Release v0.59.0

.github/workflows/agent-ci.yml

@@ -43,7 +43,7 @@ jobs:
  run: go vet ./...
 
  - name: Install staticcheck
- run: GOBIN=$PWD/.bin go install honnef.co/go/tools/cmd/staticcheck@latest
+ run: GOBIN=$PWD/.bin go install honnef.co/go/tools/cmd/staticcheck@v0.7.0
 
  - name: Staticcheck
  run: ./.bin/staticcheck ./...

.github/workflows/cli-ci.yml

@@ -43,7 +43,7 @@ jobs:
  run: go vet ./...
 
  - name: Install staticcheck
- run: GOBIN=$PWD/.bin go install honnef.co/go/tools/cmd/staticcheck@latest
+ run: GOBIN=$PWD/.bin go install honnef.co/go/tools/cmd/staticcheck@v0.7.0
 
  - name: Staticcheck
  run: ./.bin/staticcheck ./...

.github/workflows/updater-ci.yml

@@ -45,7 +45,7 @@ jobs:
  run: go vet ./...
 
  - name: Install staticcheck
- run: GOBIN=$PWD/.bin go install honnef.co/go/tools/cmd/staticcheck@latest
+ run: GOBIN=$PWD/.bin go install honnef.co/go/tools/cmd/staticcheck@v0.7.0
 
  - name: Staticcheck
  run: ./.bin/staticcheck ./...

agent/internal/traefik/reload.go

@@ -15,6 +15,7 @@ import (
 const (
 	lastReloadSuccessMetric = "traefik_config_last_reload_success"
 	pendingReloadMarkerName = ".routing-reload-pending"
+	metricsReadyTimeout = 15 * time.Second
 )
 
 var (
@@ -30,6 +31,22 @@ func LastSuccessfulReload() (time.Time, error) {
 	return readLastSuccessfulReload()
 }
 
+func waitForMetricsReady(timeout time.Duration) error {
+	deadline := time.Now().Add(timeout)
+	var lastErr error
+	for {
+ if _, err := LastSuccessfulReload(); err == nil {
+ return nil
+ } else {
+ lastErr = err
+ }
+ if time.Now().After(deadline) {
+ return fmt.Errorf("traefik metrics did not become ready within %s: %w", timeout, lastErr)
+ }
+ time.Sleep(reloadPollInterval)
+	}
+}
+
 func fetchLastSuccessfulReload() (time.Time, error) {
 	response, err := metricsHTTPClient.Get(traefikMetricsURL)
 	if err != nil {

agent/internal/traefik/reload_test.go

@@ -20,6 +20,54 @@ traefik_config_last_reload_success 1.725e+09
 	}
 }
 
+func TestWaitForMetricsReadyRetriesTemporaryErrors(t *testing.T) {
+	originalReader := readLastSuccessfulReload
+	originalPollInterval := reloadPollInterval
+	t.Cleanup(func() {
+ readLastSuccessfulReload = originalReader
+ reloadPollInterval = originalPollInterval
+	})
+
+	attempts := 0
+	readLastSuccessfulReload = func() (time.Time, error) {
+ attempts++
+ if attempts < 3 {
+ return time.Time{}, os.ErrNotExist
+ }
+ return time.Now(), nil
+	}
+	reloadPollInterval = time.Millisecond
+
+	if err := waitForMetricsReady(50 * time.Millisecond); err != nil {
+ t.Fatal(err)
+	}
+	if attempts != 3 {
+ t.Fatalf("metrics read attempted %d times, want 3", attempts)
+	}
+}
+
+func TestWaitForMetricsReadyTimesOut(t *testing.T) {
+	originalReader := readLastSuccessfulReload
+	originalPollInterval := reloadPollInterval
+	t.Cleanup(func() {
+ readLastSuccessfulReload = originalReader
+ reloadPollInterval = originalPollInterval
+	})
+
+	readLastSuccessfulReload = func() (time.Time, error) {
+ return time.Time{}, os.ErrDeadlineExceeded
+	}
+	reloadPollInterval = time.Millisecond
+
+	err := waitForMetricsReady(5 * time.Millisecond)
+	if err == nil {
+ t.Fatal("metrics readiness wait unexpectedly succeeded")
+	}
+	if !strings.Contains(err.Error(), "traefik metrics did not become ready within 5ms") {
+ t.Fatalf("unexpected timeout error: %v", err)
+	}
+}
+
 func TestDynamicConfigReloadedRequiresReloadAtOrAfterNewestFile(t *testing.T) {
 	originalDir := dynamicConfigDir
 	originalReader := readLastSuccessfulReload

agent/internal/traefik/static.go

@@ -6,7 +6,6 @@ import (
 	"os"
 	"os/exec"
 	"reflect"
-	"time"
 
 	"gopkg.in/yaml.v3"
 )
@@ -16,6 +15,7 @@ const (
 	metricsEntryPointAddr = "127.0.0.1:9100"
 )
 
+// Whole-number buckets must be ints to match yaml.v3's decoded types.
 var prometheusLatencyBuckets = []interface{}{
 	0.005,
 	0.01,
@@ -26,12 +26,12 @@ var prometheusLatencyBuckets = []interface{}{
 	0.25,
 	0.5,
 	0.75,
-	1.0,
+	1,
 	2.5,
-	5.0,
-	10.0,
-	30.0,
-	60.0,
+	5,
+	10,
+	30,
+	60,
 }
 
 func validateStaticConfig(data []byte) error {
@@ -218,6 +218,5 @@ func ReloadTraefik() error {
  return fmt.Errorf("failed to restart traefik: %w", err)
 	}
 	log.Printf("[traefik] restarted traefik to apply static config changes")
-	time.Sleep(2 * time.Second)
-	return nil
+	return waitForMetricsReady(metricsReadyTimeout)
 }

agent/internal/traefik/static_test.go

@@ -1,6 +1,10 @@
 package traefik
 
-import "testing"
+import (
+	"testing"
+
+	"gopkg.in/yaml.v3"
+)
 
 func TestEnsurePrometheusMetricsConfigAddsPrivateMetricsEndpoint(t *testing.T) {
 	config := map[string]interface{}{
@@ -44,7 +48,17 @@ func TestEnsurePrometheusMetricsConfigIsStable(t *testing.T) {
 	if !ensurePrometheusMetricsConfig(config) {
  t.Fatal("expected first call to modify config")
 	}
-	if ensurePrometheusMetricsConfig(config) {
+
+	data, err := yaml.Marshal(config)
+	if err != nil {
+ t.Fatalf("failed to marshal config: %v", err)
+	}
+	var roundTripped map[string]interface{}
+	if err := yaml.Unmarshal(data, &roundTripped); err != nil {
+ t.Fatalf("failed to unmarshal config: %v", err)
+	}
+
+	if ensurePrometheusMetricsConfig(roundTripped) {
  t.Fatal("expected second call to be stable")
 	}
 }

web/actions/compose.ts

@@ -5,6 +5,7 @@ import { db } from "@/db";
 import { services } from "@/db/schema";
 import { requireDeveloperRole } from "@/lib/auth";
 import { parseComposeYaml } from "@/lib/compose-parser";
+import { reportServerError } from "@/lib/server-errors";
 import {
 	addServiceVolume,
 	createService,
@@ -144,6 +145,9 @@ export async function importCompose(
  volume.containerPath,
  );
  } catch (e) {
+ reportServerError(e, "compose-import.volume.add", {
+ tags: { serviceId: result.id },
+ });
  warnings.push({
  service: finalName,
  field: "volumes",
@@ -182,9 +186,16 @@ export async function importCompose(
  for (const serviceId of createdServiceIds) {
  try {
  await db.delete(services).where(eq(services.id, serviceId));
- } catch {}
+ } catch (cleanupError) {
+ reportServerError(cleanupError, "compose-import.cleanup", {
+ tags: { serviceId },
+ });
+ }
  }
 
+ reportServerError(error, "compose-import.create", {
+ tags: { projectId, environmentId },
+ });
  return {
  success: false,
  created: [],

web/actions/projects.ts

@@ -54,6 +54,7 @@ import {
 	cleanupRegistryArtifactsForService,
 	prepareRegistryArtifactCleanup,
 } from "@/lib/registry-retention";
+import { reportServerError } from "@/lib/server-errors";
 import {
 	deletePreviewService,
 	deletePreviewsForBaseService,
@@ -1748,6 +1749,9 @@ export async function abortRollout(serviceId: string) {
  try {
  await inngest.send(inngestEvents.rolloutCancelled.create({ rolloutId }));
  } catch (error) {
+ reportServerError(error, "rollout.cancellation.dispatch", {
+ tags: { rolloutId, serviceId },
+ });
  console.error(
  `[rollout:${rolloutId}] failed to send cancellation:`,
  error,

web/app/api/builds/[buildId]/logs/route.ts

@@ -1,5 +1,6 @@
 import { type NextRequest, NextResponse } from "next/server";
 import { invalidLogQueryResponse, normalizeLogSearch } from "@/lib/log-query";
+import { reportServerError } from "@/lib/server-errors";
 import { isLoggingEnabled, queryLogsByBuild } from "@/lib/victoria-logs";
 
 export async function GET(
@@ -28,6 +29,7 @@ export async function GET(
 
  return NextResponse.json({ logs });
 	} catch (error) {
+ reportServerError(error, "logs.build.query", { tags: { buildId } });
  console.error("Failed to fetch build logs:", error);
  return NextResponse.json(
  { message: "Failed to query build logs" },

web/app/api/deployments/[id]/logs/route.ts

@@ -5,6 +5,7 @@ import {
 	normalizeLogCursor,
 	parseLogLimit,
 } from "@/lib/log-query";
+import { reportServerError } from "@/lib/server-errors";
 import { isLoggingEnabled, queryLogsByDeployment } from "@/lib/victoria-logs";
 
 export async function GET(
@@ -50,6 +51,9 @@ export async function GET(
  hasMore: result.hasMore,
  });
 	} catch (error) {
+ reportServerError(error, "logs.deployment.query", {
+ tags: { deploymentId },
+ });
  console.error("[logs:deployment] failed to query logs:", error);
  return Response.json(
  { message: "Failed to query deployment logs" },

web/app/api/github/repos/route.ts

@@ -4,6 +4,7 @@ import { db } from "@/db";
 import { githubInstallations } from "@/db/schema";
 import { eq } from "drizzle-orm";
 import { getInstallationRepositories } from "@/lib/github";
+import { reportServerError } from "@/lib/server-errors";
 
 export async function GET() {
 	const session = await auth.api.getSession({
@@ -48,6 +49,9 @@ export async function GET() {
  });
  }
  } catch (error) {
+ reportServerError(error, "github.repositories.list", {
+ tags: { installationId: installation.installationId },
+ });
  console.error(
  `[github:repos] failed to fetch repos for installation ${installation.installationId}:`,
  error,

web/app/api/github/setup/route.ts

@@ -5,6 +5,7 @@ import { type NextRequest, NextResponse } from "next/server";
 import { db } from "@/db";
 import { githubInstallations } from "@/db/schema";
 import { requireRequestDeveloperRole } from "@/lib/api-auth";
+import { reportServerError } from "@/lib/server-errors";
 
 async function getInstallationDetails(installationId: number): Promise<{
 	account: { login: string; type: "User" | "Organization" };
@@ -40,6 +41,13 @@ async function getInstallationDetails(installationId: number): Promise<{
 	);
 
 	if (!response.ok) {
+ reportServerError(
+ new Error(
+ `GitHub installation lookup failed with status ${response.status}`,
+ ),
+ "github.installation.get",
+ { tags: { installationId } },
+ );
  console.error(
  `[github:setup] failed to get installation ${installationId}:`,
  await response.text(),

web/app/api/inngest/route.ts

@@ -25,6 +25,7 @@ import {
 	rolloutWorkflow,
 	scheduledBackupsCheck,
 	scheduledDeploymentsCheck,
+	sentryFailureWorkflow,
 	serviceDeletionWorkflow,
 	serviceCommandRetention,
 	serviceCronDispatcher,
@@ -52,6 +53,7 @@ export const { GET, POST, PUT } = serve({
  agentUpgradeTimeoutCheck,
  registryArtifactRetention,
  migrationWorkflow,
+ sentryFailureWorkflow,
  backupWorkflow,
  restoreWorkflow,
  onRestoreFailed,