have been customized (for example, by adding, removing, or rearranging columns) no longer break after the next page refresh.

Files can be saved to a network share.

If you have a high number of Access Points (APs), you no longer experience intermittent, slow web UI performance caused by excessive query executions on the Wireless Index table. This inefficiency has been optimized, resulting in improved performance, particularly on Wireless Monitoring pages.

The SolarWinds.Topology.Calculator.exe process no longer fails with an unhandled exception during the removal of duplicate connections. Redundant Layer 3 (L3) connections are no longer stored in the database, increasing SQL Server load and extending topology calculation processing times.

The setting RemoveRedundancyMaxThreads in SolarWinds.Topology.Calculatorx64.exe.config has been updated to improve stability and prevent failures.

Interfaces with InterfaceIndex=-1 are no longer polled, resolving remapping issues. The fix applies to SNMPv2, SNMPv3, and WMI interfaces, improving accuracy and preventing incorrect polling requests.

In environments with multiple F5 nodes, the F5 API Status poller is no longer enabled without the required settings (F5.API.CredentialID, F5.API.Port, F5.API.UseSsl) configured in NodeSettings, causing repeated errors in the Collector logs and leading to log flooding and potential disruptions in F5 monitoring.

The F5 API Status poller is only enabled when the required settings are present, preventing log errors and improving stability.

When SQL Express is used as the Flow Storage database, NTA pages display a banner about size and performance limits. Dismissing this banner no longer causes NTA pages to display an error and fail to load.

The SolarWinds Query has been optimized to improve the performance of the Components widget on the Application Details page. The optimization affected also the performance of the Processes and Services, Top XX Components by Statistic Data widgets, the general Application Details page load, and the overall database performance.

The ODBC probe works correctly on AIX DB2 devices.

Incorrect SQL logic for APM_SqlBbConnection no longer leads to issues during upgrades in certain scenarios.

Tables with the same name in different DB schemas no longer break AppInsight for SQL polling.

SAM WIM/WinRM polling no longer ignores the Core per-node/global Connection Mode setting.

Filters are displayed as expected on the SCM Monitored Nodes page.

Topology/UDT/connections are no longer broken.

When upgrading to version 2025.1 or later, the Configuration wizard completes successfully and the log no longer contains any errors related to VNQM.