Known issues and limitations

Learn about the known issues in Flink and SQL Stream Builder, the impact or changes to the functionality, and the workaround in Cloudera Streaming Analytics 1.6.2.

SQL Stream Builder🔗

FLINK-18027: ROW value constructor cannot deal with complex expressions: When querying data from a table or a view with a ROW() function an exception is thrown due to a Calcite parsing issue. For example, the following query will return an error:
CREATE VIEW example AS SELECT col1, ROW(col2) FROM table; SELECT * FROM example;; Add a second SELECT layer to the SQL query as shown in the following example:
CREATE VIEW example AS SELECT col1, ROW(col2) FROM (SELECT col1, col2 FROM table); SELECT * FROM example;
Cannot access API Explorer: The API Explorer page of SSB REST API cannot be accessed when using Apache Knox as authentication method. This issue is not present when using SPNEGO authentication.; None
Uploading connector files fail: When trying to upload a new connector JAR with a size file more than 1 MB, the upload process fails with an error.; Set the server.tomcat.max-swallow-size in Cloudera Manager using the following steps:

Open your cluster in Cloudera Manager.

Select SQL Stream Builder from the list of services.

Select Configuration.

Search for Streaming SQL Engine Advanced Configuration Snippet (Safety Valve) for ssb-conf/application.properties in the search bar.

Add server.tomcat.max-swallow-size=2000MB to the Safety Valve.

Click Save.

Restart the SQL Stream Builder service.
CSA-3537: Catalogs are deleted after Streaming SQL Engine restart: When the Streaming SQL Engine instance is restarted on the cluster, the already registered catalogs are removed.; None
CSA-3464: Overwriting new changes for SQL Job with edit: A previous version of a job setting is applied when the SQL query is changed for an existing job after reloading it to the Compose page. The issue can be experienced when clicking on the Edit Job button on the SQL Jobs page.; None
CSA-3308: Changing Primary Key for Materialized View without recreating table causes job failure: The SQL job fails when directly changing the primary key of Materialized View and restarting the SQL job due to schema mismatch. The SQL job silently keeps retrying and overloads the Postgres database when the primary key of a Materialized View got changed without deleting existing data.; None
CSA-2729: DLQ topic is filled with unexpected results: Some messages that are left in the buffer can be sent to the Dead Letter Queue (DLQ) topic after handling a deserialization failure for Kafka tables.; None
CSA-2016: Deleting table from other teams: There is a limitation when using the Streaming SQL Console for deleting tables. It is not possible to delete a table that belongs to another team using the Delete button on the User Interface.; Use DROP TABLE statement from the SQL window.
CSA-1454: Timezone settings can cause unexpected behavior in Kafka tables: You must consider the timezone settings of your environment when using timestamps in a Kafka table as it can affect the results of your query. When the timestamp in a query is identified with from_unixtime, it returns the results based on the timezone of the system. If the timezone is not set in UTC+0, the timestamp of the query results will shift in time and will not be correct.; Change your local timezone settings to UTC+0.
CSA-1231: Big numbers are incorrectly represented on the Streaming SQL Console UI: The issue impacts the following scenarios in Streaming SQL Console:

When having integers bigger than 2^53-1 among your values, the Input transformations and User Defined Functions are considered unsafe and produce incorrect results as these numbers will lose precision during parsing.

When having integers bigger than 2^53-1 among your values, sampling to the Streaming SQL Console UI produces incorrect results as these numbers will lose precision during parsing.; None

Flink🔗

FLINK-18027: ROW value constructor cannot deal with complex expressions: When querying data from a table or a view with a ROW() function an exception is thrown due to a Calcite parsing issue. For example, the following query will return an error:
CREATE VIEW example AS SELECT col1, ROW(col2) FROM table; SELECT * FROM example;; Add a second SELECT layer to the SQL query as shown in the following example:
CREATE VIEW example AS SELECT col1, ROW(col2) FROM (SELECT col1, col2 FROM table); SELECT * FROM example;
FLINK-27441: Scrollbar is missing for particular UI elements on Flink Dashboard: The angular version bump introduced a bug that causes the scrollbar on the Flink Dashboard invisible.

In Cloudera Streaming Analytics, the following SQL API features are in preview:

Match recognize
Top-N
Stream-Table join (without rowtime input)

DataStream conversion limitations

Converting between Tables and POJO DataStreams is currently not supported in CSA.
Object arrays are not supported for Tuple conversion.
The java.time class conversions for Tuple DataStreams are only supported by using explicit TypeInformation: LegacyInstantTypeInfo, LocalTimeTypeInfo.getInfoFor(LocalDate/LocalDateTime/LocalTime.class).
Only java.sql.Timestamp is supported for rowtime conversion, java.time.LocalDateTime is not supported.

Kudu catalog limitations

CREATE TABLE
- Primary keys can only be set by the kudu.primary-key-columns property. Using the PRIMARY KEY constraint is not yet possible.
- Range partitioning is not supported.
When getting a table through the catalog, NOT NULL and PRIMARY KEY constraints are ignored. All columns are described as being nullable, and not being primary keys.
Kudu tables cannot be altered through the catalog other than simply renaming them.

Schema Registry catalog limitations

Currently, the Schema Registry catalog / format only supports reading messages with the latest enabled schema for any given Kafka topic at the time when the SQL query was compiled.
No time-column and watermark support for Registry tables.
No CREATE TABLE support. Schemas have to be registered directly in the SchemaRegistry to be accessible through the catalog.
The catalog is read-only. It does not support table deletions or modifications.
By default, it is assumed that Kafka message values contain the schema id as a prefix, because this is the default behaviour for the SchemaRegistry Kafka producer format. To consume messages with schema written in the header, the following property must be set for the Registry client: store.schema.version.id.in.header: true.