Known Issues

The following issues are known, but were not yet fixed in this release.

KPI formats may change after upgrade

If you installed Incorta version 4.5, 4.5 Maintenance Pack 1, or 4.5 Maintenance Pack 2, you may encounter an issue with KPIs with default formats changing to another format when you upgrade to 4.5 Maintenance Pack 3 or higher. For example, you may see a KPI with a percent rounded format changed to percent format. To fix this, you must manually change the formats.

Bookmarks for dashboards with presentation variables

If you create a bookmark on a dashboard that has a presentation variable, the bookmark may not apply correctly unless you delete the bookmark, refresh the dashboard, then recreate the bookmark.

Retrieve favorites

There was an issue where dashboard favorites were lost when upgrading from 4.3.x to 4.5.2. With 4.5.3, you can export favorites from a prior version and import it into 4.5.3 as a manual step.

Below are the steps needed to export the favorites from Incorta release 4.3.x to Incorta release 4.5.3:

  1. Install Incorta release 4.5.3.
  2. Find the tmt.jar file in the <INSTALLATION_PATH>/cmc/tmt folder.
  3. Copy tmt.jar from the Incorta release 4.5.3 <INSTALLATION_PATH>/cmc/tmt folder and replace the tmt.jar file in the <INSTALLATION_PATH>/cmc/tmt folder in the Incorta release 4.3.x installation.
  4. Restart all analytics and loaders service for Incorta release 4.3.x.
  5. In a terminal, navigate to <INSTALLATION_PATH>/cmc/tmt.
  6. Run the following command to generate a file called <TENANT_NAME>_favorite_dashboards.zip: ef,-export-all-favorite-dashboards <TENANT> ./tmt.sh -clnm <CLUSTER_NAME> -ef <TENANT_NAME>
  7. Export the tenant from Incorta release 4.3.x.
  8. Import the tenant for Incorta release 4.3.x into the Incorta release 4.5.3 instance.
  9. In a terminal, navigate to <INSTALLATION_PATH>/cmc/tmt.
  10. Run the following command to import the demofavoritedashboards.zip: ./tmt.sh -clnm <CLUSTER_NAME> -if <TENANT_NAME> <TENANT_NAME>_favorite_dashboards.zip.
  11. Sign in to the tenant to verify the favorite dashboards display as expected.

Schemas with dependencies load sequentially

More than one schema can load at a time, but if there is a dependency between any two schemas, the schemas will run sequentially in a queue and not in parallel.

Load filter in Arabic

You cannot create a load filter in Arabic.

Materialized views in English

Sometimes creating a materialized view table with a non-English schema and table name causes an error. You can only create materialized views in SQL for schemas or tables with English words.

Multiple Data Sources Causing Overrides

Starting from version 4.3, data from more than once source can cause an issue with overriding data. When you load records from Data Source 1, then load records from Data Source 2, with the same keys, data from Data Source 2 overrides the records with the same key as Data Source 1.

Replace Spark

You must replace Spark with the new version bundled with Incorta and reapply any custom configurations you set to upgrade to Incorta release 4.5.

Update JDBC driver to use SQL Server 2008

To use SQL Server 2008 with OpenJDK 11, you must update the JDBC driver to version 7.4.

Deleting a schema causes Incorta to freeze

Replicate this error with the following steps:

  1. Create two or more schemas.
  2. Create overlapping scheduled jobs.
  3. Select two schemas to delete, including one schema with a scheduled job.

Expected behavior: The schemas are deleted and no exceptions happen.

Actual behavior: The schemas are deleted and Incorta freezes.

Troubleshooting: Materialized view does not read all columns

A materialized view may not discover and read all columns when it is built on top of a table that is sourced by using Incorta as a data source.

To solve this issue: Add the following property inside the spark-defaults.conf file and set it to true: spark.sql.parquet.mergeSchema.

Troubleshooting: Old Connections Do Not Close As Expected, Cluster Won’t Restart

Issue: The number of connections increases when you open new connections without closing the old connections. This can cause the number of connections to reach the maximum number set for MySQL and prevent the cluster from starting.

Resolution: Upgrade the MySQL connector driver to 5.1.48 or later.

Troubleshooting: Incremental option toggle

If you toggle an incremental option before you add a query, a list of new incremental columns do not update and display until you save and close the table, then re-open it. Replicate this error with the following steps:

  1. Edit a schema.
  2. Select +NEW to create an SQL Database table.
  3. Select a datasource.
  4. In the Data Source modal, toggle Incremental to on.
  5. In Incremental Extract Using: select Last Successful Extract Time.
  6. In the Query box, add a query.
  7. The Incremental Extract Using: dropdown does not change.

Workaround:

  1. Select Add, then Yes to save and close the modal.
  2. Open the SQL table again to see options in the Incremental Extract Using: dropdown.

Troubleshooting: Formula Joins

Incorta allows joins only on a formula column that are referencing fields in the same table.

Replicate this error with the following steps:

  1. Create a schema.
  2. Create a table in the schema, for example, Table 1.
  3. Create another schema that contains another table, for example Table 2.
  4. Create a formula column in Table 1.
  5. Join the formula column of Table 1 to key column of Table 2. Table 2 is now the parent table and Table 1 is now the child table.
  6. Load both schemas.
  7. Create a new dashboard to observe the query plan.
  8. Load the first schema only.
  9. The join breaks in the query plan.

To solve this issue:

  1. Create and load a materialized view with the formula column in Table 1 and the key of Table 1.
  2. Break the join between the formula column of Table 1 to the key column of Table 2.
  3. Create a join from Table 1 to the materialized view on the key column.
  4. Create a join from the formula column in the materialized view to the key of Table 2.

© Incorta, Inc. All Rights Reserved.