If the National Institute of Standards and Technology (NIST) does not include data about a device type in the Common Platform Enumeration (CPE) Dictionary, the missing data no longer prevents SolarWinds Observability Self-Hosted from identifying firmware vulnerabilities that affect the device.

Searching and filtering on the All Active Alerts page works as expected.

The percent usage (disk, volume, memory) of monitored components no longer shows values over 100%.

When the delayed trigger condition is used, false positive alerts are no longer triggered after the Alerting Engine service is restarted.

When a dashboard has a limitation defined, the Intelligent Map widget displays the entities as expected.

After you edit an anomaly-based alert, the Anomaly-based alerts Details page no longer fails to open, and previously selected entities and entity type are displayed as expected.

The Management resource for Node Details and Volume Details widgets no longer disregards custom LP translations.

Configuring alert actions no longer requires different permission levels for setting up credentials.

Intelligent Map highlights the traffic value that exceeds a configured threshold.

The SNMP fast status poll no longer incorrectly updates the PercentLoss value.

The out-of-the-box alert Node not polled in last 10 minutes works as expected and the column LastSync is no longer updated if polling does not work.

If a polling engine is upgraded to 2024.4 or later but agents remain on older version, CPU data is not missing for agent-monitored Windows servers.

Device details for Westermo devices show the correct vendor information.

The zoom function in the Intelligent Map works as expected.

When you configure an alert to send an email where the variable ${N=SwisEntity;M=NodeStatusRootCauseWithLinks} is used, the Child Entities section in the email notification includes URLs for the detail pages.

Members of Windows Group Account can change the Hardware Health Sensor temperature units.

Exporting the Orion Group custom property no longer fails with an error Error: One or more errors occurred.

Account limitations no longer break the Active Alert Details page for global alerts.

Database types are no longer mismatched and causing performance issues.

When the polling method changes, CPU and memory pollers are automatically imported into the database as expected.

When you select the WMI protocol as the preferred polling method, SolarWinds Platform no longer tests all SNMP credentials against WMI nodes, causing network discovery to take longer than expected.

When adding Nodes with SNMPv3 with an MD5/AES256 configuration, the Test connection no longer fails with an error INFO SolarWinds.Orion.Core.BusinessLayer.CoreBusinessLayerService - (null) SNMP credential test finished. Success: False.

When installing scalability engines, the Configuration Wizard no longer fails with the following error related to the existence of Rabbit MQ service: Error starting Service (RabbitMQ), Service RabbitMQ was not found on computer '.'.

Details for value and custom property in Modern Dashboards no longer overlap.

List Resources on Agent node show the interface selections as expected.

The Active Diagnostics test no longer calculates the memory size incorrectly for the 64-bit SQL server.

When you uninstall an additional polling engine, RabbitMQ queues are deleted as expected.

Existing Intelligent Maps created without a container are available in the Entity Library, and they can be updated, saved, and added to other maps.

The date and time are formatted correctly in custom reports.

The Set of objects filter in the Add Alert Wizard works as expected.

When you have groups over entities with the status of Up and Inactive with group status calculation of Best, the status calculation is reflected correctly.

When you open the Deployment Health tab from an additional web server, health checks run correctly.

WMI job issues that caused JobEngine to use too much CPU and memory in environments with heavy WMI polling were addressed by configuring WMI polling jobs to run outside of in-process workers by default.

Upgrading the SolarWinds Platform log database no longer fails with an error similar to the following if the database is contained:

Cannot resolve the collation conflict between "SQL_Latin1_General_CP1_CI_AS" and "Latin1_General_100_CI_AS_KS_WS_SC" in the equal to operation.

Using a date format other than MM/DD/YYYY no longer prevents the discovery of a Power Control Unit (PCU).

When a subgroup is added to a map, the All Groups widget and Manage Groups page no longer display the subgroup as both a root level group and a subgroup.

The Incident Number column on the All Active Alerts page displays the number of currently active incidents associated with an alert. It no longer includes incidents that occurred in the past and are no longer active.

When you select the $Name variable under Map name format, you can save the map definition.

Information that should not be available to a user because of limitations is no longer exposed in certain situations, such as a failed service initialization on an unstable database connection.

Duplicate device types for FortinetFortiManager Orchestrator devices in the SolarWinds Platform database are removed. This situation no longer causes upgrades to fail with the following message:

Database configuration failed: Error while executing script- Violation of PRIMARY KEY constraint 'PK_DeviceTypeViews'. Cannot insert duplicate key in object 'dbo.ViewsByDeviceType'. The duplicate key value is (FortinetFortiManager Orchestrator). The statement has been terminated.

The custom Trap rule action Change the status of an interface changes the status as expected when the trap is received.

When the certificate store contains more than one SWI-Orion-Agent certificate that can be used by the Job Engine to encrypt WCF communications, the Job Engine no longer fails with the following error:

FATAL SolarWinds.JobEngine.ServiceProcess.SolarWindsService - (null) An error happened in Start System.ArgumentException: It is likely that certificate 'CN=SWI-Orion-Agent (22409EE3-F4B9-4B88-8957-677BA409422C)' may not have a private key that is capable of key exchange or the process may not have access rights for the private key. Please see inner exception for detail.

When a device with a removable disk (floppy or compact disk) is polled and no disk is inserted, a warning is not logged. In addition, the Volume Space Used, Volume Space Used, and Volume Space Available are shown as 0 instead of -2.

When the Job Engine starts, it ensures that multiple SWI-Agent-JobEngineService certificate do not exist. If multiple certificates exist, they are deleted and a new self-signed certificate is generated. Multiple certificates can cause issues such as metrics not being collected.

A condition that caused the following exception no longer occurs:

Exception Info: System.NullReferenceException
   at SolarWinds.Orion.Feature.Client.NonBlockingFeatureClient.RefreshCacheOnTimer()
   at SolarWinds.Orion.Feature.Client.NonBlockingFeatureClient.<.ctor>b__8_0(System.Object)
   at System.Threading.TimerQueueTimer.CallCallbackInContext(System.Object)
   at System.Threading.ExecutionContext.RunInternal(System.Threading.ExecutionContext, 
      System.Threading.ContextCallback, System.Object, Boolean)
   at System.Threading.ExecutionContext.Run(System.Threading.ExecutionContext, 
      System.Threading.ContextCallback, System.Object, Boolean)
   at System.Threading.TimerQueueTimer.CallCallback()
   at System.Threading.TimerQueueTimer.Fire()
   at System.Threading.TimerQueue.FireNextTimers()
   at System.Threading.TimerQueue.AppDomainTimerCallback(Int32)

Forwarding a significant number of syslog and trap messages per second with spoofing no longer results in some messages not being sent, high CPU and RAM usage, and errors such as the following in logs:

ERROR SolarWinds.Orion.LogMgmt.RuleProcessing.LogRuleTester - Failed action result
ERROR SolarWinds.Orion.LogMgmt.LogServices.Common.Util.UdpPacketPcapSender - Failed to forward spoofed packet using destination ip xx.xxx.xx.xx and spoofed source ip xx.xxx.xx.xx

If the language for the browser is changed, reports with a relative time period selected (for example, Last Week) show data from the correct date range whether you view them in the browser or export the report to a PDF document.

Locking issues that could result in macros not being parsed correctly or other alert failures have been resolved.

The Event log no longer includes erroneous audit event messages saying that a user changed an account.

Running the SolarWinds Platform installer for versions 2024.3 and earlier no longer returns the error Object reference not set to an instance of an object.

The Node Details widget is no longer blank when the node category is null.

Exporting a modern dashboard no longer fails with the following message:

Invalid object name 'DashboardGroupItem'.

When you edit an Intelligent Map, the points connect to the correct locations.

If a map is added to a group and a user account has a limitation associated with that group, the user is no longer prevented from viewing the map or from seeing it in the list of available maps.

When a custom property has a value of Unknown, adding a limitation that uses the custom property no longer fails.

For subscription licenses, the License Details page shows the correct number of days left in the subscription.

When upgrading ETS, the Configuration Wizard no longer fails due to a duplicate DF_Accounts_ToolsetIntegration constraint.

The SolarWinds Diagnostic tool no longer continuously increases the size of the Toolset_BusinessLayerPlugin_tracelog.svclog file. The large file could consume all disk space and also affect performance.

The Default database and archive maintenance job purges the Config Archive folder as expected and the log file no longer shows 0 files deleted.

When the option Direct URLs to JSON vulnerability announcement data files to be automatically downloaded and imported into database is disabled, firmware vulnerability matching updates the last date run as expected.

The performance of impersonation was improved to avoid long loading times for users with large profiles.

The SRD calculation no longer takes into account established ACL rules.

VLANs Ports polling no longer fails due to a broken SNMP VLAN context on a node.

When you add the Search for Nodes widget, and select Status from the Search by drop-down, the search no longer fails.

After adding Aruba Orchestrator and Edge devices, the Job Engine service process memory usage is stable, and the SdWan.Collector.Jobs and Collector.Service.log log files no longer include an error An item with the same key has already been added.

When the response from a Juniper Mist device does not include a MAC address, polling the device no longer fails and the device is not shown as down.

Members of Windows Group Account can change the Hardware Health Sensor temperature units.

After the migration to Timeseries, the FirstUpdate column of Wireless_Clients historical database tables is being properly summarized.

When the settings InterfacesCorrelationEventsRequestRetrievedFields and InterfacesRequestRetrievedFieldsPalo are empty, API polling no longer fails for Alto Prisma SD-WAN devices, resulting in missing VPN connection data.

When the response from an Extreme Networks Wireless Access Point does not include an IP address, polling no longer fails.

The SWIP query to retrieve SD-WAN orchestrator statistics no longer causes SQL Server performance issues, including rapid growth of the tempdb table.

When an Aruba API node has multiple interfaces with duplicate ifIndexes, the interfaces are discovered successfully, and logs no longer include messages such as:

ERROR SolarWinds.Orion.Core.BusinessLayer.JobSchedulerEventService - (null) Processing of discovery result for profile 4 failed for plugin SolarWinds.Interfaces.DiscoveryPlugin.InterfacesDiscoveryPlugin System.Data.SqlClient.SqlException (0x80131904): Violation of PRIMARY KEY constraint 'PK_DiscoveredInterface'. Cannot insert duplicate key in object 'dbo.DiscoveredInterfaces'.

The Aruba EdgeConnect REST Walker no longer fails with an error similar to the following:

Walk unexpectedly finished with exception. The process cannot access the file 'C:\Users\nsell\Desktop\silverpeak_thgrp_com__08222024180247\CompressedPathsMapping.map' because it is being used by another process.

When processing v9 NetFlow from Fortigate devices, NTA no longer displays an error:

ERROR SolarWinds.Netflow.Processing.Workflow.PacketProcessingWorkflow - error processing packet System.ArgumentOutOfRangeException: Specified argument was out of the range of valid values.

Unmanaged interfaces no longer cause large amounts of DiscoverInterfacesOnNode:Orion.NPM.Interfaces verb invokes.

NTA logs no longer cause high CPU utilization due to a high number of incoming PDU warning messages.

When upgrading NTA, interfaces are no longer labeled as deleted.

When the Autonomous System resource attempts to display information about Juniper MX480 devices, it no longer fails with the message Error while loading legend.

When a JMX node is added, the connection is established and the list of mBeams is displayed in the wizard.

The Cortex REST API server is limiting TLS renegotiation attempts as expected.

The aggregation type LAST is no longer declared for some AWS and Azure database instance properties. This incorrect aggregation type resulted in errors such as the following in the OrionWeb.log file:

Property 'Orion.Cloud.Azure.SqlDatabase:AllocatedDataStorage' metadata 'defaultAggregation' definition in SWIS schema is invalid. Cannot parse value 'LAST'.

The SqlDatabaseContextProviderBuilder class is thread-safe and SAM no longer displays errors about the wrong user trying to log in.

All parameters for AWS VPN connections are displayed as expected.

The password variable in a connection string for the PostgreSQL template is resolved as expected.

API Pollers no longer trigger false positive error messages about licensing.

The Performance Analysis (PerfStack) dashboard no longer displays incorrect units on charts for SAM applications.

Azure polling no longer fails with an error ERROR SolarWinds.VIM.CloudMonitoring.Pollers.Azure.AccountAutomonitoringPoller - Unable to get azure VMs.

The account automonitoring job no longer tries to find instances in regions the account does not have access to.

SAM made improvements to prevent slow performance in deployments with a large number of containers.

JMX Monitor on Linux no longer returns an error Java version , even though the correct version is used.

Assignment of bulk templates no longer causes duplicates.

Monitoring a large number of cloud accounts no longer causes high CPU utilization on the main polling engine.

After an upgrade, all AppInsight for SQL statistics are retained and any custom retention settings do not revert to their default values.

Attempts to poll Azure load balancers for Azure US Government accounts no longer fail with the following error:

SolarWinds.Orion.Contract.Exceptions.RestClientException: Failed to request /0887f83c-eb3f-44a1-8ceb-5088172c5750/oauth2/token due to StatusCode: BadRequest

{"error":"invalid_request","error_description":"AADSTS900382: Confidential Client is not supported in Cross Cloud request.

Performance has been improved to reduce timeout errors and slow response times when adding nodes to groups.

When more than one entity has the same Resource Name, the Compute Entities resource on the Azure Overview tab and the Virtual Machines resource on the Azure Compute tab display data correctly.

The maintenance of CDR files no longer prevents VNQM from starting.

When you want to check credentials by certificate, VNQM checks AXL credentials by domain name as expected.

The IP SLA Operation Details page no longer times out if there are many operations.

Job Engines on different polling engines no longer interact.

Operation frequency validation works as expected.

In large deployments with the main polling engine alongside multiple additional polling engines, artificial limitations no longer slow down parallel processing of IP SLA jobs.

Large values in the destSpan_value no longer result in call data being dropped and errors such as the following appearing in the IPSLA business layer logs:

Exception occured on GetFilesInDirectory Method: System.OutOfMemoryException: Exception of type 'System.OutOfMemoryException' was thrown. - 2024-09-24 14:41:10,093 [13] ERROR BusinessLayer.Task.CdrCmrCsvFileReader (null) - Cannot convert int value [column:destSpan,rawValue:4134469638]”.

When you are monitoring a gateway, the Calls by Region widget in the VoIP Summary View no longer displays an error message.

*The issue where JobEngine failed on start due to an unhandled exception related to System.ValueTuple was addressed.