Installing local galaxy failed at step: run miniconda installer

Hi,
I am following the galaxy with ansible installation tutorial (Galaxy Installation with Ansible) but fail when running the playbook (Ubunutu 22.04.2). It stops at the step “run miniconda installer” and prints the following error that I don’t understand. What can I do?

fatal: [localhost]: FAILED! => changed=true 
  cmd:
  - /bin/sh
  - /tmp/ansible-miniconda-installer.rnn9nixs.sh
  - -b
  - -p
  - /srv/galaxy/var/dependencies/_conda
  delta: '0:00:01.715882'
  end: '2023-08-08 08:05:18.684320'
  msg: non-zero return code
  rc: 1
  start: '2023-08-08 08:05:16.968438'
  stderr: |2-
...
stderr_lines: <omitted>
  stdout: |-
    PREFIX=/srv/galaxy/var/dependencies/_conda
    Unpacking payload ...
  stdout_lines: <omitted>

Hi @Markus_Konkol

I don’t recognize what this is except for some potential permissions problem.

Scroll down in this section of the tutorial to the last step (step 10). What content was created so far? In particular, does the content of this file appear to match what you expect? You could post that back here for more context.

/srv/galaxy/config/galaxy.yml

And, let’s cross post over to the Admin chat. They may reply here or there, and feel free to join the chat. You're invited to talk on Matrix

Hi @jennaj
I checked step 10:
Yes, Galaxy has been deployed under /srv/galaxy/server and the configuration file is in /srv/galaxy/config/galaxy.yml. I took some screenshots to show the setup and will also refer to this thread in the admin chat.

Or does anyone know how to get a more meaningful error message? The output doesn’t state what is actually going wrong…

In case anyone else has a simlar issues: This PR solved the issue for me: Proposed changes to job-destinations and ansible-galaxy admin tutorials to prevent ansible-playbook errors. by Edmontosaurus · Pull Request #4308 · galaxyproject/training-material · GitHub
Should result in the following output:

1 Like