Troubleshooting

Common issues and solutions for Hridaya

Configuration Issues

Error: Item missing required field

Error Message:

❌ CONFIG ERROR: Item 1 missing required field 'apiid'
   Item: {'market_hash_name': 'AK-47 | Redline (Field-Tested)', ...}

Cause: Your item configuration is missing a required parameter.

Solution:

Ensure every item has all required fields:

config.yaml

TRACKING_ITEMS:
  - market_hash_name: "AK-47 | Redline (Field-Tested)"  # Required
    appid: 730                                          # Required
    apiid: 'priceoverview'                              # Required
    polling-interval-in-seconds: 30                      # Required
    currency: 1                                          # Optional
    country: 'US'                                        # Optional
    language: 'english'                                  # Optional

Required fields for all endpoints:

• market_hash_name
• appid
• apiid
• polling-interval-in-seconds

Additional requirements:

• item_nameid is required for itemordershistogram and itemordersactivity

Error: Invalid apiid

Error Message:

❌ CONFIG ERROR: Item 1 has invalid apiid 'pricedata'
   Valid options: priceoverview, itemordershistogram, itemordersactivity, pricehistory

Cause: You’ve specified an API endpoint that doesn’t exist.

Solution:

Use one of the four supported endpoint names:

• priceoverview - Current prices and 24h volume
• itemordershistogram - Order book data
• itemordersactivity - Recent trading activity
• pricehistory - Historical hourly data

config.yaml

TRACKING_ITEMS:
  - market_hash_name: "AK-47 | Redline (Field-Tested)"
    appid: 730
    apiid: 'priceoverview'  # ✓ Valid
    # apiid: 'pricedata'    # ✗ Invalid
    polling-interval-in-seconds: 30

Error: Missing item_nameid

Error Message:

❌ CONFIG ERROR: Item 1 missing 'item_nameid' (required for itemordershistogram)
   Item: Dreams & Nightmares Case

Cause: You’re using itemordershistogram or itemordersactivity without providing the item_nameid.

Solution:

Find the item_nameid using browser DevTools:

1. Open item on Steam Market

   Navigate to your item’s market page.

2. Open DevTools Console

   Press F12 and go to the Console tab.

3. Extract item_nameid

   Type: g_rgAssets[730] (replace 730 with your appid)

   Look for your item in the output. The key is the item_nameid.

4. Add to config

   config.yaml

   TRACKING_ITEMS:
     - market_hash_name: "Dreams & Nightmares Case"
       appid: 730
       item_nameid: 175982336  # Add this!
       apiid: 'itemordershistogram'
       polling-interval-in-seconds: 30
   

Error: Infeasible configuration

Error Message:

❌ CONFIG ERROR: Infeasible configuration
   Calculated: 20 requests per 60s
   Limit: 15 requests per 60s
   Adjust polling intervals or reduce tracked items

Cause: Your polling intervals would exceed the rate limit.

Calculation:

# For each item: window_seconds ÷ polling-interval-in-seconds
# Example with 60s window:
# - 4 items at 10s intervals: 60÷10 = 6 req/item × 4 = 24 req/60s
# - Limit is 15 req/60s
# - Result: Infeasible (24 > 15)

Solutions:

1. Increase polling intervals:

TRACKING_ITEMS:
  - market_hash_name: "Item 1"
    polling-interval-in-seconds: 30  # Changed from 10

2. Reduce tracked items: Comment out or remove some items from TRACKING_ITEMS.

3. Increase rate limit (if safe):

LIMITS:
  REQUESTS: 20  # Increased from 15 (use cautiously!)
  WINDOW_SECONDS: 60

Increasing rate limits too high may result in temporary API bans from Steam.

Error: Unable to load config.yaml

Error Message:

FileNotFoundError: [Errno 2] No such file or directory: 'config.yaml'

Cause: The config.yaml file doesn’t exist or is in the wrong location.

Solution:

1. Ensure config.yaml is in the same directory as cerebro.py
2. Check file name spelling (case-sensitive on Linux/Mac)
3. If running from a subdirectory, use the full path:

python cerebro.py
# OR specify config path
python cerebro.py --config /path/to/config.yaml

Authentication Issues

Error: 401 Unauthorized (pricehistory endpoint)

Symptoms:

• pricehistory endpoint returns 401 errors
• Historical data is not being fetched

Cause: Missing or invalid Steam session cookies.

Solution:

1. Create .env file

   Create a .env file in your project root:

   touch .env
   
2. Extract fresh cookies

   1. Log into Steam in your browser
   2. Open DevTools (F12) → Application/Storage → Cookies
   3. Navigate to https://steamcommunity.com
   4. Copy these cookie values:
      • sessionid
      • steamLoginSecure
3. Add to .env

   .env

   sessionid=your_session_id_here
   steamLoginSecure=76561198XXXXX%7C%7CeyAid...
   
4. Restart Hridaya

   python cerebro.py
   

   Cookies are hot-swappable - you can update .env without restarting.

Steam session cookies expire after ~1 year. You’ll need to refresh them when they expire.

Cookies expired

Symptoms:

• pricehistory was working, now returns 401
• No recent changes to .env file

Cause: Steam session cookies have expired (typically after ~1 year).

Solution:

Extract fresh cookies using the same method:

1. Log into Steam in your browser
2. DevTools → Application → Cookies
3. Copy new sessionid and steamLoginSecure values
4. Update .env file
5. Hridaya will automatically use new cookies on next request (hot-swappable)

Rate Limiting Issues

Error: 429 Too Many Requests

Error Message:

aiohttp.ClientResponseError: 429, message='Too Many Requests'

Cause: You’ve exceeded Steam’s rate limits.

Solution:

1. Check your rate limit settings:

config.yaml

LIMITS:
  REQUESTS: 15  # Steam generally allows ~15-25 req/min
  WINDOW_SECONDS: 60

2. Increase polling intervals: Reduce request frequency:

TRACKING_ITEMS:
  - market_hash_name: "Item 1"
    polling-interval-in-seconds: 60  # Increased from 10

3. Wait before retrying: If you’re rate-limited, wait 1-2 minutes before restarting.

Hridaya’s RateLimiter should prevent 429 errors. If you’re seeing them, you may have multiple instances running or external scripts also hitting the API.

Requests are queuing/slow

Symptoms:

• Long delays between requests
• Items not updating at expected intervals

Cause: Rate limiter is working as designed - you’re at capacity.

Diagnosis:

Check startup output:

✓ Config feasible: 14 req/60s (93.3% capacity)

If capacity is >90%, requests will queue during peak times.

Solution:

1. Increase polling intervals for low-priority items:

TRACKING_ITEMS:
  - market_hash_name: "High Priority Item"
    polling-interval-in-seconds: 10  # Fast updates
  
  - market_hash_name: "Low Priority Item"
    polling-interval-in-seconds: 300  # 5 minutes

2. Reduce tracked items: Remove or comment out items you don’t actively monitor.

3. Increase rate limit (cautiously):

LIMITS:
  REQUESTS: 20  # From 15
  WINDOW_SECONDS: 60

Database Issues

Error: Database is locked

Error Message:

sqlite3.OperationalError: database is locked

Cause: Multiple processes are trying to write to the database simultaneously.

Solution:

1. Check for multiple instances:

ps aux | grep cerebro
# Kill extra instances
kill <pid>

2. Close SQLite connections: If you have SQLite CLI or database browsers open, close them.

3. Wait and retry: SQLite locks are usually temporary. Wait a few seconds and try again.

No data in database

Symptoms:

• market_data.db exists but tables are empty
• Queries return no results

Diagnosis:

1. Check if Hridaya is running:

ps aux | grep cerebro

2. Verify configuration:

python cerebro.py
# Look for: "✓ Started HIGH frequency tracking"

3. Check tables exist:

sqlite3 market_data.db
.tables
# Should show: orders_activity  orders_histogram  price_history  price_overview

Solution:

• Ensure Hridaya is running: python cerebro.py
• Wait for the first polling interval to complete
• Check for error messages in console output

Query returns 'no such column'

Error Message:

Error: no such column: market_name

Cause: Typo in column name.

Solution:

Check exact column names:

.schema price_overview

Common mistakes:

• market_name → market_hash_name ✓
• item_id → item_nameid ✓
• time vs timestamp (depends on table)

Refer to the Database Schema for correct column names.

Network Issues

Error: Connection timeout

Error Message:

aiohttp.ClientConnectorError: Cannot connect to host steamcommunity.com
OR
TimeoutError: 

Cause: Network connectivity issues or Steam is down.

Solution:

1. Check internet connection:

ping steamcommunity.com

2. Check Steam status: Visit steamstat.us or try accessing Steam Market in your browser.

3. Check firewall: Ensure your firewall allows outbound HTTPS connections.

4. Increase timeout (if on slow network): Edit cerebro.py:steamAPIclient.py:36:

self.session = aiohttp.ClientSession(
    timeout=aiohttp.ClientTimeout(total=60)  # Increased from 30
)

Error: SSL certificate verification failed

Error Message:

aiohttp.ClientConnectorCertificateError: SSL certificate verification failed

Cause: SSL certificate issues (often corporate proxies or outdated certificates).

Solution:

1. Update certificates:

# Ubuntu/Debian
sudo apt-get update && sudo apt-get install --reinstall ca-certificates

# macOS
brew install openssl

# Or update Python's certifi
pip install --upgrade certifi

2. If behind corporate proxy: You may need to configure proxy settings or install corporate root certificates.

Data Quality Issues

Prices seem incorrect

Symptoms:

• Prices are much higher/lower than expected
• Decimal points in wrong place

Cause: Currency/regional formatting differences.

Diagnosis:

1. Check currency setting:

TRACKING_ITEMS:
  - market_hash_name: "Item"
    currency: 1  # 1=USD, 3=EUR, etc.

2. Verify in database:

SELECT market_hash_name, currency, lowest_price
FROM price_overview
ORDER BY timestamp DESC LIMIT 5;

Solution:

Ensure your currency setting matches your expectations:

• 1 = USD ($)
• 2 = GBP (£)
• 3 = EUR (€)
• 5 = RUB (₽)
• 7 = BRL (R$)

Prices are stored as raw numbers. Apply proper currency formatting in your analysis.

Missing data points

Symptoms:

• Gaps in time series data
• Expected data points are missing

Possible Causes:

1. Hridaya was stopped: Check if there were periods when the system wasn’t running.

2. Rate limiting: If at capacity, some requests may be delayed or skipped.

3. Network issues: Temporary connectivity problems.

4. Steam API errors: Steam occasionally returns errors for specific items.

Solution:

• Keep Hridaya running continuously (use systemd/supervisor in production)
• Monitor logs for errors
• Ensure adequate rate limit capacity

Duplicate timestamps

Symptoms:

• Multiple rows with identical timestamps
• Unexpected duplicate data

Cause: This is normal behavior. Each polling interval creates a new snapshot.

Explanation:

Hridaya stores every snapshot it takes. If polling every 30 seconds, you’ll get ~2 rows per minute. This is intentional for high-fidelity tracking.

Solution:

If you want deduplicated data, use DISTINCT or aggregation:

-- Get one row per minute
SELECT strftime('%Y-%m-%d %H:%M', timestamp) AS minute,
       AVG(lowest_price) AS avg_price
FROM price_overview
WHERE market_hash_name = 'AK-47 | Redline (Field-Tested)'
GROUP BY minute;

Performance Issues

High CPU usage

Cause: Multiple concurrent async tasks and database writes.

Solution:

1. Reduce polling frequency:

TRACKING_ITEMS:
  - polling-interval-in-seconds: 60  # From 10

2. Reduce tracked items: Comment out low-priority items.

3. Optimize database:

sqlite3 market_data.db "VACUUM;"

Database file growing rapidly

Symptoms:

• market_data.db is very large (>1GB)
• Disk space running out

Cause: High-frequency polling generates large amounts of data.

Calculation:

# Rough estimate:
# 10 items × 6 req/min × 60 min × 24 hr × ~1KB per row = ~8.6 MB/day
# Over 30 days: ~260 MB

Solution:

1. Archive old data:

-- Export data older than 30 days
.mode csv
.output archive.csv
SELECT * FROM price_history WHERE time < datetime('now', '-30 days');
.output stdout

-- Delete old data
DELETE FROM price_history WHERE time < datetime('now', '-30 days');
VACUUM;

2. Reduce polling frequency for historical data:

TRACKING_ITEMS:
  - apiid: 'pricehistory'
    polling-interval-in-seconds: 3600  # Once per hour

Slow queries

Symptoms:

• Queries take >1 second
• Database feels sluggish

Solution:

1. Always filter by item name:

-- Fast (uses index)
WHERE market_hash_name = 'Item Name'

-- Slow (full table scan)
WHERE appid = 730

2. Use LIMIT for large results:

SELECT * FROM price_history
WHERE market_hash_name = 'Item'
ORDER BY time DESC
LIMIT 1000;  -- Don't fetch millions of rows

3. Optimize database:

sqlite3 market_data.db "ANALYZE; VACUUM;"

Getting Help

If you’re still experiencing issues:

1. Check logs

   Review console output for error messages:

   python cerebro.py 2>&1 | tee hridaya.log
   
2. Verify configuration

   Double-check config.yaml against examples in this documentation.

3. Search existing issues

   Check GitHub issues for similar problems and solutions.

4. Create an issue

   If you’ve found a bug, create a GitHub issue with:

   • Error message (full stack trace)
   • Your config.yaml (remove sensitive data)
   • Steps to reproduce
   • Hridaya version

Next Steps

Tracking Items

Learn how to configure item tracking

Querying Data

SQL query examples and tips