To enhance your DataSync integration for Snowflake, you can configure the Snowflake Bulk Load Meshlet to the configurations below.
To check out the general meshlet configurations, see General Meshlet Configurations for DataSync.
List of Meshlet Configurations for Snowflake
Directive | Default Value | Description |
---|---|---|
maxFileSize | Required Configuration. This configuration specifies the max size for temporary files created as the meshlet pushes records to Snowflake. 10000 will be used if input is over 10000 to prevent possible performance and memory issues. A suggested value is 5000. perspectium: filesubscriber: maxFileSize: 5000 | |
maxStringSize | 500 | Specifies the maximum string size for the entire collection. The maximum value you can use is 1000, if input is greater than that value, the input will default to 1000. perspectium: filesubscriber: maxStringSize: 500 |
customFileName | $table-$randomid | Names file with format of table - random id. File names MUST be unique. perspectium: filesubscriber: customFileName: $table-$randomid |
fileDirectory | /files | Directory where the locally created files get made. (In respects to where application is running) perspectium: filesubscriber: fileDirectory: /files |
postInterval | 2 | Minutes to check dead periods. Check every x minutes to compare if the in memory collection is the same as the last x minutes. If so, write records to file and push to Snowflake perspectium: snowflake: postInterval: 2 |
deleteFiles | true | Indicates whether you want to keep or delete locally created CSV files. Will not have a noticeable performance hit. perspectium: filesubscriber: deleteFiles: true |
fileCleanerInterval | 1 | How often (in minutes) the file cleaner job will run to clean up local files created by the meshlet. This job is similar to the ServiceNow Data Cleaner Scheduled Job. For example, value of 4 will run the job every four minutes. The default value is 1. perspectium: filesubscriber: fileCleanerInterval: 4 |
fileMaxAge | 1 | Any csv file in the filesDirectory older than fileMaxAge in minutes, will be automatically deleted. The default value is 1. perspectium: filesubscriber: fileMaxAge: 1 |
legacyDbViewId | false | Derivative to use the legacy version for database view tables of concatenating GUID values into a sys_id field. If false, meshlet will use the pre-constructed encoded sys_id created by ServiceNow. perspectium: filesubscriber: legacyDbViewId: false NOTE: If using the Share only selected fields option on your bulk share and you don't want the encoded sys_id to be used, you will want to set all three of the legacyDb properties to be true: perspectium: filesubscriber: legacyDbViewId: true legacyDbViewAlphaOrder: true legacyDbViewSkipNull: true
|
legacyDbViewAlphaOrder | false | By default, the Snowflake meshlet will concatenate GUID values for database view tables into a sys_id field in the order it reads other *_sys_id fields (such as the fc_sys_id and jc_sys_id fields in the flow_context_chunk database view table) which is not guaranteed to be in alphabetical order. With this configuration set to true, the *_sys_id fields will be read and concatenated in alphabetical order i.e. fc_sys_id + jc_sys_id for flow_context_chunk. This feature requires legacyDbViewId be set to true. If legacyDbViewId is not configured to true, this configuration will not be enabled. perspectium: filesubscriber: legacyDbViewAlphaOrder: false If this configuration is not specified, the meshlet will concatenate GUID values in the order it reads other *_sys_id fields. |
legacyDbViewSkipNull | false | By default, the Snowflake meshlet will concatenate all GUID values for database view tables into a sys_id field including GUID values that are null (i.e. have no value), matching how ServiceNow appends sys_id values. For example if the fc_sys_id field has a value of 12345 and jc_sys_id has a value of null, the sys_id field will be 12345null (in ServiceNow this would show as __ENC__ABCEDEER=-NULL as ServiceNow does encoding on *_sys_id values). With this configuration set to true, *_sys_id fields with values of null will be skipped and not added to the concatenated value. In the above example, the sys_id field would be 12345. This feature requires legacyDbViewId be set to true. If legacyDbViewId is not configured to true, this configuration will not be enabled. perspectium: filesubscriber: legacyDbViewSkipNull: false If this configuration is not specified, the meshlet will concatenate all GUID values including null values. |
receiptOutboundQueue | Queue receipts are returned to, typically default subscribed to on a ServiceNow instance. perspectium: message: receiptOutboundQueue: psp.out.servicenow.dev1234 | |
useConnectionProperties | false | When enabled, the user, password, and role properties on the yml are no longer being used to create the database connection. perspectium: snowflake: useConnectionProperties: true This will allow you to add these or other parameters to your connectionUrl to leverage other forms of authentication. For example: connectionUrl: jdbc:snowflake://hga12424.us-east-1.snowflakecomputing.com/?USER=xxxx&PASSWORD=xxxxx&ROLE=xxxx |
updateTimeOnInsert | true | In conjunction with Add custom fields in Snowflake, if you want the IO update column to be NULL on record inserts, set the value to false. perspectium: snowflake: updateTimeOnInsert: true By default, updateTimeOnInsert is true, and IO update columns are populated on record inserts. |
useLocalSchema | false | Enabling this directive will force the snowflake meshlet to read schemas via files exclusively from the local file storage and disable all calls to ServiceNow. Any new tables shared will require table schemas to be added to the local file storage. See Load Schema Files on how to get started. perspectium: auth: useLocalSchema: true |
loadSchemasOnStart | true | Supported in Krypton 8.0.3 Release and Newer In conjunction with the useLocalSchema configuration above, this configuration controls whether the schemas stored in the local file storage are loaded when the meshlet is started to immediately create the tables in the Snowflake database. By default, this value is true meaning that when the meshlet starts up, it will load all the schemas stored in local file storage and then use them to create the tables in Snowflake when the meshlet is starting up. This will ensure faster performance for when records are received by the meshlet to insert/update into the database table since it doesn't have to create tables at that time. Set this value to false if you want the meshlet to create the table only when it receives a record for the table. perspectium: auth: loadSchemasOnStart: true |
quotedColumns | false | Enabling this directive will add quotations around column names in the database. perspectium: snowflake: quotedColumns: true |
purgeTable | Enable this directive to truncate or delete all the records for a selected table in your database before bulk sharing. perspectium: snowflake: purgeTable: true | |
purgeTableOption | When the directive above is enabled (purgeTable), you can choose which option will be issued. The available options are:
perspectium: snowflake: purgeTable: true purgeTableOption: truncate | |
truncateColumns | true | By default, the Snowflake meshlet will create fields without specifying a length (the one exception is timestamps) so that these fields are created with the maximum size i.e. VARCHAR(16777216) to avoid issues with the content being too long for the column. If columns are created with a size where content being inserted is longer than the column's maximum length, by default the meshlet will truncate the text to fit into the column using Snowflake's TRUNCATECOLUMNS option. To change it so that the meshlet does not truncate the column (and instead will throw an error), change the configuration to false: perspectium: snowflake: truncateColumns: false If this configuration is not specified, the meshlet will treat this configuration as true and truncate content to fit into a column's length. |
extendColumnsMaxLength | true | With this configuration, if the meshlet sees content that is greater than the current column's length, it will extend the column to the max size supported for Snowflake i.e. VARCHAR(16777216) for string (VARCHAR) fields. Using this configuration along with the above truncateColumns will allow columns to be extended to their maximum size and then only truncate when the content is longer than the maximum size for the column so that there is no loss of records. For example, with both the extendColumnsMaxLength and truncateColumns configurations set to true, you will see: 1) The Snowflake meshlet receives a record of 5,000 characters for a table column that is defined as VARCHAR(4000) 2) The column is extended to VARCHAR(16777216) 3) The record is saved successfully 4) A record with 18,000,000 characters is received and is truncated to 16,777,216 characters to fit into this column and save the record into Snowflake If you prefer that columns not be extended, change the configuration to false. Note that by default all columns are created with their lengths as maximum size as mentioned in the truncateColumns configuration. So this is only applicable if you've configured it or created tables with smaller sizes. perspectium: snowflake: extendColumnsMaxLength: false |
tables | To prevent querying the entire database, comma separated list of tables to be shared. perspectium: snowflake: tables: incident | |
timestampColumn | Column name timestamp to be used for query. perspectium: snowflake: timestampColumn: yy-MM-dd HH:mm:ss | |
changesOnly | false | True/False. When true, only the columns changed will appear in the outbound message. When false, all columns will appear in the outbound message. perspectium: snowflake: changesOnly: true |
pollingInterval | How often in milliseconds to query against the Snowflake database. Suggested value: 30000 perspectium: meshlet: scheduler: pollingInterval: 30000 | |
outboundQueue | Queue messages will be shared out to. perspectium: message: outboundQueue: psp.out.servicenow.dev1234 | |
defaultEncryptionMode | 3 | Desired encryption method to be used for outbound messages. 1 - tripleDES, 2 - AES 128, 3 - base64, 4 - AES 256 perspectium: message: defaultEncryptionMode: 3 |
encryptionKey | Key to be used for encrypting messages. perspectium: message: encryptionKey: Your encryption key here | |
networkTimeout | 0 (no timeout is set) | Supported in Krypton 8.0.3 Release and Newer To specify in milliseconds (i.e. 1000 milliseconds = 1 second) how long to wait for a response from the network when interacting with Snowflake before returning an error. A value of 0 means we won't specify an explicit timeout duration. perspectium: snowflake: networkTimeout: 5000 NOTE: This configuration is used for configuring the JDBC driver the meshlet uses to connect to Snowflake. The configuration is designed to be a minimum value to pass to Snowflake. For example, if the value specified here is 5000, Snowflake will error after 5000 milliseconds (5 seconds) but may take longer. For example: 2023-07-29 13:10:19,021 ERROR [perspectium-/-psp.out.replicator.testqueue1] net.snowflake.client.log.JDK14Logger: Stop retrying since elapsed time due to network issues has reached timeout. Elapsed: 68,482(ms), timeout: 5,000(ms) See Snowflake documentation for more details. |
queryTimeout | 0 (no timeout is set) | Supported in Krypton 8.0.3 Release and Newer To specify in milliseconds (i.e. 1000 milliseconds = 1 second) how long to wait for a query to complete before returning an error. A value of 0 means we won't specify an explicit timeout duration. perspectium: snowflake: queryTimeout: 5000 NOTE: Like the networkTimeout configuration, this configuration is used for configuring the JDBC driver the meshlet uses to connect to Snowflake. The configuration is designed to be a minimum value to pass to Snowflake. For example, if the value specified here is 5000, Snowflake will error after a query has gone more than 5000 milliseconds (5 seconds) but may take longer. See Snowflake documentation for more details. |
maxHttpRetries | 7 | Supported in Krypton 8.0.3 Release and Newer To specify the maximum number of times to retry failed HTTP requests before returning an error. The default value is 7. perspectium: snowflake: maxHttpRetries: 7 NOTE: Like the networkTimeout configuration, this configuration is used for configuring the JDBC driver the meshlet uses to connect to Snowflake. Being that the connection to Snowflake is a HTTP connection, this configuration is designed to be a minimum value to pass to Snowflake. For example, if the value specified here is 5, Snowflake will return an error after a request has been attempted a minimum of 5 times but it may make additional attempts (based on a combination of other configurations like networkTimeout and queryTimeout) before it errors. See Snowflake documentation for more details. |
batchSize | 1000 | Supported in Krypton 8.0.5 Release and Newer To specify the batch size of information in messages sent from the meshlet back to ServiceNow. For example, when the meshlet is doing Table Compare and compiling a list of the sys_ids it has to see if any records should be deleted because they no longer exist in ServiceNow, the meshlet will create the list of sys_ids to match this configuration. For example, if the value is specified here is 1000, it will create a batch of 1000 sys_ids in the message it sends. By default, this value is 1000 if this configuration is not specified. This configuration only supports whole number (integer) values. perspectium: snowflake: batchSize: 1000 |
The following are the required configurations for a Snowflake to ServiceNow integration: | ||
softDelete | false | This feature allows you to add columns to a database table that gets updated when the meshlet receives a message indicating a record is to be deleted. The meshlet will update said columns appropriately and will not delete the record in Snowflake. For more information, see Soft Deletes. perspectium: snowflake: softDelete: true |
Debugging Logs
To enable debugging logs for the Snowflake Meshlet, you can configure your connection url in the application.yml with one the following:
jdbc:snowflake://xxxxxx.snowflakecomputing.com/?TRACING=CONFIG
jdbc:snowflake://xxxxxx.snowflakecomputing.com/?TRACING=FINEST
jdbc:snowflake://xxxxxx.snowflakecomputing.com/?TRACING=ALL
Where ALL will provide the most detailed level of additional logging. See tracing=<string> for more information on the different logging levels available.
A tmp log file will be created. See How to generate log file on Snowflake connectors.