Skip to main content
Having trouble with Flipside data shares? This guide covers common issues and their solutions.

Access and permissions

Share not appearing in Private Sharing

Symptom: You can’t find the Flipside share in your Snowflake account. Possible causes:
  1. Premium share not yet approved
    • Premium shares require manual approval
    • Check your email for approval notification
    • Allow 1-2 business days for approval
  2. Wrong Snowflake account/region
    • Shares are only visible in accounts located in AWS US-EAST-1
    • Check your current region:
    SELECT CURRENT_REGION();
    -- Should return: AWS_US_EAST_1
    
  3. Share not granted yet
    • For free shares: Visit the Marketplace and click “Get” again
    • For premium shares: Verify you submitted the correct data sharing identifier
Solution:
1

Verify your region

SELECT CURRENT_REGION();
If not AWS_US_EAST_1, you need to create a sub-account. See Creating Sub-Account.
2

Check Private Sharing

Navigate to Data → Private Sharing in Snowflake UI to see if the share appears.
3

Contact support if needed

Email [email protected] with:
  • Your data sharing account identifier
  • The specific share you’re trying to access
  • Screenshots of the issue

Permission denied when creating database

Symptom: Error when trying to create a database from the share.
SQL access control error: Insufficient privileges to operate on database
Cause: Your role lacks CREATE DATABASE privilege. Solution: Ask your Snowflake account administrator to grant you the necessary privilege:
-- As ACCOUNTADMIN
GRANT CREATE DATABASE ON ACCOUNT TO ROLE your_role;

-- Verify the grant
SHOW GRANTS TO ROLE your_role;
If you are the account owner, switch to ACCOUNTADMIN role to create the database.

Can’t query tables after mounting

Symptom: Database exists but queries fail with access errors.
Object 'ETHEREUM_CORE.ETHEREUM.CORE.FACT_TRANSACTIONS' does not exist
Possible causes:
  1. Incorrect database/schema/table name
  2. Missing USAGE grants
  3. Database not fully mounted
Solution:
SHOW DATABASES LIKE '%ETHEREUM%';

Other users can’t access the share

Symptom: You can query the share, but colleagues in your account cannot. Cause: Shares don’t automatically grant access to all roles in your account. Solution: Grant access to other roles:
-- Grant to a specific role
GRANT USAGE ON DATABASE ETHEREUM_CORE TO ROLE analyst_role;
GRANT USAGE ON ALL SCHEMAS IN DATABASE ETHEREUM_CORE TO ROLE analyst_role;
GRANT SELECT ON ALL TABLES IN SCHEMA ETHEREUM_CORE.ethereum TO ROLE analyst_role;

-- Repeat for other schemas (if Premium)
GRANT SELECT ON ALL TABLES IN SCHEMA ETHEREUM_CORE.defi TO ROLE analyst_role;
GRANT SELECT ON ALL TABLES IN SCHEMA ETHEREUM_CORE.nft TO ROLE analyst_role;

Query performance issues

Queries running very slowly

Symptom: Queries take minutes or time out. Cause: Likely missing time filter on block_timestamp. Solution: Always filter by block_timestamp:
SELECT *
FROM ethereum.core.fact_transactions
WHERE to_address = '0x...'
LIMIT 1000;
-- This scans BILLIONS of rows
See Query Optimization for more best practices.

”Query produced no results”

Symptom: Query runs successfully but returns zero rows when you expect data. Possible causes:
  1. Case sensitivity issues with addresses
    -- ❌ May not match
    WHERE from_address = '0xABC...'
    
    -- ✅ Always use LOWER()
    WHERE from_address = LOWER('0xABC...')
    
  2. Time range has no data
    -- Check data availability first
    SELECT
        MIN(block_timestamp) AS earliest_data,
        MAX(block_timestamp) AS latest_data
    FROM ethereum.core.fact_transactions;
    
  3. Wrong schema or table
    -- Free shares don't include DeFi schema
    SELECT * FROM ethereum.defi.ez_dex_swaps; -- Will fail on free shares
    
    -- Use Core schema instead
    SELECT * FROM ethereum.core.fact_transactions; -- Available on free shares
    

High compute costs

Symptom: Snowflake credits are being consumed quickly. Common causes:
  1. Selecting unnecessary columns
    -- ❌ Expensive: Scans all columns
    SELECT * FROM ethereum.core.fact_transactions
    WHERE block_timestamp >= CURRENT_DATE - 30;
    
    -- ✅ Better: Only select what you need
    SELECT
        block_timestamp,
        tx_hash,
        from_address,
        to_address
    FROM ethereum.core.fact_transactions
    WHERE block_timestamp >= CURRENT_DATE - 30;
    
  2. Open-ended time ranges
    -- ❌ Expensive: Scans years of data
    WHERE block_timestamp >= '2020-01-01'
    
    -- ✅ Better: Use narrow ranges
    WHERE block_timestamp >= CURRENT_DATE - 30
    
  3. No LIMIT on exploratory queries
    -- ❌ May return millions of rows
    SELECT * FROM ethereum.core.fact_transactions
    WHERE block_timestamp >= CURRENT_DATE - 7;
    
    -- ✅ Add LIMIT for exploration
    SELECT * FROM ethereum.core.fact_transactions
    WHERE block_timestamp >= CURRENT_DATE - 7
    LIMIT 1000;
    
Solution: Review Query Optimization and monitor query costs:
-- Check recent query costs
SELECT
    query_id,
    query_text,
    execution_time / 1000 AS seconds,
    bytes_scanned / POWER(1024, 3) AS gb_scanned,
    rows_produced
FROM TABLE(INFORMATION_SCHEMA.QUERY_HISTORY())
WHERE start_time >= DATEADD(hour, -24, CURRENT_TIMESTAMP())
  AND query_text ILIKE '%ethereum%'
ORDER BY bytes_scanned DESC
LIMIT 20;

Data issues

Data seems stale or outdated

Symptom: Latest data is older than expected. Solution: Check the actual freshness:
-- Check latest block
SELECT
    MAX(block_number) AS latest_block,
    MAX(block_timestamp) AS latest_timestamp,
    DATEDIFF('minute', MAX(block_timestamp), CURRENT_TIMESTAMP()) AS minutes_behind
FROM ethereum.core.fact_blocks;
Expected latency:
  • Core tables: 30 minutes to 1 hour behind chain head
  • DeFi/NFT/Price tables: 1-24 hours depending on complexity
See Data Freshness for detailed schedules. If data is significantly stale (greater than 24 hours behind), contact [email protected].

Missing or NULL values

Symptom: Expected columns contain NULL or are missing. Common scenarios:
  1. Transaction without value transfer
    -- eth_value is NULL when no ETH was transferred
    SELECT * FROM ethereum.core.fact_transactions
    WHERE eth_value IS NULL; -- This is expected for many transactions
    
  2. Failed transactions
    -- Failed transactions may have NULL or 0 for certain fields
    SELECT * FROM ethereum.core.fact_transactions
    WHERE success = FALSE;
    
  3. Contract creation transactions
    -- to_address is NULL for contract creation
    SELECT * FROM ethereum.core.fact_transactions
    WHERE to_address IS NULL; -- These are contract deployments
    
Solution: Check the table structure and understand the data model:
-- View column definitions
DESCRIBE TABLE ethereum.core.fact_transactions;

-- Check for NULL patterns
SELECT
    COUNT(*) AS total_rows,
    COUNT(eth_value) AS non_null_eth_value,
    COUNT(to_address) AS non_null_to_address
FROM ethereum.core.fact_transactions
WHERE block_timestamp >= CURRENT_DATE - 1;

Unexpected query results

Symptom: Results don’t match expectations or external sources. Debugging steps:
1

Verify time range

Ensure your time filter matches the period you’re analyzing.
SELECT
    MIN(block_timestamp) AS start_time,
    MAX(block_timestamp) AS end_time,
    COUNT(*) AS row_count
FROM ethereum.core.fact_transactions
WHERE block_timestamp >= CURRENT_DATE - 7;
2

Check filters

Verify address filters use LOWER() and are correct. sql SELECT LOWER('0xYourAddress') AS normalized_address;
3

Sample the data

Look at raw data to understand what’s being returned. sql SELECT * FROM ethereum.core.fact_transactions WHERE block_timestamp >= CURRENT_DATE - 1 LIMIT 10;
4

Compare with block explorers

Cross-reference a few transactions with Etherscan or similar explorers.

Connection and availability

”Share is unavailable” error

Symptom: Share was working but now shows as unavailable. Possible causes:
  1. Temporary Snowflake outage
  2. Share revoked (rare)
  3. Account issues
    • Verify your Snowflake account is active and in good standing
Solution: Try remounting the share:
-- Drop the database
DROP DATABASE IF EXISTS ETHEREUM_CORE;

-- Remount from Private Sharing in UI
-- Or contact support if issue persists

Tables or schemas missing after update

Symptom: Previously available tables are now missing. Cause: Share structure may have been updated by Flipside. Solution: Refresh your view of the share:
-- Check available schemas
USE DATABASE ETHEREUM_CORE;
SHOW SCHEMAS;

-- Check tables in a schema
SHOW TABLES IN SCHEMA ethereum.core;
If expected tables are truly missing, contact [email protected].

Edition and feature limitations

”Feature not supported” errors

Symptom: Certain queries or features fail with feature support errors. Cause: Some advanced Snowflake features require higher editions. Required edition: Enterprise or higher Verify your edition:
SELECT CURRENT_VERSION();
SHOW PARAMETERS LIKE 'EDITION' IN ACCOUNT;
If you’re on a lower edition, you’ll need to upgrade to Enterprise to use data shares.

Getting help

If you’ve tried the solutions above and still have issues:

Dedicated Slack channel

Get direct access to blockchain data experts (not just support tickets)Chat with us anytime. Direct access to our data engineers who can help with:
  • Query optimization and debugging
  • Data quality questions
  • Schema navigation
  • Custom use cases
Contact [email protected] to request Slack access

Email support

[email protected]Include:
  • Your data sharing account identifier
  • The specific share you’re using
  • Error messages or screenshots
  • Queries that reproduce the issue

Try FlipsideAI

FlipsideAIGet instant help with:
  • Query writing and debugging
  • Data exploration
  • Understanding results

What to include in support requests

  • Data sharing account identifier
  • Snowflake edition (Enterprise, Business Critical, etc.)
  • Region (should be AWS US-EAST-1)
  • What you’re trying to accomplish - What’s happening vs. what you expected - When the issue started - Whether it’s consistent or intermittent
  • Full error messages
  • Sample queries that reproduce the issue
  • Query IDs from Snowflake (if applicable)
  • Screenshots showing the problem

Next steps