hsmdebugger - igor-krechetov/hsmcpp Wiki

Contents

Overview

Hsmdebugger provides a visual way to check how HSM was executed and identify potential issues with it. To operate it requires HSM defined in scxml format and log file generated by your application.

hsmdebugger demo

Dependencies

Installing dependencies (Ubuntu)

apt install python3
apt install plantuml
pip3 install PySide6
pip3 install PyYaml

Note: Currently (2021.06) plantuml package on Ubuntu is tool old (V1.2018.13). Hopefully, this will be fixed one day, but for now you'll need to download plantuml manually. You can either:

Installing dependencies (Windows)

pip install PySide6
pip install PyYaml

How to use

Configuring environment

Before starting to use hsmdebugger please check that it's environment is correctly configured.

Go to File -> Settings:

Log loaded

Make sure that path to scxml2gen.py is correctly specified. Normally you don't need to change anything if you run hsmdebugger from it's original folder.

For Windows: it's necessary to specify full path to plantuml binary.

Log loaded

Enable debugging in your application

Make sure that HSMBUILD_DEBUGGING build option is set to ON.

Call enableHsmDebugging() API to start recording HSM activity into a log file.

// By default log will be written to ./dump.hsmlog file.
// This location can be overwritten by setting ENV_DUMPPATH environment variable with desired path.
enableHsmDebugging()

// Path has to be provided as an argument; no default path is used.
enableHsmDebugging(const std::string& dumpPath)

Log generation can be disabled at any time using:

disableHsmDebugging()

There is no limitation on how often you can enable/disable logging.

Loading log in hsmdebugger

After you have generated log file from your application you can load it in hsmdebugger to analyze:

Log loaded

Navigating the log

There are a couple of options available to navigate the log file:

Log loaded

Log loaded