aboutsummaryrefslogtreecommitdiff
path: root/secrets/terraform/terraform.tfstate (unfollow)
Commit message (Collapse)AuthorFilesLines
2020-11-18Remove default.nix and shell.nixEuAndreh1-0/+0
2020-11-18Remove most Nix filesEuAndreh1-0/+0
2020-11-16Add base Guix configuration and start switching to itEuAndreh1-0/+0
- remove NixOS stateVersion from .envrc; - add guix-reconfigure.sh; - add vps.scm with initial Guix system configuration; - update vps.tf to use the new "base-guix" snapshot. The "base-guix" image doesn't need a password. The "andreh" users has one, but it is configured for not requiring it when running commands as "sudo". The expected minimal steps one has to go through for privilege scalation is via the SSH private key, and accessing the VPS via SSH. Since password login is disabled and root can't login via SSH either, only the private SSH key allows access to the server. After that, the attacker will be able to run commands as root.
2020-11-16Forget existing resource to start working on new VPSEuAndreh1-0/+0
2020-11-02Update terraform generated filesEuAndreh1-0/+0
2020-09-06Refactor vps-configuration.nix: Split secrets from config and envsubst varsEuAndreh1-0/+0
Also rename thingTLD to thingDomain.
2020-08-29Build new VPS server from snapshot using bigger machineEuAndreh1-0/+0
I'm using the snapshot here because I don't have any backup system yet, ¯\_(ツ)_/¯ This should be reverted on vps.tf after applying, and I should get down to doing automatic backups.
2020-08-25Enable automatic backup for VPS serverEuAndreh1-0/+0
2020-08-23Chage DNS record of prosodyEuAndreh1-0/+0
Terraform file changes were du to me initially trying to do this via creating a SRV DNS record. However this is not required, because Prosody is already on the $TLD sever, the only difference being that it is listening on a different port.
2020-08-22Add songbooks documentation DNS addressEuAndreh1-0/+0
2020-08-22WIP: reenable prosody and matterbridge, and add PDFsEuAndreh1-0/+0
2020-08-16Use NGINX to handle the creation of certificates for prosodyEuAndreh1-0/+0
After a terraform state rm '...' of the server and domain configuration, and recreating the instance again while leaving the dettached one running. I had to do this because I was experimenting too much with the domain and I hit the Let's Encrypt rate limit[0]. Because of that I'll recreate certificates less often so that doesn't happen in the new domain during development of the VPS. I'm not shure if this solution works, but I'll commit just as a checkpoint. I had to change the custom DNS nameservers from Digital Ocean to Vultr and that may take a while, so it's worth having this as a checkpoint in time while I'm off to other things. [0]: https://letsencrypt.org/docs/rate-limits/
2020-08-15Use new image with ownership of /etc/nixos/configuration.nix by userEuAndreh1-0/+0
Useful reference: - https://discourse.nixos.org/t/can-i-move-etc-nixos-to-my-dotfiles-and-symlink-it-back-to-etc-nixos/4833/10
2020-08-14Add generated Terraform filesEuAndreh1-0/+0
2020-08-14Add updated terraform filesEuAndreh1-0/+0
2020-08-11vps-configuration.env.nix: Finish working Nextcloud installationEuAndreh1-0/+0
2020-08-10Update Terraform infrastructureEuAndreh1-0/+0
2020-08-10Semi working setup: Terraform and LetsEncrypt workingEuAndreh1-0/+0
2020-08-10Interactive Terraform plan -> apply cycleEuAndreh1-0/+0
2020-08-10WIP: Move to Vultr and NixOSEuAndreh1-0/+0