Contents
Spotfire® Data Streams 11.1.1 adds the following updates and new features:
- Spotfire Streaming 11.1 is an LTS Release
-
Spotfire Streaming 11.1 is retroactively designated as an LTS release. It is also clarified that 11.1.0 is an LTS release.
- Spotfire® Streaming Reorganizes Windows Start Menu Items
-
On Windows, the Spotfire Streaming installation places all Start Menu items under
. All item names include the release version, so that searching the Start Menu will find them more easily in side-by-side install situations. - LVCONF Editor Supports Auto-increment Primary Key Fields
-
You can set auto-increment for LONG type primary key fields in the lvconf Primary Indexes tab. Simply go to the lvconf Primary Indexes tab, enable the auto-increment by selecting the relevant field and clicking the "++" button in the toolbar. Auto-incrementing Primary Indexes are given a "++" overlay in the Indexes tab and the Schema tab.
- Spotfire® Streaming Now Bundles Java 17.0.12
-
The Spotfire Streaming installation kit provides a copy of Eclipse Adoptium JDK 17.0.12 in a subdirectory of the Streaming installation directory, to run Streaming program fragments and applications from within StreamBase Studio and other Streaming development tools. Streaming detects and prefers a system-installed JDK, and uses this bundled JDK as a fallback. See further details about the JDK requirements of this release in the JDK entry of the Supported Configurations page.
- Spotfire Streaming Supports Red Hat Enterprise Linux Server 9
-
Red Hat Enterprise Linux Server 9 is now a supported operating system for deployment.
- Spotfire Streaming Supports Debian Linux 12
-
Debian Linux version 12 is now a supported operating system for deployment.
- Spotfire Streaming Bundles Maven 3.9.8
-
Spotfire Streaming 11.1.1 includes a standalone installation of Maven 3.9.8 for customer command-line use as described in the Maven entry of the Supported Configurations page.
- New Property added in Decision Table Operator
-
A new property Exclude Input Tuple From Output Tuple is added in Decision Table operator. When checked, the Decision Table operator does not add an in output field with the same schema as the input tuple, and does not copy any input to the output.
This release includes the following changes in functionality:
- The Default Artifacts in ADS are Local to the EventFlow Engine that Defined the Default Artifact
-
In the Artifact Distribution Service, default artifacts are now local to the EventFlow engine that defined the default artifact. Default artifacts are not added to the cluster-wide artifact cache and cannot be managed with the
activate artifact
,deactivate artifact
,register artifact
, andremove artifact
commands. - Endpoints do not Output Stack Traces by Default
-
For security reasons, endpoints no longer output stack traces by default. If requested by Support, customers must set
liveview.suppress.error.detail=false
. - Files Containing EventFlow Tests do not Support Running as JUnit Test
-
EventFlow tests must be executed using Run As>EventFlow Fragment Unit Test. Normal Junit tests can still be executed using Run As>JUnit Test.
The following features are deprecated as of Spotfire Streaming 11.1.1.
- Spotfire Streaming No Longer Bundles TERR on macOS
-
Due to lack of support for newer versions of macOS, TIBCO Data Streams no longer bundles TERR on macOS.
- Spotfire Streaming TERR Operator for Spotfire® Enterprise Runtime for R (a/k/a TERR™) is now Deprecated
-
The Spotfire Streaming TERR operator is deprecated and will be removed in future versions.
- Spotfire Streaming TERR Predict Operator for Spotfire® Enterprise Runtime for R (a/k/a TERR™) is now Deprecated
-
The Spotfire Streaming TERR Predict operator is deprecated and will be removed in future versions.
This section describes known limitations as of the current release of Spotfire Streaming. Each item includes a tracking number, description, and workarounds when possible.
Known limitations for using the EventFlow debugger are described separately, in the Limitations, Tips, and Tricks section of the Using the EventFlow Debugger page.
This section describes the known limitations in the current release of Spotfire LiveView™. Each item includes a tracking number, description, and whenever possible, one or more workarounds.
ClusterAlert | Cluster-wide Alert Group Restriction for Heterogeneous Clusters | |
Description | Nodes with only EventFlow engines in LiveView clusters must be in their own alert groups. | |
Workaround | For more information, see Cluster-Wide Alerting guidelines | |
SB-54280 | Deadlock when nodes start simultaneously. | |
Description | Starting nodes simultaneously may lead to a race condition and subsequent deadlock during start-up. | |
Workaround | Stagger the launch of each node in a multi-node cluster. | |
SB‑46418 |
Removing the lvweb.war dependency from a
project's POM does not remove the file, and vice versa.
|
|
Description |
New LiveView Fragment projects are created by default with a dependency on
the lvweb.war file that implements LiveView Web
functionality. There are two cases: (1) If you later edit the project's
pom.xml file to remove that <dependency> element, in which doing so does not
automatically remove the lvweb.war file from
the project's src/main/liveview/lv-user-webapps
folder. (2) If you remove the lvweb.war file
manually, the pom.xml is not automatically
edited to remove the dependency.
|
|
Workaround |
In both cases, you must both remove the lvweb.war file and edit the pom.xml file to remove a project's support for LiveView
Web.
The correct way to create a new
LiveView project without LiveView Web support is to change the |
|
SB‑45720 | Failure to independently run or debug EventFlow modules in LiveView projects in 10.4.0 through 10.4.2. | |
Description | Many LiveView projects have StreamBase EventFlow modules that provide functionality such as publishers or transformers of data. In releases 10.4.0, 10.4.1, and 10.4.2, right-clicking such modules and selecting | > (or ) to run or debug the module independent of the enclosing LiveView project resulted in an error. This feature has been restored in release 10.4.3. This limitation did not prevent running the project as a LiveView fragment with all EventFlow modules working. The limitation was only on running the EventFlow modules by themselves.|
Workaround |
If you are using one of the affected releases, you can:
|
|
SB‑44494 | Metadata store restriction for clusters containing LiveView and StreamBase fragments. | |
Description |
The transactional memory metadata store type is only supported on homogeneous LiveView clusters. This means the
cluster cannot contain a mix of LiveView and StreamBase fragments when using
a transactional memory metadata store.
|
|
Workaround | Use either the JDBC or H2 metadata store type when your cluster contains both LiveView and StreamBase fragments. | |
CQS-5503 | Controlling the output for aggregation queries against no rows. | |
Description |
As documented, aggregate queries that execute against no result return
null (to indicate unknown ), except for counting functions (that
return 0). If you wish to instead return no row at all, see the workaround.
|
|
Workaround | Append GROUP BY 1 to the query. | |
CQS-4800 | Row level security authentication not performed on results of already-registered queries and query-based alerts. | |
Description | Row level security (RLS) authentication and authorization is performed at the time when new query-based alerts and queries are registered with the LiveView server. LiveView does not perform authentication on the results of already-registered queries and query-based alerts; to get the expected results based on RLS authentication, you must re-register the queries with the server. | |
Workaround | None. | |
CQS-4748 |
liveview.project.out system property is not used
for deployed applications.
|
|
Description |
The liveview.project.out system property was
recommended in previous releases as a way to manage the placement of the
LiveView server's compilation working directory. However, if you redirect
this working directory in StreamBase 10 applications, the StreamBase
application might fail to start correctly.
|
|
Workaround | None. | |
CQS-4641 | Issuing queries with literal non-printing characters is not supported in query strings. | |
Description | Query strings with literal non-printing characters can cause issues. Non-printing characters are those below 0x20, excepting tab, carriage return, and line feed. The tables and query result sets are free to have any characters as values. | |
Workaround |
Instead use StreamBase Unicode escape sequences, which looks like
\uXXXX . For example, a bell is \u0007 .
|
|
CQS-3855 |
Remote data source tables with sole primary key CQSInternalID cannot be opened.
|
|
Description |
When querying a remote data source's tables, all LiveView clients fail to
open any remote table that uses a single primary key named CQSInternalID.
This issue affects all LiveView clients, including Spotfire LiveView Desktop, Spotfire LiveView Web, the lv-client command line utility, and any custom client written using LiveView Client APIs. The issue occurs when querying any remote data source, including: Spotfire StreamBase Query Tables, other LiveView servers, JDBC databases, Spotfire ActiveSpaces tables, and custom table providers. |
|
Workaround |
To enumerate the remote tables that you must avoid, use one of the following
suggestions:
|
This section provides a list of errors corrected in release 11.1.1 of Spotfire® Data Streams.
Fixed in 11.1.1 | |
---|---|
Number | Resolution |
Spotfire Streaming 11.1.1 incorporates all fixes resolved in the release 10.0 series through 11.1.0. | |
SB-53943 |
Backward compatible conversion of 0/1 to a JSON boolean is supported. JSON boolean type supports true/false,T/F, or 0/1 values. |
SB-53829 | The Clojure dependency with a security vulnerability was removed from the release. This eliminates the security vulnerability caused by this dependency. |
SB-53783 | Windows StreamBase Command Prompt fails when Streaming is installed under the Program Files folder. |
SB-52708 | Cluster aware operators that were configured to use the Active on a single node in the cluster or Active on multiple nodes in the cluster were incorrectly started, and then immediately stopped, on nodes on which they should not have been started. This behavior has been fixed so that operators are only started on nodes on which they should be active. |
SB-50314 | JMS editor reconnect Unlimited setting results in no reconnect attempts. |
SB-42083 | Loading Excel files with the epadmin load artifact command were previously corrupted during the load. This has been resolved and Excel files can now be loaded without corruption using the epadmin load artifact command. |