53 Commits

Author SHA1 Message Date
8b30257b65 Update mcr.microsoft.com/k8s/csi/smb-csi Docker tag to v1.17.0
All checks were successful
CI / Terraform fmt + validate (pull_request) Successful in 18s
CI / Nomad job spec validate (pull_request) Successful in 17s
2026-04-19 05:26:06 +00:00
1b73fda11f ci: add nomad job validate step (Phase 1c)
All checks were successful
CI / Nomad job spec validate (pull_request) Successful in 16s
CI / Terraform fmt + validate (pull_request) Successful in 20s
CI / Terraform fmt + validate (push) Successful in 23s
CI / Nomad job spec validate (push) Successful in 16s
2026-04-19 14:52:02 +10:00
96ddfa6ec5 Add Gitea act-runner and test actions for the repo
All checks were successful
CI / Terraform fmt + validate (push) Successful in 24s
2026-04-19 14:20:18 +10:00
6c0b1c9281 Add CI/CD plan documentation outlining phases for validation and deployment 2026-04-18 17:34:11 +10:00
4f2934411b Add GitHub token to renovate configuration so it can pull release notes 2026-04-18 16:45:35 +10:00
54dbe0c667 Cleanup diun references from ntfy, diun was never actually committed 2026-04-18 16:44:44 +10:00
a30e60b557 Add custom managers to renovate.json for Docker image updates 2026-04-18 16:15:18 +10:00
2536e855e5 Add renovate.json 2026-04-18 16:14:43 +10:00
dbe11dc8fa Add renovate and ntfy (unrelated) 2026-04-18 16:14:43 +10:00
b27f3e58ca Add openreader 2026-04-18 11:37:18 +10:00
5b3f2cf8f4 Move frigate into the cluster and enable GPU detector 2026-03-28 17:13:09 +11:00
68cf58ead5 Remove some state moved declarations 2025-11-20 23:45:56 +11:00
d3ac8a252b Make appdata and unraid smb modules to reduce duplication 2025-11-20 23:45:16 +11:00
961ec128f1 Setup prowlarr 2025-11-19 20:49:05 +11:00
d84eb73db0 Connect sonarr to deluge 2025-11-19 20:46:38 +11:00
a3eaab5a07 Add deluge bittorrent client 2025-11-10 23:27:50 +11:00
cf4daacab5 Add jellyfin
Customised the forward auth in authelia to ignore Authorization headers as authelia was failing to parse the Authorization: MediaBrowser headers that jellyfin uses.
2025-11-10 19:24:21 +11:00
2a038e59e8 Add sonarr 2025-11-09 15:51:25 +11:00
443d614a66 Use the hostname for the unraid server rather than the IP 2025-11-07 19:42:38 +11:00
2803f694e8 Add Unifi Network 2025-11-06 19:30:42 +11:00
bbff0f6692 Remove no longer used teams status job 2025-10-22 22:05:25 +11:00
bd815e96c6 Complete migration now that the changes are applied. 2025-10-22 22:04:37 +11:00
92f60a7572 Sort parts of the nomad intra into folders
This should make finding things easier
2025-10-22 22:02:25 +11:00
8869bd1cb2 Stop terraform thinking some settings are inconsistent 2025-10-21 21:47:39 +11:00
c473ef68fd Move secrets to subfolder 2025-10-21 21:47:05 +11:00
fd7fdd00f3 Add more nodes and update them all to the latest config 2025-10-21 21:41:15 +11:00
8a375c0133 Remove noauth setting that is probably wrong as it doesn't work 2025-10-21 21:35:35 +11:00
7302842add Refine install instructions 2025-10-20 20:52:15 +11:00
50cddcd033 Add new servers to consul retry join list 2025-10-20 20:51:22 +11:00
307f27a10b Use ipv4 for nomad and consul to reduce connectivity issues on my network 2025-10-20 20:25:58 +11:00
700046cfd1 Remove unused strip-magic-token middleware from Traefik configuration 2025-10-20 20:25:19 +11:00
c3f25d1bc5 Prepare for adding LattePanda IOTA based cluster hosts.
This required the option to configure the available CPU as nomad assumes that we should only allocate based on the base CPU frequency but the N150 reports 800Mhz base but has a boost frequency more than 4x higher at 3.6GHz which would leave the CPU under utilised, instead we allocate at 1.8GHz (x4 cores).
2025-10-15 23:38:11 +11:00
cdf2a6b72e Fixup: traefik requires the certificate secrets to be protected. Made the same change on the actual storage host. 2025-10-14 22:07:48 +11:00
f7c2752192 Explicitly enable info level logging in traefik to make it easier to find where to set it to debug level 2025-10-14 20:11:56 +11:00
bf98fb84be Make sure there is always one traefik instance up when changing the configuration so we can still access nomad via it 2025-10-14 20:11:22 +11:00
2d931f9cfa Add kopia access to traefik 2025-10-14 20:10:32 +11:00
3cc6f4827d Route to traefik directly via localhost due to in ability to route via the external port 2025-10-13 21:23:42 +11:00
4a5ad8d36b Setup storage of immich resouces and add tailscale access to allow uploading files too large to go over cloudflare (<100mb) See https://github.com/immich-app/immich/issues/17729 and https://github.com/immich-app/immich/pull/22385 2025-10-13 20:48:35 +11:00
e6c6bb0c91 Resolve authelia via consul service mesh 2025-10-12 20:58:58 +11:00
e2562ce4b0 Add an immich server 2025-10-11 14:54:07 +11:00
b53bfe4075 Improve resilience by changing routing to traefik and setting up more servers.
Some changes were required to set 3 VMs as the cluster since the NUC failed and we are waiting for new hardware to arrive.
The ingest routing from the internet was changed to use cloudflared tunnel to traefik instead of via a specific host.
2025-10-11 14:46:06 +11:00
7f3161b2bb Add magic token domain for hass to allow app access 2025-10-04 14:36:58 +10:00
facc3c64b2 Route frigate.othrayte.one to internal frigate instance 2025-10-04 14:18:16 +10:00
d64fec4bc0 Disable Nomad job configuration for Teams Status 2025-10-04 13:58:44 +10:00
8b234b8322 Fix bug in teams status (was fixed long ago) 2025-10-04 13:51:26 +10:00
612d322c4f Update session management settings for Authelia: extend inactivity duration, adjust expiration time, and set remember_me period. 2025-10-04 13:46:56 +10:00
bc2bd41018 Add camera locations and IP addresses to documentation 2025-10-04 13:46:29 +10:00
786b2c6670 Switch from tailscale authkeys to an oauth client to fix issues with key expirey 2025-09-06 22:17:24 +10:00
2d497e5195 Add Nomad job configuration for Teams Status Updater 2025-06-07 09:07:48 +10:00
8920be1ba0 Fix use of wrong port in traefik service definition 2025-05-28 00:05:46 +10:00
525e0eaf9f Bind consul on all interfaces to fix issues with nodes finding each other. Also expose traefik to make diagnostics of issues easier. 2025-05-27 23:23:25 +10:00
234c6d075c Improve use of postgres from terraform 2025-05-27 23:22:08 +10:00
e52c2c1fc5 Ensure that host volumes are only on the required nodes 2025-05-25 23:35:36 +10:00
62 changed files with 2958 additions and 450 deletions

54
.gitea/workflows/ci.yml Normal file
View File

@@ -0,0 +1,54 @@
name: CI
on:
pull_request:
push:
branches:
- main
jobs:
terraform-validate:
name: Terraform fmt + validate
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: hashicorp/setup-terraform@v3
- name: fmt check — 1-nixos-node
run: terraform fmt -check -recursive
working-directory: 1-nixos-node
- name: fmt check — 2-nomad-config
run: terraform fmt -check -recursive
working-directory: 2-nomad-config
- name: validate — 2-nomad-config (no backend)
run: |
terraform init -backend=false
terraform validate
working-directory: 2-nomad-config
nomad-validate:
name: Nomad job spec validate
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Install Nomad CLI
run: |
curl -fsSL https://apt.releases.hashicorp.com/gpg | gpg --dearmor -o /usr/share/keyrings/hashicorp.gpg
. /etc/os-release
echo "deb [signed-by=/usr/share/keyrings/hashicorp.gpg] https://apt.releases.hashicorp.com ${VERSION_CODENAME} main" | tee /etc/apt/sources.list.d/hashicorp.list
apt-get update && apt-get install -y nomad
- name: Validate all job specs
env:
NOMAD_ADDR: http://jaglan-beta-m20.lan:4646
run: |
find 2-nomad-config -name '*.nomad.hcl' | while read f; do
echo "==> $f"
nomad job validate "$f"
done

5
0-hardware/cameras.md Normal file
View File

@@ -0,0 +1,5 @@
| Location | IP |
| --------- | ------------- |
| Doorbell | 192.168.1.108 |
| Side Gate | 192.168.1.109 |
| Ceiling | 192.168.1.110 |

View File

@@ -1,10 +1,10 @@
# Server Names
From https://namingschemes.com/Hitchhikers_Guide_Planets.
Was [google sheet](https://docs.google.com/spreadsheets/d/1aLb14nme9gprYzWOEo5IadG1vKEtf8VQk0tkCBGz9KA/edit#gid=0).
| Name | Hostname | Use |
| ---------------------- | ----------------------- | ------------------------------------------------------------- |
| --------------------- | ----------------------- | ----------------------------------------------------------------------- |
| Antares | antares-win | Win11 PC |
| Barteldan | barteldan-linux | Fedora Linux PC |
| Betelgeuse Seven | betelgeuse-seven-unraid | Unraid Storage Host |
@@ -17,7 +17,14 @@ Was [google sheet](https://docs.google.com/spreadsheets/d/1aLb14nme9gprYzWOEo5Ia
| Golgafrincham | | |
| Han Wavel | | |
| Jaglan Beta | | Compute Cluster Hosts |
| - Moon 1 | jaglan-beta-m01 | NUC Cluster Host<br>(Intel \| N5105 [4/4], 16GB ram, 1TB SSD) |
| - Moon 1 | jaglan-beta-m01 | NUC Cluster Host<br>(Intel \| N5105 [4/4], 16GB, 1TB SSD) |
| - Moon 2 | jaglan-beta-m02 | LattePanda IOTA Cluster Host<br>(Intel \| N150 [4/4], 16GB, 129GB eMMC) |
| - Moon 3 | jaglan-beta-m03 | LattePanda IOTA Cluster Host<br>(Intel \| N150 [4/4], 16GB, 129GB eMMC) |
| - Moon 4 | jaglan-beta-m04 | LattePanda IOTA Cluster Host<br>(Intel \| N150 [4/4], 16GB, 129GB eMMC) |
| - Moon 5 | jaglan-beta-m05 | LattePanda IOTA Cluster Host<br>(Intel \| N150 [4/4], 16GB, 129GB eMMC) |
| - Moon 20 | jaglan-beta-m20 | VM on unraid |
| - Moon 21 | jaglan-beta-m21 | VM on unraid |
| - Moon 22 | jaglan-beta-m22 | VM on unraid |
| - Moon 42 (Test moon) | jaglan-beta-m42 | Test server (VM) |
| Krikkit | | |
| Lamuella | | |

View File

@@ -1,13 +1,45 @@
Follow steps at https://nixos.org/manual/nixos/stable/#sec-installation-manual-summary
Note: run `lsblk` to determine correct device to install to.
Set the hostname when editing `/mnt/etc/nixos/configuration.nix` before the installation:
`networking.hostName = "jaglan-beta-mNN";`
Once `nixos-install` is complete and the root password is set and saved edit the `/etc/nixos/configuration.nix` again to enable ssh and run `nixos-rebuild switch`
Determine the correct device to install to
```sh
lsblk
```
services.openssh.enable = true;
services.openssh.settings.PermitRootLogin = "yes";
Steps based on https://nixos.org/manual/nixos/stable/#sec-installation-manual-summary for UEFI
```sh
parted /dev/mmcblk0 -- mklabel gpt
# If there is an existing OS you will have to type yes to proceed
parted /dev/mmcblk0 -- mkpart root ext4 512MB -8GB
parted /dev/mmcblk0 -- mkpart swap linux-swap -8GB 100%
parted /dev/mmcblk0 -- mkpart ESP fat32 1MB 512MB
parted /dev/mmcblk0 -- set 3 esp on
mkfs.ext4 -L nixos /dev/mmcblk0p1
mkswap -L swap /dev/mmcblk0p2
swapon /dev/mmcblk0p2
mkfs.fat -F 32 -n boot /dev/mmcblk0p3
mount /dev/disk/by-label/nixos /mnt
mkdir -p /mnt/boot
mount -o umask=077 /dev/disk/by-label/boot /mnt/boot
nixos-generate-config --root /mnt
nano /mnt/etc/nixos/configuration.nix
# Set hostname networking.hostName = "jaglan-beta-mNN";
nixos-install
# Set the root password
reboot
nano /etc/nixos/configuration.nix
# Enable ssh access
# services.openssh.enable = true;
# services.openssh.settings.PermitRootLogin = "yes";
nixos-rebuild switch
```
If starting from older nixos, upgrade the OS
```sh
nix-channel --list
nix-channel --add https://channels.nixos.org/nixos-25.05 nixos
nix-channel --list
nixos-rebuild switch --upgrade
```

View File

@@ -0,0 +1,8 @@
Avoid applying change to multiple hosts at once as this could take done the cluster, instead apply to each on one at a time
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m02"]'
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m03"]'
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m04"]'
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m05"]'
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m20"]'
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m21"]'
terraform apply -target='null_resource.deploy_nixos["jaglan-beta-m22"]'

View File

@@ -46,19 +46,25 @@
};
client = {
enabled = true;
preferred_address_family = "ipv4";
%{if cpu_total_compute != null ~}
cpu_total_compute = ${cpu_total_compute};
%{endif ~}
%{if node_class != null ~}
node_class = "${node_class}";
%{endif ~}
host_volume = {
traefik = {
path = "/opt/traefik";
read_only = false;
};
postgres = {
path = "/opt/postgres";
%{ for volume in host_volumes ~}
${volume} = {
path = "/opt/${volume}";
read_only = false;
};
%{ endfor ~}
};
cni_path = "$${pkgs.cni-plugins}/bin";
};
plugin.docker.config.allow_privileged = true;
plugin.docker.config.volumes.enabled = true;
};
extraPackages = with pkgs; [
cni-plugins
@@ -70,14 +76,22 @@
webUi = true;
interface.bind = "${bind_interface}";
interface.advertise = "${bind_interface}";
forceAddrFamily = "ipv4";
extraConfig = {
client_addr = "{{ GetPrivateInterfaces | exclude \"type\" \"ipv6\" | join \"address\" \" \" }} {{ GetAllInterfaces | include \"flags\" \"loopback\" | join \"address\" \" \" }}";
%{if bootstrap ~}
bootstrap_expect = 1;
%{endif ~}
server = true;
retry_join = [
"jaglan-beta-m01"
"jaglan-beta-m02"
"jaglan-beta-m03"
"jaglan-beta-m04"
"jaglan-beta-m05"
"jaglan-beta-m20"
"jaglan-beta-m21"
"jaglan-beta-m22"
];
datacenter = "jaglan-beta";
connect.enabled = true;
@@ -93,16 +107,71 @@
systemd.tmpfiles.rules = [
# Fix issue where nomad needs alloc_mounts to be writable
"d /var/lib/alloc_mounts 0755 root root -"
# Create a directory for Traefik to store its data (tls certs, etc.)
"d /opt/traefik 0755 root root -"
# Create a directory for Postgres to store its data
"d /opt/postgres 0755 root root -"
%{ for volume in host_volumes ~}
# Create a directory for ${volume} to store its data
"d /opt/${volume} 0755 root root -"
%{ endfor ~}
];
# Open ports in the firewall. 80/443 are for HTTP/HTTPS (terraform), 464X are the default ports for Nomad, 830X are the default ports for Consul.
networking.firewall.allowedTCPPorts = [ 80 443 4646 4647 4648 8300 8301 ];
networking.firewall.allowedTCPPorts = [ 80 443 8081 4646 4647 4648 8300 8301 8500 ];
networking.firewall.allowedUDPPorts = [ 8301 ];
# Ensure Docker daemon is available (Nomad enableDocker only configures Nomad, does not guarantee docker service)
virtualisation.docker.enable = true;
%{if node_class == "latte-panda-n150" ~}
# Enable Intel iGPU (N150 UHD Graphics) for OpenVINO / VA-API workloads running in Docker
hardware.graphics = {
enable = true;
extraPackages = with pkgs; [
intel-media-driver # VA-API (iHD)
intel-compute-runtime # OpenCL / oneAPI
];
};
%{endif ~}
# Proper systemd service definition for macvlan network creation
systemd.services.docker-macvlan-network = {
description = "Ensure macvlan Docker network exists";
after = [ "network-online.target" "docker.service" ];
wants = [ "network-online.target" "docker.service" ];
wantedBy = [ "multi-user.target" ];
serviceConfig = {
Type = "oneshot";
};
# Provide required binaries in PATH
path = [ pkgs.docker pkgs.bash pkgs.coreutils pkgs.iproute2 pkgs.gnugrep ];
script = ''
set -euo pipefail
NET_NAME=macvlan
if docker network inspect "$NET_NAME" >/dev/null 2>&1; then
echo "Docker network $NET_NAME already exists"
exit 0
fi
echo "Creating Docker macvlan network $NET_NAME on interface ${bind_interface}"
# We intentionally do NOT use --ip-range here to avoid allocating the
# same reserved pool on every host (which could lead to collisions if
# multiple macvlan containers are started across nodes). Instead, we
# give critical services (like UniFi) an explicit static IP via the
# Nomad job (Docker static assignment) and rely on manual DHCP
# reservations to prevent conflicts.
#
# If you later need multiple macvlan-assigned containers per host,
# consider one of these strategies:
# 1. Per-host distinct network name + ip-range slice (macvlan-m01, ...)
# 2. Parameterize an ip-range per host in Terraform and template here
# 3. Keep a registry of allocated static IPs in Consul KV / Nomad vars
docker network create -d macvlan \
--subnet=192.168.1.0/24 \
--gateway=192.168.1.1 \
-o parent=${bind_interface} \
"$NET_NAME"
echo "Docker macvlan network $NET_NAME created"
'';
restartIfChanged = false; # Don't rerun just because comment changed
};
# Copy the NixOS configuration file and link it from the resulting system
# (/run/current-system/configuration.nix). This is useful in case you
# accidentally delete configuration.nix.

View File

@@ -14,12 +14,15 @@ terraform {
provider "template" {}
variable "nodes" {
description = "Map of nodes with host, password, and bind interface"
description = "Map of nodes with host, password, bind interface, and host volumes"
type = map(object({
host = string
password = string
bind_interface = string
bootstrap = optional(bool, false) # Optional field for bootstrap nodes
cpu_total_compute = optional(number, null) # Optional field for CPU total compute
node_class = optional(string, null) # Optional Nomad node_class for scheduling constraints
host_volumes = list(string)
}))
}
@@ -29,6 +32,9 @@ locals {
hostname = v.host
bind_interface = v.bind_interface
bootstrap = v.bootstrap
cpu_total_compute = v.cpu_total_compute
node_class = v.node_class
host_volumes = v.host_volumes
})
}
}

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long

19
2-nomad-config/1-data.tf Normal file
View File

@@ -0,0 +1,19 @@
module "data" {
source = "./1-data"
}
data "nomad_plugin" "smb" {
plugin_id = "smb"
wait_for_healthy = true
}
provider "postgresql" {
host = "jaglan-beta-m21.lan"
port = 5432
database = "postgres"
username = "postgres"
password = data.sops_file.secrets.data["postgres.postgres"]
sslmode = "disable"
connect_timeout = 15
}

View File

@@ -6,7 +6,7 @@ job "csi-smb" {
driver = "docker"
config {
image = "mcr.microsoft.com/k8s/csi/smb-csi:v1.7.0"
image = "mcr.microsoft.com/k8s/csi/smb-csi:v1.17.0"
args = [
"--v=5",
"--nodeid=${attr.unique.hostname}",

View File

@@ -0,0 +1,5 @@
resource "nomad_job" "csi-smb" {
jobspec = file("${path.module}/csi-smb.nomad.hcl")
}

View File

@@ -0,0 +1,25 @@
terraform {
required_providers {
sops = {
source = "carlpett/sops"
version = "~> 0.5"
}
postgresql = {
source = "cyrilgdn/postgresql"
}
}
}
provider "nomad" {
address = "http://jaglan-beta-m20.lan:4646"
}
data "sops_file" "secrets" {
source_file = "secrets/secrets.enc.json"
}
data "nomad_plugin" "smb" {
plugin_id = "smb"
wait_for_healthy = true
}

View File

@@ -13,7 +13,10 @@ job "postgres" {
driver = "docker"
config {
image = "postgres:latest"
# Temporarily pin to v17 as v18 moved the default data directory and immich doesn't officially support it yet
# immich also needs >= 0.3.0, < 0.5.0. https://docs.immich.app/administration/postgres-standalone/#prerequisites
#image = "postgres:17"
image = "tensorchord/vchord-postgres:pg17-v0.4.3"
ports = ["db"]
volumes = [
@@ -35,7 +38,7 @@ job "postgres" {
resources {
cpu = 500
memory = 512
memory = 1024
}
template {

View File

@@ -0,0 +1,26 @@
resource "nomad_job" "postgres" {
jobspec = file("${path.module}/postgres.nomad.hcl")
rerun_if_dead = true
}
resource "nomad_job" "pgadmin" {
jobspec = file("${path.module}/pgadmin.nomad.hcl")
}
resource "nomad_job" "pgbackup" {
jobspec = file("${path.module}/pgbackup.nomad.hcl")
}
resource "nomad_variable" "postgres" {
path = "nomad/jobs/postgres"
items = {
postgres_password = data.sops_file.secrets.data["postgres.postgres"]
}
}
module "unraid_smb_database_dump" {
source = "../modules/unraid_smb"
name = "dump"
id = "unraid_database_dump"
share = "database-dump"
}

View File

@@ -0,0 +1,4 @@
module "ingress" {
source = "./2-ingress"
}

View File

@@ -28,6 +28,10 @@ job "authelia" {
"traefik.enable=true",
]
connect {
sidecar_service {}
}
check {
type = "http"
path = "/health"
@@ -59,6 +63,12 @@ job "authelia" {
data = <<EOF
server:
address: tcp://0.0.0.0:{{ env "NOMAD_PORT_http" }}/
endpoints:
authz:
forward-auth:
implementation: 'ForwardAuth'
authn_strategies:
- name: 'CookieSession'
theme: "auto"
identity_validation:
reset_password:
@@ -77,7 +87,9 @@ access_control:
session:
name: authelia_session
secret: "{{ with nomadVar "nomad/jobs/authelia" }}{{ .session_secret }}{{ end }}"
expiration: "1 day and 9 hours"
inactivity: '2 days and 9 hours'
expiration: '1 hour'
remember_me: '90 days'
cookies:
- domain: othrayte.one
authelia_url: "https://auth.othrayte.one"

View File

@@ -1,5 +1,5 @@
resource "nomad_job" "authelia" {
jobspec = file("authelia.nomad.hcl")
jobspec = file("${path.module}/authelia.nomad.hcl")
}
resource "postgresql_role" "authelia" {

View File

@@ -0,0 +1,33 @@
terraform {
required_providers {
sops = {
source = "carlpett/sops"
version = "~> 0.5"
}
cloudflare = {
source = "cloudflare/cloudflare"
version = "~> 5"
}
postgresql = {
source = "cyrilgdn/postgresql"
}
}
}
provider "nomad" {
address = "http://jaglan-beta-m20.lan:4646"
}
data "sops_file" "secrets" {
source_file = "secrets/secrets.enc.json"
}
provider "cloudflare" {
api_token = data.sops_file.secrets.data["cloudflare.api_token"]
}
data "nomad_plugin" "smb" {
plugin_id = "smb"
wait_for_healthy = true
}

View File

@@ -0,0 +1,252 @@
job "traefik" {
group "traefik" {
count = 2
network {
mode = "bridge"
port "http" {
static = 80
}
port "https" {
static = 443
}
port "api" {
static = 8081
}
}
service {
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "auth"
local_bind_port = 9091
}
}
}
}
}
service {
name = "traefik"
port = "api"
check {
name = "alive"
type = "tcp"
port = "api"
interval = "10s"
timeout = "2s"
}
}
task "traefik" {
driver = "docker"
config {
image = "traefik:v3.3"
ports = ["http", "https", "api"]
volumes = [
"local/traefik.yml:/etc/traefik/traefik.yml",
"local/configs/:/etc/traefik/configs/"
]
}
volume_mount {
volume = "unraid_appdata_traefik"
destination = "/opt/traefik"
read_only = false
}
template {
data = <<EOF
log:
level: INFO
entryPoints:
web:
address: ":80"
http:
redirections:
entryPoint:
to: websecure
scheme: https
websecure:
address: ":443"
http:
tls:
certResolver: letsencrypt
traefik:
address: ":8081"
api:
dashboard: true
insecure: true
providers:
file:
directory: "/etc/traefik/configs/"
consulCatalog:
prefix: "traefik"
exposedByDefault: false
defaultRule: {{"Host(`{{ .Name }}.othrayte.one`)"}}
endpoint:
address: "{{ env "NOMAD_HOST_IP_http" }}:8500"
scheme: "http"
certificatesResolvers:
letsencrypt:
acme:
email: "othrayte@gmail.com"
storage: "/opt/traefik/acme.json"
httpChallenge:
entryPoint: web
EOF
destination = "local/traefik.yml"
}
template {
data = <<EOF
http:
serversTransports:
ignorecert:
insecureSkipVerify: true
middlewares:
auth:
forwardAuth:
address: "http://localhost:9091/api/authz/forward-auth"
trustForwardHeader: true
auth-allow-token:
chain:
middlewares:
- auth
inject-kopia-basic-auth:
headers:
customRequestHeaders:
Authorization: "Basic {{ with nomadVar "nomad/jobs/traefik" }}{{ .kopia_basic_auth }}{{ end }}"
routers:
fallback:
rule: "HostRegexp(`^.+$`)"
entryPoints:
- websecure
middlewares:
- auth
service: noop@internal # This router just applies middleware
priority: 1
traefik:
rule: "Host(`traefik.othrayte.one`)"
service: traefik
middlewares:
- auth
nomad-ui:
rule: "Host(`nomad.othrayte.one`)"
service: nomad-ui
middlewares:
- auth
consul-ui:
rule: "Host(`consul.othrayte.one`)"
service: consul-ui
middlewares:
- auth
unraid:
rule: "Host(`unraid.othrayte.one`)"
service: unraid
middlewares:
- auth
kopia:
rule: "Host(`kopia.othrayte.one`)"
service: kopia
middlewares:
- auth
- inject-kopia-basic-auth
hass:
rule: "Host(`hass.othrayte.one`)"
service: hass
middlewares:
- auth
hass-token:
rule: "Host(`${hass_magic_token}-hass.othrayte.one`)"
service: hass
unifi-network:
rule: "Host(`network.othrayte.one`)"
service: unifi-network
middlewares:
- auth
services:
traefik:
loadBalancer:
servers:
- url: "http://localhost:8081"
nomad-ui:
loadBalancer:
servers:
- url: "http://{{ env "NOMAD_HOST_IP_http" }}:4646"
consul-ui:
loadBalancer:
servers:
- url: "http://{{ env "NOMAD_HOST_IP_http" }}:8500"
unraid:
loadBalancer:
servers:
- url: "http://betelgeuse-seven-unraid.lan:80"
kopia:
loadBalancer:
servers:
- url: "http://betelgeuse-seven-unraid.lan:51515"
hass:
loadBalancer:
servers:
- url: "http://192.168.1.234:8123"
unifi-network:
loadBalancer:
serversTransport: ignorecert
servers:
- url: "https://192.168.1.50:8443"
EOF
destination = "local/configs/nomad.yml"
}
resources {
cpu = 100
memory = 128
}
}
volume "unraid_appdata_traefik" {
type = "csi"
read_only = false
source = "unraid_appdata_traefik"
access_mode = "multi-node-multi-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["file_mode=0600", "uid=1000", "gid=1000"]
}
}
task "cloudflared" {
driver = "docker"
config {
image = "cloudflare/cloudflared:latest"
args = [
"tunnel", "--no-autoupdate", "run"
]
}
template {
data = <<EOH
TUNNEL_TOKEN="{{ with nomadVar "nomad/jobs/traefik" }}{{ .cf_tunnel_token }}{{ end }}"
EOH
destination = "secrets/tunnel.env"
env = true # Load the file as environment variables
}
}
}
}

View File

@@ -0,0 +1,39 @@
resource "cloudflare_dns_record" "othrayte-one" {
comment = "othrayte.one proxy via cloudflared tunnel to traefik"
zone_id = "2616ab2a44d0645b03fbc3106c79bd99"
type = "CNAME"
name = "othrayte.one"
content = "59ca3eb1-5f0b-45e1-97ff-e373569c6689.cfargotunnel.com"
proxied = true
ttl = 1 # Auto
}
resource "cloudflare_dns_record" "star-othrayte-one" {
comment = "*.othrayte.one proxy via cloudflared tunnel to traefik"
zone_id = "2616ab2a44d0645b03fbc3106c79bd99"
type = "CNAME"
name = "*.othrayte.one"
content = "59ca3eb1-5f0b-45e1-97ff-e373569c6689.cfargotunnel.com"
proxied = true
ttl = 1 # Auto
}
resource "nomad_variable" "traefik" {
path = "nomad/jobs/traefik"
items = {
cf_tunnel_token = data.sops_file.secrets.data["traefik.cf_tunnel_token"]
kopia_basic_auth = data.sops_file.secrets.data["traefik.kopia_basic_auth"]
}
}
resource "nomad_job" "traefik" {
jobspec = templatefile("${path.module}/traefik.nomad.hcl", {
hass_magic_token = nonsensitive(data.sops_file.secrets.data["hass.magic-token"])
})
}
module "appdata_traefik" {
source = "../modules/appdata"
name = "traefik"
access_mode = "multi-node-multi-writer"
}

View File

@@ -0,0 +1,66 @@
job "act-runner" {
group "act-runner" {
network {
mode = "bridge"
}
# Consul Connect upstream to Gitea so the runner can register and receive jobs
service {
name = "act-runner"
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "code-connect"
local_bind_port = 3000
}
}
}
}
}
task "act-runner" {
driver = "docker"
config {
image = "gitea/act_runner:latest"
volumes = ["/var/run/docker.sock:/var/run/docker.sock"]
}
env = {
GITEA_INSTANCE_URL = "https://gitea-1ef0bea6b75a4fd3e9393a9f7f7e4b02.othrayte.one"
CONFIG_FILE = "/secrets/runner-config.yml"
}
# Required SOPS key:
# act-runner.registration_token — runner registration token from Gitea
# Admin → Settings → Actions → Runners → Create new runner
template {
data = <<EOF
GITEA_RUNNER_REGISTRATION_TOKEN={{ with nomadVar "nomad/jobs/act-runner" }}{{ .registration_token }}{{ end }}
EOF
destination = "secrets/runner.env"
env = true
}
# Limit which images/labels the runner will accept so it doesn't pick up
# unrelated workloads if more runners are added later.
template {
data = <<EOF
runner:
labels:
- "ubuntu-latest:docker://node:20-bookworm"
- "ubuntu-22.04:docker://node:20-bookworm"
- "ubuntu-24.04:docker://node:20-bookworm"
EOF
destination = "secrets/runner-config.yml"
}
resources {
cpu = 200
memory = 256
memory_max = 1024
}
}
}
}

View File

@@ -0,0 +1,10 @@
resource "nomad_job" "act_runner" {
jobspec = file("act-runner.nomad.hcl")
}
resource "nomad_variable" "act_runner" {
path = "nomad/jobs/act-runner"
items = {
registration_token = data.sops_file.secrets.data["act-runner.registration_token"]
}
}

View File

@@ -0,0 +1,145 @@
job "deluge" {
group "deluge" {
network {
mode = "bridge"
port "http" {
to = 8112
}
}
task "wireguard" {
driver = "docker"
lifecycle {
hook = "prestart"
sidecar = true
}
config {
image = "thrnz/docker-wireguard-pia:latest"
privileged = true
ports = ["http"]
}
env {
LOC = "aus_melbourne"
LOCAL_NETWORK = "192.168.1.0/24"
# PORT_FORWARDING = "1" # TODO: Find a way to tell deluge the forwarded port, the wireguard container outputs it /pia-shared/port.dat
}
template {
data = <<EOH
USER="{{ with nomadVar "nomad/jobs/deluge" }}{{ .pia_user }}{{ end }}"
PASS="{{ with nomadVar "nomad/jobs/deluge" }}{{ .pia_pass }}{{ end }}"
EOH
destination = "secrets/pia.env"
env = true # Load the file as environment variables
}
resources {
cpu = 50
memory = 32
}
}
# Service for Traefik (external ingress)
service {
name = "deluge"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.deluge.middlewares=auth@file",
]
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
# Service for Consul Connect (internal mesh communication)
service {
name = "deluge-api"
port = "http"
address_mode = "alloc" # Use allocation IP for Connect as the sidecar can't access the host's published port (hairpin/loopback NAT issue)
# tags = [
# "traefik.enable=false",
# ]
connect {
sidecar_service {
//tags = ["traefik.enable=false"]
}
}
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
task "deluge" {
driver = "docker"
config {
image = "lscr.io/linuxserver/deluge:latest"
network_mode = "container:wireguard-${NOMAD_ALLOC_ID}" # Share namespace with VPN
}
env {
PUID = "1000"
PGID = "1000"
TZ = "Australia/Melbourne"
}
volume_mount {
volume = "unraid_appdata_deluge"
destination = "/config"
read_only = false
}
volume_mount {
volume = "unraid_media_deluge"
destination = "/data/downloads"
read_only = false
}
resources {
cpu = 400
memory = 2048
memory_max = 3000
}
}
volume "unraid_appdata_deluge" {
type = "csi"
read_only = false
source = "unraid_appdata_deluge"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
volume "unraid_media_deluge" {
type = "csi"
read_only = false
source = "unraid_media_deluge"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
}
}

24
2-nomad-config/deluge.tf Normal file
View File

@@ -0,0 +1,24 @@
resource "nomad_job" "deluge" {
jobspec = file("deluge.nomad.hcl")
}
resource "nomad_variable" "deluge" {
path = "nomad/jobs/deluge"
items = {
pia_user = data.sops_file.secrets.data["pia.user"]
pia_pass = data.sops_file.secrets.data["pia.pass"]
}
}
module "appdata_deluge" {
source = "./modules/appdata"
name = "deluge"
}
module "unraid_smb_deluge_media" {
source = "./modules/unraid_smb"
name = "deluge"
share = "media"
subDir = "downloads"
}

View File

@@ -0,0 +1,216 @@
job "frigate" {
# Pin to N150 LattePanda nodes - Intel UHD iGPU for OpenVINO-accelerated detection.
# hardware.graphics (intel-compute-runtime) is deployed to these nodes via configuration.nix.
constraint {
attribute = "${node.class}"
value = "latte-panda-n150"
}
group "frigate" {
count = 1
network {
port "http" {
to = 5000
}
}
# Prestart: restore Frigate's SQLite DB from the Litestream file replica on the CIFS share.
# Runs to completion before the frigate task starts. Safe on first boot (-if-replica-exists
# is a no-op when no replica exists yet).
task "litestream-restore" {
lifecycle {
hook = "prestart"
sidecar = false
}
driver = "docker"
config {
image = "litestream/litestream:0.5.9"
command = "restore"
args = ["-if-replica-exists", "-config", "/local/litestream.yml", "/alloc/data/frigate.db"]
}
# Litestream config: replicate to /config/frigate.db.litestream/ on the CIFS share.
# Litestream writes its own segment format - no SQLite advisory locking involved.
# Frigate must be configured with database.path: /alloc/data/frigate.db in config.yml.
template {
data = <<EOH
dbs:
- path: /alloc/data/frigate.db
replicas:
- url: file:///config/frigate.db.litestream
EOH
destination = "local/litestream.yml"
}
volume_mount {
volume = "unraid_appdata_frigate"
destination = "/config"
read_only = false
}
resources {
cpu = 100
memory = 64
memory_max = 256
}
}
# Sidecar: continuously stream WAL changes from /alloc/data/frigate.db to the CIFS replica.
# Runs alongside frigate for the lifetime of the allocation.
task "litestream-replicate" {
lifecycle {
hook = "poststart"
sidecar = true
}
driver = "docker"
config {
image = "litestream/litestream:0.5"
command = "replicate"
args = ["-config", "/local/litestream.yml"]
}
template {
data = <<EOH
dbs:
- path: /alloc/data/frigate.db
replicas:
- url: file:///config/frigate.db.litestream
EOH
destination = "local/litestream.yml"
}
volume_mount {
volume = "unraid_appdata_frigate"
destination = "/config"
read_only = false
}
resources {
cpu = 100
memory = 64
memory_max = 256
}
}
task "frigate" {
driver = "docker"
config {
image = "ghcr.io/blakeblackshear/frigate:0.17.1"
ports = ["http"]
privileged = true
# Shared memory for inter-process frame buffers (frigate forks detector processes).
shm_size = 268435456 # 256 MiB
# Large tmpfs for decoded frame cache - avoids wearing out any storage.
mounts = [
{
type = "tmpfs"
target = "/tmp/cache"
readonly = false
tmpfs_options = {
size = 1000000000 # 1 GiB in bytes
}
}
]
# Intel iGPU render node - Frigate's bundled OpenVINO runtime auto-detects
# GPU device and uses it for object detection without any extra env vars.
# Requires hardware.graphics.enable = true on the NixOS node (N150 nodes).
devices = [
{
host_path = "/dev/dri/renderD128"
container_path = "/dev/dri/renderD128"
}
]
}
# RTSP password injected from Nomad variables (sourced from sops secrets).
# Reference in config.yml as: {FRIGATE_RTSP_PASSWORD}
template {
data = <<EOH
FRIGATE_RTSP_PASSWORD="{{ with nomadVar "nomad/jobs/frigate" }}{{ .rtsp_password }}{{ end }}"
EOH
destination = "secrets/frigate.env"
env = true
}
service {
name = "frigate"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.frigate.middlewares=auth@file",
"traefik.http.routers.frigate-token.rule=Host(`n7gdph5cuh7bd1cakbq8s099rvrv3qhs-frigate.othrayte.one`)",
]
check {
name = "alive"
type = "http"
path = "/api/version"
port = "http"
interval = "10s"
timeout = "5s"
}
}
env {
TZ = "Australia/Melbourne"
}
# config.yml lives here (read from CIFS). SQLite DB is at /alloc/data/frigate.db
# (local NVMe, managed by Litestream). Requires in config.yml:
# database:
# path: /alloc/data/frigate.db
volume_mount {
volume = "unraid_appdata_frigate"
destination = "/config"
read_only = false
}
# Recordings, clips, and exports.
volume_mount {
volume = "unraid_media_frigate"
destination = "/media/frigate"
read_only = false
}
resources {
# GPU handles inference; CPU manages stream ingestion, motion detection, and recording.
cpu = 2000
memory = 2048
}
}
volume "unraid_appdata_frigate" {
type = "csi"
read_only = false
source = "unraid_appdata_frigate"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["nobrl", "uid=0", "gid=0"]
}
}
volume "unraid_media_frigate" {
type = "csi"
read_only = false
source = "unraid_media_frigate"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["nobrl", "uid=0", "gid=0"]
}
}
}
}

23
2-nomad-config/frigate.tf Normal file
View File

@@ -0,0 +1,23 @@
resource "nomad_job" "frigate" {
jobspec = file("frigate.nomad.hcl")
}
resource "nomad_variable" "frigate" {
path = "nomad/jobs/frigate"
items = {
rtsp_password = data.sops_file.secrets.data["frigate.rtsp_password"]
}
}
module "appdata_frigate" {
source = "./modules/appdata"
name = "frigate"
}
module "unraid_smb_frigate_media" {
source = "./modules/unraid_smb"
name = "frigate"
share = "media"
subDir = "frigate"
}

View File

@@ -27,6 +27,8 @@ job "gitea" {
tags = [
"traefik.enable=true",
"traefik.http.routers.gitea.middlewares=auth@file",
# Token subdomain — no auth middleware — used by act_runner step containers for git checkout
"traefik.http.routers.gitea-token.rule=Host(`gitea-1ef0bea6b75a4fd3e9393a9f7f7e4b02.othrayte.one`)",
]
check {
@@ -37,6 +39,17 @@ job "gitea" {
}
}
# Separate service for Consul Connect ingress (address_mode=alloc avoids hairpin NAT issue)
service {
name = "code-connect"
port = "http"
address_mode = "alloc"
connect {
sidecar_service {}
}
}
task "gitea" {
driver = "docker"
@@ -112,8 +125,9 @@ EOF
}
env = {
TS_AUTHKEY = "${ts_authkey}"
TS_HOSTNAME = "git"
TS_AUTHKEY = "${ts_oauthsecret}?ephemeral=true"
TS_EXTRA_ARGS = "--advertise-tags=tag:nomad"
}
resources {

View File

@@ -9,7 +9,7 @@ resource "cloudflare_dns_record" "git-othrayte-one" {
resource "nomad_job" "gitea" {
jobspec = templatefile("gitea.nomad.hcl", {
ts_authkey = data.sops_file.secrets.data["tailscale.auth_key"]
ts_oauthsecret = data.sops_file.secrets.data["tailscale.oauthsecret"]
})
}
@@ -33,28 +33,7 @@ resource "postgresql_database" "gitea" {
owner = postgresql_role.gitea.name
}
resource "nomad_csi_volume_registration" "unraid_appdata_gitea" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it
depends_on = [data.nomad_plugin.smb]
plugin_id = "smb"
volume_id = "unraid_appdata_gitea"
name = "unraid_appdata_gitea"
external_id = "unraid_appdata_gitea"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/appdata"
subDir = "gitea" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
module "appdata_gitea" {
source = "./modules/appdata"
name = "gitea"
}

View File

@@ -0,0 +1,209 @@
job "immich" {
group "immich" {
network {
mode = "bridge"
port "http" {
to = 2283
}
}
service {
connect {
sidecar_service {
proxy {
# TODO https://docs.immich.app/administration/postgres-standalone#prerequisites
upstreams {
destination_name = "postgres"
local_bind_port = 5432
}
}
}
}
}
task "immich-server" {
driver = "docker"
config {
image = "ghcr.io/immich-app/immich-server:release"
ports = ["http"]
}
service {
name = "immich"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.immich.middlewares=auth@file",
"traefik.http.routers.immich-token.rule=Host(`c3ll7nbevl5j4j8rcnfxnr95q48fuayz-immich.othrayte.one`)",
]
# Use the Immich server ping endpoint
# See: https://api.immich.app/endpoints/server/pingServer
check {
name = "alive"
type = "http"
path = "/api/server/ping"
method = "GET"
port = "http"
interval = "10s"
timeout = "2s"
}
}
env {
IMMICH_LOG_LEVEL = "log"
TZ = "Australia/Melbourne"
REDIS_HOSTNAME = "localhost"
DB_HOSTNAME = "localhost"
IMMICH_IGNORE_MOUNT_CHECK_ERRORS = "true" # Let immich start whilst we figure out what's wrong with the mount permissions
}
volume_mount {
volume = "unraid_appdata_immich"
destination = "/data"
read_only = false
}
volume_mount {
volume = "unraid_media_photosvideos"
destination = "/data/library"
read_only = false
}
volume_mount {
volume = "unraid_media_immich_encodedvideo"
destination = "/data/encoded-video"
read_only = false
}
volume_mount {
volume = "unraid_mediadump_photosvideos"
destination = "/data/upload"
read_only = false
}
resources {
cpu = 200
memory = 512
memory_max = 1500
}
template {
data = <<EOH
DB_USERNAME="immich"
DB_PASSWORD="{{ with nomadVar "nomad/jobs/immich" }}{{ .database_pw }}{{ end }}"
EOH
destination = "secrets/db.env"
env = true # Load the file as environment variables
}
}
volume "unraid_appdata_immich" {
type = "csi"
read_only = false
source = "unraid_appdata_immich"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
volume "unraid_media_photosvideos" {
type = "csi"
read_only = false
source = "unraid_media_photosvideos"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
volume "unraid_media_immich_encodedvideo" {
type = "csi"
read_only = false
source = "unraid_media_immich_encodedvideo"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
volume "unraid_mediadump_photosvideos" {
type = "csi"
read_only = false
source = "unraid_mediadump_photosvideos"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
task "immich-machine-learning" {
driver = "docker"
config {
image = "ghcr.io/immich-app/immich-machine-learning:release"
}
resources {
cpu = 200
memory = 500
memory_max = 2000
}
}
task "redis" {
driver = "docker"
config {
image = "docker.io/valkey/valkey:8-bookworm@sha256:fea8b3e67b15729d4bb70589eb03367bab9ad1ee89c876f54327fc7c6e618571"
}
resources {
cpu = 30
memory = 10
memory_max = 50
}
}
task "tailscale" {
driver = "docker"
config {
image = "tailscale/tailscale:latest"
}
env = {
TS_HOSTNAME = "immich"
TS_EXTRA_ARGS = "--advertise-tags=tag:nomad"
}
resources {
cpu = 100
memory = 100
memory_max = 300
}
lifecycle {
hook = "prestart"
sidecar = true
}
template {
data = <<EOH
TS_AUTHKEY="{{ with nomadVar "nomad/jobs/immich" }}{{ .tailscale_oauthsecret }}{{ end }}?ephemeral=true"
EOH
destination = "secrets/ts_oauth.env"
env = true # Load the file as environment variables
}
}
}
}

62
2-nomad-config/immich.tf Normal file
View File

@@ -0,0 +1,62 @@
resource "nomad_job" "immich" {
jobspec = file("immich.nomad.hcl")
}
resource "postgresql_role" "immich" {
name = "immich"
password = data.sops_file.secrets.data["immich.database_pw"]
login = true
}
resource "postgresql_database" "immich" {
name = "immich"
owner = postgresql_role.immich.name
}
resource "postgresql_extension" "immich_vchord" {
name = "vchord"
database = postgresql_database.immich.name
create_cascade = true
}
resource "postgresql_extension" "immich_earthdistance" {
name = "earthdistance"
database = postgresql_database.immich.name
create_cascade = true
}
resource "nomad_variable" "immich" {
path = "nomad/jobs/immich"
items = {
database_pw = data.sops_file.secrets.data["immich.database_pw"]
tailscale_oauthsecret = data.sops_file.secrets.data["tailscale.oauthsecret"]
}
}
module "appdata_immich" {
source = "./modules/appdata"
name = "immich"
}
module "unraid_smb_immich_photosvideos" {
source = "./modules/unraid_smb"
name = "photosvideos"
share = "media"
subDir = "Photos and Videos"
}
module "unraid_smb_immich_encodedvideo" {
source = "./modules/unraid_smb"
name = "immich_encodedvideo"
share = "media"
subDir = "immich/encoded-video"
}
module "unraid_smb_immich_mediadump_photosvideos" {
source = "./modules/unraid_smb"
name = "photosvideos"
id = "unraid_mediadump_photosvideos"
share = "media-dump"
subDir = "Photos and Videos"
}

View File

@@ -0,0 +1,88 @@
job "jellyfin" {
group "jellyfin" {
count = 1
network {
port "http" {
to = 8096
}
}
task "jellyfin" {
driver = "docker"
config {
image = "lscr.io/linuxserver/jellyfin:latest"
ports = ["http"]
}
service {
name = "jellyfin"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.jellyfin.middlewares=auth@file",
"traefik.http.routers.jellyfin-token.rule=Host(`c3ll7nbevl5j4j8rcnfxnr95q48fuayz-jellyfin.othrayte.one`)",
]
check {
name = "alive"
type = "tcp"
port = "http"
interval = "10s"
timeout = "2s"
}
}
env {
PUID = 1000
PGID = 1000
TZ = "Australia/Melbourne"
JELLYFIN_PublishedServerUrl = "https://jellyfin.othrayte.one"
}
volume_mount {
volume = "unraid_appdata_jellyfin"
destination = "/config"
read_only = false
}
volume_mount {
volume = "unraid_media_jellyfin"
destination = "/data"
read_only = false
}
resources {
cpu = 500
memory = 2048
}
}
volume "unraid_appdata_jellyfin" {
type = "csi"
read_only = false
source = "unraid_appdata_jellyfin"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
volume "unraid_media_jellyfin" {
type = "csi"
read_only = false
source = "unraid_media_jellyfin"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["nobrl", "uid=1000", "gid=1000"]
}
}
}
}

View File

@@ -0,0 +1,15 @@
resource "nomad_job" "jellyfin" {
jobspec = file("jellyfin.nomad.hcl")
}
module "appdata_jellyfin" {
source = "./modules/appdata"
name = "jellyfin"
}
module "unraid_smb_jellyfin_media" {
source = "./modules/unraid_smb"
name = "jellyfin"
share = "media"
}

View File

@@ -21,22 +21,24 @@ terraform {
}
provider "nomad" {
address = "http://jaglan-beta-m01.lan:4646"
address = "http://jaglan-beta-m20.lan:4646"
}
data "sops_file" "secrets" {
source_file = "secrets.enc.json"
source_file = "secrets/secrets.enc.json"
}
provider "cloudflare" {
api_token = data.sops_file.secrets.data["cloudflare.api_token"]
}
resource "nomad_job" "csi-smb" {
jobspec = file("csi-smb.nomad.hcl")
resource "nomad_scheduler_config" "config" {
memory_oversubscription_enabled = true
preemption_config = {
batch_scheduler_enabled = false
service_scheduler_enabled = false
sysbatch_scheduler_enabled = false
system_scheduler_enabled = false
}
}
data "nomad_plugin" "smb" {
plugin_id = "smb"
wait_for_healthy = true
}

View File

@@ -0,0 +1,62 @@
terraform {
required_providers {
sops = {
source = "carlpett/sops"
version = "~> 0.5"
}
}
}
variable "name" {
description = "Name of the application, also used as subdir on the unraid appdata share"
type = string
}
variable "id" {
description = "ID to use for the volume registration, defaults to name with - replaced by _"
type = string
default = null
}
variable "access_mode" {
description = "CSI volume access mode"
type = string
default = "single-node-writer"
validation {
condition = contains(["single-node-writer", "multi-node-multi-writer"], var.access_mode)
error_message = "access_mode must be either 'single-node-writer' or 'multi-node-multi-writer'"
}
}
data "nomad_plugin" "smb" {
plugin_id = "smb"
wait_for_healthy = true
}
data "sops_file" "secrets" {
source_file = "secrets/secrets.enc.json"
}
resource "nomad_csi_volume_registration" "this" {
depends_on = [data.nomad_plugin.smb]
plugin_id = "smb"
volume_id = var.id != null ? var.id : "unraid_appdata_${replace(var.name, "-", "_")}"
name = var.id != null ? var.id : "unraid_appdata_${replace(var.name, "-", "_")}"
external_id = var.id != null ? var.id : "unraid_appdata_${replace(var.name, "-", "_")}"
capability {
access_mode = var.access_mode
attachment_mode = "file-system"
}
context = {
source = "//betelgeuse-seven-unraid.lan/appdata"
subDir = var.name
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
}

View File

@@ -0,0 +1,62 @@
terraform {
required_providers {
sops = {
source = "carlpett/sops"
version = "~> 0.5"
}
}
}
variable "name" {
description = "Name of the volume registration"
type = string
}
variable "id" {
description = "ID to use for the volume registration, defaults to name with - replaced by _"
type = string
default = null
}
variable "share" {
description = "Name of the SMB share on the unraid server"
type = string
}
variable "subDir" {
description = "Subdirectory within the SMB share"
type = string
default = null
}
data "nomad_plugin" "smb" {
plugin_id = "smb"
wait_for_healthy = true
}
data "sops_file" "secrets" {
source_file = "secrets/secrets.enc.json"
}
resource "nomad_csi_volume_registration" "this" {
depends_on = [data.nomad_plugin.smb]
plugin_id = "smb"
volume_id = var.id != null ? var.id : "unraid_${var.share}_${replace(var.name, "-", "_")}"
name = var.id != null ? var.id : "unraid_${var.share}_${replace(var.name, "-", "_")}"
external_id = var.id != null ? var.id : "unraid_${var.share}_${replace(var.name, "-", "_")}"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = merge({
source = "//betelgeuse-seven-unraid.lan/${var.share}"
}, var.subDir == null ? {} : { "subDir" = var.subDir })
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
}

View File

@@ -0,0 +1,89 @@
job "ntfy" {
group "ntfy" {
network {
mode = "bridge"
port "http" {
to = 80
}
}
# Consul Connect sidecar with upstream to postgres
service {
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "postgres"
local_bind_port = 5432
}
}
}
}
}
service {
name = "ntfy"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.ntfy.middlewares=auth@file",
# Token subdomain bypasses Authelia — ntfy's own token auth is sufficient for API access
"traefik.http.routers.ntfy-token.rule=Host(`ntfy-2e30e5869ab6bfde4961012b48761a9b.othrayte.one`)",
]
check {
type = "http"
path = "/healthz"
interval = "10s"
timeout = "2s"
}
}
# Users and tokens are provisioned declaratively via auth-users / auth-tokens in server.yml.
# ntfy reads and applies them on every startup — no poststart task, no race conditions.
#
# Bcrypt hashes are not secrets and are hardcoded below (same as /etc/shadow — safe to commit).
# Generate with: docker run --rm -it binwiederhier/ntfy user hash
# or: echo "mypassword" | docker run --rm -i binwiederhier/ntfy user hash
# Required SOPS keys:
# ntfy.database_pw — postgres password for the ntfy role
task "ntfy" {
driver = "docker"
config {
image = "binwiederhier/ntfy:latest"
ports = ["http"]
command = "serve"
volumes = [
"local/server.yml:/etc/ntfy/server.yml",
]
}
env = {
TZ = "Australia/Melbourne"
}
template {
data = <<EOF
base-url: "https://ntfy.othrayte.one"
listen-http: ":80"
database-url: "postgres://ntfy:{{ with nomadVar "nomad/jobs/ntfy" }}{{ .database_pw }}{{ end }}@localhost:5432/ntfy"
auth-default-access: "deny-all"
behind-proxy: true
enable-login: true
auth-users:
- "admin:$2a$10$rLp4qagJnsA8Es5hQlISH.WrlzwMrXE2MBaEgz7zdd2lkAVu30lMy:admin"
EOF
destination = "local/server.yml"
}
resources {
cpu = 50
memory = 64
memory_max = 128
}
}
}
}

21
2-nomad-config/ntfy.tf Normal file
View File

@@ -0,0 +1,21 @@
resource "nomad_job" "ntfy" {
jobspec = file("ntfy.nomad.hcl")
}
resource "nomad_variable" "ntfy" {
path = "nomad/jobs/ntfy"
items = {
database_pw = data.sops_file.secrets.data["ntfy.database_pw"]
}
}
resource "postgresql_role" "ntfy" {
name = "ntfy"
password = data.sops_file.secrets.data["ntfy.database_pw"]
login = true
}
resource "postgresql_database" "ntfy" {
name = "ntfy"
owner = postgresql_role.ntfy.name
}

View File

@@ -0,0 +1,116 @@
job "openreader" {
group "openreader" {
network {
mode = "bridge"
port "http" {
to = 3003
}
}
# Consul Connect sidecar with upstream to postgres
service {
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "postgres"
local_bind_port = 5432
}
}
}
}
}
service {
name = "openreader"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.openreader.middlewares=auth@file",
]
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
service {
name = "openreader-api"
port = "http"
address_mode = "alloc" # Use allocation IP for Connect as the sidecar can't access the host's published port (hairpin/loopback NAT issue)
connect {
sidecar_service {}
}
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
task "openreader" {
driver = "docker"
config {
image = "ghcr.io/richardr1126/openreader:v2.1.2"
ports = ["http"]
}
env = {
TZ = "Australia/Melbourne"
# Use embedded SeaweedFS for blob storage (data lives in /app/docstore/seaweedfs).
# Port 8333 is not exposed; browser uploads/downloads fall back through the app API.
USE_EMBEDDED_WEED_MINI = "true"
S3_ENDPOINT = "http://localhost:8333"
S3_FORCE_PATH_STYLE = "true"
# Auth is intentionally disabled (no BASE_URL / AUTH_SECRET set).
# Access is controlled by the Authelia middleware on the Traefik router above.
# To enable server-side library import from an Unraid share, add a second CSI volume
# mount for the share (e.g. unraid_media_books → /app/docstore/library:ro) and set:
# IMPORT_LIBRARY_DIR = "/app/docstore/library"
}
template {
data = <<EOF
POSTGRES_URL=postgresql://openreader:{{ with nomadVar "nomad/jobs/openreader" }}{{ .database_pw }}{{ end }}@localhost:5432/openreader
EOF
destination = "secrets/openreader.env"
env = true
}
volume_mount {
volume = "unraid_appdata_openreader"
destination = "/app/docstore"
read_only = false
}
resources {
cpu = 200
memory = 750
memory_max = 1024
}
}
volume "unraid_appdata_openreader" {
type = "csi"
read_only = false
source = "unraid_appdata_openreader"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
}
}

View File

@@ -0,0 +1,26 @@
resource "nomad_job" "openreader" {
jobspec = file("openreader.nomad.hcl")
}
resource "nomad_variable" "openreader" {
path = "nomad/jobs/openreader"
items = {
database_pw = data.sops_file.secrets.data["openreader.database_pw"]
}
}
resource "postgresql_role" "openreader" {
name = "openreader"
password = data.sops_file.secrets.data["openreader.database_pw"]
login = true
}
resource "postgresql_database" "openreader" {
name = "openreader"
owner = postgresql_role.openreader.name
}
module "appdata_openreader" {
source = "./modules/appdata"
name = "openreader"
}

View File

@@ -1,54 +0,0 @@
resource "nomad_job" "postgres" {
jobspec = file("postgres.nomad.hcl")
}
resource "nomad_job" "pgadmin" {
jobspec = file("pgadmin.nomad.hcl")
}
resource "nomad_job" "pgbackup" {
jobspec = file("pgbackup.nomad.hcl")
}
resource "nomad_variable" "postgres" {
path = "nomad/jobs/postgres"
items = {
postgres_password = data.sops_file.secrets.data["postgres.postgres"]
}
}
provider "postgresql" {
#host = "jaglan-beta-m01.othrayte.one"
host = "192.168.1.235"
port = 5432
database = "postgres"
username = "postgres"
password = data.sops_file.secrets.data["postgres.postgres"]
sslmode = "disable"
connect_timeout = 15
}
resource "nomad_csi_volume_registration" "unraid_database_dump" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it
depends_on = [data.nomad_plugin.smb]
plugin_id = "smb"
volume_id = "unraid_database_dump"
name = "unraid_database_dump"
external_id = "unraid_database_dump"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/database-dump"
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
}

View File

@@ -0,0 +1,119 @@
job "prowlarr" {
group "prowlarr" {
network {
mode = "bridge"
port "http" {
to = 9696
}
}
service {
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "postgres"
local_bind_port = 5432
}
upstreams {
destination_name = "sonarr-api"
local_bind_port = 8989
}
}
}
}
}
service {
name = "prowlarr"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.prowlarr.middlewares=auth@file",
]
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
service {
name = "prowlarr-api"
port = "http"
address_mode = "alloc" # Use allocation IP for Connect as the sidecar can't access the host's published port (hairpin/loopback NAT issue)
connect {
sidecar_service {}
}
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
task "prowlarr" {
driver = "docker"
config {
image = "lscr.io/linuxserver/prowlarr:latest"
ports = ["http"]
}
env {
PUID = 1000
PGID = 1000
TZ = "Australia/Melbourne"
# https://wiki.servarr.com/prowlarr/postgres-setup
# Disable internal auth to use Traefik + Authelia
PROWLARR__AUTH__REQUIRED = "Enabled"
PROWLARR__AUTH__METHOD = "External"
PROWLARR__POSTGRES__USER = "prowlarr"
PROWLARR__POSTGRES__HOST = "localhost"
PROWLARR__POSTGRES__PORT = "5432"
PROWLARR__POSTGRES__MAINDB = "prowlarr-main"
PROWLARR__POSTGRES__LOGDB = "prowlarr-log"
}
volume_mount {
volume = "unraid_appdata_prowlarr"
destination = "/config"
read_only = false
}
resources {
cpu = 150
memory = 512
}
template {
data = <<EOH
PROWLARR__POSTGRES__PASSWORD="{{ with nomadVar "nomad/jobs/prowlarr" }}{{ .database_pw }}{{ end }}"
EOH
destination = "secrets/db.env"
env = true # Load the file as environment variables
}
}
volume "unraid_appdata_prowlarr" {
type = "csi"
read_only = false
source = "unraid_appdata_prowlarr"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
}
}

View File

@@ -0,0 +1,32 @@
resource "nomad_job" "prowlarr" {
jobspec = file("prowlarr.nomad.hcl")
}
resource "nomad_variable" "prowlarr" {
path = "nomad/jobs/prowlarr"
items = {
database_pw = data.sops_file.secrets.data["prowlarr.database_pw"]
}
}
# https://wiki.servarr.com/prowlarr/postgres-setup
resource "postgresql_role" "prowlarr" {
name = "prowlarr"
password = data.sops_file.secrets.data["prowlarr.database_pw"]
login = true
}
resource "postgresql_database" "prowlarr_main" {
name = "prowlarr-main"
owner = postgresql_role.prowlarr.name
}
resource "postgresql_database" "prowlarr_log" {
name = "prowlarr-log"
owner = postgresql_role.prowlarr.name
}
module "appdata_prowlarr" {
source = "./modules/appdata"
name = "prowlarr"
}

View File

@@ -1,7 +1,11 @@
# Terraform State
Mount the state on the fileshare to 2-nomad-config/.tfstate/
`sudo mount -t cifs //192.168.1.192/appdata/terraform /home/othrayte/Code/infra/2-nomad-config/.tfstate/ -o rw,username=othrayte,password=<pw>,uid=$(id -u),gid=$(id -g)`
`sudo mount -t cifs //betelgeuse-seven-unraid.lan/appdata/terraform /home/othrayte/Code/infra/2-nomad-config/.tfstate/ -o rw,username=othrayte,password=<pw>,uid=$(id -u),gid=$(id -g)`
# Tailscale Oauth Client
We use a Tailscale oauth client secret to allow our containers to connect to tailscale. We created an oauth client called `nomad` with the `auth_keys` (write) scope for the tag `nomad` and stored the secret in our secrets file.
# Secrets
@@ -11,4 +15,18 @@ Put the age keys in /home/<user>/.config/sops/age/keys.txt
## Adding Secrets
Edit the secrets using `sops secrets.enc.json`
Edit the secrets using `sops secrets/secrets.enc.json`
# Bootstrapping (starting without PostgreSQL running)
terraform apply -target=module.data
terraform apply -target=module.ingress
## Restoring PostgreSQL DBs
`psql -h jaglan-beta-m21 -p 5432 -U postgres -f ~/Downloads/all_databases.sql postgres`
# Deploying and testing changes
Sometimes the nomad job fails but the solution is to fix another job and so we need to tell nomad to retry the unchanged job.
`nomad job eval -force-reschedule glance`

View File

@@ -0,0 +1,67 @@
job "renovate" {
type = "batch"
periodic {
cron = "0 4 * * *" # Daily at 4am
prohibit_overlap = true
}
group "renovate" {
network {
mode = "bridge"
}
# Consul Connect sidecar with upstream to Gitea (service: code-connect, port 3000)
service {
name = "renovate"
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "code-connect"
local_bind_port = 3000
}
}
}
}
}
task "renovate" {
driver = "docker"
config {
image = "renovate/renovate:latest"
}
env = {
RENOVATE_PLATFORM = "gitea"
RENOVATE_ENDPOINT = "http://localhost:3000"
RENOVATE_GIT_URL = "endpoint"
RENOVATE_REPOSITORIES = "othrayte/infra"
RENOVATE_GIT_AUTHOR = "Renovate Bot <renovate@othrayte.one>"
LOG_LEVEL = "debug"
}
# Required SOPS key:
# renovate.gitea_token — PAT for the renovate bot account in Gitea
# Create a dedicated 'renovate' user in Gitea with these token scopes:
# repo (read+write), user (read), issue (read+write), organization (read)
# renovate.github_token — read-only GitHub PAT (any account) for
# fetching changelogs and avoiding github.com API rate limits
template {
data = <<EOF
RENOVATE_TOKEN={{ with nomadVar "nomad/jobs/renovate" }}{{ .gitea_token }}{{ end }}
RENOVATE_GITHUB_COM_TOKEN={{ with nomadVar "nomad/jobs/renovate" }}{{ .github_token }}{{ end }}
EOF
destination = "secrets/renovate.env"
env = true
}
resources {
cpu = 500
memory = 512
memory_max = 1024
}
}
}
}

View File

@@ -0,0 +1,11 @@
resource "nomad_job" "renovate" {
jobspec = file("renovate.nomad.hcl")
}
resource "nomad_variable" "renovate" {
path = "nomad/jobs/renovate"
items = {
gitea_token = data.sops_file.secrets.data["renovate.gitea_token"]
github_token = data.sops_file.secrets.data["renovate.github_token"]
}
}

View File

@@ -1,38 +0,0 @@
{
"unraid": {
"nomad": "ENC[AES256_GCM,data:FCGEs+XCSuunLxVPyzE=,iv:j8Ey+l8iJiPY7CbE5IoT0ZgNklnv+4odSZkorJQ/nr8=,tag:7PoizENid+vgWC/eb5MOaQ==,type:str]"
},
"cloudflare": {
"api_token": "ENC[AES256_GCM,data:445wM+3yHRnMfiAHuBg3dWzLA3jB0dpNBaHrxl1bb036sFZnzN+gOg==,iv:g8tMdxY8XFTPA2W8/RtMtDhnyCzNLY6dJDWWC2ZeIZQ=,tag:04uf/y3DWY3HIXOJ2HenJw==,type:str]",
"direct_ip6": "ENC[AES256_GCM,data:E/V1pFjBp7c0PRhUa4cxqAVl8xZKsZzn,iv:Gw0qz2x1pMaieZaCcp4dD9sEVtQfcuEqRP3UpA2Bj/0=,tag:LpsPH3cJAlPCFX6EPabWnQ==,type:str]"
},
"tailscale": {
"auth_key": "ENC[AES256_GCM,data:gzh4nqEOQLijp5DTGHHSn0aO1mFQUB3sVSdAVDLG+a2H6XJ0BtJJGU55oLJURy7E/um7gzwDofP5mwZGTA==,iv:yl8lHqnNLB2AXlBfMyw/0CAR7+KmyKKDFc7kxbo9S6c=,tag:CunYd62x3omji6ozqmhgOg==,type:str]"
},
"authelia": {
"session_secret": "ENC[AES256_GCM,data:eSpAwX/KPzed/Y0oi6QvBwB7Gv5Kiml4FJS5RyuJ7A0plAd8acNThNXi3H4=,iv:RmH0wB3smlSF+CYs4x1w2V9ixdxgdav4dAQntjO0S5g=,tag:Vo5eHiU+1/dep/IUryN/XQ==,type:str]",
"jwt_secret": "ENC[AES256_GCM,data:XGDV2+SbMPYxhzv8S/6SjfA0MZeelRNjgIR10+qcTFYs2IW+IZjkCExLpQ==,iv:hv1b2Dddm21vObwQBUb3LZFfYjAkIm2/ZE1Syt3//YI=,tag:TojRWFctm1H72oPfq62Y2g==,type:str]",
"encryption_key": "ENC[AES256_GCM,data:D5F7eScWxCQ8G7pU8khi8aj8/p8ZKSErROhrqKS569fYUQpsHt6+3QQfeH7/naMvJ45r/5oVGCGeeFcEqlY0lEnbFLJEZ/tSOcm4RcIigPcx4a+8H7s=,iv:sf+TdLzacFaDgYjYhw4RKExLu6XfpewKiklt/q7VVzw=,tag:Zu3kCJfCZ7ae7HneXF6jVA==,type:str]",
"database_pw": "ENC[AES256_GCM,data:w5TmJwjeFa8tgTXDBI7doNfbBnDBUoWyZ0Qetp4M5JpwyRv06kAj2sAKOCY=,iv:rJubsGeyxSXkOxyTjzTo1GJRgLNWbAIMy1sS74MiuHc=,tag:Sbi4gVZgRcJLriTxm2ebeQ==,type:str]"
},
"postgres": {
"postgres": "ENC[AES256_GCM,data:lKuLcVTuUbfrlVhRdCs=,iv:TsbtAbXYTysxuiCi08F0hJsgoolzzgE2EPdFdPMQ+NQ=,tag:9oNua06hHdeCzE7nB22c0g==,type:str]"
},
"gitea": {
"internal_token": "ENC[AES256_GCM,data:teIsV+6nUPWO9/amas3FmK6uv44YEZNpV780ncTwUkQDygDvQRr7A3KEbk/rYFcTjfxK6Kw8nmqi0rBrcBNX1bSVNg8jwfYHhY2TxFMgCo4tkQxLf3eSBUhlPGsfpsskACKIPnZ1RQ2m,iv:NAKPw0YVNtLlyEp7wld9ml4zQlVxo/takiOid6YQlfA=,tag:QIk+USh8MLZDzJkQsglJ+w==,type:str]",
"jwt_secret": "ENC[AES256_GCM,data:/dPDqJdn4Af3Wo005V7lU9b8RbN/wyF0Tx66827cdyaZfi4QPOSj23wNqw==,iv:yJW2PiAGGr97q0DoBr64X88eFNpuVPZX0SPyNDp5QjQ=,tag:p27XTUbMC0WDMTNJCscmGQ==,type:str]",
"database_pw": "ENC[AES256_GCM,data:EzGPKdsX3Ib2zWrz09kUdegIxGNwg1j4msbOKUmvCGy6R9/EG1nvOC9Z5Oo=,iv:msek112FxmVAwFume6b7RnSICL/sw5CK3XzgCq9Sp1s=,tag:UcxUi2hySv54liN+Ddodpw==,type:str]"
},
"sops": {
"age": [
{
"recipient": "age1zuj9ssd0kkfeefjmyz82t9h3dfjq8degqm2l5eszhu5zhazpgsys54rq2n",
"enc": "-----BEGIN AGE ENCRYPTED FILE-----\nYWdlLWVuY3J5cHRpb24ub3JnL3YxCi0+IFgyNTUxOSByUWM4ZDVVbGFrUGdMRHBX\nUFBmU3Nlc0RBSzhFK0tHNHpkQXUvUVdiZUZJCmpRN1lFdENpWW0rcThjVlVQNUl6\nWnlLU0RnQ3FZby81Ly8xTFBrek9nMncKLS0tIFQ4UTRNOC9CRmx4OFJWem1wckZz\nUDFTSzdWZldFK3FqcTNWTWRyNDhHQ2MKS811mR5xn7qiC/aVgPFYJ5c6Q3zxRfcr\nHcvxUvB01vNJKZpRg92vvKPkV6lQO3DXCT98OdfwiymlEOvYxg71Pg==\n-----END AGE ENCRYPTED FILE-----\n"
}
],
"lastmodified": "2025-05-25T11:33:40Z",
"mac": "ENC[AES256_GCM,data:HvNkVDm3HOcYSAvDNFs0/w/QmiKFTTy0d+Onl/pFXEgdH/bBLqbeOwZV0tsaZYwNJluOH8EiU4gSBZ5EaCh4JrUTpHiiug4p5UXgRSva9sZ5D+9vzvfncqTdQVXKL6gdLMRVJQjz8lZVx0jV1czFES+4AECNgSq7lNRUHhau3eU=,iv:K33uicZwQyscLr1DUEAKLWPkFSH+aIntyceKB1KTu+M=,tag:mrTSWWlv5ZkN4K4HuIE/zw==,type:str]",
"encrypted_regex": "^(.*)$",
"version": "3.10.2"
}
}

View File

@@ -0,0 +1,74 @@
{
"unraid": {
"nomad": "ENC[AES256_GCM,data:FCGEs+XCSuunLxVPyzE=,iv:j8Ey+l8iJiPY7CbE5IoT0ZgNklnv+4odSZkorJQ/nr8=,tag:7PoizENid+vgWC/eb5MOaQ==,type:str]"
},
"cloudflare": {
"api_token": "ENC[AES256_GCM,data:445wM+3yHRnMfiAHuBg3dWzLA3jB0dpNBaHrxl1bb036sFZnzN+gOg==,iv:g8tMdxY8XFTPA2W8/RtMtDhnyCzNLY6dJDWWC2ZeIZQ=,tag:04uf/y3DWY3HIXOJ2HenJw==,type:str]",
"direct_ip6": "ENC[AES256_GCM,data:E/V1pFjBp7c0PRhUa4cxqAVl8xZKsZzn,iv:Gw0qz2x1pMaieZaCcp4dD9sEVtQfcuEqRP3UpA2Bj/0=,tag:LpsPH3cJAlPCFX6EPabWnQ==,type:str]"
},
"tailscale": {
"oauthsecret": "ENC[AES256_GCM,data:c2GtA+FaDcAKqUtQquP35W650lo1soivNCJc7KzCoQws0hTkt3zICFomOArhIfpHQMnCG4SpNvnXalarKKKxVw==,iv:Pnf8+9wBGNooPl4sKX5aGXITQt7/qfpn+mWyKk8YLXo=,tag:mXL+bz0gESj18qjpdksldA==,type:str]"
},
"authelia": {
"session_secret": "ENC[AES256_GCM,data:eSpAwX/KPzed/Y0oi6QvBwB7Gv5Kiml4FJS5RyuJ7A0plAd8acNThNXi3H4=,iv:RmH0wB3smlSF+CYs4x1w2V9ixdxgdav4dAQntjO0S5g=,tag:Vo5eHiU+1/dep/IUryN/XQ==,type:str]",
"jwt_secret": "ENC[AES256_GCM,data:XGDV2+SbMPYxhzv8S/6SjfA0MZeelRNjgIR10+qcTFYs2IW+IZjkCExLpQ==,iv:hv1b2Dddm21vObwQBUb3LZFfYjAkIm2/ZE1Syt3//YI=,tag:TojRWFctm1H72oPfq62Y2g==,type:str]",
"encryption_key": "ENC[AES256_GCM,data:D5F7eScWxCQ8G7pU8khi8aj8/p8ZKSErROhrqKS569fYUQpsHt6+3QQfeH7/naMvJ45r/5oVGCGeeFcEqlY0lEnbFLJEZ/tSOcm4RcIigPcx4a+8H7s=,iv:sf+TdLzacFaDgYjYhw4RKExLu6XfpewKiklt/q7VVzw=,tag:Zu3kCJfCZ7ae7HneXF6jVA==,type:str]",
"database_pw": "ENC[AES256_GCM,data:w5TmJwjeFa8tgTXDBI7doNfbBnDBUoWyZ0Qetp4M5JpwyRv06kAj2sAKOCY=,iv:rJubsGeyxSXkOxyTjzTo1GJRgLNWbAIMy1sS74MiuHc=,tag:Sbi4gVZgRcJLriTxm2ebeQ==,type:str]"
},
"postgres": {
"postgres": "ENC[AES256_GCM,data:lKuLcVTuUbfrlVhRdCs=,iv:TsbtAbXYTysxuiCi08F0hJsgoolzzgE2EPdFdPMQ+NQ=,tag:9oNua06hHdeCzE7nB22c0g==,type:str]"
},
"gitea": {
"internal_token": "ENC[AES256_GCM,data:teIsV+6nUPWO9/amas3FmK6uv44YEZNpV780ncTwUkQDygDvQRr7A3KEbk/rYFcTjfxK6Kw8nmqi0rBrcBNX1bSVNg8jwfYHhY2TxFMgCo4tkQxLf3eSBUhlPGsfpsskACKIPnZ1RQ2m,iv:NAKPw0YVNtLlyEp7wld9ml4zQlVxo/takiOid6YQlfA=,tag:QIk+USh8MLZDzJkQsglJ+w==,type:str]",
"jwt_secret": "ENC[AES256_GCM,data:/dPDqJdn4Af3Wo005V7lU9b8RbN/wyF0Tx66827cdyaZfi4QPOSj23wNqw==,iv:yJW2PiAGGr97q0DoBr64X88eFNpuVPZX0SPyNDp5QjQ=,tag:p27XTUbMC0WDMTNJCscmGQ==,type:str]",
"database_pw": "ENC[AES256_GCM,data:EzGPKdsX3Ib2zWrz09kUdegIxGNwg1j4msbOKUmvCGy6R9/EG1nvOC9Z5Oo=,iv:msek112FxmVAwFume6b7RnSICL/sw5CK3XzgCq9Sp1s=,tag:UcxUi2hySv54liN+Ddodpw==,type:str]"
},
"hass": {
"magic-token": "ENC[AES256_GCM,data:3mKbPFgvtX+hWYEZ0q4jBjnR8KM+E/1DqmkVzoV6ROY=,iv:9L748apqK1TcsW0Y0HvU9QHVD/eSh56c/uN/K4KNct4=,tag:ZmXiaPz7MEvaQ0yu3byiKQ==,type:str]"
},
"traefik": {
"cf_tunnel_token": "ENC[AES256_GCM,data:IgrmKwdeipix1dIXNuXnTWN5rCZjClbKZQJfgr5c2IP/n8bcc/nG5Wb42WL2C4hTeVqhG5p62ZXoz0j4dNAjxvuzcW/P0XeSYaiDRXMNWKhNIcK7jOexgswio0sUC+F7f3fa6HH4C02Mx8dWoFZChYtM5EhGdcEwVwspyBlMhTSHTz+/w5T9OqH18o132ZTM6kMQY85sgH36azWoSw73N+aC4ANhgybuok06z6R5D2jMdDX47Bo5bg==,iv:yOcUDTYHh58iejbl0wxNJO1hcDypcBq6KlHKyqnMSVk=,tag:CMyHKgahkIGdXItMJ1/hOg==,type:str]",
"kopia_basic_auth": "ENC[AES256_GCM,data:GKJKTtFqW2f8L6VYsBIuNsssUk8vBn74A4TIFw==,iv:rjV0o+CKUtZi8nVsVv0m17OPkYW5ymje9QoWvlRHa7g=,tag:CqEf6n5xgc2RWddbZoNqBQ==,type:str]"
},
"immich": {
"database_pw": "ENC[AES256_GCM,data:SUyMGqu7deZyZpVt,iv:asZehOvn/JamwFyS+Xl9Xpr4JFkKlJjHVw7LywYOxTc=,tag:plRvuv7+ievfEhxurBl7YQ==,type:str]"
},
"sonarr": {
"database_pw": "ENC[AES256_GCM,data:TN381ZYJLeUHX9U3Jnd9+w==,iv:lKaMYHeaSGXJd0/EGxkDY2l2v62xG3xs8TVC0HwXL94=,tag:3z5rK+2RfJHJdQc7KC9KmA==,type:str]"
},
"pia": {
"user": "ENC[AES256_GCM,data:kniAs2gCTq4=,iv:1Oaht02fFSQwzWmWEtjsJZCJChPJsZhwRyux8dMY2CU=,tag:NqWaUhuYTSFZZK/CpSisdg==,type:str]",
"pass": "ENC[AES256_GCM,data:c8qWGcaI0p7MyQ==,iv:/3ehYrgdDwjzFdXyX/vKTK+zt6u7gWNRZBIdWDG1KiE=,tag:jqfIMnB1OKchBZ4U2s1o4g==,type:str]"
},
"prowlarr": {
"database_pw": "ENC[AES256_GCM,data:FkW5LPoyn8bh0UfWcFq3og==,iv:SFq4Xsdz3FfCDyPjIaAmz5nsC/SPdFrR03GCr3KE/nw=,tag:PVYj7hSWDnfeE7igSXGBSA==,type:str]"
},
"frigate": {
"rtsp_password": "ENC[AES256_GCM,data:8vq06/IkNOUgpHmf,iv:lj8buuIC0ub0YOUiOiaN6tokkIT2/+bBwFNz2QXmCd4=,tag:EMm/bIHdJSAtjYAlrNOCMw==,type:str]"
},
"openreader": {
"database_pw": "ENC[AES256_GCM,data:2Ey9Ypb2Ked/LP/ApJhCqhKWuzognxVK7ku60nERp7I=,iv:KdLFD+fuNpYmPEU5G96SvFcQeZB0XlnOh/6uf7OfFqI=,tag:h7DQlqx5fxhiHuWyFd7svQ==,type:str]"
},
"ntfy": {
"database_pw": "ENC[AES256_GCM,data:79c2KFs3tcbet1dSGnkSDlAeKLCZrh4aMYLXTROM8w==,iv:eZ4limyjl++nsvHUzPKy82hfLZEOc+XQYpO6Czo/8os=,tag:iX9SiEACQ5IM8f1jhZh5Qw==,type:str]"
},
"renovate": {
"gitea_token": "ENC[AES256_GCM,data:/J3CDMgWZLe20oQ+ENKBMi8fs/+jgsARV7xihMq0OLmRk8C8ae/IXg==,iv:e7WYOanSOCZ/LhN6SKrH0VrR3xLPTTppOKpGpSl+oAc=,tag:XBAilRdK3jL7WtM+92Fsmg==,type:str]",
"github_token": "ENC[AES256_GCM,data:omZpdsTV1aFgQ9PjIApITEyIRKk6Z8QyvD2Kp5tJnBWzFCm4v2lRAg==,iv:cKL7z+CSChzF9eZEcske2lbmx9KV6CrWw0tn7rmP/10=,tag:gon3Sc1d3ntNSbWwenHuOw==,type:str]"
},
"act-runner": {
"registration_token": "ENC[AES256_GCM,data:RnDvcNh69lLlL/ms+sMPKhhc+ECtc5hUHSkAQZv8e77iTD/QPd356Q==,iv:sl2Aua8rTe6cKYQAUC7O4UyHajGy1LgG/ZNLTVP4SyE=,tag:JjdaQqZ4PaWjfoiVmBl6lQ==,type:str]"
},
"sops": {
"age": [
{
"recipient": "age1zuj9ssd0kkfeefjmyz82t9h3dfjq8degqm2l5eszhu5zhazpgsys54rq2n",
"enc": "-----BEGIN AGE ENCRYPTED FILE-----\nYWdlLWVuY3J5cHRpb24ub3JnL3YxCi0+IFgyNTUxOSByUWM4ZDVVbGFrUGdMRHBX\nUFBmU3Nlc0RBSzhFK0tHNHpkQXUvUVdiZUZJCmpRN1lFdENpWW0rcThjVlVQNUl6\nWnlLU0RnQ3FZby81Ly8xTFBrek9nMncKLS0tIFQ4UTRNOC9CRmx4OFJWem1wckZz\nUDFTSzdWZldFK3FqcTNWTWRyNDhHQ2MKS811mR5xn7qiC/aVgPFYJ5c6Q3zxRfcr\nHcvxUvB01vNJKZpRg92vvKPkV6lQO3DXCT98OdfwiymlEOvYxg71Pg==\n-----END AGE ENCRYPTED FILE-----\n"
}
],
"lastmodified": "2026-04-18T07:41:42Z",
"mac": "ENC[AES256_GCM,data:+HhhsiZXok4BZI05tG3p9veZaj51kELSQlWFYMSInv7bGfEadmOrJqCxaGrFcNkMmgVPx80jWQFrILfVLW5MUvEsHAhD4Vza2TSWeUq1HuL9DbMxsK2G9Y1fbthd12r/++dDcXxVnTUf/rCD70in/+g/zRObocAnUcFEcIqx1JE=,iv:pS+aj+47J4bYZYGlMVniQVTlLt4jtCLUT7oROJLUkZo=,tag:+lznxDhs2C3bcz5quxfHjA==,type:str]",
"encrypted_regex": "^(.*)$",
"version": "3.10.2"
}
}

View File

@@ -0,0 +1,140 @@
job "sonarr" {
group "sonarr" {
network {
mode = "bridge"
port "http" {
to = 8989
}
}
service {
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "postgres"
local_bind_port = 5432
}
upstreams {
destination_name = "deluge-api"
local_bind_port = 8112
}
upstreams {
destination_name = "prowlarr-api"
local_bind_port = 9696
}
}
}
}
}
service {
name = "sonarr"
port = "http"
tags = [
"traefik.enable=true",
"traefik.http.routers.sonarr.middlewares=auth@file",
]
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
service {
name = "sonarr-api"
port = "http"
address_mode = "alloc" # Use allocation IP for Connect as the sidecar can't access the host's published port (hairpin/loopback NAT issue)
connect {
sidecar_service {}
}
check {
type = "http"
path = "/"
interval = "10s"
timeout = "2s"
}
}
task "sonarr" {
driver = "docker"
config {
image = "lscr.io/linuxserver/sonarr:latest"
ports = ["http"]
}
env {
PUID = 1000
PGID = 1000
TZ = "Australia/Melbourne"
# https://wiki.servarr.com/sonarr/environment-variables
# Disable internal auth to use Traefik + Authelia
SONARR__AUTH__REQUIRED = "Enabled"
SONARR__AUTH__METHOD = "External"
SONARR__POSTGRES__USER = "sonarr"
SONARR__POSTGRES__HOST = "localhost"
SONARR__POSTGRES__PORT = "5432"
SONARR__POSTGRES__MAINDB = "sonarr-main"
SONARR__POSTGRES__LOGDB = "sonarr-log"
}
volume_mount {
volume = "unraid_appdata_sonarr"
destination = "/config"
read_only = false
}
volume_mount {
volume = "unraid_media_sonarr"
destination = "/data"
read_only = false
}
resources {
cpu = 150
memory = 1024
}
template {
data = <<EOH
SONARR__POSTGRES__PASSWORD="{{ with nomadVar "nomad/jobs/sonarr" }}{{ .database_pw }}{{ end }}"
EOH
destination = "secrets/db.env"
env = true # Load the file as environment variables
}
}
volume "unraid_appdata_sonarr" {
type = "csi"
read_only = false
source = "unraid_appdata_sonarr"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=1000", "gid=1000"]
}
}
volume "unraid_media_sonarr" {
type = "csi"
read_only = false
source = "unraid_media_sonarr"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["nobrl", "uid=1000", "gid=1000"]
}
}
}
}

38
2-nomad-config/sonarr.tf Normal file
View File

@@ -0,0 +1,38 @@
resource "nomad_job" "sonarr" {
jobspec = file("sonarr.nomad.hcl")
}
resource "nomad_variable" "sonarr" {
path = "nomad/jobs/sonarr"
items = {
database_pw = data.sops_file.secrets.data["sonarr.database_pw"]
}
}
# https://wiki.servarr.com/sonarr/postgres-setup#schema-creation
resource "postgresql_role" "sonarr" {
name = "sonarr"
password = data.sops_file.secrets.data["sonarr.database_pw"]
login = true
}
resource "postgresql_database" "sonarr_main" {
name = "sonarr-main"
owner = postgresql_role.sonarr.name
}
resource "postgresql_database" "sonarr_log" {
name = "sonarr-log"
owner = postgresql_role.sonarr.name
}
module "appdata_sonarr" {
source = "./modules/appdata"
name = "sonarr"
}
module "unraid_smb_sonarr_media" {
source = "./modules/unraid_smb"
name = "sonarr"
share = "media"
}

View File

@@ -1,165 +0,0 @@
job "traefik" {
group "traefik" {
network {
port "http" {
static = 80
}
port "https" {
static = 443
}
port "api" {
static = 8081
}
}
service {
name = "traefik"
tags = [
"traefik.enable=true",
"traefik.http.routers.traefik.rule=Host(`traefik.othrayte.one`)",
"traefik.http.routers.traefik.service=traefik",
"traefik.http.routers.traefik.middlewares=auth@file",
"traefik.http.services.traefik.loadbalancer.server.port=8081",
]
check {
name = "alive"
type = "tcp"
port = "http"
interval = "10s"
timeout = "2s"
}
}
volume "traefik" {
type = "host"
read_only = false
source = "traefik"
}
task "traefik" {
driver = "docker"
config {
image = "traefik:v3.3"
network_mode = "host"
volumes = [
"local/traefik.yml:/etc/traefik/traefik.yml",
"local/configs/:/etc/traefik/configs/"
]
}
volume_mount {
volume = "traefik"
destination = "/opt/traefik"
read_only = false
}
template {
data = <<EOF
entryPoints:
web:
address: ":80"
http:
redirections:
entryPoint:
to: websecure
scheme: https
websecure:
address: ":443"
http:
tls:
certResolver: letsencrypt
traefik:
address: ":8081"
api:
dashboard: true
insecure: true
providers:
file:
directory: "/etc/traefik/configs/"
consulCatalog:
prefix: "traefik"
exposedByDefault: false
defaultRule: {{"Host(`{{ .Name }}.othrayte.one`)"}}
endpoint:
address: "127.0.0.1:8500"
scheme: "http"
certificatesResolvers:
letsencrypt:
acme:
email: "othrayte@gmail.com"
storage: "/opt/traefik/acme.json"
httpChallenge:
entryPoint: web
EOF
destination = "local/traefik.yml"
}
template {
data = <<EOF
http:
middlewares:
auth:
forwardAuth:
address: "http://192.168.1.235:9091/api/authz/forward-auth"
trustForwardHeader: true
routers:
fallback:
rule: "HostRegexp(`^.+$`)"
entryPoints:
- websecure
middlewares:
- auth
service: noop@internal # This router just applies middleware
priority: 1
nomad-ui:
rule: "Host(`nomad.othrayte.one`)"
service: nomad-ui
middlewares:
- auth
consul-ui:
rule: "Host(`consul.othrayte.one`)"
service: consul-ui
middlewares:
- auth
unraid:
rule: "Host(`unraid.othrayte.one`)"
service: unraid
middlewares:
- auth
services:
nomad-ui:
loadBalancer:
servers:
- url: "http://127.0.0.1:4646"
consul-ui:
loadBalancer:
servers:
- url: "http://127.0.0.1:8500"
unraid:
loadBalancer:
servers:
- url: "http://192.168.1.192:80"
EOF
destination = "local/configs/nomad.yml"
}
resources {
cpu = 100
memory = 128
}
}
}
}

View File

@@ -1,23 +0,0 @@
resource "cloudflare_dns_record" "othrayte-one" {
comment = "othrayte.one proxy to internal IP for traefik"
zone_id = "2616ab2a44d0645b03fbc3106c79bd99"
type = "AAAA"
name = "othrayte.one"
content = data.sops_file.secrets.data["cloudflare.direct_ip6"]
proxied = true
ttl = 1 # Auto
}
resource "cloudflare_dns_record" "star-othrayte-one" {
comment = "*.othrayte.one proxy to internal IP for traefik"
zone_id = "2616ab2a44d0645b03fbc3106c79bd99"
type = "AAAA"
name = "*.othrayte.one"
content = data.sops_file.secrets.data["cloudflare.direct_ip6"]
proxied = true
ttl = 1 # Auto
}
resource "nomad_job" "traefik" {
jobspec = file("traefik.nomad.hcl")
}

View File

@@ -86,10 +86,7 @@ job "transfer" {
"address": "",
"log": "stdout",
"database": "/database/filebrowser.db",
"root": "/srv",
"auth": {
"method": "noauth"
}
"root": "/srv"
}
EOF

View File

@@ -2,53 +2,14 @@ resource "nomad_job" "transfer" {
jobspec = file("transfer.nomad.hcl")
}
resource "nomad_csi_volume_registration" "unraid_transfer" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it
depends_on = [data.nomad_plugin.smb]
plugin_id = "smb"
volume_id = "unraid_transfer"
name = "unraid_transfer"
external_id = "unraid_transfer"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/transfer"
}
secrets = {
"username" = "anon"
"password" = ""
}
module "unraid_smb_transfer" {
source = "./modules/unraid_smb"
name = "transfer"
id = "unraid_transfer"
share = "transfer"
}
resource "nomad_csi_volume_registration" "unraid_appdata_transferfilebrowser" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it
depends_on = [data.nomad_plugin.smb]
plugin_id = "smb"
volume_id = "unraid_appdata_transferfilebrowser"
name = "unraid_appdata_transferfilebrowser"
external_id = "unraid_appdata_transferfilebrowser"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/appdata"
subDir = "transferfilebrowser" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
module "appdata_transferfilebrowser" {
source = "./modules/appdata"
name = "transferfilebrowser"
}

View File

@@ -0,0 +1,50 @@
job "unifi-network" {
group "unifi-network" {
count = 1
task "unifi-controller" {
driver = "docker"
config {
image = "jacobalberty/unifi:v9.5.21"
// Fixed IP on the actual network so that devices can find it
network_mode = "macvlan"
ipv4_address = "192.168.1.50"
}
env {
TZ = "Australia/Melbourne"
SYSTEM_IP = "192.168.1.50"
JVM_INIT_HEAP_SIZE = "1024M"
JVM_MAX_HEAP_SIZE = "1024M"
UNIFI_STDOUT = "true"
}
volume_mount {
volume = "unraid_appdata_unifi_network"
destination = "/unifi" # Expected root directory (contains data, log, cert subdirs)
read_only = false
}
resources {
cpu = 200
memory = 1850
memory_max = 2500
}
}
# CSI volume for UniFi Controller persistent data/logs
volume "unraid_appdata_unifi_network" {
type = "csi"
read_only = false
source = "unraid_appdata_unifi_network"
access_mode = "single-node-writer"
attachment_mode = "file-system"
mount_options {
mount_flags = ["uid=0", "gid=0"]
}
}
}
}

9
2-nomad-config/unifi.tf Normal file
View File

@@ -0,0 +1,9 @@
resource "nomad_job" "unifi_network" {
jobspec = file("unifi.nomad.hcl")
}
module "appdata_unifi_network" {
source = "./modules/appdata"
name = "unifi-network"
}

305
cicd-plan.md Normal file
View File

@@ -0,0 +1,305 @@
# CI/CD Plan
## Overview
Three distinct problems, tackled in phases:
1. **Does the config parse/validate without errors?** (static, no credentials)
2. **Does the new Docker image actually exist and start?** (pre-merge, needs Docker)
3. **Does the running service stay healthy through a deployment?** (post-merge, needs Nomad canary)
The goal is: Renovate opens a PR → CI runs checks → you review → merge → canary starts automatically → you promote (or it auto-reverts).
---
## Phase 1 — Static Validation (proves the runner works)
No secrets needed. Runs on every PR.
### Infrastructure required
- `act_runner` Nomad job (see below) with a Gitea runner token
- `.gitea/workflows/ci.yml` in this repo
### Checks
| Check | Command | Notes |
| --------------------- | ----------------------------------------------------- | ------------------------------------------------------------------- |
| HCL formatting | `terraform fmt -check -recursive` | Fails on whitespace/style drift |
| Terraform syntax | `terraform init -backend=false && terraform validate` | Catches wrong resource types, missing required args, bad references |
| Nomad job spec syntax | `nomad job validate <file>` | Catches Nomad-specific issues; needs `NOMAD_ADDR` + read token |
`terraform validate -backend=false` is the most valuable: it catches ~90% of real mistakes with zero secret exposure. The Nomad validate step requires a low-privilege read token — worth adding once the runner is trusted.
### Workflow sketch
```yaml
# .gitea/workflows/ci.yml
on: [pull_request]
jobs:
validate:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: hashicorp/setup-terraform@v3
- name: fmt check
run: terraform fmt -check -recursive
working-directory: 2-nomad-config
- name: init + validate (no backend)
run: |
terraform init -backend=false
terraform validate
working-directory: 2-nomad-config
- name: fmt check (nixos-node)
run: terraform fmt -check -recursive
working-directory: 1-nixos-node
nomad-validate:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Install Nomad CLI
run: |
curl -fsSL https://apt.releases.hashicorp.com/gpg | sudo gpg --dearmor -o /usr/share/keyrings/hashicorp.gpg
echo "deb [signed-by=/usr/share/keyrings/hashicorp.gpg] https://apt.releases.hashicorp.com $(lsb_release -cs) main" | sudo tee /etc/apt/sources.list.d/hashicorp.list
sudo apt-get update && sudo apt-get install -y nomad
- name: validate all job specs
env:
NOMAD_ADDR: ${{ secrets.NOMAD_ADDR }}
NOMAD_TOKEN: ${{ secrets.NOMAD_TOKEN }} # read-only policy sufficient
run: |
find 2-nomad-config -name '*.nomad.hcl' | while read f; do
echo "==> $f"
nomad job validate "$f"
done
```
### act_runner Nomad job
```hcl
# act-runner.nomad.hcl
job "act-runner" {
group "act-runner" {
network {
mode = "bridge"
}
# Connect upstream to Gitea
service {
name = "act-runner"
connect {
sidecar_service {
proxy {
upstreams {
destination_name = "code-connect"
local_bind_port = 3000
}
}
}
}
}
task "act-runner" {
driver = "docker"
config {
image = "gitea/act_runner:latest"
volumes = ["/var/run/docker.sock:/var/run/docker.sock"]
}
env = {
GITEA_INSTANCE_URL = "http://localhost:3000"
}
template {
data = <<EOF
GITEA_RUNNER_REGISTRATION_TOKEN={{ with nomadVar "nomad/jobs/act-runner" }}{{ .registration_token }}{{ end }}
EOF
destination = "secrets/runner.env"
env = true
}
resources {
cpu = 200
memory = 256
memory_max = 512
}
}
}
}
```
**Security note**: mounting `/var/run/docker.sock` gives the runner root-equivalent access to the host. Acceptable for a home server. Alternative: use `docker:dind` sidecar or Nomad's `exec` driver — more complex, lower risk.
---
## Phase 2 — Docker Image Validation (pre-merge)
Runs on PRs that touch `.nomad.hcl` files. Catches: tag typos, deleted images, registry outages.
Requires the `act_runner` to have Docker access (same socket mount as above).
```yaml
image-pull:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Pull changed images
run: |
# Extract image tags added or changed vs main
git fetch origin main
git diff origin/main...HEAD -- '*.nomad.hcl' \
| grep '^\+\s*image\s*=' \
| grep -oP '"[^"]+:[^"]+"' \
| tr -d '"' \
| sort -u \
| while read image; do
echo "==> Pulling $image"
docker pull "$image"
done
```
This intentionally only tests _changed_ images — no value in pulling everything on every PR.
---
## Phase 3 — Nomad Canary Deployments (post-merge gate)
Makes "merge" mean "start canary" rather than "go live". The old allocation keeps running until you promote.
### Which jobs get canaries
Most jobs already have Consul health checks — these can use `health_check = "checks"` for automatic revert gating.
| Job | Health check | Shared writable volume | Canary safe? |
| ---------- | ------------- | ----------------------- | --------------------------------------------------------------------------------- |
| ntfy | ✅ `/healthz` | no | ✅ yes |
| gitea | ✅ `/` | ✅ `single-node-writer` | ⚠️ volume blocks 2nd alloc from mounting — needs `max_parallel=1` rolling instead |
| jellyfin | ✅ | ✅ `single-node-writer` | ⚠️ same — rolling |
| immich | ✅ | ✅ `single-node-writer` | ⚠️ same — rolling |
| sonarr | ✅ | ✅ `single-node-writer` | ⚠️ same — rolling |
| prowlarr | ✅ | ✅ `single-node-writer` | ⚠️ same — rolling |
| deluge | ✅ | ✅ `single-node-writer` | ⚠️ same — rolling |
| frigate | ✅ | ✅ `single-node-writer` | ⚠️ same — rolling |
| glance | ✅ | no | ✅ yes |
| transfer | ✅ | ✅ `single-node-writer` | ⚠️ rolling |
| openreader | ❌ | ✅ `single-node-writer` | ⚠️ add check first, then rolling |
| unifi | ❌ | ✅ `single-node-writer` | ⚠️ add check first, then rolling |
| traefik | (ingress) | ✅ | ⚠️ rolling — downtime risk, promote quickly |
| authelia | (ingress) | ✅ | ✅ stateless config, canary fine |
| renovate | batch job | n/a | n/a — no deployment model |
| postgres | (data layer) | ✅ | ❌ never canary — single-writer DB |
### Canary stanza (stateless jobs with no volume conflict)
```hcl
update {
canary = 1
auto_promote = false
auto_revert = true
health_check = "checks"
healthy_deadline = "5m"
min_healthy_time = "30s"
}
```
### Rolling stanza (jobs with single-node-writer volumes)
```hcl
update {
max_parallel = 1
auto_revert = true
health_check = "checks"
healthy_deadline = "5m"
min_healthy_time = "30s"
}
```
Rolling with `max_parallel=1` still gives auto-revert but doesn't attempt to run two allocations simultaneously — the old one stops before the new one mounts the volume.
---
## Phase 4 — Automated terraform apply + Deployment Promotion
Full CD: merge triggers apply, which creates the canary, CI then watches it and promotes or reverts.
### Flow
```
PR merged to main
Gitea Actions (on: push, branches: [main])
- terraform init
- terraform apply -auto-approve
Nomad canary starts (old allocation still live)
CI polls `nomad deployment list` for the new deployment ID
CI waits for canary allocation to reach "healthy" in Consul
│ healthy within deadline
CI runs: nomad deployment promote <id>
│ or unhealthy → nomad deployment fail <id> (auto_revert fires)
ntfy notification: "deployment promoted" or "deployment reverted"
```
### Secrets required for full CD
| Secret | Used by | Risk level |
| ---------------------- | ----------------------------------- | ---------------------------------- |
| `NOMAD_ADDR` | validate + apply + promote | Low (internal LAN addr) |
| `NOMAD_TOKEN` | terraform apply (write) + promote | **High** — grants full infra write |
| `CLOUDFLARE_API_TOKEN` | terraform apply | **High** — DNS write |
| `SOPS_AGE_KEY` | terraform apply (decrypt secrets) | **High** — decrypts all secrets |
| `PG_PASSWORD` | terraform apply (postgres provider) | High |
Full CD requires all of these in Gitea Actions secrets. This is acceptable for a self-hosted, non-public Gitea instance where you control runner access — but it's the trust boundary to be deliberate about. A reasonable middle ground: **Phase 1-3 are fully automated; Phase 4 (apply + promote) runs automatically but requires a manual re-trigger or approval step** (Gitea supports required reviewers on environments).
### Promote/revert script sketch
```bash
# In CI, after terraform apply completes:
DEPLOY_ID=$(nomad deployment list -json | jq -r '[.[] | select(.JobID == "$JOB" and .Status == "running")] | first | .ID')
echo "Watching deployment $DEPLOY_ID..."
for i in $(seq 1 30); do
STATUS=$(nomad deployment status -json "$DEPLOY_ID" | jq -r '.Status')
HEALTHY=$(nomad deployment status -json "$DEPLOY_ID" | jq -r '.TaskGroups[].HealthyAllocs')
echo "[$i] status=$STATUS healthy=$HEALTHY"
if [ "$STATUS" = "successful" ]; then exit 0; fi
if [ "$STATUS" = "failed" ]; then exit 1; fi
# Check if canary is healthy enough to promote
CANARY_HEALTHY=$(nomad deployment status -json "$DEPLOY_ID" | jq -r '.TaskGroups[].DesiredCanaries == .TaskGroups[].HealthyAllocs')
if [ "$CANARY_HEALTHY" = "true" ]; then
nomad deployment promote "$DEPLOY_ID"
exit 0
fi
sleep 10
done
nomad deployment fail "$DEPLOY_ID"
exit 1
```
---
## Implementation Order
- [x] **Phase 1a**: Create `act-runner.nomad.hcl` + Terraform wrapper, register runner token in Gitea, get a hello-world workflow green
- [x] **Phase 1b**: Add `terraform fmt` + `terraform validate -backend=false` workflow — no secrets needed
- [x] **Phase 1c**: Add Nomad validate step — add `NOMAD_ADDR` + read-only `NOMAD_TOKEN` to Gitea secrets
- [ ] **Phase 2**: Add image pull validation step to the workflow
- [ ] **Phase 3a**: Add `update` stanzas to ntfy and glance (simplest, no volume conflict)
- [ ] **Phase 3b**: Add rolling `update` stanzas to remaining service jobs (jellyfin, sonarr, etc.)
- [ ] **Phase 3c**: Add health checks to openreader and unifi before adding update stanzas
- [ ] **Phase 4a**: Add on-push workflow that runs `terraform apply -auto-approve` using full credential set
- [ ] **Phase 4b**: Add deployment promotion/revert polling script
- [ ] **Phase 4c**: Wire ntfy notifications for promote/revert outcomes

15
renovate.json Normal file
View File

@@ -0,0 +1,15 @@
{
"$schema": "https://docs.renovatebot.com/renovate-schema.json",
"extends": ["config:recommended"],
"customManagers": [
{
"description": "Update Docker image tags in Nomad job files",
"customType": "regex",
"fileMatch": ["\\.nomad\\.hcl$"],
"matchStrings": [
"image\\s*=\\s*\"(?<depName>[^:\"]+):(?<currentValue>[^\"]+)\""
],
"datasourceTemplate": "docker"
}
]
}