Scanning with Multi-Module Analysis
Overview
The Multi-Module Analysis feature enables you to analyze large Project structures comprising multiple Project modules. The analysis results for each module are displayed in the application within an independent Project corresponding to the analyzed module; by default, the relevant Projects are located under a single Product corresponding to the multi‑module structure.
This feature enables you to analyze the sub modules, referenced from a main module, by referencing parameters for the main module without being required to explicitly specify parameters for each of the 'sub' modules. It provides the user with a command that inspects the configuration for a specified multi-module and creates a setup file featuring the appPath
and Project name references for the submodules reflecting the inspection. The resulting setup file is referenced within a Mend Unified Agent command.
Prerequisites
Multi-Module Analysis is supported for Java and JavaScript Projects. Also supported are Projects of a wider scope that feature any mixture of Java, Kotlin, Scala & JavaScript.
The following Unified Agent parameters must be set to True:
gradle.aggregateModules
maven.aggregateModules
enableImpactAnalysis
The following Unified Agent parameters must be set to False:
gradle.runPreStep
maven.runPreStep
npm.runPreStep
Running the Unified Agent with the '-analyzeMultiModule' Parameter
-analyzeMultiModule
inspects the configuration for a specified multi-module and creates a setup file featuring application path (AppPath) references that reflect the inspection.
A Unified Agent command with the following parameters is required to create the setup file for the multi-module definitions:
Parameter | Description |
---|---|
-analyzeMultiModule | Instructs the command to inspect the structure of a specified multi-module (rather than run Effective Usage Analysis) and save the project name for each sub-module in a setup file |
-d | The argument text value specifies the path for the main module's pom (e.g., c:\test1) |
-analyzeMultiModuleExclusions | Multi-module analysis can include exclusion rules to support automatic/default and manual exclusion of files that should not be considered by the Unified Agent as valid ‘appPath’ candidates. This parameter enables users to specify patterns for file names that should not be analyzed. For example:
The following example would exclude the two jar files: myapp-v8.1.0-internal.jar and myapp-v8.1.0.jar.original: |
-overrideExistingSetup | When set to True, enables users to override an existing multi-module setup file when running the first step of the multi-module Mend Prioritize (default is False). |
The following is the command that inspects the structure of a specified multi-module and generate a file ('setupFile.txt' or any other file name that the user defines) for multi-module analysis:
java -jar wss-unified-agent.jar -d <Path For The Main Module's Project Directory> -analyzeMultiModule setupfileName
A sample command:
CODE
|
The following message is displayed after the setup file was created successfully:
[INFO] <time_stamp>] - The multi-module analysis setup file was created successfully. |
If any of the values were left blank, the following warning message is also displayed:
[WARN][<time_stamp>] - Analysis found multiple candidates for one or more appPath settings that are listed in the multi-module analysis setup file. Please review the setup file and set the appropriate appPath parameters. |
In such cases, the 'appPath' value(s) must be entered manually. See the next section for more information on the AppPath parameter.
Setup File Structure & Contents
The following is the structure of the setup file contents:
CODE
|
The file includes the following contents:
DependencyManagerFilePath: The path to the dependency manager file folder. The first line of the generated file includes this parameter and its value. There is only one instance of this parameter in the file and it recommended not to manually modify its value.
ProjectFolderPath<number of instance>: The folder name associated with a Project (e.g., c:\a\project\project1). Its value is automatically obtained based on analysis.
The setup file can include multiple instances of the ProjectFolderPath parameter. The parameter must be specified in an independent line and have a corresponding 'appPath' parameter specified in the subsequent line. It can be removed only if the corresponding 'appPath' is removed as well.
AppPath: The filename associated with a Project (e.g., c:\a\project\project1\tester.jar). The value is automatically obtained based on analysis.
There may be cases where the AppPath value is left blank, and in such cases, the value must be manually entered.
To prevent cases where inadvertently running multi-module Mend Prioritize with duplicate Project name entries result in ignoring processes, the system includes these two parameters that, when manually entered, eliminate the scenario where multiple Projects have the same name. By default. both of these parameters have the value of the Project name.
defaultName: Default name of the file Project (editable)
altName: Alternative name of the file Project (editable)
The following is a sample of the setup file:
CODE
|
The following is a sample of the setup file that has missing 'AppPath' values:
CODE
|
In the above sample, 'AppPath2' and 'AppPath4' must be entered manually.
Downloading & Running the xModuleAnalyzer
Overview
The xModuleAnalyzer for Mend Prioritize enables you to specify the .jar path of the Unified Agent, the configuration file of the Unified Agent, and the relevant setup file that was created in the previous step.
The output analyzes each submodule listed in the multi-module setup file in a threshold mode display (also the default display status when no value is provided) or a dynamic mode display.
Downloading the xModuleAnalyzer
The latest version of our xModuleAnalyzer and its download can also be accessed by visiting our Product Downloads page:
JAR File | Features |
---|---|
Align xModuleAnalyzer with the latest Unified Agent |
Previous versions of the xModuleAnalyzer can be found here.
NOTE: The xModuleAnalyzer versions will be available and supported for one year after their release.
Running the xModuleAnalyzer
Command Line Parameters
Parameter | Type | Description | Required? | Default |
---|---|---|---|---|
-xModulePath | String | Setup file pathname | Yes | |
-fsaJarPath | String | Unified Agent .jar pathname | Yes | |
-c | String | Configuration file for Unified Agent. | Yes | |
-aggregateModules | Boolean | If True - aggregates the results of all scanned modules to a single unified Project. | No | False |
-statusDisplay | String | Screen display mode. See Threshold Mode Display and Dynamic Mode Display below for details. | No | threshold |
-logPath | String | Path for storing logs created by the analysis. | No | <DependencyManagerFilePath>\mend-Logs |
-productName | String | Product name. Overrides the -product value of Unified Agent. | No | |
-runInParallel | Boolean | Whether to run the processes of the Unified Agent in parallel (faster) or serial (less prone to errors) mode. | No | true |
-EuaMaxTotalMemAlloc | String | EuaMaxTotalMemAlloc [custom_val] | DEFAULT | MAXIMIZE Governs the allowed memory allocation for Mend Prioritize.
| No | The value provided by mend |
-ignoreEuaNotices | String | Specifies how Mend Prioritizes codes will be handled. This parameter governs the ability to ignore specified Prioritizes codes for analyzed modules.
| No | informational |
Command Structure
CODE
|
<x.x.x> represents the latest version number of the specified Jar file.
Multi-Module Prioritize - Local Scan
This video demonstrates how to perform a local scan using the “-analyzeMultiModule” command line flag, followed by running xModuleAnalyzer using the relevant setup file created by the scan.
Multi-Module Prioritize - CI/CD Scan
This video demonstrates how to configure your AZDO YAML file to perform a pipeline scan using the “-analyzeMultiModule” command line flag, followed by running xModuleAnalyzer using the relevant setup file created by the scan.
Returned Codes for xModuleAnalyzer
xModuleAnalyzer can return the following codes:
(Success) 0: All multi-module analyses completed successfully (all modules returned [EUA000], or all codes were ignored based on a supported setting).
(Error) -100: Some modules were not analyzed successfully (one or more modules returned a non-[EUA000] code and codes were not ignored based on a supported setting).
(Failure) -200: It was not possible to successfully meet pre-conditions for xModuleAnalyzer.
Also, if one or more modules in a multi-module setting failed to run or complete (not due to an EUA-related error), a corresponding informational message will be returned by xModuleAnalyzer. Note that the returned xModuleAnalyzer code will not be influenced by the conditions leading to the message (e.g., a return code of 0 (success) will not change if one or more modules did not complete processing).
The message will be formatted as follows (it will be captured in the designated xModuleAnalyzer log - see Log Files below):
The processing of the following modules did not complete:
Module module_identifier
Module module_identifier
…
Module module_identifier
Threshold Mode Display (Default)
In this mode, the screen display is only refreshed whenever analysis progress for each Project crosses a specified threshold (25%, 50%, 75%, and 100% thresholds), and only if a specified interval of one minute has elapsed. The following is a sample command for a threshold mode display command that stores the logs in the path 'C:\logs\':
CODE
|
Dynamic Mode Display
In this mode, the screen display is refreshed every 10 seconds to show the analysis progress for each Project. The following is a sample command for a dynamic mode display command:
CODE
|
Each analyzed Project has one of the following statuses for each phase of analysis: '
Pending
In-Progress
Success
Failure
Log Files
After the process completes, a log file is generated for each module. Log files are located in the following path: <DependencyManagerFilePath>\WhiteSource-Logs.
This path can be modified by using the optional -logPath parameter. For example, in the following command, the log path is C:\logs\.
CODE
|
When the analysis finishes with errors, an indication to examine the related log file is displayed. All completion statuses that include the '( * )' characters indicate that errors were detected.
Regardless of the above log files, the following log files will be created:
xModuleAnalyzerLog_ddmmyyyy_hhmm.log
- This file will include all the console log of the multi-module analysis.projectDetails{ddmmyyyy_hhmmss}.json
- This file will include the output of the Unified Agent scan (similar to the fsa.json) in order to clarify how the Unified Agent divided the modules.