Below are some common issues you may come across while using DataSync for Snowflake.  Contact support@perspectium.com if your issue is not listed below or you have any other questions.


Connection Issues

When starting the meshlet, you may encounter connection issues showing the following error messages. See below for some areas to check to troubleshoot these issues.

Verify if the ACCOUNT connection parameter was specified. Depending on your version of Snowflake, this parameter can be either ACCOUNT or ACCOUNT_NAME and should be specified in the connectionUrl configuration.

connectionUrl: jdbc:snowflake://<account>.<region>.snowflakecomputing.com/?ACCOUNT=<account>
connectionUrl: jdbc:snowflake://<account>.<region>.snowflakecomputing.com/?ACCOUNT_NAME=<account>

Verify the database is valid and one the user has access to.

Verify the warehouse is valid and one the user has access to.

Verify the schema is valid and one the user has access to.

Verify the role is valid and one that has access to the database, warehouse and schema specified.


General Issues

When the meshlet is running, you may encounter issues showing the following error messages. See below for some areas to check to troubleshoot these issues.

Verify databaseConfig.json file exist in ./static/config/databaseConfig.json and has required fields.

Verify the ServiceNow credentials are correct and the user specified has the proper role.

Verify the encryption key in the yaml configuration file (.yml) matches the key from the sharing source.

The meshlet uses SELECT statements to determine if a table you are trying to replicate exists in the Snowflake database as well as to get the table's columns to build the SQL statements to copy data into the database properly. However, if you see in Snowflake's history that there are SELECT * FROM <table_name> (i.e. SELECT * FROM incident) taking a long time to return, you can optimize it by updating the databaseConfig.json as follows:

Change

"selectAll":"SELECT * FROM %s",

To

"selectAll":"SELECT * FROM %s LIMIT 1",

Other Questions/Issues

Other questions and issues related you may encounter related to the Snowflake meshlet.

In the data folder where the meshlet is installed there is a schemaColumns.dat file that holds the schemas of the tables it has created.

The meshlet creates columns with data types with the full length to prevent any data integrity issues. That is, a VARCHAR data type (what is generally created for a string field) is created as VARCHAR without a specified column length as you would in other databases. This is the equivalent of creating it as the maximum VARCHAR(16777216) since there is no performance impact whether you use the full length declaration of a column data type or a smaller size. As well, this means that when column lengths are extended in ServiceNow, this does not need to be changed in Snowflake.

To ensure there isn't any data loss, the meshlet will not alter columns to have different data types. Data will continued to be saved in the original column data type as when the column was first created.


Can't find what you're looking for?  

See the FAQ or browse the Perspectium Community Forum.