Log Analyzer

Parse and summarize application logs to find errors, patterns, and anomalies.

Use when analyzing log files, debugging errors, or summarizing system events.

Requirements

• •Standard Unix tools: grep, awk, sort, uniq
• •Optional: jq (for JSON logs)
• •No API keys needed

Instructions

1. •

   Get log source — common locations:

   bash

   # System logs
   /var/log/syslog
   /var/log/auth.log
   journalctl -u <service> --since "1 hour ago" --no-pager
   
   # Web servers
   /var/log/nginx/error.log
   /var/log/nginx/access.log
   
   # Docker
   docker logs --since 1h <container> 2>&1
   
   # Application logs
   # Ask user for path
   

2. •

   Quick analysis — run these in sequence:

   bash

   # File overview
   wc -l logfile
   head -1 logfile && tail -1 logfile   # time range
   
   # Error counts by level
   grep -ciE '(error|fatal|critical)' logfile
   grep -ciE '(warn|warning)' logfile
   
   # Top error patterns (deduplicated)
   grep -iE '(error|exception|fatal)' logfile | \
     sed 's/[0-9]\{4\}-[0-9]\{2\}-[0-9]\{2\}[T ][0-9:.]*//g' | \
     sed 's/\[.*\]//g' | sort | uniq -c | sort -rn | head -20
   
   # Error timeline (per hour)
   grep -iE '(error|fatal)' logfile | \
     grep -oE '[0-9]{4}-[0-9]{2}-[0-9]{2}[T ][0-9]{2}' | \
     sort | uniq -c
   
   # Recent errors with context
   grep -iE '(error|exception|fatal)' logfile | tail -30
   

3. •

   JSON logs (structured logging):

   bash

   cat logfile | jq -r 'select(.level == "error") | .message' | sort | uniq -c | sort -rn
   cat logfile | jq -r 'select(.level == "error") | .timestamp' | head -5  # first errors
   

4. •

   Output format:

   code

   📊 Log Analysis — filename.log (42,531 lines)
   Time range: 2025-01-15 00:00 → 2025-01-15 14:30
   
   ## Summary
   | Level | Count | % |
   |-------|-------|---|
   | ERROR | 67 | 0.16% |
   | WARN | 234 | 0.55% |
   | INFO | 42,230 | 99.3% |
   
   ## Top Errors
   | Count | Error Pattern |
   |-------|--------------|
   | 42 | Connection refused to db:5432 |
   | 18 | TimeoutError: request exceeded 30s |
   | 7 | OutOfMemoryError: heap space |
   
   ## Error Timeline
   | Hour | Errors | |
   |------|--------|-|
   | 14:00 | 3 | ▪ |
   | 15:00 | 47 | ▪▪▪▪▪▪▪▪▪▪▪▪▪ ⚠️ Spike |
   | 16:00 | 5 | ▪▪ |
   
   ## 🔍 Recommendations
   - [ ] Check database connectivity (42 connection refused errors)
   - [ ] Review timeout settings (18 timeouts at 30s)
   - [ ] Increase JVM heap size (7 OOM errors)
   

Edge Cases

• •Large files (>100MB): Use tail -n 10000 or --since filtering first. Never read entire file into memory.
• •Non-standard log format: Ask user about the timestamp and level format, then adapt grep patterns.
• •Binary/compressed logs: zgrep for .gz files, journalctl for systemd binary logs.
• •Multi-line stack traces: Use grep -A 5 to capture context lines after errors.
• •No timestamps: Fall back to line-number-based analysis.

Real-time Monitoring

For live monitoring, suggest:

tail -f logfile | grep --line-buffered -iE '(error|fatal)'

Security Considerations

• •Log files may contain sensitive data (IPs, emails, tokens). Redact before sharing analysis.
• •Don't pipe log contents to external services.