Add health checks and runtime monitoring #3

New issue

Closed

opened 2026-05-14 04:47:39 +00:00 by Mark · 0 comments

Mark commented

2026-05-14 04:47:39 +00:00

Owner

The container currently has a basic Docker healthcheck (curl -f http://localhost:6080/vnc.html), but lacks:

Process health: Verify Xvfb, openbox, x11vnc, and noVNC are all running via supervisorctl
Display health: Verify the X11 display is actually accepting connections (xdpyinfo)
Automation health: Verify pyautogui can take a screenshot successfully
Structured metrics: Expose widget counts, screenshot latency, action log rate

Files to modify:

entrypoint.sh — add readiness probe
testenv/automation.py — add healthcheck() method
New testenv/health.py — health check aggregator

Why: In CI and long-running test sessions, silent failures (e.g. Xvfb crash) are hard to diagnose without explicit health signals.

The container currently has a basic Docker healthcheck (`curl -f http://localhost:6080/vnc.html`), but lacks: - **Process health:** Verify Xvfb, openbox, x11vnc, and noVNC are all running via supervisorctl - **Display health:** Verify the X11 display is actually accepting connections (`xdpyinfo`) - **Automation health:** Verify pyautogui can take a screenshot successfully - **Structured metrics:** Expose widget counts, screenshot latency, action log rate Files to modify: - `entrypoint.sh` — add readiness probe - `testenv/automation.py` — add `healthcheck()` method - New `testenv/health.py` — health check aggregator **Why:** In CI and long-running test sessions, silent failures (e.g. Xvfb crash) are hard to diagnose without explicit health signals.