Frequently asked questions

How do I use Inmanta with a http/https proxy?

Use the http_proxy and https_proxy environment variables to specify the proxy server to use. For the server installed from our RPMs, add the environment variable to the systemd unit file. Copy inmanta-server.service from /lib/systemd/systemd/system to /etc/systemd/system and add the following lines to the [Service] section with the correct proxy server details:

Environment=http_proxy=1.2.3.4:5678
Environment=https_proxy=1.2.3.4:5678

Afterwards run systemctl daemon-reload and restart the inmanta server.

I get a click related error/exception when I run inmanta-cli.

The following error is shown:

Traceback (most recent call last):
    File "/usr/bin/inmanta-cli", line 11, in <module>
        sys.exit(main())
    File "/opt/inmanta/lib64/python3.4/site-packages/inmanta/main.py", line 871, in main
        cmd()
    File "/opt/inmanta/lib64/python3.4/site-packages/click/core.py", line 722, in __call__
        return self.main(*args, **kwargs)
    File "/opt/inmanta/lib64/python3.4/site-packages/click/core.py", line 676, in main
        _verify_python3_env()
    File "/opt/inmanta/lib64/python3.4/site-packages/click/_unicodefun.py", line 118, in _verify_python3_env
        'for mitigation steps.' + extra)
RuntimeError: Click will abort further execution because Python 3 was configured to use ASCII as encoding for the environment.  Consult http://click.pocoo.org/python3/for mitigation steps.

This error occurs when the locale are not set correctly. Make sure that LANG and LC_ALL are set. For example:

export LC_ALL=en_US.utf8
export LANG=en_US.utf8

The model does not compile and exits with “could not complete model”.

There is an upperbound on the number of iterations used in the model transformation algorithm. For large models this might not be enough. This limit is controlled with the environment variable INMANTA_MAX_ITERATIONS The default value is set to 10000 iterations.

Shell command caching

If you experience the issue that pytest doesn’t find a package that is installed in your venv or if it picks up a package from an unexpected location, this might be caused by the shell command cache. You can resolve this issue by executing hash -r (bash) or rehash (zsh).

Explanation:

When you execute a command (like pytest) on the commandline, your shell looks up the path to the corresponding binary and caches it. As such you can get into the following scenario:

• Create a new venv that will be used to run the test suite of an Inmanta project.

• You execute the pytest command and notice that you forgot to install the Inmanta project into that venv. If pytest is installed in your global python environment, your shell will have resolved and cached the path to that binary.

• You then install the Inmanta project into your venv. This installs pytest into your venv with a different version than the one installed in the global python environment.

• If you now run the pytest command again, it will execute the pytest binary from the global Python environment instead of the one from your new venv, because it’s still present in the shell cache.

Situations as mentioned above usually don’t happen, because the shell cache is cleared automatically when a new venv is activated. In this scenario, the cache was populated before the venv’s pytest was installed, so the shell still points to the old binary even though the correct one is now available.