BCards/src/BCards.Web/HealthChecks/SystemResourcesHealthCheck.cs

using Microsoft.Extensions.Diagnostics.HealthChecks;
using System.Diagnostics;

namespace BCards.Web.HealthChecks;

public class SystemResourcesHealthCheck : IHealthCheck
{
    private readonly ILogger<SystemResourcesHealthCheck> _logger;
    private static readonly DateTime _startTime = DateTime.UtcNow;

    public SystemResourcesHealthCheck(ILogger<SystemResourcesHealthCheck> logger)
    {
        _logger = logger;
    }

    public Task<HealthCheckResult> CheckHealthAsync(
        HealthCheckContext context,
        CancellationToken cancellationToken = default)
    {
        var stopwatch = Stopwatch.StartNew();

        try
        {
            // Informações de memória
            var totalMemory = GC.GetTotalMemory(false);
            var workingSet = Environment.WorkingSet;
            
            // Informações do processo atual
            using var currentProcess = Process.GetCurrentProcess();
            var cpuUsage = GetCpuUsage(currentProcess);
            
            // Uptime
            var uptime = DateTime.UtcNow - _startTime;
            var uptimeString = FormatUptime(uptime);

            // Thread count
            var threadCount = currentProcess.Threads.Count;

            stopwatch.Stop();
            var duration = stopwatch.ElapsedMilliseconds;

            var data = new Dictionary<string, object>
            {
                { "status", "healthy" },
                { "duration", $"{duration}ms" },
                { "memory", new Dictionary<string, object>
                    {
                        { "total_managed_mb", Math.Round(totalMemory / 1024.0 / 1024.0, 2) },
                        { "working_set_mb", Math.Round(workingSet / 1024.0 / 1024.0, 2) },
                        { "gc_generation_0", GC.CollectionCount(0) },
                        { "gc_generation_1", GC.CollectionCount(1) },
                        { "gc_generation_2", GC.CollectionCount(2) }
                    }
                },
                { "process", new Dictionary<string, object>
                    {
                        { "id", currentProcess.Id },
                        { "threads", threadCount },
                        { "handles", currentProcess.HandleCount },
                        { "uptime", uptimeString },
                        { "uptime_seconds", (int)uptime.TotalSeconds }
                    }
                },
                { "system", new Dictionary<string, object>
                    {
                        { "processor_count", Environment.ProcessorCount },
                        { "os_version", Environment.OSVersion.ToString() },
                        { "machine_name", Environment.MachineName },
                        { "user_name", Environment.UserName }
                    }
                }
            };

            _logger.LogInformation("System resources health check completed in {Duration}ms - Memory: {Memory}MB, Threads: {Threads}",
                                 duration, Math.Round(totalMemory / 1024.0 / 1024.0, 1), threadCount);

            // Definir thresholds para status
            var memoryMb = totalMemory / 1024.0 / 1024.0;
            
            if (memoryMb > 1000) // > 1GB
            {
                data["status"] = "degraded";
                return Task.FromResult(HealthCheckResult.Degraded($"High memory usage: {memoryMb:F1}MB", data: data));
            }

            if (threadCount > 500)
            {
                data["status"] = "degraded"; 
                return Task.FromResult(HealthCheckResult.Degraded($"High thread count: {threadCount}", data: data));
            }

            return Task.FromResult(HealthCheckResult.Healthy($"System resources normal (Memory: {memoryMb:F1}MB, Threads: {threadCount})", data: data));
        }
        catch (Exception ex)
        {
            stopwatch.Stop();
            var duration = stopwatch.ElapsedMilliseconds;

            _logger.LogError(ex, "System resources health check failed after {Duration}ms", duration);

            var data = new Dictionary<string, object>
            {
                { "status", "unhealthy" },
                { "duration", $"{duration}ms" },
                { "error", ex.Message }
            };

            return Task.FromResult(HealthCheckResult.Unhealthy($"System resources check failed: {ex.Message}", ex, data));
        }
    }

    private static double GetCpuUsage(Process process)
    {
        try
        {
            return process.TotalProcessorTime.TotalMilliseconds;
        }
        catch
        {
            return 0;
        }
    }

    private static string FormatUptime(TimeSpan uptime)
    {
        if (uptime.TotalDays >= 1)
            return $"{(int)uptime.TotalDays}d {uptime.Hours}h {uptime.Minutes}m";
        if (uptime.TotalHours >= 1)
            return $"{uptime.Hours}h {uptime.Minutes}m";
        if (uptime.TotalMinutes >= 1)
            return $"{uptime.Minutes}m {uptime.Seconds}s";
        return $"{uptime.Seconds}s";
    }
}