ssmithx
e559ccb767
Both had regressed back to the pre-2026-06-28-incident state: GPU_FLAGS used --enable-gpu-rasterization on any node with a GPU, and CPUQuota was 75%. On archy-x250-exp (Intel HD 5500 under X11) this spins a dedicated GPU process at 55-92% CPU (falls back to software compositing anyway), and the 75% cgroup quota throttled the kiosk unit ~81% of the time (nr_throttled 4856/6005) — the exact CPU-starvation signature documented as the choppy- audio root cause. Restores the validated fix: --in-process-gpu, --num-raster-threads=1, GpuRasterization disabled via --disable-features, CPUQuota=200%. Verified on archy-x250-exp: no separate gpu-process, throttling dropped to ~3% (nr_throttled 14/503). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
38 lines
1.7 KiB
Desktop File
38 lines
1.7 KiB
Desktop File
[Unit]
|
|
Description=Archipelago Kiosk (X11 + Chromium)
|
|
After=archipelago.service systemd-user-sessions.service network-online.target
|
|
Wants=archipelago.service network-online.target
|
|
ConditionPathExists=/usr/local/bin/archipelago-kiosk-launcher
|
|
Conflicts=getty@tty1.service
|
|
|
|
[Service]
|
|
Type=simple
|
|
# Wait up to 5 min for archipelago to serve /health. On slow hardware
|
|
# first-boot is dominated by the FileBrowser pull (unbundled ISO),
|
|
# initial archipelago state sync, and frontend settle — .198 took
|
|
# longer than 120s and chromium launched against an empty backend,
|
|
# producing a white window that only recovered on reboot. 300s gives
|
|
# slow-but-functional hardware enough headroom; TimeoutStartSec is
|
|
# bumped in lockstep so systemd doesn't kill us mid-wait.
|
|
ExecStartPre=/bin/bash -c 'for i in $(seq 1 150); do curl -sf http://localhost/health >/dev/null 2>&1 && break; sleep 2; done'
|
|
ExecStart=/usr/local/bin/archipelago-kiosk-launcher
|
|
TimeoutStartSec=360
|
|
Restart=always
|
|
RestartSec=5
|
|
|
|
# Resource guardrail (#36). On GPU-less / headless hardware chromium could spin
|
|
# software compositing at ~92% of a core, saturating the node and starving the
|
|
# backend (it caused the .198 receive timeout + deploy storms). Cap CPU + memory
|
|
# so a runaway kiosk can never take the whole machine down; Delegate so the cap
|
|
# also binds the chromium/Xorg children in this unit's cgroup.
|
|
# CPUQuota=75% (0.75 cores) was too tight even for normal playback — the kiosk
|
|
# was throttled ~40% of the time, which is what caused choppy HDMI audio on
|
|
# archy-x250-exp (2026-06-28 incident). 200% (2 cores) gives enough headroom.
|
|
Delegate=yes
|
|
CPUQuota=200%
|
|
MemoryMax=1500M
|
|
MemoryHigh=1200M
|
|
|
|
[Install]
|
|
WantedBy=multi-user.target
|