1import subprocess
2import json
3import time
4import csv
5from datetime import datetime
6import argparse
7import base64
8
9# Parse command-line arguments for data logging parameters
10parser = argparse.ArgumentParser(description='Parameters for data logging')
11parser.add_argument('-mj', '--maxjobs', type=int, required=True, help="Max Jobs parameter value")
12parser.add_argument('-c', '--churn', type=int, required=True, help="Churn parameter value")
13parser.add_argument('-i', '--interval', type=int, required=True, help="Interval parameter value")
14parser.add_argument('-u', '--username', type=str, required=True, help="Username for authentication")
15parser.add_argument('-p', '--password', type=str, required=True, help="Password for authentication")
16parser.add_argument('-db', '--filterByDatabase', type=str, required=False, help="Filter by specific database ID (optional)")
17parser.add_argument('-s', '--serverURL', type=str, default="http://localhost:5984/", help="Server URL (optional, default is http://localhost:5984/)")
18args = parser.parse_args()
19
20# Encode the username and password into base64
21authentication_token = base64.b64encode(f"{args.username}:{args.password}".encode()).decode()
22
23# Send a curl request to get scheduler data from CouchDB
24cmd = ["curl", "--location", f'{args.serverURL}_scheduler/docs',
25 "--header", f'Authorization: Basic {authentication_token}']
26
27while True:
28 result = subprocess.run(cmd, stdout=subprocess.PIPE)
29 data = json.loads(result.stdout)
30
31 with open('output.csv', 'a', newline='') as f:
32 writer = csv.writer(f)
33 for doc in data['docs']:
34 doc_id = doc['doc_id']
35 if args.filterByDatabase:
36 if doc_id != args.filterByDatabase:
37 continue
38 else:
39 print("doc_id is", doc_id)
40
41 docs_written = doc['info']['docs_written']
42 start_time = datetime.strptime(doc['start_time'], "%Y-%m-%dT%H:%M:%SZ")
43 time_elapsed = -(datetime.now() - start_time).total_seconds()
44 speed = docs_written/time_elapsed
45
46 # Log the speed of document processing
47 print("Speed is", speed, "documents per second")
48 writer.writerow([datetime.now(), doc_id, docs_written, speed, args.maxjobs, args.churn, args.interval])
49
50 # Adjust time between iterations here
51 time.sleep(15)View as plain text