Python crashes when running role

I’m not sure if this is related to this particular role, or perhaps Python, but I thought it would be a good idea to document the bug in the event anyone else runs into this, and perhaps the maintainers of this repository have recommendations on how to debug this.

I installed this role by doing the following:

requirements.yml

ansible-galaxy install -r requirements.yml

site.yml

- name: Install node_exporter on all bare-metal hosts
  hosts: bare-metal
  roles:
    - cloudalchemy.node-exporter

And then attempted to do a dry run with:

ansible-playbook -i ansible/hosts.yml --ask-become-pass --check ansible/site.yml

The output looks like this:

PLAY [Install node_exporter on all bare-metal hosts] ******************************************************************************************************************

TASK [Gathering Facts] ************************************************************************************************************************************************
ok: [nuc7i5b.brooks.network]
ok: [nuc7i5a.brooks.network]
ok: [pi.brooks.network]

TASK [cloudalchemy.node-exporter : Gather variables for each operating system] ****************************************************************************************
fatal: [nuc7i5a.brooks.network]: FAILED! => {"msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"}
fatal: [nuc7i5b.brooks.network]: FAILED! => {"msg": "No file was found when using with_first_found. Use the 'skip: true' option to allow this task to be skipped if no files are found"}
ok: [pi.brooks.network] => (item=/Users/brooks/.ansible/roles/cloudalchemy.node-exporter/vars/debian.yml)

TASK [cloudalchemy.node-exporter : Naive assertion of proper listen address] ******************************************************************************************
ok: [pi.brooks.network] => {
    "changed": false,
    "msg": "All assertions passed"
}

TASK [cloudalchemy.node-exporter : Fail on unsupported init systems] **************************************************************************************************
skipping: [pi.brooks.network]

TASK [cloudalchemy.node-exporter : Check collectors] ******************************************************************************************************************

TASK [cloudalchemy.node-exporter : Get checksum list from github] *****************************************************************************************************
objc[93522]: +[__NSPlaceholderDate initialize] may have been in progress in another thread when fork() was called.
objc[93522]: +[__NSPlaceholderDate initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork()
child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug.

And I get this from OSX:

I’m running this from MacOS 10.13.6, with Python version 2.7.15 installed from brew.

[bswinnerton]

Sure! I use node_exporter to monitor the hosts in my home lab. I’m starting to migrate over to Ansible as a configuration management tool and wanted to avoid re-inventing the wheel, so I came across this role.

[bswinnerton]

I was able to resolve this by exporting the OBJC_DISABLE_INITIALIZE_FORK_SAFETY variable to YES.

Via https://github.com/ansible/ansible/issues/31869#issuecomment-337769174.