769 lines
40 KiB
Bash
769 lines
40 KiB
Bash
#!/bin/bash
|
||
# ==============================================================================================
|
||
# ========================== HOST1 CONFIGURATION — unRAID-Gmer4Lfe ============================
|
||
# ==============================================================================================
|
||
# HOST1-specific variables — credentials, container names, share paths, failover lists.
|
||
# Sourced after master.conf — values here extend shared profile arrays and add HOST1-specific
|
||
# identity, credentials, and container configuration.
|
||
#
|
||
# Sparse checkout (git) ensures HOST2 never receives this file.
|
||
# HOST2 never sees HOST1 credentials — clean separation at the file level.
|
||
#
|
||
# DO NOT put shared config here — thresholds, toggles, profiles belong in master.conf.
|
||
# DO NOT put HOST2 variables here — they belong in host2.conf.
|
||
#
|
||
# ── INDEX ─────────────────────────────────────────────────────────────────────────────────────
|
||
#
|
||
# ── IDENTITY & CONNECTIVITY ────────────────────────────────────────────────────────────────
|
||
# IDENTITY hostname, SSH key
|
||
# EMBY container name, URL, API key
|
||
# NOTIFICATIONS Discord webhook
|
||
# PARTNERSHIP auth containers, backup paths
|
||
#
|
||
# ── RSYNC ──────────────────────────────────────────────────────────────────────────────────
|
||
# DAILY SYNC SHARES media shares HOST1 owns and pushes to HOST2
|
||
# WEEKLY SYNC SHARES appdata shares synced weekly (Sunday 2:30am)
|
||
# CRITICAL SYNC SHARES appdata shares synced every 30 minutes
|
||
# BACKUP VERIFY shares for checksum verification against remote
|
||
# HOST1 RSYNC PROFILE host1-appdata profile for HOST1-specific appdata syncs
|
||
#
|
||
# ── DOCKER ─────────────────────────────────────────────────────────────────────────────────
|
||
# DOCKER DAILY RESTART containers restarted daily
|
||
# DOCKER WEEKLY RESTART containers restarted weekly
|
||
# DOCKER WATCHDOG memory limits, health URLs, required containers, ignore list
|
||
# DOCKER NETWORK CONNECT networks and containers for docker_network_connect.sh
|
||
#
|
||
# ── FALLBACK ───────────────────────────────────────────────────────────────────────────────
|
||
# DDNS DDNS containers managed by HOST1
|
||
# INTERNET LOSS containers stopped when internet is lost
|
||
# FALLBACK TIERS what HOST1 runs for HOST2 per tier
|
||
# TIER DELAYS how long HOST1 must be down before each tier activates on HOST2
|
||
# RSYNC WRITEBACK HOST1 appdata synced back on handback
|
||
#
|
||
# ── MEDIA ──────────────────────────────────────────────────────────────────────────────────
|
||
# MEDIA PERMISSIONS share list for media_shares_permissions.sh
|
||
# MEDIA CLEANER folder lists for media_cleaner.sh
|
||
#
|
||
# ── MONITORS ───────────────────────────────────────────────────────────────────────────────
|
||
# CERTIFICATE MONITOR domains checked for SSL expiry
|
||
# SMART HEALTH drives to skip in SMART monitoring
|
||
# ZFS REPORT pools to exclude from ZFS health report
|
||
#
|
||
# ── TRANSCODES ─────────────────────────────────────────────────────────────────────────────
|
||
# TRANSCODES ramdisk size, thresholds, SSD path, server array
|
||
#
|
||
# ── ARR STACK ──────────────────────────────────────────────────────────────────────────────
|
||
# DOWNLOADERS slskd, SABnzbd, qBittorrent credentials and URLs
|
||
# LIDARR URL, API key, path map
|
||
# SONARR URL, API key, path map
|
||
# RADARR URL, API key, path map
|
||
# ARR RECOVERY per-arr recovery toggles
|
||
#
|
||
# ── SYSTEM WATCHDOG ────────────────────────────────────────────────────────────────────────
|
||
# SYSTEM WATCHDOG per-host check toggles and NIC configuration
|
||
#
|
||
# ── RESOURCE MANAGER ───────────────────────────────────────────────────────────────────────
|
||
# RESOURCE MANAGER containers paused/stopped under memory pressure
|
||
#
|
||
# ==============================================================================================
|
||
|
||
# ==============================================================================================
|
||
# ── IDENTITY & CONNECTIVITY ───────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# ━━━ Identity ━━━
|
||
# HOST1 hostname lives in master.conf (not a credential — safe for all servers).
|
||
# SSH key used for all server-to-server operations — rsync, failover container commands.
|
||
# Must be in /root/.ssh/ and authorised in HOST2's /root/.ssh/authorized_keys.
|
||
HOST1_SSH_KEY="/root/.ssh/gmer4lfe_rsync_automation"
|
||
HOST1_OWNER="gmer4lfe"
|
||
HOST1_OWNER_EMAIL="gmer4lfe@gmail.com"
|
||
|
||
# ━━━ Emby ━━━
|
||
# Referenced by transcode_manager.sh, emby_session_report.sh, emby_database_repair.sh,
|
||
# weekly_sync_maintenance.sh, and HOST1_TRANSCODE_SERVERS below.
|
||
# API key: Emby Dashboard → API Keys → + New Key
|
||
HOST1_EMBY_CONTAINER="Emby"
|
||
HOST1_EMBY_URL="http://localhost:8096"
|
||
HOST1_EMBY_API_KEY="0c27448d93a7431f9ac63569f7655829"
|
||
|
||
# ━━━ Jellyfin ━━━
|
||
# API key: Jellyfin Dashboard → Administration → API Keys → + New Key
|
||
HOST1_JELLYFIN_CONTAINER="Jellyfin"
|
||
HOST1_JELLYFIN_URL="http://localhost:8095"
|
||
HOST1_JELLYFIN_API_KEY="4e820e7df74c4933acec212b1996314e"
|
||
|
||
# ━━━ Gitea ━━━
|
||
# Personal access token for gitea_ssh_setup.sh — registers this server's SSH public key
|
||
# with Gitea so git operations use key auth instead of passwords.
|
||
# Create in Gitea: Settings → Applications → Generate Token → scope: write:user
|
||
HOST1_GITEA_API_TOKEN=""
|
||
|
||
# ━━━ Notifications ━━━
|
||
# Discord webhook — leave blank to disable.
|
||
# Per-host so HOST1 and HOST2 can post to different channels or only one server notifies.
|
||
HOST1_DISCORD_WEBHOOK=""
|
||
|
||
# ━━━ Partnership ━━━
|
||
# HOST1 is always the owner (source of truth) unless --transfer has been run.
|
||
# See README-Partnership.md and master.conf PARTNERSHIP section for full lifecycle docs.
|
||
|
||
# Auth containers reconfigured on onboard/offboard.
|
||
# Format: "ContainerName|WebUIPort"
|
||
# On onboard → WebUI pointed at owner's Tailscale IP (mirror clicks NPM, gets owner's auth)
|
||
# On offboard → WebUI pointed back at localhost
|
||
HOST1_PARTNERSHIP_AUTH_WEBUIS=(
|
||
"NginxProxyManager|81"
|
||
"Lldap-Gmer4Lfe|17170"
|
||
"Authelia|9091"
|
||
"Authelia-Secondary|9092"
|
||
)
|
||
|
||
# XML templates (from this server's templates-user/) pushed to mirror during onboard.
|
||
# These become the mirror's active auth stack, backed by the rsync-synced appdata.
|
||
# Update filename if Lldap is renamed to drop the host suffix.
|
||
HOST1_PARTNERSHIP_AUTH_STACK=(
|
||
# Dependencies first — Mariadb/Redis must be healthy before Authelia starts
|
||
"my-Mariadb-Authelia.xml"
|
||
"my-Mariadb-Authelia-Secondary.xml"
|
||
"my-Redis-Authelia.xml"
|
||
"my-Redis-Authelia-Secondary.xml"
|
||
# Auth apps — deployed after their deps are confirmed healthy
|
||
"my-Authelia.xml"
|
||
"my-Authelia-Secondary.xml"
|
||
"my-NginxProxyManager.xml"
|
||
"my-Lldap-Gmer4Lfe.xml"
|
||
# Source of truth — must be available on HOST2 independently of the auth stack
|
||
"my-Gitea.xml"
|
||
)
|
||
|
||
# XML templates pushed to mirror for the arr stack during onboard.
|
||
# Deps (e.g. databases) first if any — same ordering rule as auth stack.
|
||
HOST1_PARTNERSHIP_ARR_STACK=(
|
||
# "my-Sonarr.xml"
|
||
# "my-Radarr.xml"
|
||
# "my-Lidarr.xml"
|
||
# "my-Prowlarr.xml"
|
||
# "my-Bazarr.xml"
|
||
)
|
||
|
||
# Paths HOST2 should collect during the grace window after offboard.
|
||
# Notified on offboard — no auto-deletion, HOST2 must collect manually within PARTNERSHIP_GRACE_HOURS.
|
||
HOST1_PARTNERSHIP_MIRROR_BACKUPS=(
|
||
# "/mnt/user/appdata-Fallback/Jayred365-Emby"
|
||
)
|
||
|
||
# Containers parked on this server when partnership is active.
|
||
# Stopped on onboard (owner deploys its stack instead), restarted on offboard.
|
||
HOST1_PARTNERSHIP_OWN_CONTAINERS=(
|
||
# "Emby"
|
||
# "NginxProxyManager"
|
||
)
|
||
|
||
# Emby admin provisioning — toggle is owner-only, credentials are per-host.
|
||
# Owner enables/disables the feature. Each host sets the account they want on the shared Emby.
|
||
# On onboard: owner reads mirror's HOST*_PARTNERSHIP_EMBY_ADMIN_* and creates that account.
|
||
# On offboard: account is deleted. Username collision → onboard exits with error.
|
||
HOST1_PARTNERSHIP_PROVISION_EMBY_ADMIN=false # owner controls whether Emby is shared
|
||
HOST1_PARTNERSHIP_EMBY_PORT=8096
|
||
HOST1_PARTNERSHIP_EMBY_ADMIN_USER="" # this server's desired Emby username
|
||
HOST1_PARTNERSHIP_EMBY_ADMIN_PASS="" # this server's desired Emby password
|
||
|
||
# ==============================================================================================
|
||
# ── RSYNC ─────────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# ━━━ Daily Sync Shares ━━━
|
||
# Shares HOST1 pushes to all other nodes every night (1am via daily_sync_maintenance.sh).
|
||
# Mesh model: every node pushes every media share — no ownership, no mirrors.
|
||
# arr_sync ensures all arr libraries converge (union). rsync spreads files (additive, no --delete).
|
||
# arr_cleanup removes true orphans based on local arr state.
|
||
# Any node can download content to any share — it propagates to all nodes on the next cycle.
|
||
# Nextcloud is intentionally one-directional (HOST1→HOST2 offsite backup — not arr-managed).
|
||
# Uses DEFAULT_RSYNC_OPTS from master.conf — no profile needed.
|
||
# For shares needing container stops or custom options — add a profile in master.conf.
|
||
HOST1_DAILY_SYNC_SHARES=(
|
||
/mnt/user/Books
|
||
/mnt/user/Intros
|
||
/mnt/user/Kids_Movies
|
||
/mnt/user/Kids_Tv_Shows
|
||
/mnt/user/Movies
|
||
/mnt/user/Music
|
||
/mnt/user/Music_Videos
|
||
/mnt/user/Nextcloud
|
||
/mnt/user/stand-up_comedy
|
||
/mnt/user/Sports
|
||
/mnt/user/Tv_Shows
|
||
/mnt/user/Anime_Shows-Old
|
||
/mnt/user/Anime_Movies-Old
|
||
/mnt/user/Anime_Movies
|
||
/mnt/user/Anime_Shows
|
||
)
|
||
|
||
# Personal encrypted shares — synced for offsite backup, independent of media shares.
|
||
# ZFS encrypted at dataset level — remote receives encrypted blocks, cannot read content.
|
||
# See README-Rsync_Setup.md for ZFS encryption setup before uncommenting.
|
||
HOST1_PERSONAL_SHARES=(
|
||
# /mnt/user/HOST1-Personal # uncomment after creating encrypted dataset
|
||
)
|
||
|
||
# ━━━ Weekly Sync Shares ━━━
|
||
# Appdata shares synced during the weekly maintenance window (Sunday 2:30am).
|
||
# Containers stopped both sides before sync — full clean state guaranteed.
|
||
# Profiles drive container stops, excludes, and options — configured in master.conf RSYNC section.
|
||
# Order matters — Emby first (larger transfer), then Critical-Data (auth stack).
|
||
HOST1_WEEKLY_SYNC_SHARES=(
|
||
"/mnt/user/Media_Server/Emby" # emby profile — full clean mirror
|
||
"/mnt/user/appdata-Fallback/Critical-Data" # critical-data profile — auth stack
|
||
)
|
||
|
||
# ━━━ Intermediate Sync Shares ━━━
|
||
# Shares synced every 4 hours by intermediate_sync_maintenance.sh.
|
||
# Uses DEFAULT_RSYNC_OPTS (no --delete) — for sub-daily propagation of metadata or watch state.
|
||
# Full media share sync stays in the daily window. Leave empty to skip mid-day rsync entirely.
|
||
HOST1_INTERMEDIATE_SYNC_SHARES=(
|
||
# Add shares here to enable mid-day rsync
|
||
# Example: "/mnt/user/Emby_Metadata"
|
||
)
|
||
|
||
# ━━━ Critical Sync Shares ━━━
|
||
# Appdata shares synced every 30 minutes by critical_sync_maintenance.sh.
|
||
# Format: "/path/to/share" or "/path/to/share|profile-name"
|
||
# Order matters — Critical-Data first (auth stack), then Emby dirty sync.
|
||
HOST1_CRITICAL_SYNC_SHARES=(
|
||
"/mnt/user/appdata-Fallback/Critical-Data|critical-fallback" # auth dirty sync — stays running
|
||
"/mnt/user/Media_Server/Emby|emby-fallback" # Emby dirty sync — stays running
|
||
)
|
||
|
||
# ━━━ Backup Verify ━━━
|
||
# Shares verified by backup_verify.sh — random file checksum comparison against remote.
|
||
# Leave empty to use HOST1_DAILY_SYNC_SHARES automatically.
|
||
# Sample size and minimum file size defined in master.conf.
|
||
HOST1_BACKUP_VERIFY_SHARES=(
|
||
# leave empty to use HOST1_DAILY_SYNC_SHARES automatically
|
||
)
|
||
|
||
# ━━━ HOST1 Rsync Profile — host1-appdata ━━━
|
||
# HOST1-specific appdata sync profile — extends the shared PROFILE_* arrays in master.conf.
|
||
# Use for appdata unique to HOST1 (Organizrv2, VaultWarden, UptimeKuma etc.)
|
||
# Shared appdata (auth stack, Emby) use dedicated profiles defined in master.conf.
|
||
# Run manually: bash Rsync/rsync.sh /mnt/user/appdata-Fallback/HOST1-Appdata --profile=host1-appdata
|
||
PROFILE_RSYNC_OPTS[host1-appdata]="-av --info=progress2 --bwlimit=${PROFILE_BW_LIMIT[host1-appdata]:-8000}"
|
||
PROFILE_BW_LIMIT[host1-appdata]=8000
|
||
PROFILE_RETRY_COUNT[host1-appdata]=3
|
||
PROFILE_SLEEP[host1-appdata]=300
|
||
PROFILE_CRITICAL_CONTAINER_NAMES[host1-appdata]="Organizrv2-Gmer4Lfe UptimeKuma-Gmer4Lfe VaultWarden-Gmer4Lfe"
|
||
PROFILE_DELAYED_CONTAINERS[host1-appdata]=""
|
||
PROFILE_CONTAINER_DELAY[host1-appdata]=5
|
||
PROFILE_EXCLUDE_DIRS[host1-appdata]="logs *.tmp"
|
||
|
||
# ==============================================================================================
|
||
# ── DOCKER ────────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# ━━━ Docker Daily Restart ━━━
|
||
# Containers restarted every day via DAILY_MAINTENANCE_SCRIPTS.
|
||
# Dispatcharr degrades over time without restart — daily is intentional, not just housekeeping.
|
||
# Order matters — auth stack first, then media services.
|
||
HOST1_DAILY_RESTART_CONTAINERS=(
|
||
"NginxProxyManager"
|
||
"Lldap-Gmer4Lfe"
|
||
"Authelia"
|
||
"Authelia-Secondary"
|
||
"Dispatcharr-Iptv-Users"
|
||
"Dispatcharr" # Live TV scheduler — degrades without daily restart
|
||
"Dispatcharr-Basic"
|
||
"ErsatzTV-Emby"
|
||
"Slskd" # Soulseek connection drops after extended uptime; restart refreshes share index
|
||
)
|
||
|
||
# ━━━ Docker Weekly Restart ━━━
|
||
# Less critical services restarted weekly via WEEKLY_MAINTENANCE_SCRIPTS (Sunday 2:30am).
|
||
# Containers already stopped for weekly sync — restart adds zero extra downtime.
|
||
HOST1_WEEKLY_RESTART_CONTAINERS=(
|
||
"NextCloud"
|
||
"Organizrv2-Gmer4Lfe"
|
||
"AdGuard-Home"
|
||
"Immich-Gmer4Lfe"
|
||
)
|
||
|
||
# ━━━ Docker Watchdog ━━━
|
||
# Per-HOST1 container configuration for docker_watchdog.sh.
|
||
# Shared thresholds and toggles live in master.conf.
|
||
|
||
# Memory hard limits in MB — immediate restart if exceeded.
|
||
# Set at "container is clearly broken" not "container is busy".
|
||
# 20GB=20480 18GB=18432 16GB=16384 12GB=12288 8GB=8192 4GB=4096 2GB=2048 1GB=1024
|
||
declare -A HOST1_WATCHDOG_CONTAINERS=(
|
||
["Emby"]=18432 # 18GB — large library + active transcodes
|
||
["LidaTube"]=6144 # 6GB — memory leak over time
|
||
["Tdarr"]=6144 # 6GB — encoding is memory intensive
|
||
["Code-Server"]=1024 # 1GB — should never need more
|
||
)
|
||
|
||
# HTTP health check URLs — checked every cycle, strike system before restart.
|
||
# Only add containers with a meaningful web interface to check.
|
||
declare -A HOST1_WATCHDOG_CONTAINER_URLS=(
|
||
["Emby"]="http://localhost:8096"
|
||
)
|
||
|
||
# Required containers — must always be running on HOST1.
|
||
# Strike system before restart — repeated failures go on skip list, auto-clears on recovery.
|
||
# Listed in dependency order — dependencies before dependents.
|
||
HOST1_WATCHDOG_REQUIRED_CONTAINERS=(
|
||
"NginxProxyManager"
|
||
"Lldap-Gmer4Lfe"
|
||
"Mariadb-Authelia"
|
||
"Mariadb-Authelia-Secondary"
|
||
"Redis-Authelia"
|
||
"Redis-Authelia-Secondary"
|
||
"Authelia"
|
||
"Authelia-Secondary"
|
||
)
|
||
|
||
# Containers to skip in Tier 2 global scan — legitimately stopped or frequently restarting.
|
||
# Watchdog leaves these alone entirely — no restart attempts, no crash loop tracking.
|
||
HOST1_WATCHDOG_SCAN_IGNORE=(
|
||
"DashGate"
|
||
"PIA-WG-Config-Generator"
|
||
"Aperture"
|
||
"Aperture-Kids"
|
||
"pgvector-18-Apeture-Kids"
|
||
"Pgvector18-Aperture"
|
||
)
|
||
|
||
# Dependency ordering — skip restarting a container if its dependency is also down.
|
||
# Prevents watchdog from restarting Authelia before Mariadb is back up.
|
||
# SPACE-SEPARATED STRINGS — converted to array at runtime.
|
||
declare -A HOST1_WATCHDOG_DEPENDENCIES=(
|
||
["Authelia"]="Mariadb-Authelia Redis-Authelia"
|
||
["Authelia-Secondary"]="Mariadb-Authelia Redis-Authelia-Secondary"
|
||
["NextCloud"]="Postgres-NextCloud"
|
||
)
|
||
|
||
# Per-container appdata growth suppress ceilings in MB.
|
||
# ONLY needed in specific cases — growth rate detection covers all containers automatically.
|
||
# Use this when a container legitimately has large stable data and you want to guarantee
|
||
# it never triggers a false-positive growth alert. Growth warnings are suppressed while the
|
||
# container's dir stays below this ceiling; above it, warnings resume as normal.
|
||
# 50GB=51200 25GB=25600 20GB=20480 15GB=15360 10GB=10240 5GB=5120
|
||
declare -A HOST1_WATCHDOG_APPDATA_SIZES=(
|
||
["Tdarr"]="25600" # 25GB — transcode cache grows legitimately during active jobs
|
||
["7dtd"]="20480" # 20GB — game server world data, expected to be large
|
||
)
|
||
|
||
# ━━━ Network Watchdog ━━━
|
||
# Host-specific connectivity config for Watchdogs/System/network_watchdog.sh.
|
||
HOST1_NETWORK_WATCHDOG_DDNS_DOMAIN="gmer4lfe.com"
|
||
HOST1_NETWORK_WATCHDOG_DDNS_CONTAINER="Gmer4Lfe.com"
|
||
HOST1_NETWORK_WATCHDOG_NPM_URL="https://gmer4lfe.com"
|
||
|
||
# ━━━ Docker Network Connect ━━━
|
||
# Containers connected to custom networks at array start by docker_network_connect.sh.
|
||
# Networks created if they don't exist — idempotent, safe to re-run.
|
||
HOST1_NETWORK_CONNECT_CONTAINERS=(
|
||
"memcached"
|
||
"Npm-CrowdSec"
|
||
)
|
||
|
||
HOST1_NETWORK_CONNECT_NETWORKS=(
|
||
"high-availability"
|
||
)
|
||
|
||
# ==============================================================================================
|
||
# ── FALLBACK ──────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# ━━━ DDNS ━━━
|
||
# DDNS containers HOST1 manages — started/stopped by fallback.sh per DDNS absolute rules:
|
||
# Internet loss → stop immediately
|
||
# Failover → HOST2 starts HOST1's DDNS as Tier 1 (before any other containers)
|
||
# Handback → stop HOST1's DDNS on HOST2 → rsync → start containers → start local DDNS last
|
||
HOST1_DDNS_CONTAINERS=(
|
||
"Gmer4Lfe.com"
|
||
)
|
||
|
||
# ━━━ Internet Loss ━━━
|
||
# Containers stopped immediately on HOST1 when internet connection is lost.
|
||
# Prevents external-facing services from operating without connectivity.
|
||
FALLBACK_HOST1_STOP_ON_NO_NET=(
|
||
"Gmer4Lfe.com"
|
||
)
|
||
|
||
# ━━━ Fallback Tiers — HOST1 Runs for HOST2 ━━━
|
||
# Containers HOST1 starts when HOST2 goes down.
|
||
# Tier 1 is always immediate — vital services cannot wait.
|
||
# Higher tiers activate after HOST2_TIER*_DELAY minutes (set in host2.conf).
|
||
FALLBACK_HOST1_COVERS_HOST2_TIER1=(
|
||
"Gmer4Lfe.us"
|
||
"VaultWarden-Jayred365"
|
||
)
|
||
|
||
FALLBACK_HOST1_COVERS_HOST2_TIER2=(
|
||
# "container-placeholder"
|
||
)
|
||
|
||
FALLBACK_HOST1_COVERS_HOST2_TIER3=(
|
||
# "container-placeholder"
|
||
)
|
||
|
||
FALLBACK_HOST1_COVERS_HOST2_TIER4=(
|
||
# "container-placeholder"
|
||
)
|
||
|
||
# ━━━ Tier Delays — HOST1's Containers on HOST2 ━━━
|
||
# How long HOST1 must be down before each tier activates on HOST2 — in minutes.
|
||
# Tier 1 is always immediate — no delay var needed.
|
||
HOST1_TIER2_DELAY=240 # 4 hours — NextCloud, Immich
|
||
HOST1_TIER3_DELAY=720 # 12 hours — secondary services
|
||
HOST1_TIER4_DELAY=1440 # 24 hours — arrs + downloaders
|
||
|
||
# ━━━ Rsync Writeback — HOST1 Appdata Back on Handback ━━━
|
||
# Syncs HOST1 appdata BACK to HOST1 when it comes back online after a failover.
|
||
# Containers stopped before writeback — clean source, no competing writes.
|
||
#
|
||
# HOST1_TIER1_WRITEBACK_DELAY: short outages skip Tier 1 writeback — primary state
|
||
# is more reliable than dirty sync data for brief outages.
|
||
HOST1_TIER1_WRITEBACK_DELAY=60 # skip Emby writeback if outage under 1hr
|
||
|
||
# Tier 4 automatically syncs HOST1_DAILY_SYNC_SHARES — only list paths NOT in that array.
|
||
FALLBACK_HOST1_WRITEBACK_TIER1=(
|
||
"/mnt/user/Media_Server/Emby" # watch states built up during outage
|
||
)
|
||
|
||
FALLBACK_HOST1_WRITEBACK_TIER2=(
|
||
"/mnt/user/appdata-Fallback/Important-Data" # NextCloud + Postgres — files added during outage
|
||
)
|
||
|
||
FALLBACK_HOST1_WRITEBACK_TIER3=(
|
||
# "location-placeholder"
|
||
)
|
||
|
||
FALLBACK_HOST1_WRITEBACK_TIER4=(
|
||
"/mnt/user/appdata-Fallback/Arrs_Stack" # arr databases — downloads queued during outage
|
||
)
|
||
|
||
# ==============================================================================================
|
||
# ── MEDIA ─────────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# ━━━ Media Permissions ━━━
|
||
# Shares that media_shares_permissions.sh applies PERMISSIONS_MODE and PERMISSIONS_OWNER to.
|
||
# Runs first in DAILY_MAINTENANCE_SCRIPTS — arr cleanup depends on correct ownership.
|
||
HOST1_MEDIA_PERMISSION_SHARES=(
|
||
/mnt/user/Anime_Movies
|
||
/mnt/user/Anime_Movies-Old
|
||
/mnt/user/Anime_Shows
|
||
/mnt/user/Anime_Shows-Old
|
||
/mnt/user/appcache
|
||
/mnt/user/Books
|
||
/mnt/user/Downloads
|
||
/mnt/user/Games
|
||
/mnt/user/Intros
|
||
/mnt/user/Kids_Movies
|
||
/mnt/user/Kids_Tv_Shows
|
||
/mnt/user/Movie_Recordings
|
||
/mnt/user/Movies
|
||
/mnt/user/Music
|
||
/mnt/user/Music_Videos
|
||
/mnt/user/Photo
|
||
/mnt/user/Sports
|
||
/mnt/user/stand-up_comedy
|
||
/mnt/user/Temp_Storage
|
||
/mnt/user/Tv_Recordings
|
||
/mnt/user/Tv_Shows
|
||
/mnt/user/YouTube
|
||
)
|
||
|
||
# ━━━ Media Cleaner ━━━
|
||
# Folder lists for media_cleaner.sh — two profiles: anime and media.
|
||
# File patterns shared across all servers — defined in master.conf.
|
||
# Called via DAILY_MAINTENANCE_SCRIPTS. Run manually: Media/media_cleaner.sh anime|media
|
||
HOST1_ANIME_CLEAN_FOLDERS=(
|
||
/mnt/user/Anime_Movies
|
||
/mnt/user/Anime_Movies-Old
|
||
/mnt/user/Anime_Shows
|
||
/mnt/user/Anime_Shows-Old
|
||
)
|
||
|
||
HOST1_MEDIA_CLEAN_FOLDERS=(
|
||
/mnt/user/Kids_Movies
|
||
/mnt/user/Kids_Tv_Shows
|
||
/mnt/user/Movies
|
||
/mnt/user/Music
|
||
/mnt/user/Sports
|
||
/mnt/user/stand-up_comedy
|
||
/mnt/user/Tv_Shows
|
||
)
|
||
|
||
# ==============================================================================================
|
||
# ── MONITORS ──────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# ━━━ Certificate Monitor ━━━
|
||
# Domains checked via direct openssl connection — not relying on NPM's certificate state.
|
||
# Checks the actual certificate served, not what NPM thinks it has.
|
||
# Thresholds (CERT_WARN_DAYS, CERT_CRIT_DAYS) defined in master.conf.
|
||
HOST1_CERT_MONITOR_DOMAINS=(
|
||
"Gmer4Lfe.com"
|
||
"Gmer4Lfe.us"
|
||
)
|
||
|
||
# ━━━ SMART Health ━━━
|
||
# Drives skipped in SMART attribute monitoring — hardware is server-specific.
|
||
# Thresholds read from dynamix.cfg at runtime — fallbacks in master.conf.
|
||
HOST1_SMART_IGNORE_DRIVES=(
|
||
"sda" # boot USB — SMART not meaningful on flash drives
|
||
)
|
||
|
||
# ━━━ ZFS Report ━━━
|
||
# Pools excluded from the weekly ZFS health report — reduces noise from single-disk array pools.
|
||
# These are individual array disks formatted as ZFS — converting to XFS over time via unBalance.
|
||
# Pool health thresholds defined in master.conf.
|
||
HOST1_ZFS_REPORT_IGNORE_POOLS=(
|
||
"disk5"
|
||
"disk6"
|
||
"disk8"
|
||
"disk9"
|
||
"disk10"
|
||
)
|
||
|
||
# ==============================================================================================
|
||
# ── TRANSCODES ────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
|
||
# Ramdisk size ceiling — tmpfs only uses RAM actually needed, not the full size upfront.
|
||
# Real-world: 9 streams peaked at ~5.5GB — 10G gives generous headroom on 128GB RAM.
|
||
HOST1_RAMDISK_SIZE="10G"
|
||
|
||
# Usage thresholds — coupled to HOST1_RAMDISK_SIZE, adjust all three together if size changes.
|
||
# Hysteresis gap (8.5 - 7 = 1.5GB) prevents flip-flop between ramdisk and SSD.
|
||
HOST1_RAMDISK_WARN_GB=8.5 # flip to SSD when ramdisk usage reaches this
|
||
HOST1_RAMDISK_LOW_GB=7 # flip back to ramdisk when usage drops to this
|
||
|
||
# SSD fallback path — where transcodes land when ramdisk exceeds HOST1_RAMDISK_WARN_GB.
|
||
# Must be on cache pool — array disks too slow for active transcode writes.
|
||
HOST1_TRANSCODE_SSD="/mnt/cache/Temp_Storage/Emby/Transcodes/"
|
||
|
||
# Media servers sharing the ramdisk transcode space on HOST1.
|
||
# Format: "ContainerName|URL|APIKey|Type" — Type: emby | jellyfin | plex
|
||
# Entries with placeholder API keys are skipped automatically.
|
||
# ⚠️ Tdarr does NOT belong here — keep Tdarr on SSD, not ramdisk.
|
||
HOST1_TRANSCODE_SERVERS=(
|
||
"${HOST1_EMBY_CONTAINER}|${HOST1_EMBY_URL}|${HOST1_EMBY_API_KEY}|emby"
|
||
"${HOST1_JELLYFIN_CONTAINER}|${HOST1_JELLYFIN_URL}|${HOST1_JELLYFIN_API_KEY}|jellyfin"
|
||
)
|
||
|
||
# ==============================================================================================
|
||
# ── ARR STACK ─────────────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
# Used by arr cleanup scripts and arrs_failed_stalled_recovery.sh.
|
||
# detect_hosts() selects HOST1 vars when running on HOST1.
|
||
#
|
||
# PATH MAPS — container path → host path translation.
|
||
# Arr stores file paths using container-internal paths — scripts need host paths to scan.
|
||
# Add one entry per root folder in arr Settings → Media Management → Root Folders.
|
||
|
||
# ━━━ Downloaders ━━━
|
||
# Used by downloaders_reset.sh — runs every 30min via CRITICAL_MAINTENANCE_SCRIPTS.
|
||
# Clears stuck states, purges old history, prepares each client for a clean cycle.
|
||
|
||
# slskd — clears stuck searches, dead transfers, purges expired failed imports.
|
||
# SLSKD_FAILED_IMPORTS_DIR: where Soularr moves albums Lidarr rejected.
|
||
HOST1_SLSKD_URL="http://localhost:8980"
|
||
HOST1_SLSKD_API_KEY="4bF9kL2mNpQrT7vWxYz1A3dEgHjKoRsU"
|
||
HOST1_SLSKD_FAILED_IMPORTS_DIR="/mnt/user/Temp_Storage/Slskd/completed/failed_imports"
|
||
|
||
# SABnzbd
|
||
HOST1_SABNZBD_URL="http://localhost:8180"
|
||
HOST1_SABNZBD_API_KEY="8bfefe41d83b4d50883e32859b55ca9a"
|
||
|
||
# qBittorrent — deleteFiles=false removes torrent from qBit but leaves files on disk.
|
||
# Radarr/Sonarr manage actual files independently.
|
||
HOST1_QBIT_URL="http://localhost:8080"
|
||
HOST1_QBIT_USERNAME="root"
|
||
HOST1_QBIT_PASSWORD="Stay0utD!ck"
|
||
|
||
# ━━━ Lidarr — HOST1 only ━━━
|
||
# HOST2 does not run Lidarr — HOST1_LIDARR_RECOVERY flag handles the exit cleanly.
|
||
HOST1_LIDARR_URL="http://localhost:8686"
|
||
HOST1_LIDARR_API_KEY="b2977e71ef074bc0a0529d9fcce3b2dc"
|
||
HOST1_LIDARR_MUSIC_ROOT="/mnt/user/Music-New"
|
||
HOST1_FANART_API_KEY="Yd7147a43b692df0b364b94dc47efb81"
|
||
HOST1_LASTFM_API_KEY="be6dc169c33ae263e690c30d18b7491d"
|
||
|
||
declare -A HOST1_LIDARR_PATH_MAP=(
|
||
["/ext-music"]="/mnt/user/Music-New"
|
||
)
|
||
|
||
# ━━━ Sonarr ━━━
|
||
HOST1_SONARR_URL="http://localhost:8989"
|
||
HOST1_SONARR_API_KEY="130decd3db5b4c25afad64864cd03f9f"
|
||
HOST1_SONARR_TV_ROOT="/mnt/user/Tv_Shows"
|
||
|
||
# Note: stand-up_comedy in both Sonarr + Radarr — TV specials and movie specials, one folder
|
||
declare -A HOST1_SONARR_PATH_MAP=(
|
||
["/tv"]="/mnt/user/Tv_Shows"
|
||
["/ext-standup-comedy"]="/mnt/user/stand-up_comedy"
|
||
["/kids tv"]="/mnt/user/Kids_Tv_Shows"
|
||
["/ext-anime-shows"]="/mnt/user/Anime_Shows-Old"
|
||
)
|
||
|
||
# ━━━ Radarr ━━━
|
||
HOST1_RADARR_URL="http://localhost:7878"
|
||
HOST1_RADARR_API_KEY="d43a3ec6cf1549edb4af0cc63f98b2a9"
|
||
HOST1_TMDB_API_KEY="3dac5e2e49b5540472d2eafec4f01260"
|
||
HOST1_RADARR_MOVIES_ROOT="/mnt/user/Movies"
|
||
|
||
# Note: stand-up_comedy in both Radarr + Sonarr — movie specials and TV specials, one folder
|
||
declare -A HOST1_RADARR_PATH_MAP=(
|
||
["/movies"]="/mnt/user/Movies"
|
||
["/kids movies"]="/mnt/user/Kids_Movies"
|
||
["/ext-stand-up-comedy"]="/mnt/user/stand-up_comedy"
|
||
["/anime-movies"]="/mnt/user/Anime_Movies-Old"
|
||
)
|
||
|
||
# ━━━ Arr Recovery Toggles ━━━
|
||
# false = skip that arr on this host — exits cleanly without error
|
||
HOST1_SONARR_RECOVERY=true
|
||
HOST1_RADARR_RECOVERY=true
|
||
HOST1_LIDARR_RECOVERY=true # HOST1 only — exits cleanly on HOST2
|
||
|
||
# ==============================================================================================
|
||
# ── SYSTEM WATCHDOG ───────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
# Per-host check toggles and NIC config for system_watchdog.sh.
|
||
# Aliased by detect_hosts() — script uses unprefixed SYS_WATCHDOG_* names.
|
||
# HOST1: TR1950X 128GB — full media server, active transcoding, ZFS cache pools.
|
||
#
|
||
# Three-tier response — all critical checks enabled by default on HOST1:
|
||
# Tier 1 (bypass strikes, reboot now): docker daemon, rootfs full, kernel oops, FD, /boot
|
||
# Tier 2 (bypass strikes with OOM): RAM critical + OOM kills in cycle
|
||
# Tier 3 (standard strike system): everything else
|
||
#
|
||
# RAM tiers, OOM limits, and reboot loop settings in master.conf System Watchdog section.
|
||
|
||
# ━━━ Primary NIC ━━━
|
||
# Network interface for NIC state check — verify with: ip link show | grep "^[0-9]"
|
||
# Common values: eth0, bond0, br0, eno1
|
||
HOST1_SYS_WATCHDOG_NIC="eth0"
|
||
|
||
# ━━━ Tier 1 — Critical Checks ━━━
|
||
# These bypass the strike system — a single hit triggers immediate reboot.
|
||
# Disabling any of these is not recommended — they protect against acute system failure.
|
||
|
||
# Docker daemon unresponsive → try restart, reboot if restart fails.
|
||
# Without a working daemon docker_watchdog.sh is blind and containers cannot be managed.
|
||
HOST1_SYS_WATCHDOG_CHECK_DOCKER_DAEMON=true
|
||
|
||
# rootfs at critical threshold (ROOTFS_CRITICAL_PCT=99) → reboot immediately.
|
||
# At 99% rootfs writes fail silently — logs stop, Docker errors out, SSH may stop working.
|
||
# Standard 95% threshold still uses strike system — only 99%+ is critical tier.
|
||
HOST1_SYS_WATCHDOG_CHECK_ROOTFS=true
|
||
|
||
# Kernel BUG/Oops in dmesg delta since last cycle → reboot immediately.
|
||
# A kernel oops means the kernel ran with a corrupted state — stability is not guaranteed.
|
||
HOST1_SYS_WATCHDOG_CHECK_KERNEL_OOPS=true
|
||
|
||
# File descriptor exhaustion at FD_CRITICAL_PCT (95%) → reboot immediately.
|
||
# At 95% FD: new connections fail, Docker can't spawn processes, SSH drops.
|
||
HOST1_SYS_WATCHDOG_CHECK_FD=true
|
||
|
||
# /boot read-only detected → reboot immediately.
|
||
# Unexpected read-only /boot means state files and config writes are silently failing.
|
||
# Fallback state, watchdog reboot log, and lock files all go stale silently.
|
||
HOST1_SYS_WATCHDOG_CHECK_BOOT=true
|
||
|
||
# ━━━ Tier 2 — Urgent OOM Check ━━━
|
||
# Bypass strikes when RAM is critically low AND OOM kill rate confirms active crisis.
|
||
# Both must be enabled for Tier 2 bypass to function — disable either to always use strikes.
|
||
|
||
# Track kernel OOM kills each cycle via /proc/vmstat oom_kill delta.
|
||
# Also provides diagnostic context in reboot messages (which processes were killed).
|
||
HOST1_SYS_WATCHDOG_CHECK_OOM=true
|
||
|
||
# Free RAM check — required for both Tier 2 bypass and RAM tier logic.
|
||
# Tiers: MEM_WARN_GB(10) → notify | MEM_SHUTDOWN_GB(6) → stop containers | MEM_GB(4) → strikes
|
||
HOST1_SYS_WATCHDOG_CHECK_RAM=true
|
||
|
||
# ━━━ Tier 3 — Standard Checks (strike system) ━━━
|
||
# Each check must fail SYS_WATCHDOG_STRIKE_LIMIT consecutive cycles before action is taken.
|
||
# Single spikes are ignored — sustained problems trigger reboot.
|
||
|
||
# /var/log filesystem usage above SYS_WATCHDOG_LOG_PCT.
|
||
# Log spam (Docker log storms, syslog loops) fills rootfs — indicates something broken.
|
||
HOST1_SYS_WATCHDOG_CHECK_LOG=true
|
||
|
||
# ZFS ARC memory pinned above SYS_WATCHDOG_ARC_PINNED_PCT after cache drop.
|
||
# Enabled on HOST1 — ZFS cache pools actively used. Disable on hosts without ZFS.
|
||
HOST1_SYS_WATCHDOG_CHECK_ARC=true
|
||
|
||
# CPU temperature above SYS_WATCHDOG_CPU_TEMP_MAX (95°C).
|
||
# Sustained high temp causes kernel throttling or panic. Requires lm-sensors.
|
||
HOST1_SYS_WATCHDOG_CHECK_CPU_TEMP=true
|
||
|
||
# Load average above SYS_WATCHDOG_LOAD_MULTIPLIER × core count.
|
||
# DISABLED on HOST1 — Tdarr and Emby cause legitimate sustained load spikes during encoding.
|
||
# Enable on idle servers or adjust SYS_WATCHDOG_LOAD_MULTIPLIER if load is always high.
|
||
HOST1_SYS_WATCHDOG_CHECK_LOAD=false
|
||
|
||
# Zombie process count above SYS_WATCHDOG_ZOMBIE_LIMIT (50).
|
||
# Large zombie counts indicate serious process management failure — something is stuck.
|
||
HOST1_SYS_WATCHDOG_CHECK_ZOMBIES=true
|
||
|
||
# Check docker_watchdog.sh persistent skip list — required containers on skip list.
|
||
# Cross-watchdog coordination: if docker_watchdog gave up, system_watchdog escalates.
|
||
# ENABLED — HOST1 fully built and operational, skip list is meaningful.
|
||
HOST1_SYS_WATCHDOG_CHECK_CONTAINERS=true
|
||
|
||
# /tmp filesystem usage above SYS_WATCHDOG_TMP_PCT with auto-clear attempt.
|
||
# Script tries to clear aged /tmp files first — only strikes if clear fails.
|
||
# Lock files, rsync temp files, and Docker ops use /tmp — 100% means lock failures.
|
||
HOST1_SYS_WATCHDOG_CHECK_TMP=true
|
||
|
||
# Array disk error count delta in /proc/mdstat — accumulating errors = disk failing now.
|
||
# Triggers on SYS_WATCHDOG_MDSTAT_ERROR_LIMIT new errors in one cycle.
|
||
HOST1_SYS_WATCHDOG_CHECK_MDSTAT=true
|
||
|
||
# Primary NIC operstate — detects NIC going down (physical or driver failure).
|
||
# Uses HOST1_SYS_WATCHDOG_NIC above. Strike system — brief flaps don't trigger reboot.
|
||
HOST1_SYS_WATCHDOG_CHECK_NETWORK=true
|
||
|
||
# sshd running check — attempts restart before escalating.
|
||
# sshd down = no remote access. Script tries rc.sshd start, notifies, strikes on failure.
|
||
HOST1_SYS_WATCHDOG_CHECK_SSHD=true
|
||
|
||
# Runaway process detection — single process above SYS_WATCHDOG_RUNAWAY_CPU_PCT sustained.
|
||
# DISABLED — Tdarr encoding and Emby transcoding legitimately peg CPU for extended periods.
|
||
# Enable only if HOST1 has no CPU-intensive workloads.
|
||
HOST1_SYS_WATCHDOG_CHECK_RUNAWAY=false
|
||
|
||
# ==============================================================================================
|
||
# ── RESOURCE MANAGER ──────────────────────────────────────────────────────────────────────────
|
||
# ==============================================================================================
|
||
# Containers to manage under pressure — see master.conf RW_CRITICAL_CONTAINERS for exclusions.
|
||
|
||
# docker pause at medium pressure (RAM < RW_RAM_MEDIUM_GB or load > medium threshold)
|
||
# Suspended in-place — instant to pause/unpause, no state lost, no restart delay.
|
||
HOST1_RW_PAUSE_CONTAINERS=(
|
||
"Huntarr" # arr search automation — safe to suspend
|
||
"Cleanuparr" # download cleanup — safe to suspend
|
||
"Healarr" # arr health checks — safe to suspend
|
||
"Soularr" # Slskd automation — background only
|
||
"ChannelTube" # YouTube archiver — background only
|
||
"Pinchflat" # YouTube archiver — background only
|
||
)
|
||
|
||
# docker stop at hard pressure (RAM < RW_RAM_HARD_GB)
|
||
# Full stop — these are optional/heavy services that free significant RAM when stopped.
|
||
# resource_watchdog.sh restarts them when pressure fully clears (RAM >= RW_RAM_RECOVER_GB).
|
||
HOST1_RW_STOP_CONTAINERS=(
|
||
"LocalAI" # GPU/CPU heavy — largest RAM consumer when idle
|
||
"7DaysToDie" # game server — optional
|
||
"V-Rising" # game server — optional
|
||
"Code-Server" # IDE — not needed during pressure events
|
||
)
|
||
|
||
# ==============================================================================================
|
||
# ──────────────────────── End Of HOST1 Variables ──────────────────────────────────────────────
|
||
# ============================================================================================== |