| |
| |
| This section will list some of the operations that we have to automate to deploy/maintain/migrate/decommission services and/or infrastructure components used in the CentOS infra. |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| Red Hat Enterprise Linux |
| We also started to deploy RHEL in parallel for some services that will have to stay longer available (starting from CentOS Stream, expected life time is ~5y, versus 10y for RHEL). |
| We point to an internal mirror for deployments and also use some TLS certs (subscription) to access Red Hat CDN to get updates (or point also to internal mirror) |
| |
| Deployed and maintained versions (in ansible roles and repositories) : |
| |
| * RHEL 8 |
| * RHEL 9 |
| |
| !!! note |
| See internal ansible inventory `group_vars/all` doc/notes/snippet around the `rhel_*` variables about how to use this, not covered in public documentation for obvious reasons |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| In the centos infra, we can either use internal mirror (to speed up updates accross internal fleet) or just use the classical Red Hat CDN. It's all defined through ansible (see note before about group_vars/host_vars) |
| The internal mirrors have distributed script/template that would fetch/reposync repositories for all used versions/releases/architectures |
| |
| |