Skip to main content

Monitoring Logs and Analytics

View real-time logs from your pods and use AI-powered analysis to understand issues.

Pod Logs Overview​

Pod logs show output from your running applications in the cluster.

What are Pod Logs?​

Pod logs are text output generated by applications:

  • Application startup messages
  • Errors and warnings
  • Debug information
  • Info messages
  • Business events

Why View Logs?​

  • Troubleshooting: Find out what went wrong
  • Debugging: Understand application behavior
  • Monitoring: Watch real-time activity
  • Auditing: Record what happened
  • Analysis: Find patterns and issues

Accessing Pod Logs​

Option 1: Via Dashboard​

  1. Go to Clusters page
  2. Select cluster with agent
  3. Click Pod Logs tab
  4. Select application from dropdown
  5. Click Fetch Logs or Stream Logs

Option 2: Via Command Line​

# View recent logs
kubectl logs <pod-name> -n <namespace>

# Stream live logs
kubectl logs -f <pod-name> -n <namespace>

# View last 100 lines
kubectl logs <pod-name> -n <namespace> --tail=100

# View logs from specific time
kubectl logs <pod-name> --since=1h

Fetching Logs​

One-Time Fetch​

Get a snapshot of recent logs:

  1. Select Application

    • Choose app from dropdown
    • Select how many lines:
      • 50 lines: Last few seconds
      • 100 lines: Last minute
      • 500 lines: Last 5 minutes
      • 1000 lines: Last 10 minutes

  2. Click Fetch Logs

    • Wait for logs to load
    • Results appear in viewer

  3. View Results

    • See log entries with timestamps
    • Each line shows log level and message

Log Format​

[Timestamp] [Pod Name] [Level] Message
[2024-01-15 14:32:45] [api-server-xyz] [INFO] Request processed
[2024-01-15 14:32:46] [api-server-xyz] [ERROR] Database connection failed

Streaming Logs​

Watch logs in real-time as they're generated:

Start Streaming​

  1. Select Application

    • Choose the app to monitor
    • Only one stream at a time

  2. Click Stream Logs

    • Live logs start appearing
    • New entries appear at bottom
    • Stream indicator shows status

Streaming Features​

Auto-Scroll:

  • Toggle "Auto-scroll" on/off
  • When on: Jumps to newest entry
  • When off: Stay at current position

Search:

  • Type in search box
  • Filters logs in real-time
  • Shows matching entries
  • Case-insensitive

Filter by Level:

  • All Levels: Show everything
  • Error: Only error messages
  • Warn: Warnings and errors
  • Info: Info and above
  • Debug: All messages

Stop Streaming​

  1. Click Stop Stream button
  2. Live updates stop
  3. View your captured logs
  4. Can export or analyze

Filtering and Searching​

Search Logs​

Find specific messages:

  1. Enter search term

    • Type what you're looking for
    • Search is real-time

  2. Results update

    • Only matching logs shown
    • Count shows matches found

  3. Clear search

    • Delete search text
    • All logs appear again

Filter by Level​

Show only certain severity:

ERROR: Application errors
  └─ Indicates something failed

WARN: Warnings
  └─ Indicates potential issue

INFO: Information messages
  └─ General status updates

DEBUG: Debug information
  └─ Detailed technical info

Example Searches​

Find database errors:

Search: "database"
Shows: All lines mentioning database

Find timeout errors:

Search: "timeout"
Shows: All timeout-related errors

Find a user's activity:

Search: "[email protected]"
Shows: Everything that user did

Exporting Logs​

Save logs for analysis or archival:

Export Options​

Export as Text (.txt)

  • Plain text format
  • Easy to read
  • Good for sharing
  • Use for: Documentation, emails

Export as JSON (.json)

  • Structured format
  • Includes metadata
  • Machine-readable
  • Use for: Analysis tools, automation

How to Export​

  1. Load or filter logs

    • Fetch or stream logs first
    • Filter to what you want

  2. Click Export

    • Choose format (TXT or JSON)
    • File downloads automatically

  3. Use exported logs

    • Analyze offline
    • Share with team
    • Import to analysis tools
    • Archive for compliance

AI-Powered Log Analysis​

Use AI to automatically analyze logs and find issues.

What AI Analysis Does​

The AI analyzes your logs to:

  • Detect Issues: Find errors and problems
  • Identify Patterns: Spot recurring issues
  • Provide Recommendations: Suggest fixes
  • Explain Problems: Describe what went wrong

Using AI Analysis​

Step 1: Prepare Logs​

Option A: Use Current Pod Logs

  1. Toggle "Use current pod logs" ON
  2. Make sure logs are loaded
  3. Shows how many entries will be analyzed

Option B: Paste Logs Manually

  1. Toggle "Use current pod logs" OFF
  2. Paste your logs in the text area
  3. Any log format is fine

Step 2: Run Analysis​

  1. Click Analyze with AI
  2. Wait for analysis (usually 10-30 seconds)
  3. Hourglass icon shows progress
  4. Results appear when ready

Step 3: Review Results​

AI provides:

  • Summary: What happened overall
  • Issues Found: Specific problems detected
  • Severity: How serious each issue is
  • Recommendations: How to fix

Understanding AI Analysis Results​

Analysis Summary​

Brief overview of what AI found:

"Application experienced 5 errors in the last hour, 
mostly related to database timeouts. Performance 
degraded after 14:30 UTC."

Issues Detected​

Specific problems found:

IssueSeverityDescription
Database Connection TimeoutHighCould not connect to database
Memory LeakMediumMemory usage growing over time
Slow QueryHighQuery taking 5+ seconds

Recommendations​

How to fix each issue:

Database Connection Timeout
Recommendations:
1. Increase database connection pool size
2. Check database server health
3. Verify network connectivity
4. Review query timeouts

Patterns Found​

Recurring issues and trends:

Error Pattern 1: Database timeouts spike at 14:00-15:00 UTC
  → Coincides with backup jobs
  → Recommendation: Schedule backups at off-peak hours

Error Pattern 2: Memory usage grows 100MB per hour
  → Suggests memory leak in application
  → Recommendation: Profile application memory usage

Interpreting AI Insights​

Issue Severity Levels​

Critical: 🔴

  • Application is down or failing
  • Immediate action required
  • Fix immediately

High: 🟠

  • Performance degraded
  • Users affected
  • Fix very soon

Medium: 🟡

  • Minor issues
  • Should be addressed
  • Fix when convenient

Low: 🔵

  • Informational
  • Good to know
  • Can defer

Common Troubleshooting Scenarios​

Scenario 1: Application Keeps Crashing​

Logs show:

Starting application...
OutOfMemoryError: Java heap space
Application terminated

AI Analysis suggests:

  • Insufficient memory allocated
  • Possible memory leak
  • Large data processing causing spike

Solutions:

  1. Increase pod memory limit
  2. Check for memory leaks in code
  3. Process data in smaller chunks
  4. Enable memory profiling

Scenario 2: Database Errors​

Logs show:

ERROR: Cannot connect to database
Connection timeout after 30s
Failed to execute query

AI Analysis suggests:

  • Database server unreachable
  • Network connectivity issues
  • Connection pool exhausted

Solutions:

  1. Verify database is running
  2. Check firewall rules
  3. Increase connection pool
  4. Review network configuration

Scenario 3: High Latency​

Logs show:

Request received
Processing...
Completed in 5000ms (expected: 100ms)

AI Analysis suggests:

  • Slow queries
  • External API delays
  • Resource contention

Solutions:

  1. Optimize database queries
  2. Add caching
  3. Use CDN for external assets
  4. Scale cluster resources

Best Practices​

1. Regular Monitoring​

  • Check logs daily
  • Monitor trends
  • Act on warnings
  • Review errors

2. Use Streaming for Live Issues​

  • Stream when troubleshooting
  • Watch real-time behavior
  • Easier than fetching later
  • See issue as it happens

3. Use AI Analysis Regularly​

  • Run weekly analysis
  • Track recurring issues
  • Monitor for patterns
  • Proactive problem finding

4. Export and Archive​

  • Export important logs
  • Keep for compliance
  • Analyze historical patterns
  • Document issues

5. Set Up Alerts​

  • Alert on error rates
  • Alert on specific errors
  • Alert on performance degradation
  • Set up escalation

Log Retention​

How Long are Logs Kept?​

  • Real-time logs: 7 days
  • Archived logs: 30 days
  • Compliance logs: 1 year (if enabled)

Exporting Before Expiration​

If you need logs longer:

  1. Export before expiration date
  2. Store in your own system
  3. Archive as needed
  4. Use for analysis later

Limitations​

Maximum Log Entries​

  • Fetch: Up to 1000 lines
  • Stream: Keeps last 500 lines
  • Export: Based on what's loaded

Network Requirements​

  • Stable internet connection
  • 1+ Mbps bandwidth for streaming
  • Some browsers better than others

App Requirements​

  • Application must have agent deployed
  • Agent must have Logging capability
  • Application must output logs to stdout/stderr

Next Steps​

  1. View Security Findings - Check cluster security
  2. Manage Resources - Monitor cluster health
  3. Deploy Applications - Use your cluster

Support​

Questions about logs?

  • Check the scenarios above
  • Review AI analysis suggestions
  • Contact support: [email protected]

Logs not appearing?

  • Verify agent is deployed
  • Check agent status
  • Ensure Logging capability enabled
  • Verify application outputs logs