introduce monitoring (#2503)

* Bring back docs on local development

* Improve local dev docs

* Agent task for moving from callbacks to webhooks

* Move from callback to webhooks, again

* Make repos webhook async; dont comment erorr when installed to all repos

* add staging deploy

* fix docker service

* fix projects refresh service

* add monitoring logs for ui

* update request logging

* copy file

* further attempt to log

* dont log unnecessary stuff

* fix syntax error

---------

Co-authored-by: Igor Zalutski <izalutski@gmail.com>

Author: motatoes

Dockerfile_ui

@@ -38,6 +38,7 @@ RUN --mount=type=cache,target=/root/.npm \
 # Bring in built assets and server entry from builder
 COPY --from=builder /app/dist ./dist
 COPY --from=builder /app/server-start.js ./
+COPY --from=builder /app/request-logging.js ./
 
 EXPOSE 3030
 

backend/controllers/github.go

@@ -66,20 +66,20 @@ func (d DiggerController) GithubAppWebHook(c *gin.Context) {
 			"installationId", *event.Installation.ID,
 		)
 
-		if *event.Action == "deleted" {
-			err := handleInstallationDeletedEvent(event, appId64)
-			if err != nil {
-				slog.Error("Failed to handle installation deleted event", "error", err)
-				c.String(http.StatusAccepted, "Failed to handle webhook event.")
-				return
-			}
-		} else if *event.Action == "created" || *event.Action == "unsuspended" || *event.Action == "new_permissions_accepted" {
-			if err := handleInstallationUpsertEvent(c.Request.Context(), gh, event, appId64); err != nil {
-				slog.Error("Failed to handle installation upsert event", "error", err)
-				c.String(http.StatusAccepted, "Failed to handle webhook event.")
-				return
+		// Run in goroutine to avoid webhook timeouts for large installations
+		go func(ctx context.Context) {
+			defer logging.InheritRequestLogger(ctx)()
+			if *event.Action == "deleted" {
+				if err := handleInstallationDeletedEvent(event, appId64); err != nil {
+					slog.Error("Failed to handle installation deleted event", "error", err)
+				}
+			} else if *event.Action == "created" || *event.Action == "unsuspended" || *event.Action == "new_permissions_accepted" {
+				// Use background context so work continues after HTTP response
+				if err := handleInstallationUpsertEvent(context.Background(), gh, event, appId64); err != nil {
+					slog.Error("Failed to handle installation upsert event", "error", err)
+				}
 			}
-		}
+		}(c.Request.Context())
 	case *github.InstallationRepositoriesEvent:
 		slog.Info("Processing InstallationRepositoriesEvent",
 			"action", event.GetAction(),

backend/controllers/github_comment.go

@@ -162,7 +162,7 @@ func handleIssueCommentEvent(gh utils.GithubClientProvider, payload *github.Issu
 	}
 
 	diggerYmlStr, ghService, config, projectsGraph, prSourceBranch, commitSha, changedFiles, err := getDiggerConfigForPR(gh, orgId, prLabelsStr, installationId, repoFullName, repoOwner, repoName, cloneURL, issueNumber)
-	if err != nil {		
+	if err != nil {
 		slog.Error("Error getting Digger config for PR",
 			"issueNumber", issueNumber,
 			"repoFullName", repoFullName,

backend/controllers/github_helpers.go

@@ -830,7 +830,7 @@ func getDiggerConfigForPR(gh utils.GithubClientProvider, orgId uint, prLabels []
 			"branch", prBranch,
 			"error", err,
 		)
-		return "", nil, nil, nil, nil, nil, nil, fmt.Errorf("error loading digger.yml: %v", err)
+		return "", nil, nil, nil, nil, nil, nil, fmt.Errorf("error loading digger.yml: %w", err)
 	}
 
 	return diggerYmlStr, ghService, config, dependencyGraph, &prBranch, &prCommitSha, changedFiles, nil
@@ -893,7 +893,7 @@ func GetDiggerConfigForBranchOrSha(gh utils.GithubClientProvider, installationId
 			"branch", branch,
 			"error", err,
 		)
-		return "", nil, nil, nil, fmt.Errorf("error cloning and loading config %v", err)
+		return "", nil, nil, nil, fmt.Errorf("error cloning and loading config: %w", err)
 	}
 
 	projectCount := 0

backend/utils/github.go

@@ -408,6 +408,59 @@ func GetGithubHostname() string {
 	return githubHostname
 }
 
+// IsAllReposInstallation checks if the GitHub App installation is configured to access all repositories
+// (as opposed to a selected subset). Returns true if installation is for "all" repos.
+// Note: This requires app-level JWT authentication, not installation token authentication.
+func IsAllReposInstallation(appId int64, installationId int64) (bool, error) {
+	githubAppPrivateKey := ""
+	githubAppPrivateKeyB64 := os.Getenv("GITHUB_APP_PRIVATE_KEY_BASE64")
+	if githubAppPrivateKeyB64 != "" {
+		decodedBytes, err := base64.StdEncoding.DecodeString(githubAppPrivateKeyB64)
+		if err != nil {
+			slog.Error("Failed to decode GITHUB_APP_PRIVATE_KEY_BASE64", "error", err)
+			return false, fmt.Errorf("error decoding private key: %v", err)
+		}
+		githubAppPrivateKey = string(decodedBytes)
+	} else {
+		githubAppPrivateKey = os.Getenv("GITHUB_APP_PRIVATE_KEY")
+		if githubAppPrivateKey == "" {
+			return false, fmt.Errorf("missing GitHub app private key")
+		}
+	}
+
+	// Use app-level transport (JWT) instead of installation token
+	atr, err := ghinstallation.NewAppsTransport(net.DefaultTransport, appId, []byte(githubAppPrivateKey))
+	if err != nil {
+		slog.Error("Failed to create GitHub app transport",
+			"appId", appId,
+			"error", err,
+		)
+		return false, fmt.Errorf("error creating app transport: %v", err)
+	}
+
+	client := github.NewClient(&net.Client{Transport: atr})
+
+	installation, _, err := client.Apps.GetInstallation(context.Background(), installationId)
+	if err != nil {
+		slog.Error("Failed to get GitHub installation details",
+			"installationId", installationId,
+			"error", err,
+		)
+		return false, fmt.Errorf("error getting installation details: %v", err)
+	}
+
+	repositorySelection := installation.GetRepositorySelection()
+	isAllRepos := repositorySelection == "all"
+
+	slog.Debug("Checked installation repository selection",
+		"installationId", installationId,
+		"repositorySelection", repositorySelection,
+		"isAllRepos", isAllRepos,
+	)
+
+	return isAllRepos, nil
+}
+
 func GetWorkflowIdAndUrlFromDiggerJobId(client *github.Client, repoOwner string, repoName string, diggerJobID string) (int64, string, error) {
 	slog.Debug("Looking for workflow for job",
 		"diggerJobId", diggerJobID,

libs/digger_config/digger_config.go

@@ -18,6 +18,9 @@ import (
 	"gopkg.in/yaml.v3"
 )
 
+// ErrDiggerConfigNotFound is returned when neither digger.yml nor digger.yaml exists in the repository
+var ErrDiggerConfigNotFound = errors.New("digger config file not found")
+
 type DirWalker interface {
 	GetDirs(workingDir string, config DiggerConfigYaml) ([]string, error)
 }
@@ -39,7 +42,7 @@ func ReadDiggerYmlFileContents(dir string) (string, error) {
 			slog.Error("could not read digger config file",
 				"error", err,
 				"dir", dir)
-			return "", fmt.Errorf("could not read the file both digger.yml and digger.yaml are missing: %v", err)
+			return "", fmt.Errorf("%w: both digger.yml and digger.yaml are missing: %v", ErrDiggerConfigNotFound, err)
 		}
 	}
 	diggerYmlStr := string(diggerYmlBytes)

taco/Dockerfile_token_service

@@ -7,6 +7,7 @@ WORKDIR /go/src/github.com/diggerhq/digger/taco
 
 # Copy go.mod/go.sum first for better layer caching
 COPY cmd/token_service/go.mod cmd/token_service/go.sum ./cmd/token_service/
+COPY internal/go.mod internal/go.sum ./internal/
 RUN cd cmd/token_service && go mod download
 
 # Copy source code

ui/request-logging.js

@@ -0,0 +1,74 @@
+import { unsealData } from 'iron-session';
+import { decodeJwt } from 'jose';
+
+// Request logging utilities
+export async function extractUserInfoFromRequest(req) {
+  try {
+    const cookieName = process.env.WORKOS_COOKIE_NAME || 'wos-session';
+    const cookiePassword = process.env.WORKOS_COOKIE_PASSWORD;
+    
+    if (!cookiePassword) {
+      return { userId: 'anonymous', orgId: 'anonymous' };
+    }
+    
+    const cookieHeader = req.headers?.cookie || req.getHeader?.('cookie');
+    if (!cookieHeader) {
+      return { userId: 'anonymous', orgId: 'anonymous' };
+    }
+    
+    const cookies = cookieHeader.split(';').reduce((acc, cookie) => {
+      const [key, value] = cookie.trim().split('=');
+      acc[key] = decodeURIComponent(value);
+      return acc;
+    }, {});
+    
+    const sessionCookie = cookies[cookieName];
+    if (!sessionCookie) {
+      return { userId: 'anonymous', orgId: 'anonymous' };
+    }
+    
+    const session = await unsealData(sessionCookie, {
+      password: cookiePassword,
+    });
+    
+    if (!session?.user?.id || !session?.accessToken) {
+      return { userId: 'anonymous', orgId: 'anonymous' };
+    }
+    
+    // Decode JWT to get organization ID
+    let orgId = 'anonymous';
+    try {
+      const decoded = decodeJwt(session.accessToken);
+      orgId = decoded.org_id || 'anonymous';
+    } catch (error) {
+      // If JWT decode fails, just use anonymous
+    }
+    
+    return { userId: session.user.id, orgId };
+  } catch (error) {
+    return { userId: 'anonymous', orgId: 'anonymous' };
+  }
+}
+
+export function logRequestInit(method, path, requestId, userId, orgId) {
+  console.log(JSON.stringify({
+    event: 'request_initialized',
+    method,
+    path,
+    requestId,
+    userId,
+    orgId,
+  }));
+}
+
+export function logResponse(method, path, requestId, latency, statusCode) {
+  console.log(JSON.stringify({
+    event: 'response_sent',
+    method,
+    path,
+    requestId,
+    latency,
+    statusCode,
+  }));
+}
+

ui/server-start.js

@@ -6,6 +6,10 @@ import { fileURLToPath } from 'node:url';
 import { Readable } from 'node:stream';
 import { createGzip } from 'node:zlib';
 import serverHandler from './dist/server/server.js';
+import { extractUserInfoFromRequest, logRequestInit, logResponse } from './request-logging.js';
+
+// Verify logging functions are loaded
+console.log('✅ Request logging module loaded');
 
 const __dirname = fileURLToPath(new URL('.', import.meta.url));
 const PORT = process.env.PORT || 3030;
@@ -68,6 +72,42 @@ const server = createServer(async (req, res) => {
   const requestId = req.headers['x-request-id'] || `ssr-${Math.random().toString(36).slice(2, 10)}`;
   const requestStart = Date.now();
 
+  // Parse URL early for logging
+  const url = new URL(req.url, `http://${req.headers.host}`);
+  const pathname = url.pathname;
+  const method = req.method;
+  
+  // Debug: Log that request handler is being called
+  console.log(`[DEBUG] Request received: ${method} ${pathname} [${requestId}]`);
+  
+  // Extract user ID and org ID and log request initialization
+  // Always log, even if extraction fails
+  let userId = 'anonymous';
+  let orgId = 'anonymous';
+  try {
+    const userInfo = await extractUserInfoFromRequest(req);
+    userId = userInfo.userId;
+    orgId = userInfo.orgId;
+  } catch (error) {
+    console.error(`User info extraction error [${requestId}]:`, error);
+  }
+  
+  // Always log request initialization
+  try {
+    logRequestInit(method, pathname, requestId, userId, orgId);
+  } catch (error) {
+    console.error(`Request logging error [${requestId}]:`, error);
+    // Fallback to direct console.log if logging function fails
+    console.log(JSON.stringify({
+      event: 'request_initialized',
+      method,
+      path: pathname,
+      requestId,
+      userId,
+      orgId,
+    }));
+  }
+  
   // Set request timeout
   req.setTimeout(REQUEST_TIMEOUT, () => {
     console.error(`⏱️  Request timeout (${REQUEST_TIMEOUT}ms): ${req.method} ${req.url} [${requestId}]`);
@@ -78,8 +118,6 @@ const server = createServer(async (req, res) => {
   });
 
   try {
-    const url = new URL(req.url, `http://${req.headers.host}`);
-    const pathname = url.pathname;
 
     // Try to serve static files from dist/client first
     // Serve: /assets/*, *.js, *.css, *.json, images, fonts, favicons
@@ -99,6 +137,13 @@ const server = createServer(async (req, res) => {
           'Cache-Control': 'public, max-age=31536000, immutable',
         });
         res.end(content);
+        // Log response for static files
+        try {
+          const latency = Date.now() - requestStart;
+          logResponse(method, pathname, requestId, latency, 200);
+        } catch (err) {
+          console.error(`Response logging error [${requestId}]:`, err);
+        }
         return;
       } catch (err) {
         // File not found, fall through to SSR handler
@@ -132,9 +177,9 @@ const server = createServer(async (req, res) => {
 
     // Log slow SSR requests
     if (ssrTime > 2000) {
-      console.log(`🔥 VERY SLOW SSR: ${req.method} ${pathname} took ${ssrTime}ms [${requestId}]`);
+      console.debug(`🔥 VERY SLOW SSR: ${req.method} ${pathname} took ${ssrTime}ms [${requestId}]`);
     } else if (ssrTime > 1000) {
-      console.log(`⚠️  SLOW SSR: ${req.method} ${pathname} took ${ssrTime}ms [${requestId}]`);
+      console.debug(`⚠️  SLOW SSR: ${req.method} ${pathname} took ${ssrTime}ms [${requestId}]`);
     }
 
     // Convert Web Standard Response to Node.js response
@@ -221,13 +266,28 @@ const server = createServer(async (req, res) => {
     } else {
       res.end();
     }
+    
+    // Log response after sending
+    try {
+      const latency = Date.now() - requestStart;
+      logResponse(method, pathname, requestId, latency, res.statusCode);
+    } catch (err) {
+      console.error(`Response logging error [${requestId}]:`, err);
+    }
   } catch (error) {
     console.error(`Server error [${requestId}]:`, error);
     if (!res.headersSent) {
       res.statusCode = 500;
       res.setHeader('Content-Type', 'text/plain');
       res.end('Internal Server Error');
     }
+    // Log error response
+    try {
+      const latency = Date.now() - requestStart;
+      logResponse(method, pathname, requestId, latency, res.statusCode || 500);
+    } catch (err) {
+      console.error(`Error response logging error [${requestId}]:`, err);
+    }
   }
 });
 

ui/src/authkit/serverFunctions.ts

@@ -137,12 +137,12 @@ export const getWidgetsAuthToken = createServerFn({method: 'GET'})
   // Check cache first
   const cached = serverCache.getWidgetToken(userId, organizationId);
   if (cached) {
-    console.log(`✅ Widget token cache hit for ${userId}:${organizationId}`);
+    console.debug(`✅ Widget token cache hit for ${userId}:${organizationId}`);
     return cached;
   }
 
   // Cache miss - generate new token
-  console.log(`❌ Widget token cache miss, generating new token for ${userId}:${organizationId}`);
+  console.debug(`❌ Widget token cache miss, generating new token for ${userId}:${organizationId}`);
   const token = await getWorkOS().widgets.getToken({
     userId: userId,
     organizationId: organizationId,