fix(scheduler): Cert-Renewal-Alerts pro Domain statt aggregierter Zählung (v1.1.143)

- certrenewer.Result.FailedDomains []string — jeder fehlgeschlagene Domain-Name
  wird erfasst (Issue/Parse/Write-Fehler)
- runRenewer: pro Domain eigener Alert + eigener Dedupe-Key statt einem
  shared "cert.renew_failed"-Key; Alert-Message nennt jetzt den Domain-Namen
  und gibt Hinweis auf ACME/DNS-Debugging

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

VERSION

@@ -1 +1 @@
-1.1.142
+1.1.143

cmd/edgeguard-api/main.go

@@ -60,7 +60,7 @@ import (
 	usersvc "git.netcell-it.de/projekte/edgeguard-native/internal/services/users"
 )
 
-var version = "1.1.142"
+var version = "1.1.143"
 
 func main() {
 	addr := os.Getenv("EDGEGUARD_API_ADDR")

cmd/edgeguard-ctl/main.go

@@ -11,7 +11,7 @@ import (
 	"git.netcell-it.de/projekte/edgeguard-native/internal/services/setup"
 )
 
-var version = "1.1.142"
+var version = "1.1.143"
 
 const usage = `edgeguard-ctl — EdgeGuard CLI
 

cmd/edgeguard-scheduler/main.go

@@ -41,7 +41,7 @@ import (
 	"git.netcell-it.de/projekte/edgeguard-native/internal/services/tlscerts"
 )
 
-var version = "1.1.142"
+var version = "1.1.143"
 
 const (
 	// renewTickInterval — how often we re-evaluate expiring certs.
@@ -1062,11 +1062,18 @@ func runRenewer(ctx context.Context, r *certrenewer.Service, a *alerts.Service,
 	slog.Info("scheduler: renewer pass complete",
 		"checked", res.Checked, "renewed", res.Renewed,
 		"failed", res.Failed, "skipped", res.Skipped)
-	if a != nil && res.Failed > 0 && d != nil && d.shouldFire("cert.renew_failed") {
+	if a != nil && d != nil {
-		_, _ = a.Fire(ctx, "cert.renew_failed", alerts.SeverityError,
+		for _, domain := range res.FailedDomains {
-			"Cert-Renewal teilweise fehlgeschlagen",
+			key := "cert.renew_failed:" + domain
-			fmt.Sprintf("Renewer-Cycle: %d checked, %d renewed, %d failed, %d skipped",
+			if !d.shouldFire(key) {
-				res.Checked, res.Renewed, res.Failed, res.Skipped))
+				continue
+			}
+			_, _ = a.Fire(ctx, "cert.renew_failed", alerts.SeverityError,
+				"Cert-Renewal fehlgeschlagen: "+domain,
+				"Let's Encrypt Erneuerung für "+domain+" ist fehlgeschlagen. "+
+					"Prüfe ACME-Konfiguration und DNS-Erreichbarkeit. "+
+					"Nächster Versuch beim nächsten Renewer-Tick (alle 6h).")
+		}
 	}
 }
 

internal/services/certrenewer/certrenewer.go

@@ -62,12 +62,14 @@ func (s *Service) log() *slog.Logger {
 }
 
 // Result counts what one Run pass did so the scheduler can log /
-// surface it.
+// surface it. FailedDomains holds the domain names that failed so
+// the scheduler can fire per-domain alerts.
 type Result struct {
-	Checked  int
+	Checked       int
-	Renewed  int
+	Renewed       int
-	Failed   int
+	Failed        int
-	Skipped  int
+	Skipped       int
+	FailedDomains []string
 }
 
 // Run scans tls_certs for LE-issued rows that expire within the
@@ -95,6 +97,7 @@ func (s *Service) Run(ctx context.Context) (Result, error) {
 		certPEM, chainPEM, keyPEM, err := s.Issuer.Issue(row.Domain)
 		if err != nil {
 			res.Failed++
+			res.FailedDomains = append(res.FailedDomains, row.Domain)
 			s.log().Error("certrenewer: issue failed", "domain", row.Domain, "error", err)
 			_ = s.Repo.MarkError(ctx, row.Domain, err.Error())
 			continue
@@ -102,12 +105,14 @@ func (s *Service) Run(ctx context.Context) (Result, error) {
 		info, err := certstore.Parse(certPEM)
 		if err != nil {
 			res.Failed++
+			res.FailedDomains = append(res.FailedDomains, row.Domain)
 			_ = s.Repo.MarkError(ctx, row.Domain, "parse: "+err.Error())
 			continue
 		}
 		path, err := certstore.WriteCombined(s.CertDir, row.Domain, certPEM, chainPEM, keyPEM)
 		if err != nil {
 			res.Failed++
+			res.FailedDomains = append(res.FailedDomains, row.Domain)
 			_ = s.Repo.MarkError(ctx, row.Domain, "write: "+err.Error())
 			continue
 		}