When you deploy exported manifests with the kubectl apply command, existing resources in the target namespace are silently overwritten. This is standard kubectl apply behavior and not specific to the mta-ops command-line interface (CLI). As a consequence, you might accidentally overwrite existing workloads on the target cluster.

To work around this problem, manually verify that the target namespace does not have conflicting resources before deployment.

MTA-7103

The mta-ops transform command does not currently support the --overwrite flag. If a custom stage directory already exists, the pipeline stops. As a consequence, you cannot automatically overwrite existing custom modifications during a pipeline re-run.

To work around this problem, manually delete the existing transform or custom stage directory before you re-run the mta-ops transform command.

MTA-7234

After authenticating to the MTA web console as a Keycloak administrator, you cannot perform user management tasks because these tasks require new scopes for the user, role, and tokens resources in MTA. A local user with admin role can create and manage users.

To work around this problem, add the following scopes for administrator users in Keycloak:

After you configure an external identity provider (IdP) for authentication, MTA does not automatically redirect to the IdP login page. As a consequence, you must click the Sign in with <identity_provider> option on the MTA login page to access the IdP login page.

To work around this problem, set the idp_primary field to true in the Tackle custom resource (CR). The MTA Hub identifies the external IdP as the primary authentication method and automatically routes to its login page for authentication.

MTA-7236

When you use a wildcard to import packages used by an annotation in your Java source code, an MTA analysis does not trigger a violation for such annotations. As a consequence, you are unable to fix all occurrences of an annotation defined in a rule because MTA triggered no incidents for some annotations.

No known workaround exists.

MTA-6530

The MTA web console completes tech discovery and analysis of certain projects with errors. As a consequence, the tech discovery does not generate all relevant tags and the analysis could miss triggering issues.

To work around this problem, ensure that you have a Gradle or a Maven build file at the root directory of your project.

MTA-6140

When you run an analysis by using a custom YAML rule for a compiled Java binary in the MTA web console, MTA does not trigger a violation.

No known workaround exists.

MTA-5199

Before this update, MTA did not generate an analysis log when you used the --http-proxy or --https-proxy option in an MTA CLI analysis. As a consequence, you could not verify if MTA routed the analysis through the proxy. With this fix, you can verify proxy use because MTA analysis records the use of a proxy.

MTA-6553

Before this update, after you enabled the option to Consume insecure artifact repositories in the MTA web console, MTA discovery for Maven applications that use self-signed certificates failed. As a consequence, you could not generate deployment assets for the affected applications. With this fix, MTA connects with Maven application repositories that use a self-signed certificate and completes application discovery.

MTA-7361

Before this update, the MTA web console did not validate the custom YAML rules you upload for an analysis or in custom targets. As a consequence, if a rule contained invalid keys or syntax errors, you did not see an error message. With this fix, MTA improves user experience by validating the custom rules you upload in the web console.

MTA-6834

Before this update, MTA did not trigger violations for open source dependency libraries when analyzing Gradle applications. As a consequence, a source+dependency analysis generated fewer incidents in the static report. With this fix, MTA includes violations on open source dependencies in the Gradle analysis results.

MTA-6211

The MTA Visual Studio Code extension 8.1 renders inconsistently in the Red Hat OpenShift Dev Spaces dashboard open in the Firefox browser version 149.0. As a consequence, you are unable to perform tasks by using the extension.

To work around the problem, use the MTA Visual Studio Code extension in Red Hat OpenShift Dev Spaces through the Chrome browser.

MTA-6735

When you request a code resolution in the but reject the suggested resolutions in the review editor, Red Hat Developer Lightspeed for MTA modifies the source code with the suggested changes. As a consequence, the source code state changes after you reject the suggestions but GitHub status does not show any state change for the file.

To work around the problem, restore the original code file by re-opening the Visual Studio Code editor.

MTA-6863

Before this update, the MTA extension showed resolution in progress if you accepted resolutions for a project in an earlier analysis and then, restored the project files. As a consequence, you were unable to perform further analysis because the extension did not restore its state based on the current state of the project. With this update, the MTA extension restores its state after running an analysis. As a result, you can accept changes suggested in an earlier analysis of a project and restore the project files, before running more analyses.

(MTA-6862)

Before this update, when analyzing C# applications, the migration toolkit for applications (MTA) incorrectly calculated the boundaries for code snippets when rule violations occurred further down in a source file. As a consequence, the codesnip section in the analysis output cut off right before the violation, failing to display the actual line of code causing the issue. With this update, the C# analyzer provider properly calculates the code snippet boundaries to include the affected lines. As a result, the generated code snippets in the analysis output correctly display the exact violation line numbers and their full context.

(MTA-6898)

Before this update, the Application discovery manifest in the migration toolkit for applications (MTA) user interface (UI) did not display YAML-formatted content. As a consequence, you could only see raw JSON data that included all database record metadata instead of formatted YAML. With this release, the Application discovery manifest includes only YAML-formatted content fields.

(MTA-6774)

Before this update, the migration toolkit for applications (MTA) user interface (UI) did not allow you to generate assets for applications without an associated discovery manifest. As a consequence, you could not generate assets for applications with only source code that had not been discovered in a source platform. With this update, MTA supports generating assets for applications without a discovery manifest. As a result, you can use the MTA UI to generate these assets, which enhances deployment flexibility.

(MTA-6788)

You can use the centralized configuration management to standardize analysis configurations and custom rules for the following migration toolkit for applications (MTA) components:

Profiles contain analysis configuration that allows organizations to standardize configuration and simplify configuration management by adopting a platform engineering approach. The architect can create, update, and delete profiles and custom rules. Migrators can sync with the MTA Hub to download available configuration bundles and use them to run an analysis in the UI, CLI, and by using the Visual Studio Code plugin.

MTA-6491

You can use the external csharp provider to run an analysis in source-only mode for C# source code in the migration toolkit for applications (MTA) command-line interface (CLI), user interface, and the MTA Visual Studio Code extension. csharp parses the source code by using tree-sitter and uses stack graph for the analysis to find references to types, methods, classes, and fields. Based on the C# rule definition, the analyzer identifies violations in your code that you must resolve before the application migration.

MTA-6492

The migration toolkit for applications (MTA) Visual Studio Code extension consists of an analyzer Remote Procedure Call (RPC) binary that runs an analysis for the following extensions:

You can download the MTA version 8.1.0 Visual Studio Code extension as an extension pack that bundles the Core extension along with all the supported language-specific extensions. You can also use this extension when you want to analyze a project in a specific programming language. Alternatively, you can download the MTA Core extension with one language-specific extension to analyze a project coded in the programming language supported by the extension.

MTA user interface (UI) does not trigger a violation for open source libraries when you run a source+dependency analysis of a Gradle application. As a result, the analysis report does not list incidents for open source dependencies.

To work around this problem, analyze Gradle applications by using the MTA command-line interface (CLI).

MTA-6211

Before this update, you could not create Source Control credentials in the MTA user interface (UI) if you entered a user credential password that exceeded the 120-character limit. With this update, MTA does not limit the maximum length for user credentials. As a result, you can create Source Control credentials with longer passwords.

MTA-6598

Before this update, after you ran the source code analysis by using the MTA command-line interface (CLI), non-UTF-8 encoded source files appeared distorted in a static report, causing readability issues. With this update, the browser correctly displays non-UTF-8 encoded source files in the static report.

MTA-5932

Before this update, the discovery process for Cloud Foundry (CF) applications with service bindings by using live connection failed in the migration toolkit for applications (MTA) command-line interface (CLI). MTA CLI did not parse the VCAP_SERVICES environment variables correctly to handle service binding information for services, for example, cf bind-service <application_name> postgres01. With this update, MTA CLI no longer fails to complete the discovery of CF applications with service bindings.

MTA-6424

Before this update, the analysis performance for Java binaries degraded because MTA relied on Maven search to analyze Java binary applications. With this enhancement, MTA has a pre-generated bundled Maven index that does not require connecting to an external index through a network connection. As a result, MTA performs fast, low-memory look-ups of Maven artifacts by using local Maven Search and you can analyze Java binaries in disconnected environments.

MTA-6231

In the MTA user interface, selecting an answer to a question in the assessment form generated from a custom questionnaire causes the form to select an option available for another question.

To work around this problem, ensure that one or more order keys listed in a section, for example answers, have a unique order number.

MTA-6262

MTA UI does not trigger a violation for open source libraries when you run a source+dependency analysis of a Gradle application. As a result, the analysis report does not list incidents for open source dependencies.

To work around this problem, analyze Gradle applications by using the MTA command-line interface (CLI).

MTA-6211

When you use the migration toolkit for applications (MTA) user interface (UI) to run an application discovery for a deployed Cloud Foundry (CF) application with a service binding, MTA UI does not parse the VCAP_SERVICES environment variables correctly to handle service binding information. As a consequence, the discovery fails.

To work around this problem, run the application discovery by using the MTA command-line interface (CLI).

MTA-6424

Search patterns in Java rules that have the METHOD_CALL and CONSTRUCTOR_CALL locations do not match the expected methods and constructors in the application source code.

For example:

    java.referenced:
      location: CONSTRUCTOR_CALL
      pattern: javax.servlet.UnavailableException(javax.servlet.Servlet,java.lang.String)

does not match

 String str = "something";
 Servlet s = exception1.getServlet();
 UnavailableException exception2 = new UnavailableException(s, str);

In addition, search patterns that do not contain Fully Qualified Names (FQNs) do not match. For example:

    java.referenced:
      location: CONSTRUCTOR_CALL
      pattern: UnavailableException(Servlet s,String str)

No known workaround exists.

MTA-6402

Before this update, when you analyzed Java binary applications, MTA incorrectly classified some open source dependencies as internal dependencies. With this update, MTA classifies embedded open source dependencies correctly after a source+dependency binary analysis.

MTA-6357

Before this update, when you imported applications from Cloud Foundry (CF), MTA did not have a provision for you to mention the CF organizations. With this update, you can configure one or more organizations, spaces, and applications for the discovery process. The spaces can belong to different organizations and the applications can belong to different spaces. You must specify at least one organization for application discovery.

MTA-6271

Before this update, the discovery process of Cloud Foundry (CF) applications with service bindings by using live connection failed in MTA CLI. The MTA CLI did not parse the VCAP_SERVICES environment variables correctly to handle service binding information for an array of services, for example, cf bind-service <application_name> postgres01. With this update, MTA CLI successfully completes the discovery of CF applications with service bindings.

MTA-6399

Before this update, when you imported applications from a Cloud Foundry (CF) instance, they lacked the Cloud Foundry tag. As a consequence, you had to manually add this tag to link an application to an archetype. With this update, MTA automatically applies the tag. As a result, you no longer have to add it manually.

MTA-6282

Before this update, an MTA Architect could not discover or import applications from the Cloud Foundry (CF) source platform on the application inventory page. As a consequence, only an MTA Administrator could import the applications on the Source platforms page in the Administrator mode. With this update, Architects can import CF applications in the Application Inventory.

MTA-6120

Before this update, concurrent usage of the Solution Server through the Red Hat Developer Lightspeed for MTA in Visual Studio Code for code resolution at scale caused race conditions for duplicate incidents. With this update, the plugin generates resolutions for the incidents identified when many users use it concurrently.

MTA-6230

Before this update, the annotation rules for Java applications did not use Fully Qualified Names (FQNs) to match with the annotation location in source code. As a consequence, the rule triggered incorrect matches or false positives for issues. For example, a rule that looked for the javax.enterprise.inject.Produces incorrectly matched with occurrences of javax.ws.rs.Produces. With this update, MTA uses the FQN in the ANNOTATION field in rules to match annotations in the source code.

MTA-6195

Before this update, when analyzing a Java application, MTA did not trigger rules to match import statements that contained a wildcard. As a consequence, MTA did not detect specific incidents in the report. With this update, MTA triggers the rules to match these statements. As a result, it successfully detects the incidents.

MTA-4027

Before this update, editing an application stored in a Subversion (SVN) repository caused issues if you disabled the Allow insecure connection setting. As a consequence, the user interface displayed a blank page with the TypeError cannot read properties of undefined (reading 'kind') error. With this update, MTA correctly handles secure SVN applications. As a result, you can successfully edit these applications in the user interface.

MTA-6263

Before this update, for a Java application source code stored in a Subversion (SVN) repository, MTA produced an error when trying to checkout the repository by using the SVN credential. As a consequence, the language and tech discovery of the application failed. With this update, you can analyze an application stored in SVN repository without any error.

MTA-6141

Before this update, MTA did not show the source repository URL in the application details view if you did not select between the Git or Subversion repository type. As a consequence, you could not run an analysis on the application source code. With this update, you can use MTA to run an analysis on the source code even if you do not configure the repository type field.

MTA-6105

Before this update, you could not proceed to the final step after filling in the responses to the custom questionnaire because MTA disabled the Next button in the web console. With this update, you can complete the questionnaire in MTA user interface.

MTA-6143

Before this update, when you uploaded more than one custom rule at a time in the web console, MTA did not detect the rules. With this update, you can upload one or more custom rules at once.

MTA-6029

You can opt to use Red Hat Developer Lightspeed for MTA features in the Visual Studio Code extension. With the Developer Lightspeed feature, you can use a large language model (LLM) of your choice to request code changes for resolving the issues found through a static code analysis of your Java application.

The Red Hat Developer Lightspeed for MTA core features are the following:

MTA-5371

You can use platform awareness to discover applications from a source platform and import them into the MTA application inventory. MTA discovers the applications by their coordinates on the source platform where you deployed them. With platform awareness, you can use the platform and runtime configurations of applications to create a discovery manifest to generate other deployment assets. In MTA 8.0.0, you can discover applications deployed in the Cloud Foundry platform.

MTA-4846

You can use the discovery manifest to generate assets in the web console to deploy applications in Red Hat OpenShift or Kubernetes. The asset generation workflow includes generators, target profiles, and the archetypes. Generators parse the application manifest configurations to create the deployment assets for a specific kind of target platform. In the target profiles, you can run one or more generators in sequence for a given archetype.

MTA-4847

With this update, you can perform a live discovery to determine applications deployed in a certain Cloud Foundry (CF) cluster. For example, you can determine how many applications are in the cluster. You can also use the live discovery if you do not have access to manifest YAML files.

For more information, see Performing a live discovery in a remote CF instance.

MTA-4847

You can conceal sensitive information, for example, services and Docker credentials, in a Cloud Foundry (CF) discovery manifest by using the mta-cli discover cloud-foundry --conceal-sensitive-data command. This command generates the following files:

For more information, see Concealing sensitive information in a discovery manifest.

MTA-4847

Analysis insights contain information about the technologies used in the application and their usage in the code. Before this update, you could only view insights in the analysis output and a static report after running an application analysis. With this enhancement, you can access analysis insights also from the migration toolkit for applications (MTA) user interface (UI).

MTA-5420

With this enhancement, you can use a new Visual Studio Code Integrated Development Environment (IDE) plugin that has the following features:

MTA-5360

Before this update, you could only assign credentials to an application manually after you added or imported this application to the migration toolkit for applications (MTA) user interface (UI). However, you might need to import a large set of applications, with each application requiring credentials for language and technology discovery tasks. With this update, instead of manually defining credentials, you can define default credentials for any application that does not have credentials.

For more information, see Setting default credentials.

MTA-5254

Solution Server is a component that allows Developer Lightspeed for MTA to build a collective memory of code changes from all analysis performed in an organization. This capability is a Technology Preview feature. The Solution Server works with the LLM to improve the pattern of resolutions from solved examples for future analysis.

The Solution Server has the following benefits:

In migration toolkit for applications (MTA) 8.2.0, MTA removes XML rules and no longer supports them in the MTA user interface (UI) and command-line interface (CLI).

MTA-5357

An application analysis by using the migration toolkit for applications (MTA) command-line interface (CLI) might fail with the following error message:

level=error msg="initialize failed" error="context canceled" provider=java
error="invalid header line "An error has occurred. See the log file"

To work around this problem, clean up metadata and cache:

MTA-6125

When you use Developer Lightspeed for MTA, the analysis results do not change if you include or remove a custom rule but do not restart the analyzer process.

To work around this problem, restart the analyzer process by clicking the Start/Stop button after making a configuration change.

MTA-6129

The ANNOTATION search location does not properly use Fully Qualified Names (FQNs) to match annotations. This leads to incorrect matches and false positives in analysis issues. For example, a rule that searches for annotations using the javax.enterprise.inject.Produces pattern might incorrectly match the occurrences of the javax.ws.rs.Produces annotation in the source code.

No known workaround exists.

MTA-6195

The Solution Server throws a connection error when concurrent connections increase in a short span of time. To work around this problem, enter the following command to allow the Developer Lightspeed for MTA database to expire idle connections:

oc -n openshift-mta exec deploy/kai-db – psql -U postgres -d postgres -c "ALTER ROLE kai SET idle_session_timeout = '1min'; ALTER ROLE kai SET idle_in_transaction_session_timeout = '1min';"

MTA-6204

Before this update, when you generated a deployment manifest by using the Helm template, MTA ignored the default values in the values.yaml file packaged with the Helm chart. As a consequence, MTA used only the values from the values.yaml file created after running the mta-cli discover command. With this update, MTA merges the default values.yaml file with the discover values.yaml file.

MTA-5793

Before this update, when you ran an application analysis for Gradle projects in containerless mode, migration toolkit for applications (MTA) command-line interface (CLI) did not detect Gradle dependencies. As a consequence, the analysis report did not list the dependencies and displayed an error message in the analysis logs instead. With this update, MTA CLI detects Gradle dependencies after the analysis is complete.

MTA-4033

Before this update, when you analyzed an application built with Gradle in Source code + dependencies mode in the migration toolkit for applications (MTA) user interface (UI), the analysis failed if you had Gradle version 7 or later installed. With this update, you can analyze applications built with Gradle 7 or later in the MTA user interface.

MTA-5907

Before this update, the mta-cli discover command only moved an application-level process with the process type explicitly defined to the processes block in the discovery manifest. As a consequence, the discovery manifest did not list implicit application-level processes in the processes block. This resulted in the violation of the recommended clarity and proper configuration practices, especially if Cloud Foundry application manifest included multiple process types. With this update, mta-cli discover moves both implicit and explicit application-level process types to the processes block in the discovery manifest.

MTA-6030