all fixs to make hashistack up and running in 0.9.58

[amayerwescale]

Upgrade hashistack from version 0.9.33 to version 0.9.58.

We have tested the install in :

• offline/online

on differents OS :

• debian12
• rocky9
• alma9.

Adding "common_vars" roles

We faced severals time to errors like undefined variable so to counter this, we have added the common_vars role in these files :

• playbooks/01_infra_scaleway.yml
• playbooks/13_core_scaleway_dns_delegation.yml
• playbooks/01_infra_offline.yml
• playbooks/00_offline_prepare.yml
• playbooks/20_vault_install.yml
• playbooks/14_selfsigned_ca.yml

Upgrade Scaleway provider version

We need to upgrade scaleway provider from version 2.37.0 to version 2.42.1 (latest) to be able to deploy infrastructure on scaleway

Multi nodes deployment by default

It seems the default deployment is multi node so we need to add hs_stage0_archi variable in playbook/init.yml file

Add duplicity python package

We got error like duplicity python libarary not found. We need to add duplicity in roles/stage1_pip/defaults/main.yml file

Variables renaming

We faced severals times to undefined variables due to name changing

From	To	In
public_ipv4	edge_public_ipv4	roles/stage0/templates/mono/out.stage0.hashistack.yml.j2
hs_nsupdate_host̀	acme_nsupdate_host	roles/stage1_rproxy/tasks/common/_configure.yml

Add missing paths

We got error like directory does not exist during the execution of init playbook. To fix we have added these missing paths in playbooks/init.yml

• {{ _output_dir }}/host_vars
• {{ _output_dir }}/host_vars/{{ hs_workspace }}-{{ hs_archi }}

Add creation of nomad volumes directories

We got error like directory does not exist during the nomad installation. So we added the missing task to create to required volumes in roles/nomad/tasks/common/_install.yml

Define missing variable

• Added missing variable hs_infra_default_user in roles/common_vars/defaults/main.yml to avoid undefined variable
• Added missing variable hs_archi in playbooks/init.yml

Remove online tag on prepare step

We need to delete online tag on few prepare steps to be able to execute prepare phase even in offline mode in :

• roles/vault/tasks/debian/main.yml
• roles/consul/tasks/debian/main.yml
• roles/nomad/tasks/debian/main.yml
• roles/envoy/tasks/debian/main.yml

Split nomad installation

To be able to download docker images during online step and execute nomad group and user creation during offline mode we need to split in two differents steps

Add tf_action variable

In ̀roles/grafana/vars/main.yml`

## Fix load dashboards

In roles/grafana/files/tf_setup/files/dashboards/consul.json

Orig	modif
netdata_system_load_load_average{dimension="load1"}	avg(node_load1)
netdata_system_load_load_average{dimension="load5"}	avg(node_load5)
netdata_system_load_load_average{dimension="load15"}	avg(node_load15)

Change roles grafana labs to support RedHat Family OS

Change also installation method from package manager to binary

• loki
• grafana
• prometheus
• alloy

Add tasks to support Redhat Family

Install librsync-devel package in:

roles/stage1_pip/tasks/redhat/_prepare.yml

Uninstall chardet if installed via RPM and Install Python packages in:

roles/stage1_pip/tasks/redhat/_install.yml

Load br_netfilter module for Nomad network configuration and ensure Docker is started and enabled in:

roles/nomad/tasks/redhat/main.yml
roles/nomad/tasks/common/_configure.yml

Install skopeo in:

roles/nomad/tasks/common/_prepare.yml

Add local rpm repo

For the RHEL offline installation we added ansible tasks in playbooks/00_offline_prepare.yml to create a local rpm repo to be able to install all required packages and their dependencies