MicroStrategy ONE
Starting in MicroStrategy 2021 Update 4, Hadoop Gateway is no longer supported.
Known Issues
Symptom
In MicroStrategy Web, users have an expansion in the variety of data sources to which they can connect. These new options are under Add External Data. A user with Web Reporter, Web Analyst, and Web Professional privileges will receive a "You do not have enough privileges to perform this operation" error when trying to import a table from a data source. This is due to the fact that the user privilege "Access data from Databases, Google BigQuery, BigData, OLAP, BI tools" is not the only privilege needed. Creating a data source object to import Tables is considered an Administrator Privilege.
In the Administration group, give users the Create and edit database and instances and connections privilege.
The "Create and edit database and instances and connections" is a Server level only privilege and can only be granted only at the project source level, so it needs to be given at a user or user group level.
Symptom
There are persisting tables in the Hive Metastore when the MicroStrategy Hadoop Gateway service starts. In the log file you can see that the MicroStrategy Hadoop Gateway service is trying to modify the structure of some Metastore tables (such as DBS tables). When it tries to commit the transaction, an error is produced and a rollback operation is generated.
MicroStrategy Hadoop Gateway requires a Hive component to load ORC files, and basic metadata needs to be created to establish the Hive component. Modifying the structure of Metastore tables is not permitted by default.
To enable Metastore modification, add the following section to /etc/hive/conf/hive-site.xml.
<property>
<name>datanucleus.autoCreateSchema</name>
<value>true</value>
</property>Alternatively, the Hive component can be disabled if you are not going to import ORC files. Set the property hgos.spark.enable value to false in the MicroStrategy Hadoop Gateway configuration file<MicroStrategy Hadoop Gateway installation path>/conf/hgos-spark.properties.
Symptom
Text columns in the parquet file created by Impala are loaded as binaries.
The Impala Parquet writer always creates an unannotated binary field for string columns. It is possible to specify that the field include the UTF-8 annotation to indicate UTF-8 string data. It was resolved in Impala 2.6 and CDH 5.11 has 2.8.
If all these columns need to be converted to UTF-8 strings, add spark.sql.parquet.binaryAsString=true into configuration file <MicroStrategy Hadoop Gateway installation path>/conf/hgos-spark.properties.
Related Topics
Introduction to the MicroStrategy Hadoop Gateway
How to Deploy the MicroStrategy Hadoop Gateway
How to Start the MicroStrategy Hadoop Gateway