Fix memory leaks in long-running Docker container streams

[Copilot]

Problem

The stream-webpage-container was experiencing memory leaks when running continuously for several days. Memory usage would gradually increase over time, eventually causing performance degradation or container crashes in long-running deployments.

Root Cause Analysis

Investigation revealed several sources of memory leaks:

1. Incomplete Process Cleanup: FFmpeg processes were killed with Process.Kill() but child processes weren't properly terminated, leading to zombie process accumulation
2. Chrome Browser Resource Leaks: Chrome contexts were cancelled but browser processes weren't explicitly cleaned up, causing memory to accumulate over restarts
3. Missing Process Group Management: No process group handling meant child processes could survive parent termination
4. Resource Accumulation: The infinite restart loop gradually accumulated resources without explicit cleanup between iterations

Solution

This PR implements comprehensive memory leak fixes:

Enhanced Process Management

• Graceful Termination: Processes now receive SIGTERM with a 3-second timeout before SIGKILL
• Process Groups: FFmpeg processes use Setpgid: true to enable complete process tree cleanup
• Child Process Cleanup: Added explicit process group termination to kill all child processes

Chrome Browser Resource Management

• Explicit Process Cleanup: New cleanupChromeProcesses() function kills remaining Chrome processes
• Memory Management Flags: Added Chrome flags like max_old_space_size=512 and memory-pressure-off
• Enhanced Context Cleanup: Improved defer blocks with proper timing for resource cleanup

Memory Monitoring and Management

• Garbage Collection: Added strategic runtime.GC() calls before stream starts and after cleanup
• Memory Statistics: New logMemoryStats() function tracks heap usage, GC cycles, and system memory
• Periodic Cleanup: Every 10 restarts, performs full cleanup with memory monitoring
• Resource Limits: Added GOMEMLIMIT="512MiB" to constrain Go memory usage

Startup Optimizations

# Added to start.sh
export GOMEMLIMIT="512MiB"
export GOGC=100
pkill -f chrome 2>/dev/null || true  # Cleanup existing processes

Testing

Added comprehensive tests:

• TestCleanupChromeProcesses() - Validates Chrome process cleanup
• TestStopStreamWithProcessGroup() - Tests process group termination
• Manual testing confirms memory statistics logging and cleanup behavior

Expected Impact

These changes should eliminate memory leaks for long-running containers by ensuring:

• Zero zombie processes through complete process tree cleanup
• Bounded memory growth via regular garbage collection and memory limits
• Proper resource cleanup of all contexts, processes, and handles
• Operational visibility through memory usage monitoring

The container should now be suitable for multi-day continuous operation without memory accumulation.

Breaking Changes

None - all changes are backward compatible and don't affect the API or configuration.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• id.twitch.tv
  • Triggering command: /tmp/go-build889050422/b343/twitch.test -test.testlogfile=/tmp/go-build889050422/b343/testlog.txt -test.paniconexit0 -test.timeout=10m0s -test.v=true (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.