Fresh provision of b@d stops with error in kernelci-backend

Issue Background

I noticed this error on Friday just retried and the error is still present on another new VM

Issue description

==> default: TASK [init-conf : Enable celery systemd service] *******************************
==> default: changed: [kernel-ci-backend]
==> default: 
==> default: TASK [init-conf : Enable celery beat systemd service] **************************
==> default: changed: [kernel-ci-backend]
==> default: 
==> default: RUNNING HANDLER [common : reload-sysctl] ***************************************
==> default: changed: [kernel-ci-backend]
==> default: 
==> default: RUNNING HANDLER [init-conf : restart-kernel-ci-backend] ************************
==> default: fatal: [kernel-ci-backend]: FAILED! => {"changed": false, "failed": true, "msg": "Unable to start service kernelci-backend: Failed to start kernelci-backend.service: Unit mongod.service failed to load: No such file or directory.\n"}
==> default: 
==> default: RUNNING HANDLER [init-conf : restart-celery] ***********************************
==> default: 
==> default: RUNNING HANDLER [init-conf : reload-systemd] ***********************************
==> default:    to retry, use: --limit @/home/vagrant/git-repos/kernelci-backend/site.retry
==> default: 
==> default: PLAY RECAP *********************************************************************
==> default: kernel-ci-backend          : ok=28   changed=23   unreachable=0    failed=1   
The SSH command responded with a non-zero exit status. Vagrant
assumes that this means the command failed. The output for this command
should be in the log above. Please read the output to determine what
went wrong.

Acceptance criteria

When a fresh instance of b@d provisions correctly