install.sh generates the sunrise, sunrise-upload, seasons, and moon jobs
as Type=oneshot with only OnFailure=notify - a transient ffmpeg/network
blip fails the whole day's job with just an alert, no retry.
Add systemd drop-in overrides (Restart=on-failure, RestartSec=60,
StartLimitBurst=3 within a 10 min window) for each of these units after
install.sh runs. Drop-ins live outside the files install.sh generates,
so they survive re-running install.sh (e.g. after editing
sky-cam.conf), unlike a direct edit to the generated unit which would
be silently overwritten next time. systemd only fires OnFailure once
retries are exhausted, so this doesn't add notification spam - just
one alert after 3 tries, 60s apart.
capture.sh/capture-watchdog.sh already have Restart=on-failure baked
into install.sh's own generation (Type=simple, long-running) and don't
need this.
capture.sh already replaces motionEye/any NVR itself - it just needs
each camera's RTSP URL. services/sky-cam.sh never actually prompted for
CAM_RTSP_<cam>, so capture/audio never had anything to connect to.
- Prompt per camera for its RTSP URL -> CAM_RTSP_<cam> in .env
- Prompt for sunrise mic / optional ambient audio library
- Fix Mattermost integration: sunrise2mm.py reads mattermost_url/
access_token/channel_id (bot-token REST upload), not the
MM_WEBHOOK_URL/MM_CHANNEL incoming-webhook scheme the installer used
to write - uploads never worked before this
- Add optional ntfy push notifications
- Auto-generate SCHEDULE_SEASONS_<cam> (staggered 30 min apart) for
every configured camera, not just the stock east/north/south, so
install.sh wires up every applicable systemd timer for any camera set
Removes services/sky-cam-frigate.sh entirely - routing sky-cam's frames
through Frigate (via export API or restream) turned out to be solving a
problem that doesn't exist; sky-cam's own capture.sh talking directly to
each camera is simpler and has no quality/resolution tradeoffs. Frigate
continues to run fully independently for NVR/detection.
The real sky-cam repo's capture.sh already replaces MotionEye/any NVR
itself (plain ffmpeg RTSP frame-grab) - it and daily_sunrise_video.sh's
optional audio capture are the only places that touch a camera's RTSP
URL directly. Every other script (4-seasons, montage-mvt, year-end-join,
moon-track, moon-phase-monthly) only reads JPEGs/audio already on disk.
So the entire motionEye->Frigate transition is pointing CAM_RTSP_<cam>
at Frigate's go2rtc restream (rtsp://<frigate-host>:8554/<cam>) instead
of the camera directly - no upstream script changes needed. Replaces
the previous frigate-retime.sh/export-API approach, which solved a
problem (matching an arbitrary recording length to music duration) that
sky-cam's own 4-seasons.sh/montage-mvt.sh already handle via JPEG frame
counts.
Duplicates services/sky-cam.sh into a Frigate-backed variant that pulls
recordings via Frigate's export API instead of a JPEG image folder.
Includes a frigate-retime.sh helper that exports a coarse timelapse,
measures its actual duration with ffprobe, and re-encodes once with a
computed setpts factor to hit an exact target length (e.g. a Four
Seasons movement's runtime).
Group membership added by 'usermod -aG docker' (in require_docker) doesn't
apply to the shell that invoked sudo — only to new logins. Users had to
manually run 'newgrp docker' or reconnect SSH after every install. Since a
child process can't change its parent shell's group list directly, the
practical fix is to exec a fresh 'su - ' login shell at the end
of the guided flow, which re-reads /etc/group and lands the user back in
the same terminal with docker access already active.
Gated on: running via sudo (SUDO_USER set), interactive (not --unattended),
docker group exists and the user is actually a member, and stdin is a real
tty — so this never fires for scripted/explicit-service/piped invocations.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Previously re-running the installer always overwrote config.yml and .env
from scratch, silently discarding any real camera credentials already on
disk. Now install_frigate parses an existing config.yml + .env (best-effort,
matching this installer's own output shape) and presents a numbered list
of detected cameras with a menu:
[1] Keep everything as-is (no changes at all)
[2] Backup existing config and start fresh
[3] Add more cameras (keep these)
[4] Remove cameras (choose numbers, or 'all'), then optionally add more
Implementation switches from building config.yml/.env as concatenated text
blocks inline in the collection loop to parallel CAM_* bash arrays
(name/ip/port/var-names/enabled/notify/substream-suffixes), so cameras can
be parsed, listed, removed, and re-rendered independently:
- _frigate_parse_existing: reads go2rtc streams + cameras: enabled/notifications
from config.yml, and credential values from .env, into the CAM_* arrays
- _frigate_review_existing: numbered menu, mutates arrays per choice
- _frigate_next_suffix_int: kept cameras retain their existing FRIGATE_RTSP_USER[N]
var names unchanged; new cameras get the next unused numeric suffix so
credentials never collide after removals
- _frigate_camera_wizard / _frigate_render_config: same prompts and output
shape as before, now array-driven so kept + new cameras render uniformly
- _frigate_backup_existing: copies config.yml/.env/docker-compose.yml to a
timestamped backup-YYYYMMDD-HHMMSS/ dir before any destructive rewrite
Verified with a 5-scenario test harness (fresh install, keep-as-is producing
byte-identical output, add-camera preserving existing credentials, remove-
by-number without var collisions, backup-and-fresh) before committing.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Previously Frigate always wrote a placeholder config.yml the operator had
to hand-edit to add cameras. Now install_frigate prompts to add cameras
one at a time (name, RTSP IP/port/user/password/path, optional sub-stream,
enabled, notifications), matching the go2rtc + cameras structure used in
production frigate configs:
- Each camera gets a go2rtc stream entry (+ optional _sub for detection)
and a cameras: block with ffmpeg inputs/roles, detect, notifications
- RTSP credentials/IPs are written to .env as FRIGATE_* variables (first
camera gets FRIGATE_RTSP_USER/PASSWORD, later cameras get numbered
suffixes _1, _2, ... to avoid collisions) and referenced in config.yml
via Frigate's {FRIGATE_VAR} substitution syntax — secrets never appear
in the YAML directly
- docker-compose.yml now includes env_file: .env so those vars actually
reach the container for substitution to work
- Skipping all camera prompts falls back to the original starter
config.yml for manual editing, preserving existing behavior
- README and DRY-RUN summary updated to reflect the new flow
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Defaults install_asterisk() to FQDN networking mode and prompts for VLAN/VPN
subnets (with host-network auto-detection to filter out noise like Docker
bridges) so phones on other networks get correct NAT/SDP handling from the
first boot.
The container now mounts Caddy's cert store read-only when Caddy is
installed, and the entrypoint syncs a matching Let's Encrypt cert for
DOMAIN_NAME automatically, re-checking every 12h to pick up renewals without
a restart. Falls back to self-signed only when no matching cert is found.
Also fixes a real bug hit in the field: a preserved/migrated pjsip.conf could
be missing the transport-udp/transport-tcp sections entirely, with no bind
error logged, silently blocking any device that registers without TLS. Adds
the same migration-injection already used for transport-tls.
The 'Install Caddy now?' prompt ran regardless of the just-answered
Caddy location question, so choosing 'remote' still asked whether to
install Caddy locally — contradicting the choice made one prompt earlier.
Gate it on CADDY_MODE being local (or unset, for configs predating the
wizard split).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Previously the Caddy-location question lived inside run_site_configure,
gated behind 'Configure site defaults now? (y/n)'. Answering 'n' (e.g.
because Caddy is on a different box and you don't care about domain/tz
autofill) meant CADDY_MODE never got set, which silently disabled Caddy
prompts for every service for the life of the install (configure_caddy_for_service
falls through to mode 'none' and returns immediately).
Split into two steps:
1. ask_caddy_location() — always runs on first setup.sh invocation,
independent of any other prompt, and persists CADDY_MODE immediately.
2. run_site_configure() — now only asks timezone/domain/Caddy-network,
and is only offered when CADDY_MODE=local (those defaults are only
useful for FQDN autofill tied to a locally-managed Caddyfile).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Reorders the site defaults wizard so 'Where does Caddy run?' comes before
timezone/domain, since it's the more fundamental choice and the answer
context matters when explaining the other prompts. Also skips the Caddy
Docker network prompt entirely when Caddy isn't running locally — that
setting is only relevant to services joining a local Caddy container's
bridge network; remote/none mode proxies via localhost:PORT + snippet
files instead.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
The get.docker.com convenience script internally wraps every step in
'sudo -E sh -c ...'. On minimal/cloud Ubuntu images that never installed
the sudo package (common when operating purely as root), those internal
sudo calls silently fail while the outer script still exits 0 — apt never
actually runs, but no error surfaces. require_docker already runs as root,
so there's no need for sudo at all.
Replaced it with Docker's documented apt-repo steps run directly: add the
keyring, add the repo (with architecture/codename detected via dpkg and
os-release), apt-get install docker-ce + compose plugin, enable the
service. Real apt/curl/systemctl failures now propagate and print to the
terminal instead of being masked by the wrapper script.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
require_docker returning non-zero was silently ignored (no set -e).
Add explicit warning so the operator sees the failure; setup.sh already
has an unconditional Docker check after base that will retry.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
The Docker check+install was inside the else branch that only runs when
base has never been installed. On re-runs (base already present) Docker
was silently skipped and only warned about. Move the check outside the
if/else so Docker is always installed if missing, regardless of whether
base was skipped.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
whiptail fix:
- bootstrap.sh: redirect stdout and stderr to /dev/tty alongside stdin so
whiptail has full terminal control for raw mode (arrow keys, highlighting)
- setup.sh: run 'stty sane' on /dev/tty before the menu loop to reset any
stale terminal state from SSH reconnections or prior sessions
Installed-service summary:
- Print a grouped list of all currently-installed services before every
menu session so the operator knows the current state at a glance
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Two fixes:
1. Export TERM (default xterm-256color) early — whiptail needs a valid
TERM to enter raw mode; when bash is started via pipe TERM may be
unset, causing keypresses to leak to the shell instead of the menu
2. Add </dev/tty to both whiptail calls so keyboard input always comes
from the controlling terminal regardless of how stdin was redirected
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Three fixes:
1. configure_caddy_for_service: remove the '!= example.com' filter that
silently dropped any valid domain matching that string; now any non-empty
SITE_DOMAIN is used as the default subdomain suggestion
2. load_site_config: trim leading/trailing whitespace from key and val so
hand-edited .config files with extra spaces still parse correctly
3. setup.sh: call load_site_config after the site wizard saves so the
in-memory values are guaranteed fresh for all subsequent service installs
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
- require_docker now runs as part of base so Docker is present on every box
- Install openssh-server, offer GitHub (gh:) and Launchpad (lp:) key import
via ssh-import-id; disable password auth only after keys are confirmed imported
- Handle Ubuntu cloud-init drop-in that re-enables PasswordAuthentication
- Offer NetBird install with optional --allow-server-ssh flag and setup key
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
command -v may miss the binary if sudo stripped PATH; check the canonical
apt install location directly as a fallback before reporting failure, and
use the same fallback when printing the installed version.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Two bugs in require_docker:
1. apt post-install hooks (needrestart etc.) block on stdin which is
at EOF when running via pipe; DEBIAN_FRONTEND=noninteractive skips them
2. bash's command hash table doesn't pick up a newly installed binary;
hash -r flushes it so command -v docker finds /usr/bin/docker
Also moved usermod and success log after the binary check so [OK] only
prints when docker is actually reachable.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
curl | bash consumes stdin from the pipe, so when bootstrap hands off
to setup.sh the script gets EOF immediately and exits with 'Cancelled'.
Redirecting </dev/tty restores keyboard input for the interactive menu.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
BASH_SOURCE[0] is unbound when bash reads from a pipe; set -u turns
this into a fatal error that no amount of :- or set +u reliably fixes
across bash versions. $0 is always set: 'bash' when piped (dirname
gives '.' where no setup.sh exists), and the correct path when run
directly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
${BASH_SOURCE[0]:-} still triggers set -u when BASH_SOURCE is entirely
unset (not just empty) in pipe mode. Temporarily disable -u for that
single assignment, then restore it.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
set -euo pipefail causes ${BASH_SOURCE[0]} to abort with 'unbound variable'
when the script is fed via curl | bash. Use ${BASH_SOURCE[0]:-} so the
variable expands to an empty string in that context, letting SCRIPT_DIR
resolve safely and the pipe path continue to the git-clone branch.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LQJBvqzXeyuhhAcAA3Q5Wq
Three root-cause fixes found by reading the Docker image source:
1. Wine/Proton page fault: Docker's default seccomp profile blocks
system calls that Wine Proton GE requires. Fix: security_opt:
seccomp=unconfined + shm_size: 256m (Xvfb needs /dev/shm for
MIT-SHM extension; 64 MB default is too small).
Added network_mode: host for game traffic (dynamic UDP ports).
2. KYBER_MAP_ROTATION exit 64: the Kyber CLI decodes base64 and
parses newline-separated "MODE;MAP_PATH" lines, not JSON objects.
Our JSON [{map:...,mode:...}] format split on semicolons into one
field → ExitCode.usage (64). Fixed builder to emit MODE;MAP_PATH\n
lines and updated .env comment + README example.
3. GPU passthrough removed: Proton GE includes DXVK which crashes
headlessly when a GPU is passed through (no Vulkan display). The
server needs no GPU; removing passthrough is the correct fix.
Also: install libgamemode0:i386 on the host (Wine/Proton dep),
add alphanumeric-password warning (special chars → INVALID_PASSWORD).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014be1aK9G8CY2msho5LjxR4
dall-e-2/dall-e-3 retired May 12 2026 and gpt-image-1 deprecates Oct 23
2026, so move every OpenAI default (config.py, both provider classes,
both compose files, .env.example, the in-app provider-settings dropdown,
README) to gpt-image-2 for both generation and edits. Also fix response
parsing in ai_provider.py's OpenAIProvider, which never sent a model
param and assumed a url response — gpt-image-1/2 only return b64_json.
Separately, .env.example shipped AI_PROVIDER=replicate by default, but
replicate has no driver in remote_provider.py, so following the
documented "cp .env.example .env" setup silently broke every AI call
and defeated the GPU quick-start (an explicit non-empty .env value
overrides docker-compose.gpu.yml's own local_gpu fallback). Default to
local_gpu instead, mark replicate/stability as not-yet-implemented, and
recommend Lykon/dreamshaper-8-inpainting as a hands/face-tuned
HF_MODEL_INPAINT override for 4-6GB cards (Quadro P2200, GTX 1060/1660).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
inpaint()/img2img()/outpaint() called /v1/images/edits without a model
field, so OpenAI defaulted every cloud edit to dall-e-2 regardless of
configuration — while txt2img used dall-e-3. Add a separate
OPENAI_EDIT_MODEL (default gpt-image-1, the only current model that
supports masked edits at ChatGPT-comparable quality), thread it through
the provider and both compose files, and handle gpt-image-1's
b64_json-only response shape alongside the url shape dall-e-2/3 return.
setup.sh's run_service() looks up install_<name> using the literal
hyphenated registered name, not an underscore-converted one. borg-backup,
calibre-web, gaming-backup, and stirling-pdf all used underscored function
names and were therefore uninstallable ("has no install_<name>"). Same
fix already applied to ai-gpu/ai-stack; this closes out the rest of the
repo-wide audit.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
PaintPlus's backend already supported invokeai/comfyui providers (generic
"self-hosted, on another machine" remote APIs) but the installer never
exposed them and the compose file never passed the URLs through. Add a
3rd provider choice — shown only when the ai-stack service is installed —
that sets AI_PROVIDER + INVOKEAI_URL/COMFYUI_URL and joins ai-stack's
Docker network (ai-stack_default) so PaintPlus can reach those containers
by name. No cloud key, no extra GPU download: it rides on ai-stack's
already-running InvokeAI/ComfyUI.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
Add gpu-mode.sh (written into ~/docker/ai-stack at install): time-shares
one small GPU between local chat (Ollama, for Open WebUI) and local
image-gen (InvokeAI + ComfyUI, backends for PaintPlus), leaving the
always-on services (Open WebUI, Gitea, RAG, MCP, Kiwix) untouched. A new
"small local GPU?" prompt enables it: when on, the GPU defaults to chat
mode after install and the user is told to swap deliberately. Cloud
models need no swap.
Document the intended split in POST-INSTALL-NOTES: Open WebUI is the
text hub (chat/research/light coding, wired to Gitea + GitHub sync + RAG
+ MCP), PaintPlus is the single image hub (cloud or local comfyui/
invokeai backend). Note web search is DuckDuckGo — this installer path
generates no SearXNG, so the SearXNG references are dropped.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
Vendor the functional source of github.com/outis1one/local-ai into
./ai-stack (22 files) and add services/ai-stack.sh, which copies the
source to ~/docker/ai-stack and hands off to the app's VRAM-aware
installer (local-ai-setup.sh). The stack bundles Ollama, Open WebUI,
RAG + MCP servers, ChromaDB, SearXNG, Kiwix, Gitea, InvokeAI, ComfyUI
and Portainer.
Cloud LLM providers (Groq/DeepInfra/OpenAI/OpenRouter) are optionally
wired into Open WebUI via the plural OPENAI_API_BASE_URLS list, with the
local RAG connection kept as the first entry so RAG keeps working. Open
WebUI ships built-in auth, so Caddy is configured without Authelia.
Excludes the upstream's two bundled copies of this very project
(ubuntu-post-install.sh, ubuntu-post-install-main.zip) — stale and
circular. Coexists with the existing ai-gpu service.
Also fix the install-function names for ai-gpu and ai-stack: the
dispatcher calls install_<raw-name>, so the function must be
install_ai-gpu / install_ai-stack (hyphen), matching the working
mail-archiver / wg-easy services. ai-gpu was previously uninstallable.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
Shows a numbered menu of 12 known SWBF2 maps (Galactic Assault, Planetary
Battles, Starfighter Assault) during install. Selected maps are encoded as
base64 JSON and written to KYBER_MAP_ROTATION in .env.
Map IDs decoded from maprotation.hive binary: Geonosis, Kamino, Endor,
Tatooine, Death Star II, Starkiller Base, Crait, Hoth, Fondor (space).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Bring the full EditmaskwithAI application into the repo under paintplus/
(429 files) so the service is self-contained — the installer copies the
vendored source to ~/docker/paintplus/src instead of cloning at runtime.
Rename to PaintPlus (service + branding; app logic untouched):
- services/editmaskwithai.sh -> services/paintplus.sh (register_service
paintplus, install_paintplus, ~/docker/paintplus, Caddy paintplus:8000,
Authelia option preserved)
- container names -> paintplus across docker-compose*.yml; dev network
-> paintplus-network
- browser <title> -> "PaintPlus - AI Image Editor"; README heading ->
PaintPlus with upstream provenance note
- README utilities table: editmaskwithai -> paintplus
Backend/frontend code (help strings referencing the old container name,
the ai_photo_edit.db filename) is intentionally left as-is to avoid
touching application logic.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
Package the self-hosted mask-based AI photo editor
(github.com/outis1one/EditmaskwithAI) as a utilities service. Clones the
repo into ~/docker/editmaskwithai and deploys on port 3080.
Two install modes:
- Cloud API (no GPU): writes .env from the repo template with the chosen
AI_PROVIDER (OpenAI gpt-image or Replicate), API key, and a generated
SECRET_KEY, then `docker compose up -d --build`.
- Local GPU (NVIDIA): runs the repo's own install-local-gpu.sh and
bring-up-local-gpu.sh (downloads ~13 GB of models).
The app has no built-in auth, so it offers Authelia SSO and reverse-proxies
via Caddy (ai-photo-edit:8000). Cloud mode joins caddy_net through a
docker-compose.override.yml; GPU mode is attached with `docker network
connect` since the GPU compose runs with an explicit -f and does not merge
overrides.
Also lists the service in the README utilities table.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
When run directly without setup.sh, write_readme is not defined.
Add a minimal implementation to the else block of both kyber-server.sh
and sunshine.sh so standalone runs no longer fail with command not found.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Generalize the optional Groq toggle into a provider picker so the GPU AI
stack can wire Groq, DeepInfra, OpenAI, and/or OpenRouter into Open WebUI
at once. All are OpenAI-compatible, so they share Open WebUI's
OPENAI_API_BASE_URLS / OPENAI_API_KEYS (semicolon-separated, matched by
position). Keys are written to llm/.env; only ${VAR} references go into
docker-compose.yml.
Updates the install prompt, .env template, compose patch, README, the
dry-run note, and the final summary accordingly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Nb2vJ8W7bHKx1JXVvpCraH
If the kyber bin symlink and desktop entry are not present, prompt to
set up the launcher. Calls install_kyber_launcher directly if already
loaded (wizard), or sources kyber-launcher.sh when run standalone.
Skips the prompt on rerun if the launcher is already installed.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
If auth.toml has no token, launch the Kyber AppImage as the actual user
(with DISPLAY/XAUTHORITY passed through), prompt the user to log in via
the EA Account button, then re-read auth.toml once they close Kyber.
Falls back to manual token paste if auth.toml still has nothing.
Skips the AppImage entirely on rerun when a token is already cached.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Running kyber_cli get_token under sudo opens a broken browser OAuth flow.
Instead, read the token directly from ~/.local/share/maxima/auth.toml which
Kyber writes after a normal GUI login. If missing, show clear instructions
to run the AppImage as yourself first, then re-run the installer.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
On rerun, if ~/.local/share/maxima exists, offer to wipe it so EA login
starts completely fresh (fixes stuck/broken auth states).
Single-quote all .env values so passwords and tokens containing $, !, &,
spaces, and other shell special characters are written safely.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
The standard Steam path search missed installs on secondary drives.
Now also reads libraryfolders.vdf to find all Steam library paths, then
falls back to a broad find across home/mnt/media. Wolf's copy is excluded
from the fallback search to prefer the native Steam install.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Adds a Moonlight app entry that launches SWBF2 directly via Steam
(steam://rungameid/1237950) — both with and without the Kyber entry.
Moonlight app list now shows: Kyber SWBF2 | SWBF2 Single Player | Desktop
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
When SWBF2 is not found locally, prompt to rsync from a remote host.
Asks for remote host, username, and path (defaults to the standard Steam
Linux path). Files land in ~/docker/kyber-server/swbf2/ and that path is
used automatically. Falls back to manual local path entry if rsync declined.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Instead of hard-failing, show an rsync example for copying game files from
another machine, then prompt for the path. Validates that starwarsbattlefrontii.exe
exists at the given location before continuing.
Also silence "register_service: command not found" in all three services when
run standalone (guard with command -v check).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Steam must be running when Kyber is used. The installer now detects whether
Steam is present (command or ~/.steam path), and if not, prompts to install
it via apt (steam-installer or steam package, with i386 multiarch enabled).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
When run directly (sudo bash services/foo.sh), bash executes the bootstrap
block before reaching the install_* function definition below it, causing
"command not found". Moving the bootstrap to the end of each file ensures
the function is defined by the time the bootstrap calls it.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Installs the Kyber community multiplayer launcher (simonlinuxcraft/kyber-
linuxport-unofficial) via the setup.sh wizard. Handles glibc check, bwrap/
userns fix, GitHub release fetch, desktop entry, and bin symlink. Warns on
Intel integrated graphics and links to Sunshine as the streaming alternative.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Installs Sunshine (LizardByte) as a Moonlight-compatible streaming host,
offset to port 48090 so it coexists with Wolf. Auto-detects the Kyber
AppImage and adds a "Kyber SWBF2" app entry. Optional virtual display for
closed-lid/headless use. uinput udev rules for controller/mouse input.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
- Auto-detects SWBF2 install path; loud error with install instructions if missing
- Downloads Kyber AppImage if not present, extracts kyber_cli, runs get_token
(opens browser for EA OAuth — one-time step, token never expires)
- Prompts for EA credentials (written to .env, chmod 600) and server name
- Auto-detects GPU (NVIDIA/Intel/AMD) and adds appropriate passthrough config
- Writes docker-compose.yml + .env to ~/docker/kyber-server/
- Map rotation left blank by default with instructions in .env comments
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
- Add troubleshooting for 'Origin Error: title installed in language not
entitled to play' — Maxima's umu-run reg commands fail silently on some
systems; manual wine64 regedit import of Origin locale keys fixes it
- Add windowed-mode / HDR fix for SWBF2 crash during level load on
integrated GPUs (Intel Iris Xe, DXVK rendering crash)
- Document fix in both README.md and script header comments
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
- Bots are set via BOTS TEAM 1 / BOTS TEAM 2 sliders in HOST panel (not automatic)
- Correct launch order: Steam open but don't click Play; Kyber launches SWBF2 itself
- Document Alt+Tab / taskbar click workaround for window focus issue after game loads
- Add bwrap unprivileged userns fix (Ubuntu 24.04 blocks by default); script applies
automatically with sudo and persists via /etc/sysctl.d/99-userns.conf
- Script now checks/fixes userns at step 0 before download
- Kyber manages its own GE-Proton; no need to configure Proton for SWBF2 in Steam
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
The Windows kyber_launcher.exe approach under Wine/Proton has two
unfixable problems on Linux: Wine's cmd.exe crashes with STATUS_ACCESS_VIOLATION
on the long EA OAuth URL, and EA's auth callback uses the eadesktop://
URI scheme which has no Linux handler.
The correct solution is the native Linux port AppImage maintained at
github.com/simonlinuxcraft/kyber-linuxport-unofficial. It bundles Maxima
(open-source EA Desktop replacement) and handles OAuth natively — no shims,
no watchers, no per-Proton-update maintenance.
Script now: fetches latest release from GitHub API, downloads AppImage,
installs desktop entry and ~/.local/bin/kyber symlink.
README: add Gaming scripts section documenting Kyber setup, private server
hosting with bots, requirements, and what does not work.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
After extensive testing, Wine/Proton loads cmd.exe exclusively from
Proton's own installation (files/lib/wine/x86_64-windows/cmd.exe).
Replacing the prefix system32 copy, IFEO registry keys, WINEDLLOVERRIDES,
and Wine DllOverrides registry entries all have no effect — Proton's copy
always takes precedence.
The working fix: compile a tiny shim (GetCommandLineA scan for http URLs,
write to C:\kyber_oauth_url.txt, ExitProcess(0)) and replace Proton's
cmd.exe directly, backing up the original as cmd.exe.bak. The script
detects size (shim ~9KB vs real cmd.exe ~1.2MB) to avoid double-replacing.
Note for users: re-run setup-kyber-linux.sh after any Proton Experimental
update since Steam restores the original cmd.exe on update.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
system.reg values use \\ for one backslash, so C:\shim\kyber_cmd.exe
must be written as C:\\shim\\kyber_cmd.exe. The previous script wrote
C:\\\\shim\\\\kyber_cmd.exe (4 backslashes = two literal backslashes),
which Wine couldn't resolve.
Also detect when the shim slot contains Wine's cmd.exe (122231 bytes)
left behind by a failed earlier attempt, and recompile automatically.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Proton's prefix-setup phase overwrites system32/cmd.exe on every game
launch, breaking the previous approach of placing a shim there. Switch to
Windows Image File Execution Options (IFEO): the shim is compiled to
drive_c/shim/kyber_cmd.exe (Proton never touches that directory) and the
IFEO Debugger key is written to system.reg so Wine intercepts cmd.exe at
CreateProcess time without touching the filesystem binary.
Also fix the wine64 path used for winetricks: correct path is
files/lib/wine/x86_64-unix/wine64 not files/bin/wine64.
The shim is simplified to scan all args for any http URL (IFEO passes the
debuggee path as an extra arg, so the /c start pattern match was fragile).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Steam sets STEAM_COMPAT_DATA_PATH to compatdata/<appid> when launching
a non-Steam shortcut. The prefix was named 'kyber' so Steam was launching
into an empty prefix, causing immediate exit. Change KYBER_COMPAT_ID to
match KYBER_APPID (9900000001) and migrate any existing 'kyber' prefix.
Also handle cmd.exe being a symlink: resolve and copy the real file as
cmd-real.exe, then remove the symlink before writing the shim so we only
affect the Kyber prefix and not Proton globally.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Large stack array WCHAR[4096] triggered a stack probe (__chkstk_ms)
unavailable in -nostdlib builds. Fix: heap-allocate the pass-through
command line buffer, and add -mno-stack-arg-probe as a safety net.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
_wcsicmp is a CRT function unavailable with -nostdlib/-ffreestanding.
Replace with CompareStringOrdinal from kernel32 (always available).
Also fix step numbering (1-2 were still labeled /5 instead of /7),
and remove 2>/dev/null from compile so errors are visible if it fails.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
- Python shortcut script now checks for existing 'Kyber Launcher' entry
and exits early instead of appending a duplicate on each re-run
- Use sudo apt-get directly for mingw-w64 install so the password prompt
works (previous chained || approach suppressed the interactive prompt)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Wine's cmd.exe crashes (0xC0000005) when Kyber calls cmd /c start with
the EA auth URL, on native Linux Steam just as in Wolf. Fix:
1. Compile a minimal cmd shim with mingw-w64 (auto-installed if missing).
The shim intercepts cmd /c start <http url>, writes the URL to
C:\kyber_oauth_url.txt, and exits 0. All other cmd invocations are
passed through to the backed-up cmd-real.exe.
Installed into the Kyber prefix's system32 only — does not affect
other games or Proton globally.
2. Install a watcher script (~/.local/bin/kyber-oauth-watcher.sh) that
polls for the URL file and calls xdg-open to open it in the system
browser. Registered as a systemd user service so it starts on login.
Login flow: Kyber → cmd shim → URL file → watcher → xdg-open → browser
→ EA login → redirect to 127.0.0.1:PORT → Kyber loopback catches code.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
On re-run the installer was trying to overwrite locked files in an existing
prefix, causing repeated Access Denied dialogs. Now check for Kyber.exe
first and skip the installer entirely if it's found.
Also kill any leftover Wine/Proton processes before a fresh install to
avoid file-lock conflicts, and move Windows path resolution to after the
installer/skip decision so it always runs regardless of path taken.
Default fallback path updated to match Kyber's actual install location:
C:\Program Files (x86)\KYBER Launcher\
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
If shortcuts.vdf exists but is not writable, attempt chmod 644 before
writing. If that also fails, print a clear instruction and skip the
shortcut step rather than crashing with a Python traceback.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
The Evergreen bootstrapper (linkid=2124703) requires a second download from
inside the Wine process, which fails in Proton. Switch to the standalone
offline installer (linkid=2135547, ~150 MB) which installs without any
Wine-internal network calls.
Also prefer winetricks if installed — it uses a local cache, handles the
WINEPREFIX env correctly, and is the most reliable method. Fall back to
the standalone download if winetricks is not available.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Instead of erroring when the installer is missing, attempt to fetch it
automatically: first scrape kyber.gg for a direct .exe link, then fall
back to a known CDN path. If download fails, print the manual instruction
and exit cleanly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Kyber uses a PKCE loopback OAuth flow (Maxima): it starts a temporary
HTTP server on 127.0.0.1 and calls cmd /c start to open the EA auth URL
in the system browser. On native Linux, Wine's cmd passes http:// URLs
to xdg-open, which works fine inside Steam's single bwrap layer.
Remove the incorrect claim that WebView2 intercepts qrc:// URIs and is
the only path to completing login. Keep the WebView2 runtime install
(harmless, prevents in-app render errors) but clarify it is not the
login mechanism.
Update end-of-script instructions to describe what the user will actually
see: browser opens EA page, redirects to 127.0.0.1:PORT, tab shows OK or
connection-refused (normal), login completes in Kyber. Add troubleshooting
notes for the two real failure modes: xdg-open not reaching the desktop
(headless/no DISPLAY), and login timeout (loopback server expired).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Mount retroarch/shaders and retroarch/overlays from the game drive into
~/.config/retroarch/{shaders,overlays} for both ES-DE and standalone
RetroArch (sibling to the existing cores mount). Anything pulled from the
Online Updater or set in the Quick Menu now persists across sessions and
survives a Wolf/OS reinstall, so a CRT shader or bezel set is configured
once. Dirs are pre-created on the game drive; storage summary and README
updated, including a fallback note to repoint Settings -> Directory if
RetroArch ever writes elsewhere.
gaming-backup: add opt-in snapshots for emulator BIOS (bios/) and the
RetroArch shaders/overlays dirs (cores excluded — re-downloadable). New
BACKUP_BIOS and BACKUP_RA_SHADERS flags wired through the prompts,
backup.conf, summary, and the worker script.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XV8mwKLFwUKt94dAKh3k79
The GoW es-de/retroarch images ship the RetroArch frontend but NO libretro
cores, and wolf.sh never mounted a BIOS directory. As a result every
libretro-based game (SNES/NES/Genesis/N64/PSX/GBA/...) failed to launch on a
fresh install, and BIOS-dependent systems had nowhere to read firmware from.
Changes (services/wolf.sh):
- Pre-download RetroArch cores at install into <game>/retroarch/cores and
bind-mount them into the ES-DE and RetroArch apps at
~/.config/retroarch/cores (where both ES-DE's bundled config and GoW's
rom_launcher.sh look). Defaults to the full libretro buildbot set.
- Add a <game>/bios dir mounted at ~/bioses (matches retroarch.cfg's
system_directory) so PSX/Saturn/Dreamcast/Neo-Geo/etc. find their BIOS.
- New './manage.sh cores [all|common] [force]' to download/refresh cores
later; wired into help text and bash tab-completion. Install reuses it.
- Also expose emulators/ at ~/Applications so ES-DE's app finder detects the
Azahar 3DS AppImage.
- Pre-create the new dirs; extend the install summary and README to document
BIOS placement, the cores command, that controllers auto-configure via SDL
(no manual input setup), and how to exit a game back to ES-DE.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XV8mwKLFwUKt94dAKh3k79
The WolfSteam app container already runs seccomp=unconfined + apparmor=unconfined
+ SYS_ADMIN, so the WebView2 CLONE_NEWUSER failure is gated at the HOST level by
kernel.apparmor_restrict_unprivileged_userns=1 (default-on since Ubuntu 23.10),
which overrides the unconfined container. This script relaxes that sysctl (and
the legacy unprivileged_userns_clone) persistently, then probes userns creation
on the host and inside the running container so login can complete without
leaving Wolf.
Co-Authored-By: Claude <noreply@anthropic.com>
Runs Kyber outside Wolf so its embedded WebView2 can initialize without
nested-bwrap blocking CLONE_NEWUSER. Builds the Wine prefix, ensures the
WebView2 Evergreen runtime is installed (the component that intercepts EA's
qrc:// OAuth redirect), registers Kyber as a non-Steam shortcut forced to
Proton Experimental, and notes the Steam Remote Play path for headless boxes.
No fake cmd.exe / Firefox scaffolding needed here.
Co-Authored-By: Claude <noreply@anthropic.com>
Documents the complete Kyber/EA OAuth investigation: root cause is EA's
qrc:// redirect URI which only Kyber's embedded WebView2 can intercept;
WebView2 cannot initialize inside Wolf due to nested bwrap blocking
CLONE_NEWUSER. Script automates all discovered workarounds (fake cmd.exe,
Firefox profile with network.process.enabled: false) and clearly documents
why the real fix is running Kyber outside the Docker container.
Co-Authored-By: Claude <noreply@anthropic.com>
With multiple WolfSteam containers, the previous head -1 pick was
arbitrary. Now matches each Steam state directory against running
container IDs so ge-proton, fix-ea-game, and other commands all
target the active session.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
With multiple WolfSteam containers running, _steam_home() picks the first
Steam directory found which may not contain the target game's prefix.
_apply_ea_fix now scans all Wolf Steam homes and selects the one that has
the AppID's Proton prefix, falling back to the original behavior if none
match.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Steam writes localconfig.vdf via atomic rename (write temp + rename over
original). chmod 444 on the file is bypassed by the rename. chmod 555 on
the directory blocks rename-into, preserving the launch option across
Steam restarts.
Also unlocks the directory at the start of step 6 so re-running the
script can update the launch option.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
On native Linux Steam, EA App self-installs via EAappInstaller.exe on first
SWBF2 launch. Link2EA.exe appears in the Wine prefix without any MSI extraction.
Script now checks for Link2EA.exe first and skips steps 2-3 (MSI locate +
extract) if EA Desktop is already installed, proceeding directly to writing
registry fix files and the launch wrapper.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
The MSI is bundled in steamapps/common with the game files. The script
was only checking drive_c (where Steam copies it after a launch), causing
a false "not found" error before the user had launched the game.
Now searches steamapps/common and all of steamapps as fallbacks, reports
which path it found the MSI at. Also remove stale references to the
"Origin is not installed" error which no longer appears.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Both scripts now detect and install msitools if missing, supporting
apt-get (Debian/Ubuntu), dnf (Fedora), and pacman (Arch). Falls back
to a clear error if the package manager is not recognised.
Remove msitools from manual prerequisites in the doc.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Both scripts now download and install GE-Proton10-34 if not already present:
- setup-swbf2-linux.sh: installs to ~/.steam/root/compatibilitytools.d/
- setup-swbf2-wolf.sh: installs into the WolfSteam container via docker cp + tar
Both pause after install and prompt the user to set GE-Proton in Steam
(Compatibility → Force) before continuing, since that step requires the GUI.
Update docs to reflect GE-Proton is now handled automatically.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
- Remove editorial content (EA server speculation, 2005 alternative, opinions)
- Add native Linux Steam section covering all six setup steps
- Add separate key paths tables for Wolf and native
- Add native Linux troubleshooting commands alongside Wolf equivalents
- Rename title to reflect both platforms
- Fresh install table rewritten as factual survival matrix
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Long-form document tracing the actual discovery process: the wrong turns,
the aha moments, and the reasoning behind each step. Covers:
- Why EA is involved in a Steam game at all (link2ea:// architecture)
- JunoConfigureRegistry failure and why DISABLEROLLBACK was a dead end
- msiextract bypass as the actual solution
- Why direct system.reg edits don't survive reboots (wineserver ownership)
- The launch wrapper approach for registry persistence
- RPC_S_SERVER_UNAVAILABLE and EALocalHostSvc
- Wolf-specific bwrap capabilities mismatch
- localconfig.vdf atomic rename problem and chmod 555 fix
Written for sharing — no identifying information, explains the "why"
for each fix so readers can adapt if EA changes their MSI structure.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Two self-contained scripts for getting SWBF2 (2017, AppID 1237950) working
with GE-Proton on Linux — no dependency on the ubuntu-post-install framework.
setup-swbf2-wolf.sh — for Wolf/Games-on-Whales + Moonlight streaming
setup-swbf2-linux.sh — for native Linux Steam (no Docker)
Both implement the same core fix:
- msiextract bypass for JunoConfigureRegistry Wine incompatibility
- EA Desktop files copied into Wine prefix from extracted MSI
- link2ea_fix.reg + ea_services.reg written to drive_c
- Launch wrapper that imports .reg files via Proton on every launch
(direct system.reg edits are overwritten by wineserver on shutdown)
Wolf version also handles Docker cp, container paths, STEAM_UNIX_SOCKET,
and localconfig.vdf directory locking inside the Wolf state folder.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Automates the complete working solution for SWBF2 (AppID 1237950) on Wolf:
1. Extracts EA Desktop from ea_app.msi using msiextract on the host,
bypassing JunoConfigureRegistry Wine incompatibility that causes MSI rollback
2. Copies EA Desktop files (Link2EA.exe, EALocalHostSvc.exe, etc.) into the
Wine prefix at the versioned path with symlink
3. Writes link2ea_fix.reg and ea_services.reg into drive_c so they persist
across Wine prefix operations
4. Installs a launch wrapper (/home/retro/ea_install.sh) that runs regedit
via Steam's sniper+GE-Proton launch chain — required because direct edits
to system.reg are overwritten when wineserver flushes on shutdown
5. Sets LaunchOptions for AppID 1237950 in localconfig.vdf and locks the
config directory (chmod 555) to prevent Steam overwriting it via
atomic rename on shutdown
Run after: Steam open in Moonlight, GE-Proton set, SWBF2 launched once
(triggers Wine prefix + ea_app.msi creation), EA account linked at ea.com.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Documents the EA account requirement, first-time login flow, credential
caching for machine transfers, community server options if EA shuts down,
and the SWBF2 2005 alternative for a launcher-free experience.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Revised approach: skip link2ea:// and EA App account requirement entirely.
Launch starwarsbattlefrontii.exe directly via the wrapper, spoof Origin
registry entries so the game's built-in launcher check passes. Documents
what doesn't work and why (msiexec/bwrap/EA account issues).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Documents the EA App installation workaround for Star Wars Battlefront II
(2017) running under Wolf/WolfSteam with GE-Proton10-34. Covers the bwrap
capability problem, Steam launch wrapper trick, msiextract bypass for the
JunoConfigureRegistry Wine incompatibility, and correct EA Desktop file layout.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
wine cmd.exe /c "timeout /t 900 > nul" failed silently (docker exec -d
hides errors), leaving no anchor process and letting wineserver die as
soon as EAappInstaller.exe exited.
Replace with wineserver -f (foreground mode) which keeps the server
alive unconditionally until explicitly killed with wineserver -k.
This is the correct primitive for holding a Wine session open.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
EAappInstaller.exe is a WiX/Burn bootstrapper that spawns the real
installer as a background Wine process and exits immediately. When the
foreground wine process exited, wineserver shut down and killed the
background installer before it could write any files.
Fix: launch a cmd.exe/timeout anchor before running the installer so
the shared wineserver stays up while background installer processes run.
Kill the anchor (wineserver -k) after EADesktop.exe is detected or the
15-minute timeout expires.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Two bugs in install-ea-app:
1. Installer search included compatdata (the Wine prefix), so it found the
temp-extracted copy at drive_c/windows/temp/{UUID}/.cr/EAappInstaller.exe
instead of the bundled copy at steamapps/common/…/__Installer/…/
Fixed: search only under steamapps/common.
2. Our fake InstallSuccessful=true (written by fix-ea-game to bypass the
install-script loop) caused EAappInstaller to detect EA App as "already
installed" and exit in ~15 seconds without doing anything.
Fixed: delete the InstallSuccessful values via wine reg delete inside the
container BEFORE launching the installer. The real EA App installer writes
them back correctly after a successful install.
Co-Authored-By: Claude <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
install-ea-app: runs EAappInstaller.exe inside the WolfSteam container
using GE-Proton's wine binary. The container has a virtual display so the
EA App installer GUI appears in Moonlight; the user clicks through it and
logs in to their EA account. After EA App installs, fix-ea-game registers
the link2ea:// handler and clears the install-script loop.
diagnose-ea: prints the current link2ea:// registry state from both
system.reg and user.reg, locates EADesktop.exe/Link2EA.exe, shows
StateFlags, and tails the relevant Proton log lines — the first thing to
run when the game returns straight to the Play screen.
_apply_ea_fix: now checks user.reg as well as system.reg for an existing
link2ea:// registration (GE-Proton writes to user.reg during prefix setup);
also searches for Link2EA.exe as a fallback handler when EADesktop.exe is
not yet installed; reports a clear next step when neither is found.
Co-Authored-By: Claude <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
GE-Proton is required for EA games but was a manual step users had to
remember after first Steam launch. This change eliminates that gap:
- install_wolf() downloads the GE-Proton tarball to
~/docker/wolf/ge-proton-cache/ during initial setup (~500 MB, runs
before the "Start Wolf now?" prompt so it's ready immediately)
- _cache_ge_proton() helper in manage.sh does the same on demand
- ge-proton subcommand checks the cache dir first and extracts from
there instead of re-downloading (~instant vs ~500 MB download)
- start subcommand auto-extracts the cached tarball into Steam's
compatibilitytools.d if Steam has already launched at least once,
so GE-Proton appears in the Compatibility dropdown without any
manual ./manage.sh ge-proton step
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
SWBF2 (AppID 1237950) launches, calls ShellExecute("link2ea://..."),
and exits. Without a link2ea:// URL protocol handler registered in the
Wine registry, ShellExecute returns error 31 and the game never plays.
_apply_ea_fix now also writes the link2ea:// Classes registry keys
pointing to EADesktop.exe when it is found in the Proton prefix. If
EADesktop.exe is not yet present (EAappInstaller runs async via
RunType=1 in installScript.vdf), the function warns and skips the
handler registration with instructions to run wait-ea-app.
New 'wait-ea-app [appid]' subcommand polls every 15 seconds (up to 20
minutes) for EADesktop.exe to appear, then calls _apply_ea_fix so the
full fix including the link2ea:// handler is applied automatically once
EA App finishes installing in the background.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
wineserver holds the Wine registry in memory and flushes it back to
system.reg on shutdown, overwriting any edits made while the container
is running. _apply_ea_fix now stops the WolfSteam container first,
patches system.reg + appmanifest StateFlags on disk, then restarts Wolf
so the keys persist when wineserver next starts.
Also sets StateFlags to 6 (fully installed) so Steam skips the
install-script phase entirely and launches the game directly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D8ckUJQtj1pH8jtAddBDZs
Claude