Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Unable to Volume Mount in Playbook Execution Container #1927

Open
3 tasks done
advanceboy opened this issue Jul 24, 2024 · 2 comments
Open
3 tasks done

Unable to Volume Mount in Playbook Execution Container #1927

advanceboy opened this issue Jul 24, 2024 · 2 comments

Comments

@advanceboy
Copy link

advanceboy commented Jul 24, 2024

Please confirm the following

  • I agree to follow this project's code of conduct.
  • I have checked the current issues for duplicates.
  • I understand that AWX Operator is open source software provided for free and that I might not receive a timely response.

Feature Summary

AWX creates a automation-job-<job_id>-<random> pod and executes the playbook within the worker container.

Using the method described in Custom Volume and Volume Mount Options, volumes can be mounted to the Web or Control Plane containers.

However, there is no way to mount a volume to the job's worker container.

apiVersion: awx.ansible.com/v1beta1
kind: AWX
metadata:
  name: awx-demo
  namespace: awx
spec:
  service_type: NodePort
  extra_volumes: |
    - name: volume-mount-hostpath-volume
      hostPath:
        path: /path/to/volume-mount/
        type: Directory
  web_extra_volume_mounts: |
    - name: volume-mount-hostpath-volume
      mountPath: /path/to/volume-mount/
      readOnly: true
  task_extra_volume_mounts: |
    - name: volume-mount-hostpath-volume
      mountPath: /path/to/volume-mount/
      readOnly: true
  ee_extra_volume_mounts: |
    - name: volume-mount-hostpath-volume
      mountPath: /path/to/volume-mount/
      readOnly: true
@YaronL16
Copy link
Contributor

Maybe I misunderstood but is the ee_extra_volume_mounts not what you are looking for?

EE means execution environment which is the container that runs your jobs

@advanceboy
Copy link
Author

@YaronL16
Thank you for your response!

I appreciate the suggestion; however, I believe my issue might not be resolved by using ee_extra_volume_mounts.
The problem is that even with this setting, I am unable to mount volumes correctly in the execution environment container.

The settings of task_extra_volume_mounts and ee_extra_volume_mounts affect the mount of the <resourcename>-task and <resourcename>-ee containers in the <resourcename>-task-<random> pod, respectively.
The <resourcename>-ee container uses the quay.io/ansible/awx-ee image, but only acts as a "receptor" and does not execute Playbooks.

$ kubectl -n awx describe pod awx-demo-task-6898d4ddbc-l82ts
Name:             awx-demo-task-6898d4ddbc-l82ts
Namespace:        awx
Priority:         0
Service Account:  awx-demo
...
Containers:
  redis:
    Image:         docker.io/redis:7
    ...
  awx-demo-task:
    Image:         quay.io/ansible/awx:24.6.1
    ...
    Mounts:
      ...
      /etc/tower/settings.py from awx-demo-settings (ro,path="settings.py")
      /path/to/volume-mount/ from volume-mount-hostpath-volume (ro)
      /var/lib/awx/projects from awx-demo-projects (rw)
      ...
  awx-demo-ee:
    Image:         quay.io/ansible/awx-ee:24.6.1
    ...
    Args:
      /bin/sh
      -c
      if [ ! -f /etc/receptor/receptor.conf ]; then
        cp /etc/receptor/receptor-default.conf /etc/receptor/receptor.conf
        sed -i "s/HOSTNAME/$HOSTNAME/g" /etc/receptor/receptor.conf
      fi
      exec receptor --config /etc/receptor/receptor.conf

    ...
    Mounts:
      ...
      /etc/receptor/work_private_key.pem from awx-demo-receptor-work-signing (ro,path="work-private-key.pem")
      /path/to/volume-mount/ from volume-mount-hostpath-volume (ro)
      /var/lib/awx/projects from awx-demo-projects (rw)
      ...
  awx-demo-rsyslog:
    Image:         quay.io/ansible/awx:24.6.1
...
Volumes:
  ...
  volume-mount-hostpath-volume:
    Type:          HostPath (bare host directory volume)
    Path:          /path/to/volume-mount/
    HostPathType:  Directory
...

EE means execution environment which is the container that runs your jobs

As far as I have tried, when running Playbooks on AWX v24, it appears that the pod automation-job-<job_id>-<random> is created and the worker container is used.
In the above settings, we can see that this worker container does not mount anything.

For example, running the following playbook below:

- hosts: all
  tasks:
    - local_action:
        module: ansible.builtin.shell
        cmd: printenv
    - local_action:
        module: ansible.builtin.shell
        cmd: ls -la /
    - name: sleep 60
      ansible.builtin.command: sleep 60

Then, you'll know the name of the pod that's running.

changed: true
stdout: >-
  ...
  HOSTNAME=automation-job-3-568rw

  HOME=/runner

  JOB_ID=3

  ...
stderr: ''
rc: 0
cmd: printenv

Also, you can see that /path/to/volume-mount/ is not mounted.

changed: true
stdout: |-
  total 72
  drwxr-xr-x   1 root root 4096 Oct 31 06:45 .
  drwxr-xr-x   1 root root 4096 Oct 31 06:45 ..
  dr-xr-xr-x   2 root root 4096 Jun 25 14:23 afs
  lrwxrwxrwx   1 root root    7 Jun 25 14:23 bin -> usr/bin
  dr-xr-xr-x   2 root root 4096 Jun 25 14:23 boot
  drwxr-xr-x   5 root root  360 Oct 31 06:45 dev
  drwxr-xr-x   1 root root 4096 Oct 31 00:25 etc
  drwxr-xr-x   2 root root 4096 Jun 25 14:23 home
  lrwxrwxrwx   1 root root    7 Jun 25 14:23 lib -> usr/lib
  lrwxrwxrwx   1 root root    9 Jun 25 14:23 lib64 -> usr/lib64
  drwx------   2 root root 4096 Oct 28 04:09 lost+found
  drwxr-xr-x   2 root root 4096 Jun 25 14:23 media
  drwxr-xr-x   2 root root 4096 Jun 25 14:23 mnt
  drwxr-xr-x   1 root root 4096 Oct 31 00:19 opt
  dr-xr-xr-x 315 root root    0 Oct 31 06:45 proc
  dr-xr-x---   1 root root 4096 Oct 31 00:24 root
  drwxr-xr-x   1 root root 4096 Oct 31 00:26 run
  drwxrwxr-x   1 root root 4096 Oct 31 06:45 runner
  lrwxrwxrwx   1 root root    8 Jun 25 14:23 sbin -> usr/sbin
  drwxr-xr-x   2 root root 4096 Jun 25 14:23 srv
  dr-xr-xr-x  13 root root    0 Oct 31 06:45 sys
  drwxrwxrwt   1 root root 4096 Oct 31 06:45 tmp
  drwxr-xr-x   1 root root 4096 Oct 28 04:09 usr
  drwxr-xr-x   1 root root 4096 Oct 28 04:09 var
stderr: ''
rc: 0
cmd: ls -la /

With the playbook above running, even if you look at the state of the corresponding pod on K8s, you can see that the container running the job has not mounted anything.

$ kubectl -n awx get pod
NAME                                              READY   STATUS      RESTARTS   AGE
automation-job-3-568rw                            1/1     Running     0          9s
awx-demo-migration-24.6.1-w6m5g                   0/1     Completed   0          111m
awx-demo-postgres-15-0                            1/1     Running     0          119m
awx-demo-task-6898d4ddbc-l82ts                    4/4     Running     0          119m
awx-demo-web-f9d668b4c-l77sv                      3/3     Running     0          119m
awx-operator-controller-manager-bb695bf5d-6lhls   2/2     Running     0          120m

$ kubectl -n awx describe pod automation-job-3-568rw
Name:             automation-job-3-568rw
Namespace:        awx
Priority:         0
Service Account:  default
...
Containers:
  worker:
    Image:         quay.io/ansible/awx-ee:latest
    ...
    Args:
      ansible-runner
      worker
      --private-data-dir=/runner
    ...
    Mounts:       <none>
...
Volumes:                      <none>

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants