Scenario

Multiple microservices generate log files with events tracked by request IDs. You need to correlate logs across services to trace individual requests through the system.

Task

Write a Python script at /home/interview/trace_request.py that reads log files from three services (frontend.log, api.log, database.log in /home/interview/), finds all log entries for a specific request ID, sorts them chronologically, and saves the trace to /home/interview/request_trace.json.

The script should accept a request_id as a command-line argument: python3 trace_request.py <request_id>

All fields from the original log entries must be preserved in the output (some entries may contain additional fields like duration_ms).

Example

Log entry format (JSON lines):

{"timestamp": "2026-02-13T10:15:30.123Z", "service": "frontend", "level": "INFO", "request_id": "req_abc123", "message": "Request received", "duration_ms": 142}

Expected output in /home/interview/request_trace.json:

[
  {
    "timestamp": "2026-02-13T10:15:30.123Z",
    "service": "frontend",
    "level": "INFO",
    "request_id": "req_abc123",
    "message": "Request received",
    "duration_ms": 142
  },
  {
    "timestamp": "2026-02-13T10:15:30.245Z",
    "service": "api",
    "level": "INFO",
    "request_id": "req_abc123",
    "message": "Processing request"
  }
]

Step 1: Examine the log files

head -n 5 /home/interview/frontend.log
head -n 5 /home/interview/api.log
head -n 5 /home/interview/database.log

Each log file contains JSON lines with timestamp, service, level, request_id, message, and optionally duration_ms fields.

Step 2: Create the Python script

nano /home/interview/trace_request.py

Write a script that reads all log files, filters by request_id, and sorts chronologically:

import json
import sys
from datetime import datetime

if len(sys.argv) != 2:
    print("Usage: python3 trace_request.py <request_id>")
    sys.exit(1)

request_id = sys.argv[1]

# Log files to read
log_files = [
    '/home/interview/frontend.log',
    '/home/interview/api.log',
    '/home/interview/database.log'
]

# Collect all log entries matching the request_id
matching_logs = []

for log_file in log_files:
    with open(log_file, 'r') as f:
        for line in f:
            log_entry = json.loads(line.strip())
            if log_entry['request_id'] == request_id:
                matching_logs.append(log_entry)

# Sort by timestamp
matching_logs.sort(key=lambda x: datetime.fromisoformat(x['timestamp'].replace('Z', '+00:00')))

# Save to output file
with open('/home/interview/request_trace.json', 'w') as f:
    json.dump(matching_logs, f, indent=2)

print(f"Found {len(matching_logs)} log entries for request_id: {request_id}")
print(f"Trace saved to /home/interview/request_trace.json")

Step 3: Get a sample request_id for testing

python3 -c "import json; print(json.loads(open('/home/interview/frontend.log').readline())['request_id'])"

Step 4: Run the script with a request_id

python3 /home/interview/trace_request.py <request_id>

Step 5: Verify the output

cat /home/interview/request_trace.json

Should show chronologically ordered log entries from all three services for the specified request.