12b2f9d43b
Replaces the previous hand-rolled MigrateWithLock + GORM AutoMigrate path,
which had two compounding problems:
- AutoMigrate ran on every pod startup (~5 min over the transatlantic
link) even when no schema changes had landed
- pg_advisory_lock is session-scoped, which silently fails through
Neon's pgbouncer transaction-mode pooler — turns out this is a
known and documented limitation that bites golang-migrate too
Goose was chosen over golang-migrate (the other heavyweight) because:
- Goose wraps each migration file in a transaction by default, so a
failure rolls back cleanly instead of leaving a "dirty" version
state requiring manual force-reset (golang-migrate's known
weakness, per its own issue tracker — see #1001 + Atlas's writeup)
- Goose's locking is opt-in. We don't opt in: migrations run as a
single Kubernetes Job, which IS the singleton process. No advisory
lock needed at all.
Layout:
- migrations/000001_init.sql — schema-only pg_dump of the live Neon
DB at adoption, stripped of psql-only directives that block goose's
bookkeeping insert. Pre-goose hand-numbered migrations 002-022 had
their effects folded into this baseline; deleted from the live tree
but preserved in git history at 58e6997.
- Dockerfile installs `goose v3.22.1` at build time and copies the
binary into the api image. The migrate Job reuses the api image with
command=goose, so no separate image to build/push/version.
- deploy-k3s/manifests/migrate/job.yaml: a one-shot Job that strips
the -pooler segment from DB_HOST (advisory lock won't survive
pgbouncer transaction-mode), runs `goose up`, exits.
- deploy-k3s/scripts/03-deploy.sh: deletes any prior Job, applies the
fresh one, `kubectl wait --for=condition=complete --timeout=10m`,
then proceeds with api/worker rollout. Job failure aborts the deploy
before any new app pod sees a stale schema.
- internal/database/database.go::RequireSchemaApplied checks
goose_db_version on startup. api/worker refuse to boot if the
table is missing or its latest row has is_applied=false — the
fail-fast for "operator forgot to run migrate."
- Makefile: migrate-up / migrate-down / migrate-status / migrate-new
for local workflow.
Production DB was bootstrapped manually:
$ goose -dir migrations postgres "$DSN" version # creates table
$ psql ... -c "INSERT INTO goose_db_version (version_id, is_applied, tstamp) VALUES (1, true, NOW());"
Smoke test against fresh Postgres locally: 50 user tables created in
284ms via `goose up`, version_id=1 + is_applied=t recorded.
Verified the local goose CLI talks to prod successfully:
$ goose ... status
Applied At Migration
=======================================
Mon Apr 27 03:43:55 2026 -- 000001_init.sql
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
257 lines
8.1 KiB
Go
257 lines
8.1 KiB
Go
package main
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"net/http"
|
|
"os"
|
|
"os/signal"
|
|
"syscall"
|
|
"time"
|
|
|
|
"github.com/rs/zerolog/log"
|
|
"gorm.io/gorm"
|
|
|
|
"github.com/treytartt/honeydue-api/internal/config"
|
|
"github.com/treytartt/honeydue-api/internal/database"
|
|
"github.com/treytartt/honeydue-api/internal/i18n"
|
|
"github.com/treytartt/honeydue-api/internal/monitoring"
|
|
"github.com/treytartt/honeydue-api/internal/push"
|
|
"github.com/treytartt/honeydue-api/internal/router"
|
|
"github.com/treytartt/honeydue-api/internal/services"
|
|
"github.com/treytartt/honeydue-api/internal/tracing"
|
|
"github.com/treytartt/honeydue-api/pkg/utils"
|
|
)
|
|
|
|
func main() {
|
|
// Load configuration
|
|
cfg, err := config.Load()
|
|
if err != nil {
|
|
fmt.Printf("Failed to load configuration: %v\n", err)
|
|
os.Exit(1)
|
|
}
|
|
|
|
// Initialize basic logger first (will be enhanced after Redis connects)
|
|
utils.InitLogger(cfg.Server.Debug)
|
|
|
|
// Initialize i18n
|
|
if err := i18n.Init(); err != nil {
|
|
log.Warn().Err(err).Msg("Failed to initialize i18n - using English only")
|
|
} else {
|
|
log.Info().Strs("languages", i18n.SupportedLanguages).Msg("i18n initialized")
|
|
}
|
|
|
|
log.Info().
|
|
Bool("debug", cfg.Server.Debug).
|
|
Int("port", cfg.Server.Port).
|
|
Str("db_host", cfg.Database.Host).
|
|
Int("db_port", cfg.Database.Port).
|
|
Str("db_name", cfg.Database.Database).
|
|
Str("db_user", cfg.Database.User).
|
|
Str("redis_url", config.MaskURLCredentials(cfg.Redis.URL)).
|
|
Msg("Starting HoneyDue API server")
|
|
|
|
// Initialize OpenTelemetry tracing — exports to obs.88oakapps.com
|
|
// (Jaeger via OTLP/HTTP) when OBS_TRACES_URL is set; otherwise installs
|
|
// a no-op tracer so call sites can use otel.Tracer() unconditionally.
|
|
tracingShutdown, err := tracing.Init(context.Background(), tracing.Config{
|
|
ServiceName: "honeydue-api",
|
|
Environment: deploymentEnvironment(cfg.Server.Debug),
|
|
EndpointURL: os.Getenv("OBS_TRACES_URL"),
|
|
BearerToken: os.Getenv("OBS_INGEST_TOKEN"),
|
|
SampleRatio: tracing.SampleRatioFromEnv(),
|
|
})
|
|
if err != nil {
|
|
log.Error().Err(err).Msg("tracing init failed — continuing without traces")
|
|
}
|
|
defer func() {
|
|
shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
|
|
defer cancel()
|
|
if err := tracingShutdown(shutdownCtx); err != nil {
|
|
log.Warn().Err(err).Msg("tracing shutdown error")
|
|
}
|
|
}()
|
|
|
|
// Connect to database (retry with backoff)
|
|
var db *gorm.DB
|
|
var dbErr error
|
|
for i := 0; i < 3; i++ {
|
|
db, dbErr = database.Connect(&cfg.Database, cfg.Server.Debug)
|
|
if dbErr == nil {
|
|
break
|
|
}
|
|
log.Warn().Err(dbErr).Int("attempt", i+1).Msg("Failed to connect to database, retrying...")
|
|
time.Sleep(time.Duration(i+1) * time.Second)
|
|
}
|
|
if dbErr != nil {
|
|
log.Error().Err(dbErr).Msg("Failed to connect to database - API will start but database operations will fail")
|
|
} else {
|
|
defer database.Close()
|
|
// Migrations are managed out-of-band by golang-migrate (see
|
|
// cmd/migrate and deploy-k3s/manifests/migrate/job.yaml) so the api
|
|
// no longer runs AutoMigrate at startup. Instead we verify the
|
|
// schema is at the expected version and refuse to start if not —
|
|
// this catches the "operator forgot to run migrate" footgun loudly,
|
|
// at boot, instead of with mysterious runtime errors.
|
|
if err := database.RequireSchemaApplied(); err != nil {
|
|
log.Fatal().Err(err).Msg("Schema precondition failed — run `kubectl -n honeydue create job --from=cronjob/honeydue-migrate` (or `make migrate-up` locally) and retry")
|
|
}
|
|
}
|
|
|
|
// Connect to Redis (optional - don't fail if unavailable)
|
|
var cache *services.CacheService
|
|
cache, err = services.NewCacheService(&cfg.Redis)
|
|
if err != nil {
|
|
log.Warn().Err(err).Msg("Failed to connect to Redis - caching disabled")
|
|
cache = nil
|
|
} else {
|
|
defer cache.Close()
|
|
if database.SeedInitialDataApplied {
|
|
if err := cache.InvalidateSeededData(context.Background()); err != nil {
|
|
log.Warn().Err(err).Msg("Failed to invalidate seeded data cache after initial seed")
|
|
} else {
|
|
log.Info().Msg("Invalidated seeded_data cache after initial seed migration")
|
|
}
|
|
}
|
|
}
|
|
|
|
// Initialize monitoring service (if Redis is available)
|
|
var monitoringService *monitoring.Service
|
|
if cache != nil {
|
|
monitoringService = monitoring.NewService(monitoring.Config{
|
|
Process: "api",
|
|
RedisClient: cache.Client(),
|
|
DB: db, // Pass database for enable_monitoring setting sync
|
|
})
|
|
|
|
// Reinitialize logger with monitoring writer
|
|
utils.InitLoggerWithWriter(cfg.Server.Debug, monitoringService.LogWriter())
|
|
|
|
// Start stats collection
|
|
monitoringService.Start()
|
|
defer monitoringService.Stop()
|
|
|
|
log.Info().
|
|
Bool("log_capture_enabled", monitoringService.IsEnabled()).
|
|
Msg("Monitoring service initialized")
|
|
}
|
|
|
|
// Initialize email service
|
|
var emailService *services.EmailService
|
|
log.Info().
|
|
Str("email_host", cfg.Email.Host).
|
|
Str("email_user", cfg.Email.User).
|
|
Str("email_from", cfg.Email.From).
|
|
Int("email_port", cfg.Email.Port).
|
|
Msg("Email config loaded")
|
|
if cfg.Email.Host != "" && cfg.Email.User != "" {
|
|
emailService = services.NewEmailService(&cfg.Email, cfg.Features.EmailEnabled)
|
|
log.Info().
|
|
Str("host", cfg.Email.Host).
|
|
Msg("Email service initialized")
|
|
} else {
|
|
log.Warn().
|
|
Str("host", cfg.Email.Host).
|
|
Str("user", cfg.Email.User).
|
|
Msg("Email service not configured - emails will not be sent")
|
|
}
|
|
|
|
// Initialize storage service for file uploads (local filesystem or S3-compatible)
|
|
var storageService *services.StorageService
|
|
if cfg.Storage.UploadDir != "" || cfg.Storage.IsS3() {
|
|
storageService, err = services.NewStorageService(&cfg.Storage)
|
|
if err != nil {
|
|
log.Warn().Err(err).Msg("Failed to initialize storage service - uploads disabled")
|
|
} else {
|
|
// Initialize file encryption at rest if configured
|
|
if cfg.Storage.EncryptionKey != "" {
|
|
encSvc, encErr := services.NewEncryptionService(cfg.Storage.EncryptionKey)
|
|
if encErr != nil {
|
|
log.Error().Err(encErr).Msg("Failed to initialize encryption service - files will NOT be encrypted")
|
|
} else {
|
|
storageService.SetEncryptionService(encSvc)
|
|
log.Info().Msg("File encryption at rest enabled")
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// Initialize PDF service for report generation
|
|
pdfService := services.NewPDFService()
|
|
log.Info().Msg("PDF service initialized")
|
|
|
|
// Initialize push notification client (APNs + FCM)
|
|
var pushClient *push.Client
|
|
pushClient, err = push.NewClient(&cfg.Push, cfg.Features.PushEnabled)
|
|
if err != nil {
|
|
log.Warn().Err(err).Msg("Failed to initialize push client - push notifications disabled")
|
|
} else {
|
|
log.Info().
|
|
Bool("ios_enabled", pushClient.IsIOSEnabled()).
|
|
Bool("android_enabled", pushClient.IsAndroidEnabled()).
|
|
Msg("Push notification client initialized")
|
|
}
|
|
|
|
// Setup router with dependencies (includes admin panel at /admin)
|
|
deps := &router.Dependencies{
|
|
DB: db,
|
|
Cache: cache,
|
|
Config: cfg,
|
|
EmailService: emailService,
|
|
PDFService: pdfService,
|
|
PushClient: pushClient,
|
|
StorageService: storageService,
|
|
MonitoringService: monitoringService,
|
|
}
|
|
e := router.SetupRouter(deps)
|
|
|
|
// Create HTTP server
|
|
srv := &http.Server{
|
|
Addr: fmt.Sprintf(":%d", cfg.Server.Port),
|
|
Handler: e,
|
|
ReadTimeout: 30 * time.Second,
|
|
WriteTimeout: 30 * time.Second,
|
|
IdleTimeout: 60 * time.Second,
|
|
}
|
|
|
|
// Start server in goroutine
|
|
go func() {
|
|
log.Info().
|
|
Str("addr", srv.Addr).
|
|
Msg("HTTP server listening")
|
|
|
|
if err := srv.ListenAndServe(); err != nil && err != http.ErrServerClosed {
|
|
log.Fatal().Err(err).Msg("Failed to start HTTP server")
|
|
}
|
|
}()
|
|
|
|
// Wait for interrupt signal for graceful shutdown
|
|
quit := make(chan os.Signal, 1)
|
|
signal.Notify(quit, syscall.SIGINT, syscall.SIGTERM)
|
|
<-quit
|
|
|
|
log.Info().Msg("Shutting down server...")
|
|
|
|
// Graceful shutdown with timeout
|
|
ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
|
|
defer cancel()
|
|
|
|
if err := srv.Shutdown(ctx); err != nil {
|
|
log.Fatal().Err(err).Msg("Server forced to shutdown")
|
|
}
|
|
|
|
log.Info().Msg("Server exited")
|
|
}
|
|
|
|
// deploymentEnvironment turns the boolean Debug flag into the conventional
|
|
// environment label spans get tagged with.
|
|
func deploymentEnvironment(debug bool) string {
|
|
if env := os.Getenv("DEPLOYMENT_ENVIRONMENT"); env != "" {
|
|
return env
|
|
}
|
|
if debug {
|
|
return "dev"
|
|
}
|
|
return "prod"
|
|
}
|