21 Commits

Author SHA1 Message Date
d0744c1658 Update ghcr.io/richardr1126/openreader Docker tag to v2.2.0
All checks were successful
CI / Terraform fmt + validate (pull_request) Successful in 18s
CI / Nomad job spec validate (pull_request) Successful in 16s
2026-04-19 05:25:52 +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
42 changed files with 1999 additions and 645 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

View File

@@ -49,6 +49,9 @@
preferred_address_family = "ipv4"; preferred_address_family = "ipv4";
%{if cpu_total_compute != null ~} %{if cpu_total_compute != null ~}
cpu_total_compute = ${cpu_total_compute}; cpu_total_compute = ${cpu_total_compute};
%{endif ~}
%{if node_class != null ~}
node_class = "${node_class}";
%{endif ~} %{endif ~}
host_volume = { host_volume = {
%{ for volume in host_volumes ~} %{ for volume in host_volumes ~}
@@ -61,6 +64,7 @@
cni_path = "$${pkgs.cni-plugins}/bin"; cni_path = "$${pkgs.cni-plugins}/bin";
}; };
plugin.docker.config.allow_privileged = true; plugin.docker.config.allow_privileged = true;
plugin.docker.config.volumes.enabled = true;
}; };
extraPackages = with pkgs; [ extraPackages = with pkgs; [
cni-plugins cni-plugins
@@ -113,6 +117,61 @@
networking.firewall.allowedTCPPorts = [ 80 443 8081 4646 4647 4648 8300 8301 8500 ]; networking.firewall.allowedTCPPorts = [ 80 443 8081 4646 4647 4648 8300 8301 8500 ];
networking.firewall.allowedUDPPorts = [ 8301 ]; 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 # Copy the NixOS configuration file and link it from the resulting system
# (/run/current-system/configuration.nix). This is useful in case you # (/run/current-system/configuration.nix). This is useful in case you
# accidentally delete configuration.nix. # accidentally delete configuration.nix.

View File

@@ -21,6 +21,7 @@ variable "nodes" {
bind_interface = string bind_interface = string
bootstrap = optional(bool, false) # Optional field for bootstrap nodes bootstrap = optional(bool, false) # Optional field for bootstrap nodes
cpu_total_compute = optional(number, null) # Optional field for CPU total compute 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) host_volumes = list(string)
})) }))
} }
@@ -32,6 +33,7 @@ locals {
bind_interface = v.bind_interface bind_interface = v.bind_interface
bootstrap = v.bootstrap bootstrap = v.bootstrap
cpu_total_compute = v.cpu_total_compute cpu_total_compute = v.cpu_total_compute
node_class = v.node_class
host_volumes = v.host_volumes 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

View File

@@ -18,27 +18,9 @@ resource "nomad_variable" "postgres" {
} }
} }
resource "nomad_csi_volume_registration" "unraid_database_dump" { module "unraid_smb_database_dump" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "../modules/unraid_smb"
depends_on = [data.nomad_plugin.smb] name = "dump"
plugin_id = "smb" id = "unraid_database_dump"
share = "database-dump"
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

@@ -2,30 +2,3 @@
module "ingress" { module "ingress" {
source = "./2-ingress" source = "./2-ingress"
} }
# traefik.tf
moved {
from = cloudflare_dns_record.othrayte-one
to = module.ingress.cloudflare_dns_record.othrayte-one
}
moved {
from = cloudflare_dns_record.star-othrayte-one
to = module.ingress.cloudflare_dns_record.star-othrayte-one
}
moved {
from = nomad_variable.traefik
to = module.ingress.nomad_variable.traefik
}
moved {
from = nomad_job.traefik
to = module.ingress.nomad_job.traefik
}
moved {
from = nomad_csi_volume_registration.unraid_appdata_traefik
to = module.ingress.nomad_csi_volume_registration.unraid_appdata_traefik
}

View File

@@ -63,6 +63,12 @@ job "authelia" {
data = <<EOF data = <<EOF
server: server:
address: tcp://0.0.0.0:{{ env "NOMAD_PORT_http" }}/ address: tcp://0.0.0.0:{{ env "NOMAD_PORT_http" }}/
endpoints:
authz:
forward-auth:
implementation: 'ForwardAuth'
authn_strategies:
- name: 'CookieSession'
theme: "auto" theme: "auto"
identity_validation: identity_validation:
reset_password: reset_password:

View File

@@ -112,6 +112,9 @@ EOF
template { template {
data = <<EOF data = <<EOF
http: http:
serversTransports:
ignorecert:
insecureSkipVerify: true
middlewares: middlewares:
auth: auth:
forwardAuth: forwardAuth:
@@ -154,11 +157,6 @@ http:
service: unraid service: unraid
middlewares: middlewares:
- auth - auth
frigate:
rule: "Host(`frigate.othrayte.one`)"
service: frigate
middlewares:
- auth
kopia: kopia:
rule: "Host(`kopia.othrayte.one`)" rule: "Host(`kopia.othrayte.one`)"
service: kopia service: kopia
@@ -173,6 +171,11 @@ http:
hass-token: hass-token:
rule: "Host(`${hass_magic_token}-hass.othrayte.one`)" rule: "Host(`${hass_magic_token}-hass.othrayte.one`)"
service: hass service: hass
unifi-network:
rule: "Host(`network.othrayte.one`)"
service: unifi-network
middlewares:
- auth
services: services:
traefik: traefik:
@@ -190,19 +193,20 @@ http:
unraid: unraid:
loadBalancer: loadBalancer:
servers: servers:
- url: "http://192.168.1.192:80" - url: "http://betelgeuse-seven-unraid.lan:80"
frigate:
loadBalancer:
servers:
- url: "http://192.168.1.192:5000"
kopia: kopia:
loadBalancer: loadBalancer:
servers: servers:
- url: "http://192.168.1.192:51515" - url: "http://betelgeuse-seven-unraid.lan:51515"
hass: hass:
loadBalancer: loadBalancer:
servers: servers:
- url: "http://192.168.1.234:8123" - url: "http://192.168.1.234:8123"
unifi-network:
loadBalancer:
serversTransport: ignorecert
servers:
- url: "https://192.168.1.50:8443"
EOF EOF
destination = "local/configs/nomad.yml" destination = "local/configs/nomad.yml"

View File

@@ -32,28 +32,8 @@ resource "nomad_job" "traefik" {
}) })
} }
resource "nomad_csi_volume_registration" "unraid_appdata_traefik" { module "appdata_traefik" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "../modules/appdata"
depends_on = [data.nomad_plugin.smb] name = "traefik"
plugin_id = "smb" access_mode = "multi-node-multi-writer"
volume_id = "unraid_appdata_traefik"
name = "unraid_appdata_traefik"
external_id = "unraid_appdata_traefik"
capability {
access_mode = "multi-node-multi-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/appdata"
subDir = "traefik" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
} }

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 = [ tags = [
"traefik.enable=true", "traefik.enable=true",
"traefik.http.routers.gitea.middlewares=auth@file", "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 { 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" { task "gitea" {
driver = "docker" driver = "docker"

View File

@@ -33,28 +33,7 @@ resource "postgresql_database" "gitea" {
owner = postgresql_role.gitea.name owner = postgresql_role.gitea.name
} }
resource "nomad_csi_volume_registration" "unraid_appdata_gitea" { module "appdata_gitea" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/appdata"
depends_on = [data.nomad_plugin.smb] name = "gitea"
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"]
}
} }

View File

@@ -34,107 +34,29 @@ resource "nomad_variable" "immich" {
} }
} }
resource "nomad_csi_volume_registration" "unraid_appdata_immich" { module "appdata_immich" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/appdata"
depends_on = [data.nomad_plugin.smb] name = "immich"
plugin_id = "smb"
volume_id = "unraid_appdata_immich"
name = "unraid_appdata_immich"
external_id = "unraid_appdata_immich"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/appdata"
subDir = "immich" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
} }
resource "nomad_csi_volume_registration" "unraid_media_photosvideos" { module "unraid_smb_immich_photosvideos" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/unraid_smb"
depends_on = [data.nomad_plugin.smb] name = "photosvideos"
plugin_id = "smb" share = "media"
subDir = "Photos and Videos"
volume_id = "unraid_media_photosvideos"
name = "unraid_media_photosvideos"
external_id = "unraid_media_photosvideos"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/media"
subDir = "Photos and Videos" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
} }
resource "nomad_csi_volume_registration" "unraid_media_immich_encodedvideo" { module "unraid_smb_immich_encodedvideo" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/unraid_smb"
depends_on = [data.nomad_plugin.smb] name = "immich_encodedvideo"
plugin_id = "smb" share = "media"
subDir = "immich/encoded-video"
volume_id = "unraid_media_immich_encodedvideo"
name = "unraid_media_immich_encodedvideo"
external_id = "unraid_media_immich_encodedvideo"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/media"
subDir = "immich/encoded-video" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
} }
resource "nomad_csi_volume_registration" "unraid_mediadump_photosvideos" { module "unraid_smb_immich_mediadump_photosvideos" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/unraid_smb"
depends_on = [data.nomad_plugin.smb] name = "photosvideos"
plugin_id = "smb" id = "unraid_mediadump_photosvideos"
share = "media-dump"
volume_id = "unraid_mediadump_photosvideos" subDir = "Photos and Videos"
name = "unraid_mediadump_photosvideos"
external_id = "unraid_mediadump_photosvideos"
capability {
access_mode = "single-node-writer"
attachment_mode = "file-system"
}
context = {
source = "//192.168.1.192/media-dump"
subDir = "Photos and Videos" # Note: Needs to be manually created on the share
}
secrets = {
"username" = "nomad"
"password" = data.sops_file.secrets.data["unraid.nomad"]
}
} }

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

@@ -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:v2.2.0"
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

@@ -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 +1,7 @@
# Terraform State # Terraform State
Mount the state on the fileshare to 2-nomad-config/.tfstate/ 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 # Tailscale Oauth Client
@@ -20,6 +20,7 @@ Edit the secrets using `sops secrets/secrets.enc.json`
# Bootstrapping (starting without PostgreSQL running) # Bootstrapping (starting without PostgreSQL running)
terraform apply -target=module.data terraform apply -target=module.data
terraform apply -target=module.ingress
## Restoring PostgreSQL DBs ## Restoring PostgreSQL DBs

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

@@ -33,6 +33,32 @@
"immich": { "immich": {
"database_pw": "ENC[AES256_GCM,data:SUyMGqu7deZyZpVt,iv:asZehOvn/JamwFyS+Xl9Xpr4JFkKlJjHVw7LywYOxTc=,tag:plRvuv7+ievfEhxurBl7YQ==,type:str]" "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": { "sops": {
"age": [ "age": [
{ {
@@ -40,8 +66,8 @@
"enc": "-----BEGIN AGE ENCRYPTED FILE-----\nYWdlLWVuY3J5cHRpb24ub3JnL3YxCi0+IFgyNTUxOSByUWM4ZDVVbGFrUGdMRHBX\nUFBmU3Nlc0RBSzhFK0tHNHpkQXUvUVdiZUZJCmpRN1lFdENpWW0rcThjVlVQNUl6\nWnlLU0RnQ3FZby81Ly8xTFBrek9nMncKLS0tIFQ4UTRNOC9CRmx4OFJWem1wckZz\nUDFTSzdWZldFK3FqcTNWTWRyNDhHQ2MKS811mR5xn7qiC/aVgPFYJ5c6Q3zxRfcr\nHcvxUvB01vNJKZpRg92vvKPkV6lQO3DXCT98OdfwiymlEOvYxg71Pg==\n-----END AGE ENCRYPTED FILE-----\n" "enc": "-----BEGIN AGE ENCRYPTED FILE-----\nYWdlLWVuY3J5cHRpb24ub3JnL3YxCi0+IFgyNTUxOSByUWM4ZDVVbGFrUGdMRHBX\nUFBmU3Nlc0RBSzhFK0tHNHpkQXUvUVdiZUZJCmpRN1lFdENpWW0rcThjVlVQNUl6\nWnlLU0RnQ3FZby81Ly8xTFBrek9nMncKLS0tIFQ4UTRNOC9CRmx4OFJWem1wckZz\nUDFTSzdWZldFK3FqcTNWTWRyNDhHQ2MKS811mR5xn7qiC/aVgPFYJ5c6Q3zxRfcr\nHcvxUvB01vNJKZpRg92vvKPkV6lQO3DXCT98OdfwiymlEOvYxg71Pg==\n-----END AGE ENCRYPTED FILE-----\n"
} }
], ],
"lastmodified": "2025-10-13T12:19:46Z", "lastmodified": "2026-04-18T07:41:42Z",
"mac": "ENC[AES256_GCM,data:QJ1Prqf37xMZbvyMvjBVxZOiOr07CmCYrWmr+5hwDsEmG4eEC9sPF/UY+/Cy2OTzsMp+cHb6C3maAo09O171wj6nJIZucg3B9fjEW2+4AoO217G4vmauMl3FFkut2CuvVV9zt2B/fLAskRg/yeYYOhjzPkWA6lyeV31sV5ZQ6Kw=,iv:5WfkmNr5vdfTqp6+INjQN/Zmc7/iJNc/2auO9h3En08=,tag:snBgJyMzBXVAkV3zERkK8g==,type:str]", "mac": "ENC[AES256_GCM,data:+HhhsiZXok4BZI05tG3p9veZaj51kELSQlWFYMSInv7bGfEadmOrJqCxaGrFcNkMmgVPx80jWQFrILfVLW5MUvEsHAhD4Vza2TSWeUq1HuL9DbMxsK2G9Y1fbthd12r/++dDcXxVnTUf/rCD70in/+g/zRObocAnUcFEcIqx1JE=,iv:pS+aj+47J4bYZYGlMVniQVTlLt4jtCLUT7oROJLUkZo=,tag:+lznxDhs2C3bcz5quxfHjA==,type:str]",
"encrypted_regex": "^(.*)$", "encrypted_regex": "^(.*)$",
"version": "3.10.2" "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

@@ -1,345 +0,0 @@
job "teamsstatus" {
group "app" {
task "teamsstatus" {
driver = "docker"
config {
image = "python:3.11-slim"
command = "/local/start.sh"
}
# Template for the startup script
template {
data = <<EOF
#!/bin/bash
set -e
cd /local
echo "Starting Teams Status Updater service..."
echo "Installing required Python packages..."
pip install msal requests
echo "Running Teams Status Updater script..."
exec python teamsstatus_standalone.py 2>&1
EOF
destination = "local/start.sh"
perms = "755"
}
# Template for the token cache
template {
data = "{{ with nomadVar \"nomad/jobs/teamsstatus\" }}{{ .token_cache_json }}{{ end }}"
destination = "local/token_cache.json"
}
# Template for the Python script
template {
data = <<EOF
import logging
import time
from datetime import datetime, timedelta, timezone
import random
import json
import msal
import requests
import os
import atexit
# Configure logging
logging.basicConfig(level=logging.INFO)
# Embedded journey data
JOURNEY_DATA = '''Day Start Time AEST End Time AEST Start Dist End Dist Start Location End Location
08/06/2025 08:00:00 19:10:00 0km 973km Melbourne Port Pirie SA
09/06/2025 07:30:00 19:35:00 973km 2119km Port Pirie SA Mundrabilla WA
10/06/2025 06:45:00 15:38:00 2119km 2916km Mundrabilla WA Kalgoorlie WA
11/06/2025 10:45:00 17:55:00 2916km 3512km Kalgoorlie WA Perth
13/06/2025 07:00:00 13:58:00 3512km 4083km Perth Kalbarri WA
15/06/2025 07:00:00 16:52:00 4083km 4862km Kalbarri WA Coral Bay WA
18/06/2025 06:00:00 16:52:00 4862km 5554km Coral Bay WA Karijini NP WA
21/06/2025 14:00:00 15:21:00 5554km 5686km Karijini NP WA Karijini NP WA
22/06/2025 06:00:00 16:23:00 5686km 6559km Karijini NP WA Broome WA
23/06/2025 06:00:00 19:10:00 6559km 7688km Broome WA Kununurra WA
27/06/2025 06:00:00 16:29:00 7688km 8593km Kununurra WA Derby WA
28/06/2025 07:00:00 16:06:00 8593km 9358km Derby WA Port Hedland WA
29/06/2025 07:00:00 16:31:00 9358km 10150km Port Hedland WA Exmouth WA
02/07/2025 07:00:00 15:13:00 10150km 10866km Exmouth WA Shark Bay WA
05/07/2025 07:00:00 17:12:00 10866km 11712km Shark Bay WA Fremantle WA
06/07/2025 07:00:00 15:27:00 11712km 12411km Fremantle WA Esperance WA
08/07/2025 06:00:00 18:09:00 12411km 13144km Esperance WA Madura WA
09/07/2025 06:45:00 16:39:00 13144km 13821km Madura WA Ceduna SA
11/07/2025 08:30:00 17:46:00 13821km 14599km Ceduna SA Adelaide
12/07/2025 08:30:00 18:52:00 14599km 15348km Adelaide Melbourne'''
# Global variables for journey segments
_segments = []
def setup_token_cache(cache_filename="token_cache.json"):
"""Set up and return a serializable token cache"""
cache = msal.SerializableTokenCache()
cache.deserialize(open(cache_filename, "r").read())
atexit.register(
lambda: open(cache_filename, "w").write(cache.serialize())
if cache.has_state_changed else None
)
return cache
def get_msal_app(client_id, authority="https://login.microsoftonline.com/organizations", cache_filename="token_cache.json"):
"""Create and return an MSAL PublicClientApplication"""
cache = setup_token_cache(cache_filename)
return msal.PublicClientApplication(
client_id,
authority=authority,
token_cache=cache,
)
def acquire_token(app, scope):
"""Acquire a token using the MSAL app"""
result = None
# Check if user account exists in cache
accounts = app.get_accounts(username=None)
if accounts:
logging.debug("Account(s) exists in cache, probably with token too. Let's try.")
logging.debug("Account(s) already signed in:")
for a in accounts:
logging.debug(a["username"])
chosen = accounts[0] # Assuming the end user chose this one to proceed
logging.debug(f"Automatically using first account: {chosen['username']}")
# Try to find a token in cache for this account
result = app.acquire_token_silent(scope, account=chosen)
# If no suitable token was found, get a new one
if not result:
logging.debug("No suitable token exists in cache. Let's get a new one from AAD.")
print("A local browser window will be open for you to sign in. CTRL+C to cancel.")
result = app.acquire_token_interactive(scope)
# Validate the result
if "access_token" not in result:
logging.error(result.get("error"))
logging.error(result.get("error_description"))
logging.debug(f"Correlation ID: {result.get('correlation_id')}")
return None
return result["access_token"]
def set_teams_status_message(access_token, user_id, status_message, expiration_date_time="2025-06-01T12:00:00", time_zone="UTC"):
"""Set the status message for a Teams user"""
url = f"https://graph.microsoft.com/v1.0/users/{user_id}/presence/microsoft.graph.setStatusMessage"
payload = {
"statusMessage": {
"message": {
"content": status_message + "<pinnednote></pinnednote>",
"contentType": "text",
}
},
"expirationDateTime": {
"dateTime": expiration_date_time,
"timeZone": time_zone
},
}
headers = {
'Authorization': f'Bearer {access_token}',
'Content-Type': 'application/json'
}
logging.debug(f"Setting status message for user {user_id}")
response = requests.post(url, json=payload, headers=headers)
if response.status_code == 200:
logging.info(f"Teams status message set to: {status_message}")
return True
else:
logging.error(f"Failed to set Teams status message: {response.status_code}")
return False
def _load_segments():
"""Load the journey segments from embedded data into memory"""
global _segments
if _segments: # Already loaded
return
aest = timezone(timedelta(hours=10))
for line in JOURNEY_DATA.split('\n')[1:]: # Skip header
day, start_time, end_time, start_dist, end_dist, start_loc, end_loc = line.strip().split('\t')
# Convert day and times to datetime in AEST
day_start = datetime.strptime(f"{day} {start_time}", "%d/%m/%Y %H:%M:%S").replace(tzinfo=aest)
day_end = datetime.strptime(f"{day} {end_time}", "%d/%m/%Y %H:%M:%S").replace(tzinfo=aest)
# Extract the numeric distance values
start_dist = int(start_dist.rstrip('km'))
end_dist = int(end_dist.rstrip('km'))
_segments.append({
'start_time': day_start,
'end_time': day_end,
'start_dist': start_dist,
'end_dist': end_dist,
'start_location': start_loc,
'end_location': end_loc
})
def get_trip_info(target_datetime):
"""Determine the distance travelled and locations for the current datetime."""
if target_datetime.tzinfo is None:
raise ValueError("target_datetime must be timezone-aware")
# Ensure data is loaded
_load_segments()
# Before journey starts
if not _segments or target_datetime < _segments[0]['start_time']:
start_loc = end_loc = _segments[0]['start_location']
return (0, start_loc, end_loc)
# During journey
for i, segment in enumerate(_segments):
# If target is before this segment starts
if target_datetime < segment['start_time']:
prev_segment = _segments[i-1]
return (prev_segment['end_dist'], prev_segment['end_location'], prev_segment['end_location'])
# If target is during this segment, interpolate
if segment['start_time'] <= target_datetime <= segment['end_time']:
# Calculate what fraction of the segment has elapsed
total_seconds = (segment['end_time'] - segment['start_time']).total_seconds()
elapsed_seconds = (target_datetime - segment['start_time']).total_seconds()
fraction = elapsed_seconds / total_seconds
# Interpolate the distance
distance_delta = segment['end_dist'] - segment['start_dist']
current_dist = segment['start_dist'] + int(distance_delta * fraction)
return (current_dist, segment['start_location'], segment['end_location'])
# Between segments
if i < len(_segments) - 1:
next_segment = _segments[i + 1]
if segment['end_time'] < target_datetime < next_segment['start_time']:
return (segment['end_dist'], segment['end_location'], segment['end_location'])
# After journey ends
return (_segments[-1]['end_dist'], _segments[-1]['end_location'], _segments[-1]['end_location'])
def build_message(distance, start_loc, end_loc):
"""Build the status message based on distance and locations"""
message = "On leave"
if distance > 13144:
message += f", driving my EV back from WA"
elif distance > 2118:
message += f", driving my EV around WA"
elif distance > 0:
message += f", driving my EV to WA"
if distance > 0:
distance += random.randint(-5, 5)
message += f", {distance}kms travelled so far"
if start_loc != end_loc:
message += f", next stop {end_loc}"
else:
message += f", near {start_loc}"
message += ", returning July 21st. Contacts {CIM: Grant Gorfine, Inserts: Daniel Pate, DevOps: Rob Duncan, else: Andrian Zubovic}"
return message
def main():
test_mode = False # Set to True to run in test mode
time_scale = 1 # 1/600 # Set to 1/60 to run at 1 second per minute, 1 for normal speed
# Set start time to 7:30 AM AEST (UTC+10) on June 8th, 2025
aest = timezone(timedelta(hours=10))
start_time = datetime.now(aest)
date_offset = datetime(2025, 6, 8, 7, 30, 0, tzinfo=aest) - start_time
if test_mode:
logging.info("Running in test mode - status messages will not actually be set")
app = get_msal_app(client_id = "e6cda941-949f-495e-88f5-10eb45ffa0e7")
last_token_refresh = 0
# Token refresh interval (60 minutes in seconds)
TOKEN_REFRESH_INTERVAL = int(60 * 60) # Scale the 1 hour refresh interval
old_distance = -1
while True:
try:
# Check if we need to refresh the token
current_time = time.time()
if current_time - last_token_refresh >= TOKEN_REFRESH_INTERVAL or last_token_refresh == 0:
logging.info("Acquiring/refreshing access token...")
access_token = acquire_token(app, scope = ["https://graph.microsoft.com/Presence.ReadWrite"])
if not access_token:
logging.error("Failed to acquire token")
exit(1)
last_token_refresh = current_time
logging.info("Token successfully refreshed")
# Set the status message
now = datetime.now(aest) # Get current time in AEST
if time_scale != 1:
# Adjust the current time based on the time scale
now = start_time + (now - start_time) / time_scale
now += date_offset # Adjust to the target start time
distance, start_loc, end_loc = get_trip_info(now) # We only need distance for comparison
if distance != old_distance:
message = build_message(distance, start_loc, end_loc)
timestamp = now.strftime("%Y-%m-%d %H:%M:%S %Z")
if not test_mode:
logging.info(f"[{timestamp}] Message: {message}")
success = set_teams_status_message(
access_token = access_token,
user_id = "1b625872-d8a8-42f4-b237-dfa6d8062360",
status_message = message,
)
else:
logging.info(f"[TEST MODE] [{timestamp}] Message: {message}")
success = True
else:
logging.debug("Status message has not changed, skipping update")
success = True
old_distance = distance
if success:
wait_time = 900 * time_scale # Scale the 15 minute wait time
logging.debug(f"Waiting {wait_time} seconds before updating status message again...")
time.sleep(wait_time)
else:
last_token_refresh = 0 # Reset token refresh time on failure
except KeyboardInterrupt:
logging.info("Status update interrupted by user. Exiting...")
break
except Exception as e:
logging.error(f"An error occurred: {e}")
time.sleep(300) # Wait 5 minutes before retrying
return 0
if __name__ == "__main__":
exit(main())
EOF
destination = "local/teamsstatus_standalone.py"
}
resources {
cpu = 500
memory = 256
}
}
restart {
attempts = 3
interval = "5m"
delay = "15s"
mode = "fail"
}
}
}

View File

@@ -1,12 +0,0 @@
# Disabled
# resource "nomad_job" "teamsstatus" {
# jobspec = file("${path.module}/teamsstatus.nomad.hcl")
# }
# resource "nomad_variable" "teamsstatus" {
# path = "nomad/jobs/teamsstatus"
# items = {
# token_cache_json = file("${path.module}/token_cache.json")
# }
# }

View File

@@ -2,53 +2,14 @@ resource "nomad_job" "transfer" {
jobspec = file("transfer.nomad.hcl") jobspec = file("transfer.nomad.hcl")
} }
resource "nomad_csi_volume_registration" "unraid_transfer" { module "unraid_smb_transfer" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/unraid_smb"
depends_on = [data.nomad_plugin.smb] name = "transfer"
plugin_id = "smb" id = "unraid_transfer"
share = "transfer"
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" = ""
}
} }
resource "nomad_csi_volume_registration" "unraid_appdata_transferfilebrowser" { module "appdata_transferfilebrowser" {
#Note: Before chaning the definition of this volume you need to stop the jobs that are using it source = "./modules/appdata"
depends_on = [data.nomad_plugin.smb] name = "transferfilebrowser"
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"]
}
} }

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"
}
]
}