spiff-arena/docs/Support/Running_Server_Locally.md

3.4 KiB

Troubleshooting: Running Server Locally

When setting up the SpiffWorkflow backend project locally, you might encounter issues related to the sample-process-models directory. This documentation aims to address those concerns.

Problem

While following the instructions provided in the repository to set up the SpiffWorkflow backend project locally, you may find that the script ./bin/run_server_locally expects the sample-process-models directory to be present. However, this directory might not be immediately available in the repository.

Solutions

1. Clone the sample-process-models Repository

The sample-process-models directory refers to a separate repository. To resolve the issue:

2. Use Any Git Repository

If you prefer not to use the sample-process-models directory or want to start from scratch:

  • Locate or create a git repository.

  • Run the following command with the path to your chosen repository

    SPIFFWORKFLOW_BACKEND_BPMN_SPEC_ABSOLUTE_DIR=/path/to/any/git/repo ./bin/run_server_locally
    

3. Use Docker Compose

For a potentially faster setup:

4. Mac install issues

Please follow the README.md at the root of the spiff-arena repo first. If you're on a Mac and trying to run natively (it might translate elsewhere) and get errors when running poetry install, many of the issues are related to relational database engine libraries. You will find lots of people having pain related to these on the internet, so hopefully the solution to your particular problem is easy to find, but this is not always the case. You may need to install additional system dependencies.

brew install mysql-client pkg-config
export PKG_CONFIG_PATH="$(brew --prefix)/opt/mysql-client/lib/pkgconfig"
pip install mysqlclient
pip install psycopg2

One person decided that mysqlclient and psycopg2 were more trouble than they were worth and removed them from the pyproject.toml, while doing poetry add pymysql instead. If you are using mysql, psycopg2 is not necessary, and pymysql is a pure python implementation of the MySQL client library. In that case, python won't recognize MySQLdb as a module, so after the above installs (which you would do pre-Poetry), add these lines to init.py in the backend project:

import pymysql;
pymysql.install_as_MySQLdb()

5. Access Hosted Version of Spiff

If you prefer not to install anything locally:

Setting up the SpiffWorkflow backend project locally can be straightforward once you're aware of the dependencies and options available. Whether you choose to clone the sample-process-models repository, use a different git repository, or opt for Docker Compose, the solutions provided should help you get started without any hitches.

If you encounter further issues, always refer back to the repository's README or seek assistance from our discord community.