Table of Contents

General Main Release 10.6.0 CU3

Note

For known issues with this version, refer to Known issues.

Tip

Prerequisites

Before you upgrade to this DataMiner version:

Changes

Breaking changes

Protocols: As many SLScripting processes as SLProtocol processes by default [ID 44420]

Up to now, one SLScripting process was used by default. From now on, by default, there will be as many SLScripting processes as SLProtocol processes.

Note that is possible to configure the number of simultaneously running SLScripting processes. See Setting the number of simultaneously running SLScripting processes.

Important

If you are using multiple SLScripting processes, it is important that elements running the same protocol are not sharing/exchanging data with each other through static fields. More information can be found in the QAction documentation.

Enhancements

Protocols: Elements will now restart automatically when an SLScripting process has disappeared [ID 42306]

Up to now, when an SLScripting process disappeared, elements relying on that process could become unstable, requiring manual intervention to restore functionality.

From now on, when an SLScripting process disappears, a new process instance will be started automatically, and any elements that depended on the process that disappeared will be restarted to maintain consistency across SLProtocol, SLScripting, and other related components. This will ensure that lost SLScripting data is properly reinitialized and remains in sync with other processes.

When an SLScripting process disappears, the following notice alarm will be generated:

Process disappearance of SLScripting.exe with PID <processId>; <x> elements affected by the disappearance have been restarted.

Also, the SLElementsInProtocol.txt log file has been updated to track restart reasons more accurately.

  • The restart reason column will now indicate either "SLScriptingCrashRestart" or "SLProtocolCrashRestart" (if everything is OK, NormalStart will be shown instead).
  • A new counter will now indicate the number of times the element was started due to a SLScripting process disappearance.

If SLProtocol requests an SLScripting process that is no longer valid, the system will now detect this, and trigger the same element restart flow.

Note

There will be a one-minute delay between the disappearance of an SLScripting process and the creation of a new SLScripting process and the subsequent element restarts. However, when one of the elements that was hosted in the SLScripting process that disappeared tries to trigger a QAction within that one-minute delay, the new SLScripting process will be created when that QAction is triggered.

Enhanced performance when retrieving service state information [ID 44392]

Because of a number of enhancements, performance has increased when retrieving service state information.

From now on, a GetServiceStateMessage will no longer be forwarded to the Agent hosting the service. Instead, the service state information will be retrieved from the local SLNet cache.

STaaS: Enhanced performance when closing alarm trees [ID 45081]

Because of a number of enhancements, on STaaS systems, overall performance has increased when closing alarm trees.

BrokerGateway installer will now give a clear indication when .NET is missing [ID 45169]

When you install the BrokerGateway DxM on a server that does not have the Microsoft .NET hosting bundle installed yet, from now on, a message will appear, saying that .NET has to be installed first.

APIGateway: Enhanced handling of files in use during upgrades [ID 45230]

From now on, the Windows Restart Manager will be enabled during installations. This will prevent unnecessary delays caused by files still in use by the APIGateway instance that is being upgraded, increasing overall performance and reliability of the upgrade process.

Aliases can now be configured for DaaS systems [ID 45327]

If you want to configure an alias for a DataMiner Agent, in the DataMiner.xml file, the mode attribute of the DMAName element has to be set to "manual". If this attribute is not set to "manual", configuring an alias will also change the computer name, which will cause unintended behavior.

As it is not possible for a user to make changes to the DataMiner.xml file of a DaaS system, up to now, it would not be possible to configure an alias for such a system. From now on, in the DataMiner.xml file of a DaaS system, the mode attribute of the DMAName element will be set to "manual" by default. This will allow users to also configure aliases for DaaS systems.

SLLogCollector packages now include hosts and lmhosts files [ID 45369]

From now on, SLLogCollector packages will also include the following files:

File Description
C:\Windows\System32\drivers\etc\hosts Maps hostnames to IP addresses for DNS name resolution. Used to override or supplement DNS locally.
C:\Windows\System32\drivers\etc\lmhosts Maps NetBIOS names to IP addresses for legacy Windows networking. Rarely used in modern systems.

Both files will be added to the Network Information folder.

Fixes

BPA tests could incorrectly not be run on DMAs that were not connected to the internet [ID 45040]

Up to now, it would not be possible to run BPA tests on DataMiner Agents that were not connected to the internet because they were not able to fetch the necessary certificates.

From now on, it will be possible to run BPA tests on DataMiner Agents that are not connected to the internet.

STaaS: Problem when retrieving data to be visualized in 'State timeline' components or element and parameter heatlines [ID 45043]

When, on systems using STaaS, the GetReportTimelineDataMessage was used to retrieve data to be visualized in State timeline components or element and parameter heatlines, in some cases, the data that was visualized would not be correct.

Dynamic units: A unit would incorrectly be assigned to date/time parameters [ID 45047]

When the dynamic units feature is enabled in, for example, DataMiner Cube, many numerical values are automatically converted into more readable formats with appropriate units. Up to now, the dynamic units feature would incorrectly also try to assign a unit to parameters of type date/time.

Alarm squashing: Alarm would incorrectly not show up in the Alarm Console when the element had been restarted and the AlarmsPerParameter limit was exceeded [ID 45063]

When, on a system with alarm squashing enabled, the AlarmsPerParameter limit in the AlarmSettings section of the MaintenanceSettings.xml file was exceeded, an alarm of which all alarms in the alarm tree were squashable would incorrectly not show up in the Alarm Console after the element associated with the alarm had been restarted.

SLWatchdog incorrectly interpreted processes being stopped during a DataMiner shutdown as a crash [ID 45115]

When DataMiner is shutting down, processes are expected to be stopped. However, up to now, the tracking in SLWatchdog would not be cleaned correctly, causing certain processes to still be monitored. When these processes eventually stopped, SLWatchdog would incorrectly interpret this as a crash, which would then trigger a number of crash detection features, such as notifying CDMR or attempting to perform a full restart.

SLAnalytics: Problem due to a large number of state icon events being added to the SLNet queue simultaneously [ID 45145]

In some cases, timeouts could occur due to a large number of state icon events being added to the SLNet queue simultaneously.

From now on, the number of state icon events sent by SLAnalytics instances will be reduced to a maximum of 10,000 events per minute. Also, they will be sent in batches of maximum 1000 events.

Additionally, state icon events, alarm focus events and RAD parameter aggregation events will now all have a smaller weight, meaning that more of them will be allowed in the queue.

Problem with alarm forwarding after an element had been edited [ID 45167]

When an element had been edited, in some cases, alarm updates would incorrectly no longer be forwarded to northbound SNMP managers.

Note

This fix will not work if the element in question has the Enable SNMP agent option enabled, and has a virtual IP address and a subnet mask configured.

Up to now, during NATS migrations or NATS-related BPA test runs, establishing connections to NATS servers of other DataMiner Agents in the DMS would be slow due to reverse DNS lookups.

SLNet: Problems with user picture requests would cause other messages to get blocked [ID 45210]

When SLNet was not able to respond to user picture requests from a web app, up to now, those requests could stay active within SLNet for several minutes, causing other messages to get blocked.

From now on, user picture requests will time out after 10 seconds.

Problem when a connector was using the makeCommandByProtocol option while in slow poll mode [ID 45217]

Up to now, a problem could occur when a connector was using the makeCommandByProtocol option while in slow poll mode. The slow poll timer triggered the ping pair command as soon as the slow poll group was added to the execution queue. However, this caused a problem when regular groups were still being processed. If a valid response arrived at that moment, the system attempted to stop the slow poll timer. This failed because the timer was waiting for the response to release its resources, which led to an issue.

From now on, the slow poll timer will no longer make the ping command when adding the slow poll group to the queue. The ping command will be made when the group is executed, and the makeCommandByProtocol setting will be disregarded.

SLAnalytics: Problem due to incorrect internal state in Automatic Incident Tracking [ID 45220]

In some cases, SLAnalytics could stop working due to an incorrect internal state in Automatic Incident Tracking.

SLDataMiner: Problem when trying to fetch information about loopback adapters [ID 45285]

When the SLDataMiner process starts, it tries to fetch information about all network adapters. However, up to now, when it tried to fetch information about a loopback adapter, an error resembling the following one would be thrown:

CIPSettings::Init|ERR|-1|Opening device failed for adapter {14763620-5D53-11EA-90D5-806E6F6E6963} (Software Loopback Interface 1): The system cannot find the file specified. (2)

As loopback adapters are not part of any communication flow, from now on, SLDataMiner will no longer try to fetch information about those adapters.

DataMiner upgrade: Incorrect .NET 6 error would be thrown during the upgrade process [ID 45374]

Up to now, the first time a newly installed DataMiner Agent was upgraded, an incorrect .NET 6 error would be thrown during the upgrade process.

Improve this page
Advanced search syntax