July 2017
Version: 4.0.5.30000
New Features
- Advanced Datasource filters for 'out-of-the-box' SQL Server User Logons Datasource Template within Alignia Business Security module.
- A new Database Generic Query Errors Check for Business Applications.
Fixes
- 2-60132590 - When deleting a used credential it was not possible to reconfigure the asset. This has been fixed.
- 2-60570816 / 2-60182141 - Stopping the Orchestrator service caused a message to be displayed stating that the Orchestrator service had ended unexpectedly. This was caused by exceptions of the type: Unhandled exception at 0x58DE7963 (MDB_Server.dll) in MdbServer_MiniDump_20170322.dmp: 0xC0000005: Access violation reading location 0x00000004. The issue has been corrected and the Orchestrator service can be stopped normally.
- 2-60250631 - Some Online Business Services monitors in Alignia ThinkServer caused an unlimited growth of database user connections as the number of connections was not limited. This meant the number of connections could grow unchecked resulting in an impact on the performance of the database engine. This has been fixed.
- 2-58419492 - Support has been added for writing multi-byte characters to databases with multi-byte collations. This is disabled by default to improve performance. To enable this feature, add the parameter <supportMultiByteCharacterDatabases> to T4BD.config. Messages collected by EventLog monitors that contained multi-byte characters would sometimes fail to be written to databases with collations that support multi-byte characters.
- Some Device elements were monitored as having 'No Core' assigned even if they were configured as 'Role Core'. Therefore, potential issues with Device Core health were not reported as they should have been. This has been fixed.
- The availability percentage shown in the Business Applications Availability Manager page was different to the one shown in the report created by using the link "Create Report" from the same page. For periods not ended, the report availability percentage was being calculated using the whole time period rather than just the period up to the current time. This has been fixed.
- A 'Maximum request length exceeded' error message was generated when importing a file greater than 20 MBs into Alignia Business Applications. Furthermore, in some cases the import process failed due to a Microsoft SQL Server deadlock. The issue was caused by the web.config file having a maximum file size limit of 20 MB. This has now been increased. The SQL Server deadlock problem was caused by some queries that were blocking the database.
- The creation and/or modification of Business Views in SmartConsole and of Monitors in ThinkServer has been improved whenever the user performs any change in an asset. Issues that arose when synchronizing monitors with remote ThinkServer have been resolved. Currently the changes are evaluated sequentially in a queue. The process has been improved to synchronize each asset and evaluate when there are multiple pending evaluations.
- Online Business Services alerts did not work correctly for KPI, Real & Simulated Experience definitions with some "All" values defined for Channel, Function and Consumers. If the summary definitions contained at least one value of 'All' and aggregate data from more than one specific business instance, this aggregation was not performed correctly. This has been fixed.
- 2-61778871 - When editing a macro, if the user clicked on a date in the calendar, the "Save" button did not appear. The issue has been fixed.
- 2-61792221 Online Business Services response time alerts may not have reported correct values for the number of transactions and average response time for both Real & Simulated Experience data. For example, zero transactions and response times may have been reported when transactions actually existed. This issue has been fixed.
- When selecting columns to be displayed or hidden in the OBS Summary table, the change was not processed and after saving, the selection did not take effect. The selection engine of the columns in the OBS summary table reported a problem when selecting columns to display or hide. This has been fixed.
- The data recollection process for Linux physical and virtual memory has been changed. The Linux monitors for physical/virtual memory and CPU usage would sometimes fail if the monitored asset was running a very old version of procps.
- Windows EventLog Agent. When using Automatic Recollection to retrieve EventLog messages and the Datasource recollection fails, the reported error (in Alignia Web and ThinkServer Configurator) was always: 'Exception in EvLgDataSource::RecollectData. Operation: RecollectEvents Error: Exception while retrieving EventLog events'. In order to see the full details you needed to open the actual log file. The error was correctly recorded in the log file but instead of including the actual information in the generated exception, a generic message was always used instead. This has been fixed.
- The Job name did not appear in any of the normalized variables of the "User Activity > User Statement > Command execution" event retrieved during the audit of IBM i systems CD journal entries. This information has been added into the Application normalized variable.
- 2-56177007 - The links generated for the "Execute again" feature did not contain the original start time of the macro. This resulted in the reports, whose date range was relative to the current date, being executed with a different date range to the one that originally failed. The issue has been fixed and now the links contain the original start time of the macro.
- 2-62069203 - When using Database Settings Administrator and loading the PMDB tab (under the Maintenance section), an error page was sometimes displayed. This issue has been fixed.
- 2-61945153 - If the user resets the filters in the grid on the Forensic Analysis page, the headers of the columns to which filters were applied are displayed as though the filter is still active. This issue has been fixed.
- A fix has been applied so that the correct labels for the Architecture section are displayed when adding or editing a Connectivity Group within Business Applications. The issue was caused by the text labels not being updated.
- Copying an asset that is being monitored creates the copied asset as active. This results in the same asset being monitored twice, creating lot of duplicate events and data. Furthermore, there was an issue with Security assets because the enabling of the asset monitors was not performed, so the monitors in ThinkServer were not created. When copying an asset, only Processes were created in a deactivated state. From this version onward, all copied assets will use this behavior. Furthermore, another issue existed whereby the copied asset was not then processed.
- When importing an XML that included an asset with an unexpected attribute, the full import process failed with a message error such as: Exception importing Entities. Error: The attribute type 'attribute_type' is not applicable to assets of type: 'Asset_type'. When importing assets, a check is made that all the attribute types can be imported. If any of them failed then the full import process was canceled to avoid an inconsistent status. Now, if the attribute type is not valid for an asset, it is ignored and the import of the remaining attributes and assets continues.
- When importing an asset, existing templates assignments were not removed but templates defined in the imported file were added. The correct behavior should be that in cases where the imported file indicates that an asset has templates assigned only these templates should be included after the importing process is complete. Any previous template assignment should be removed. Previously, imported template assignments were added to existing assignments. Any existing template assignments are now overridden.
- Sometimes when starting Orchestrator, and the SQL Server was temporarily unavailable or starting, Orchestrator didn't operate properly even when the SQL Server started. The easiest way to check if this was happening was by trying to use Alignia Business Applications Configuration because the list of Devices or Applications was never retrieved from the server. In some cases there was a deadlock when the ODBC connection was reset so it never recovered from this status condition. This has been fixed.
- The Analysis tab of websites in Alignia, now shows the Transfer Start Time graph instead of the Transfer Total Time. The description of the graphs has been changed to avoid confusion.
- 2-57624461 - Creating a report from the Availability Manager page resulted in the percentage value in the report being different to the one displayed on-screen. This error occurred because the report didn't use the calendar associated with the Asset. The SLA calculation now uses the appropriate asset calendar.
- 2-61322801 - There was an error accessing "Management > General Settings" from the SmartConsole Web Client. The Security model for SmartConsole Web Client was not being created correctly. This has been fixed.
- Orchestrator failed to start if was unable to connect to SharedObjects. Orchestrator would start even if a connection to SharedObjects could not be made. It would then crash at the first attempt to access any object that needed to be loaded from SharedObjects. This has been fixed.
- The user custom templates for ThinkServer Monitors messages were lost after an upgrade due to a defect in the installer. This has been fixed.
- The content of the folders "ThinkServer\ts_shared_referentials\Business Consumers" and "ThinkServer\ts_shared_referentials\Integrations" were not maintained after an upgrade due to an issue in the installation routine. This has been fixed.
- 2-61617502 - When editing a Security Asset, switching to the "Actions to Audit" tab and enabling monitoring of the asset while the tab is still loading resulted in the monitoring request not being actioned. This issue has been fixed.
- A bug in Inspector, the Event Control Engine, caused operations-related events (including Business Processes events) to be stored out-of-sequence. This caused SmartConsole based products to behave unexpectedly as the status of a monitored asset depends on sequentially generated events. This has been fixed.
- An error in the VMWare 'out of the box' template in Alignia for Business Security caused the permissions addition to a role to remain undetected. This issue has been corrected.
- The single sign-on utility did not work on SmartConsole WebClient and SharedObjects. The product asked for credentials even when the user had previously signed-on to another Alignia Web application. This has been fixed.
- Architecture Load Balancer is not shown when editing an Application or a Connectivity Group. This was due to an issue in the Application and Connectivity Group edition and the asset was always shown as "Non Cluster".
- 2-62257181 - Attempting to change a scheduled task macro that was running resulted in a programming error message being displayed. The issue has been fixed and now the correct error message is now displayed.
- A single event can match multiple security controls. When a controlled event was opened on the Forensic Analysis page in order to view its details, the subsequent Event Details page displayed the details for the first controlled event instead of displaying the details for the event that had a higher security Impact. A code error resulted in the first event always being selected. This has been fixed.
- A Python code error on Alignia for Business Security Oracle 'out-of-the-box' template:Health has changed to Monitoring error due to some errors in the script. This has been corrected.
- Business Security. When the user changes the Security Regulations of an asset, the existing ThinkServer Monitors change to monitor "LocalHost" instead of the correct Device. Furthermore, these monitors were not deleted when Security monitoring is deactivated. The change in Security Regulation was not processed as a change in the Device. which caused issues during the creation of ThinkServer Monitors. This has been fixed.
- PostgreSQL standalone application metrics were not displayed (No data was displayed instead) as the metrics definition was not created properly. This has been fixed.
- 2-63112701 - Business Applications IBM i Job Activity Monitor was not detecting an inactive job due to an error in the active jobs count. This has been fixed.
- Events were not correctly arriving in the SmartConsole from environments with multi-byte characters. This has been fixed.
- Devices created within Business Security contained a wrong value if the Device was then renamed. Devices created in previous versions would have a wrong value assigned in the 'NetworkInterfaceUsedForMonitoring' attribute if they were later renamed because the previous value was retained. Both issues have now been fixed. In some instances, the processing of the 'NetworkInterfaceUsedForMonitoring' attribute was wrong. Additionally, possible wrong values in the database were not corrected.
- 2-62450541 - Business Security Windows Eventlog Custom collector was not filtering by subaction. The subaction filter was not being executed and therefore all the events were bypassing all of the subaction filters. This has been fixed.
- After creating a Connectivity Group Cluster, a renaming action displayed a pop up window asking if you wanted to change the asset model. A bad reference to the entity tree made the editor think that the type had changed. This has been fixed.
- Sometimes, reports based on the Custom Event List template were displaying graphs that were unreadable. This behavior was caused by the values of the X axis being too large to properly fit the display and has now been fixed.
- Errors were found when importing assets whose relationships with another assets were a subset of the relationships already defined in the Database. In these instances, the existing dependencies were used instead of being overridden by the imported ones. For example, if an exported security asset has a security regulation "A" assigned and is imported when the existing asset had regulations "A" and "B" assigned, then the existing values are used, instead of just "A". There was an issue processing asset relationships that has now been corrected.
- 2-63299204: When using Internet Explorer, the Forensic Analysis grid was unfiltered when opened. This has been fixed.
- After upgrading Alignia from pre-Business Security versions, the Event Control Service was not able to insert events in the database. When upgrading from previous versions that used a database schema version older than v7 (i.e. Pre-Alignia Business Security versions), the database model upgrade procedure failed. As the database model was not up-to-date, events could not be inserted. This has been fixed.
- 2-62137937 - There was an inconsistency of SLA percentage between Availability Manager, Dashboards and Reports. Availability Manager was calculating the SLA percentage using only the period up until the current time and Dashboards and Reports were using the full period. This has been fixed.
- In Business Processes, when a Task had a Datasource assigned (as defined in the Data Collection tab) that used a specific credential, it was not read from the configuration. When the Process was reloaded, the credential used was "Monitoring node default credential (inherited from process)". The configuration routine didn't read the credential section from the configuration file. It always used the default setting "Monitoring node default credential (inherited from process)". This has been fixed.
- It was not possible to modify the interface where the Syslog listener was attached. Sometimes the datasource was retrieving wrong values from the DeviceIP, and returning the following error: "Error in UDP syslog listener. ErrorCode: -1 Error message: Error opening UDP port. Probably the UDP port is being used by another application". A Syslog credential selection has been added for the datasource. It now works in exactly the same way as the basic Syslog ThinAgent, which takes a device and a credential. Selecting the proper listening interface, port and protocol in the credential now resolves the above issue. For existing installations; apply this patch, and then assign a proper Syslog credential.
- A timeout error in Business Process Configurator existed when resetting a Business Process. In the PMDB log file, an error was logged containing the message "An expression of non-boolean type specified in a context where a condition is expected, near ')'.". The SQL sentence used to delete records was incorrect and furthermore, the Orchestrator code continued to wait for the operation to end, even when the error had been already reported. This has been fixed.