General Main Release 10.3.0 CU5

Note

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

Tip

For release notes related to DataMiner Cube, see DataMiner Cube Main Release 10.3.0 CU5.
For release notes related to the DataMiner web applications, see DataMiner web apps Main Release 10.3.0 CU5.
For information on how to upgrade DataMiner, see Upgrading a DataMiner Agent.

Enhancements

Security enhancements [ID 36294]

A number of security enhancements have been made.

Service & Resource Management: Enhanced performance of GetEligibleResources call [ID 36430]

Because of a number of enhancements, overall performance of the GetEligibleResources call has increased.

DataMiner Agents joining a cluster will now synchronize their ProtocolScripts\DllImport folder [ID 36494]

When a DataMiner Agent joins a cluster, it will now synchronize its ProtocolScripts\DllImport folder.

Also, when processing a protocol, a DataMiner Agent will now synchronize

the files in the ProtocolScripts/DllImport folder, and
the files in the folders mentioned in the *QAction@dllImport* attribute.

Cassandra & Amazon Keyspaces: 'analytics_changepointalarmentries_v2' table renamed to 'ai_cpalarms' [ID 36503]

In a Cassandra Cluster and an Amazon Keyspaces database, the analytics_changepointalarmentries_v2 table has now been renamed to ai_cpalarms.

As this new table name is quite a bit shorter, for both types of databases, keyspace prefixes can now have a maximum length of 20 characters instead of 11 characters.

Stream Viewer will now display parameter IDs in decimal format instead of octal format [ID 36525]

Stream Viewer will display an error message when an SNMP poll group contained either an invalid parameter or a parameter that did not have its SNMP setting enabled.

Up to now, that error message would contain the ID of the parameter in octal format. From now on, it will contain the ID of the parameter in decimal format instead.

Factory reset tool will no longer try to delete non-existing folders [ID 36550]

Up to now, the factory reset tool SLReset.exe would log an exception each time it had tried to delete a non-existing folder. From now on, when it has to delete a folder, it will first check whether that folder exists. If not, it will not try to delete it.

SNMP tables: Columns of type 'retrieved' can now be placed in between columns of type 'snmp' [ID 36559]

Up to now, when an SNMP table had columns of type "retrieved" in between columns of type "snmp", problems could occur. All columns of type "retrieved" had to be grouped and placed at the right of the columns of type "snmp".

From now on, in an SNMP table, columns of type "retrieved" can be placed in between columns of type "snmp", providing the primary key column is a column of type "snmp" and not a column of type "retrieved".

Service & Resource Management: Enhanced performance [ID 36568]

Because of a number of enhancements with regard to fetching LinkerTableEntries of function resources, overall performance has increased.

SLAnalytics - Behavioral anomaly detection: Enhanced anomaly labelling for periodically returning behavioral changes [ID 36664]

A behavioral change in the trend data of a parameter is considered an anomaly if there have not been similar behavioral changes that occurred regularly or frequently in the historical behavior of the parameter. This anomaly labelling has now been enhanced for periodically returning behavioral changes.

Smart baselines: Information event generation at 5-minute intervals has been disabled [ID 36691]

When smart baselines were configured, by default information events would be generated every 5 minutes. This information event generation has now been disabled to avoid information event floods in e.g. EPM environments.

Fixes

In some rare cases, the following NATS-related error would be thrown:

Failed to copy credentials from [IP address] - corrupt zip file

SLAnalytics: Incorrect trend predictions in case of incorrect data ranges set in the protocol [ID 36521]

If, in the protocol, a data range is specified for a parameters for which trend data prediction is required, the trend prediction algorithm will cap the prediction values to the data range. For example, if a parameter has a rangeLow value equal to 0 and a rangeHigh value equal to 100, the prediction will not contain values lower than 0 or higher than 100.

From now on, if the trend data contains values outside of the specified data range, the trend prediction algorithm will no longer consider the data range values to be valid or reliable, and will not limit the prediction to this range.

Problem with protocol.SendToDisplay API call [ID 36528]

When the following protocol API call was used to update specific matrix crosspoints, in some cases, the API call could ignore the physical size of the matrix. Also, the API call could change the dimensions of future ParameterChangeEventMessages.

protocol.SendToDisplay(matrixReadParameterId, changedInputs, changedOutputs);

Problem when requesting alarms on a system with Cassandra Cluster and Elasticsearch [ID 36549]

On systems with a Cassandra Cluster and an Elasticsearch database, the following issues could occur:

When alarms were requested via a query with a service filter, no alarms would be returned.
When alarms were requested via a query with a view filter, no alarms would be returned when that view or any of its subviews contained services. Also, when a view was enhanced with an element, that element would not be queried.

Problem with SLElement due to timeout actions of an element being overwritten [ID 36591]

In some rare cases, an error could occur in SLElement when a timeout action of an element with multiple connections would overwrite another timeout action of the same element.

SLAnalytics - Behavioral anomaly detection: False change point could be generated before a gap in a trend graph [ID 36605]

When there was a gap in a trend graph that showed a perfectly increasing line, in some cases, a false change point could be generated right before that gap.

SLAnalytics - Automatic incident tracking: Attempt to clear an alarm group that had already been cleared [ID 36654]

In some rare cases, the system would incorrectly try to clear an alarm group that had already been cleared.

NATSMaxPayloadException could be thrown when a client requested large amounts of data [ID 36655]

When a client requested large amounts of data, in some cases, a NATSMaxPayloadException could be thrown.

Cassandra Cluster: DVE properties would be cleared when an update was sent to the database [ID 36658]

DVE properties would be cleared each time an update was sent to a database of type Cassandra Cluster.

SLAnalytics would shut down immediately when it failed to get information from SLNet at start-up [ID 36692]

When starting up, SLAnalytics requests information from SLNet. Up to now, when it failed to get that information, it would immediately shut down. From now on, when the first request fails, it will try again once before shutting down.

Table of Contents