Google Cloud Error Logs Analysis

Fetch and analyze error logs from Google Cloud Logging for Cloud Functions. This skill helps identify error patterns, group similar errors, and provide root cause analysis.

Trigger Phrases

• "get GCP errors"
• "analyze Cloud Functions logs"
• "check Google Cloud logs"
• "debug Cloud Functions"
• "investigate GCP errors"
• "show error logs from GCP"

Prerequisites

Verify gcloud CLI is installed and authenticated:

# Check gcloud is installed
gcloud --version

# Check authentication
gcloud auth list

# If not authenticated
gcloud auth login

# Set default project (optional)
gcloud config set project PROJECT_ID

Workflow: Fetch and Analyze Errors

Step 1: Fetch Error Logs

Use the fetch script to retrieve error logs from Cloud Logging:

bash "${SKILL_DIR}/scripts/fetch-errors.sh" --project=PROJECT_ID [OPTIONS]

Required parameter:

• --project=PROJECT_ID: GCP project ID

Optional parameters:

• --function=FUNCTION_NAME: Specific Cloud Function name
• --hours=24: Time range in hours (default: 24)
• --severity=ERROR: Log severity level (ERROR, WARNING, CRITICAL)
• --limit=100: Maximum number of log entries (default: 100)
• --output=errors.json: Output file path (default: /tmp/gcp-errors.json)

Examples:

# Get all errors from last 24 hours
bash scripts/fetch-errors.sh --project=my-project

# Get errors for specific function
bash scripts/fetch-errors.sh --project=my-project --function=processPayment

# Get critical errors from last hour
bash scripts/fetch-errors.sh --project=my-project --hours=1 --severity=CRITICAL

# Get last 500 errors
bash scripts/fetch-errors.sh --project=my-project --limit=500

Step 2: Analyze Error Patterns

Use the analysis script to group and analyze errors:

python3 "${SKILL_DIR}/scripts/analyze-errors.py" /tmp/gcp-errors.json

The analysis script:

• Groups similar errors together
• Calculates error frequency
• Identifies time distribution patterns
• Detects common error types
• Suggests potential root causes

Step 3: Review Analysis Results

The analysis output includes:

Error Summary

• Total error count
• Unique error patterns
• Time range covered
• Most common error types

Error Groups

Each group shows:

• Error pattern (normalized message)
• Occurrence count
• First and last seen timestamps
• Affected resources (function names)
• Sample stack trace
• Root cause analysis

Time Distribution

• Errors per hour histogram
• Peak error times
• Trend analysis (increasing/decreasing)

Output Format

Present findings using this template:

## GCP Error Analysis Results

**Time Range**: [START] to [END]  
**Total Errors**: [COUNT]  
**Unique Patterns**: [COUNT]  
**Project**: [PROJECT_ID]

### Top Error Patterns

#### 1. [Error Type]
- **Occurrences**: [COUNT] ([PERCENTAGE]%)
- **First Seen**: [TIMESTAMP]
- **Last Seen**: [TIMESTAMP]
- **Affected Functions**: [FUNCTION_NAMES]

**Error Message**:

[NORMALIZED_ERROR_MESSAGE]

**Root Cause Analysis**:
[ANALYSIS_FROM_COMMON_PATTERNS]

**Recommended Actions**:
- [ACTION_1]
- [ACTION_2]

---

[REPEAT FOR TOP 5-10 ERROR PATTERNS]

### Time Distribution

[ERROR_FREQUENCY_CHART_OR_DESCRIPTION]

### Summary & Recommendations

[OVERALL_ANALYSIS_AND_NEXT_STEPS]

Common Cloud Functions Error Patterns

Authentication & Authorization

Error: Permission denied or 403 Forbidden

• Cause: IAM permissions missing
• Check: Service account permissions, API enablement
• Fix: Grant required roles to service account

Error: Unauthenticated or 401 Unauthorized

• Cause: Missing or invalid auth token
• Check: Token generation, expiration
• Fix: Implement proper authentication flow

Resource Limits

Error: Memory limit exceeded or Container killed

• Cause: Function using more memory than allocated
• Check: Memory allocation in function config
• Fix: Increase memory limit or optimize code

Error: Function execution timeout

• Cause: Function takes longer than timeout setting
• Check: Timeout configuration, blocking operations
• Fix: Increase timeout or optimize slow operations

Error: Quota exceeded

• Cause: API quota or rate limit reached
• Check: API quotas in GCP console
• Fix: Request quota increase or implement rate limiting

Networking

Error: Connection refused or ECONNREFUSED

• Cause: Unable to connect to external service
• Check: VPC configuration, firewall rules
• Fix: Configure VPC connector or check service availability

Error: DNS lookup failed or ENOTFOUND

• Cause: Cannot resolve hostname
• Check: DNS configuration, hostname spelling
• Fix: Verify DNS settings or use IP address

Error: SSL certificate error

• Cause: Invalid or expired SSL certificate
• Check: Certificate validity, chain completeness
• Fix: Update certificate or disable verification (dev only)

Code Errors

Error: TypeError, ReferenceError, SyntaxError

• Cause: JavaScript/Node.js code error
• Check: Recent code changes, null checks
• Fix: Fix code logic, add error handling

Error: ModuleNotFoundError or Cannot find module

• Cause: Missing dependency
• Check: package.json dependencies, deployment
• Fix: Install dependency and redeploy

Error: Unhandled promise rejection

• Cause: Async error not caught
• Check: Try-catch blocks, .catch() handlers
• Fix: Add proper error handling

Cold Start Issues

Error: Function cold start timeout

• Cause: Initialization taking too long
• Check: Global scope initialization, dependencies
• Fix: Optimize imports, reduce initialization work

Configuration

Error: Environment variable not set

• Cause: Missing required environment variable
• Check: Function configuration
• Fix: Set environment variable in deployment

Error: Invalid configuration

• Cause: Malformed config file or settings
• Check: YAML/JSON syntax, required fields
• Fix: Validate and correct configuration

Troubleshooting

Script Fails to Authenticate

# Reauthenticate
gcloud auth login

# Or use service account
gcloud auth activate-service-account --key-file=KEY_FILE.json

No Logs Found

• Verify project ID is correct
• Check time range (increase --hours)
• Verify function name spelling
• Check if logs exist in GCP Console

Analysis Script Errors

# Ensure Python 3 is installed
python3 --version

# The script uses only standard library, no pip install needed

Rate Limiting

If you hit API rate limits:

• Reduce --limit parameter
• Increase time between requests
• Use multiple time ranges instead of one large query

Advanced Usage

Custom Filters

Modify the fetch script filter for specific queries:

# Edit the FILTER variable in scripts/fetch-errors.sh
# Examples:
# - Filter by specific text: textPayload=~"payment failed"
# - Filter by HTTP status: httpRequest.status>=500
# - Filter by user: labels.user_id="12345"

Export for Further Analysis

# Save to JSON for processing
bash scripts/fetch-errors.sh --project=my-project --output=errors.json

# Convert to CSV (requires jq)
cat errors.json | jq -r '.[] | [.timestamp, .severity, .message] | @csv' > errors.csv

Continuous Monitoring

# Set up cron job for hourly checks
0 * * * * cd /path/to/skill && bash scripts/fetch-errors.sh --project=my-project --hours=1 --output=/tmp/errors-$(date +\%Y\%m\%d-\%H).json

Additional Resources

For detailed Cloud Logging filter syntax and API reference, see reference.md.

Quick Reference

Command	Purpose
gcloud auth list	Check authentication
gcloud projects list	List available projects
gcloud functions list	List Cloud Functions
gcloud logging read --help	View logging command help