Repo to aid with reproducing issues on Nomad

Nomad infra

Cluster	Nomad Regions	Nomad Datacenters	Nomad Servers	Nomad Clients	Nomad Allocations	Nomad Total CPU	Nomad Total Memory	Consul Datacenters
Development	1	1	3	2	27	9 Ghz	16 GiB	1
Staging/UAT	2	4	9	7	190	200 Ghz	240 GiB	2
Production	2	3	9	17	410	2655 Ghz	1873 GiB	2

Ubuntu (22.04)
servers/masters -> Nomad + Consul + Unbound (service.consul) + Traefik + Keepalived (vm)
- keepalived for centralized UI and APP differentation.
  - example
    - nomad-development.company.com IN A -> nomad-development-keepalived-VIP *.nomad-development.company.com IN A -> nomad-development-keepalived-VIP
    - nomad-staging.company.com IN A -> nomad-staging-keepalived-VIP *.nomad-staging.company.com IN A -> nomad-staging-keepalived-VIP
    - nomad-production.company.com IN A -> nomad-production-keepalived-VIP *.nomad-production.company.com IN A -> nomad-production-keepalived-VIP
      - why?
      - OIDC / Oauth like to have consistent callback-URL's, so if you login to Nomad-UI with OIDC, and a server fails, users can just keep using same login-url.
      - in jobspecs, you can work with templated traefik-rules to more easily automate a staging-job to get templated to jobname.nomad-staging.company.com
      - and if you leverage Letsencrypt (natively supported by Traefik) less hassle with SSL certificates.
        
        of course, traefik also supports including custom certificates if preferred.
      - developers can work without too much manual interference needed.
Clients/Workers -> Nomad/Consul + Dockerd
- DNS resolvers set to masters (service.consul)
Job deployments through Gitlab CI/CD
- custom templating on gitlab-side & on nomad side with Levant