Accessing the ServiceNow® data in Snowflake¶
The Snowflake Connector for ServiceNow® V2 is subject to the Snowflake Connector Terms.
This topic describes how to access ServiceNow® data from your Snowflake account.
For each table in ServiceNow® that is configured for synchronization, the connector creates the following tables and views:
A table with the same name that contains the data in raw form, where each record is contained in a single VARIANT column.
A table named
table_name__event_log
that contains the history of changes made to ServiceNow® records.A view named
table_name__view
that contains the data in flattened form, where the view contains a column for each column in the original table and a row for each record that is present in the original table.A view named
table_name__view_with_deleted
that contains the same data astable_name__view
as well as rows for records that have been deleted in ServiceNow®.
Note
After starting the connector, it may take some time for the views to be created.
View creation depends on data in the ServiceNow®
sys_db_object
, sys_dictionary
and sys_glide_object
tables.
The connector loads metadata from these tables after a business table is enabled for synchronization.
When the metadata tables are ingested, a background task will create flattened views of the enabled tables.
The task is run as often as the schedule of the most frequent table ingestion. After the metadata tables are synced,
the task also captures any table schema changes and updates the already created views accordingly (only the views
with the suffixes __view
and __view_with_deleted
, but not with __view_with_display_values
).
As it’s not an immediate process, status of view creation process is available under the CONFIGURED_TABLES
view.
If the view creation takes too long, the CONNECTOR_ERRORS
view can also be checked for any related errors.
Warning
If you plan to set ROW ACCESS POLICIES on the tables
and views created by the connector, make sure they do not block access to the role with the same name as the connector application.
For example, if your connector application instance is called MY_CONNECTOR_SERVICENOW
, then your policies cannot
block a role named MY_CONNECTOR_SERVICENOW
. Otherwise, the policies will interfere with the data ingestion process.
The following sections explain how to grant the privileges to access this data and how to access the data from these tables and views.
Granting Privileges for Accessing the ServiceNow® Data in Snowflake¶
After the Snowflake Connector for ServiceNow®V2 synchronizes the data with Snowflake, to access the ServiceNow® data a role needs:
USAGE privilege on the database and schema that contain the ServiceNow® data in Snowflake, and
Snowflake recommends creating a dedicated role with these privileges that can be granted to users who need access to the ingested ServiceNow® data. If the connector has been installed with Snowsight then the role provided during Configure step already has the necessary privileges.
For example, if you configured the connector application called my_connector_servicenow
to store the ServiceNow®
data in the dest_db
database and dest_schema
schema, you can create a role named
servicenow_data_reader_role
and grant the privileges for accessing the data to that role.
The following example shows how to grant these privileges:
CREATE ROLE servicenow_data_reader_role; GRANT USAGE ON DATABASE dest_db TO ROLE servicenow_data_reader_role; GRANT USAGE ON SCHEMA dest_db.dest_schema TO ROLE servicenow_data_reader_role; GRANT APPLICATION ROLE my_connector_servicenow.DATA_READER to role servicenow_data_reader_role;
Note
Do not run
GRANT OWNERSHIP ON FUTURE TABLES IN SCHEMA
on the schema that contains the ServiceNow® data in Snowflake. Also, do not change the ownership of the tables that are already created by the connector. Changing the ownership prevents the connector from ingesting the data to the table.Do not change the ownership of the views in the schema that contains the ServiceNow® data in Snowflake. Changing the ownership prevents the connector from updating the views when changes occur in the ServiceNow® table schema.
Accessing the Raw Data¶
For each ServiceNow® table that you synchronize, the Snowflake Connector for ServiceNow®V2 creates a new table with the same name in the database and schema for the ServiceNow® data in Snowflake.
For example, if you configured the connector to store the ServiceNow® data in the dest_db
database and
dest_schema
schema, and if you configured the connector to synchronize the incident
table in
ServiceNow®, the connector creates the table named dest_db.dest_schema.incident
.
This table contains raw data ingested from ServiceNow®. This table contains the following columns:
Column |
Data Type |
Description |
---|---|---|
|
VARCHAR |
The value of the |
|
VARIANT |
The data for the record in raw form. |
|
BOOLEAN |
Specifies whether or not the record was deleted in ServiceNow®. |
|
TIMESTAMP_NTZ |
The last time the record was updated in Snowflake. Note that the displayed timestamp is provided in the UTC timezone with no offset, which may differ from the timezone of dates displayed in the ServiceNow instance. |
The following is an example of the output for a SELECT statement that retrieves the data for the
dest_db.dest_schema.incident
table:
SELECT * FROM DEST_DB.DEST_SCHEMA.INCIDENT LIMIT 5; +----------------------------------+-------------------------+-------------+--------------------------+ | SYS_ID | RAW:ACTIVE | IS_DELETED | LAST_UPDATE_DATE | +----------------------------------+-------------------------+-------------+--------------------------+ | caa04d36db8ba0106e9643c81396197b | {"active": "true", ...} | FALSE | 2021-08-24 12:59:23.932 | | cea045be1b03e010eac562c4bd4bcbb2 | {"active": "true", ...} | FALSE | 2021-08-24 12:59:23.932 | | caa0c9bedb8be010f9f19c41ba961934 | {"active": "true", ...} | FALSE | 2021-08-24 12:59:23.932 | | caa0c9bedb8be010f9f19c41ba961969 | {"active": "true", ...} | FALSE | 2021-08-24 12:59:23.932 | | b9a0c53adb436410d6fa2b691396190a | {"active": "true", ...} | FALSE | 2021-08-24 12:59:23.932 | +----------------------------------+-------------------------+-------------+--------------------------+
Accessing the Flattened Data¶
For each table that contains data, the connector creates two flattened views over the raw data.
The names of the views are the names of the table with the suffixes __view
and
__view_with_deleted
. For example, for the ServiceNow® table named incident
, the connector creates
the following views:
dest_db.dest_schema.incident__view
dest_db.dest_schema.incident__view_with_deleted
The view with the __view
suffix contains the records that are in the ServiceNow® table. The view with
the __view_with_deleted
suffix includes these same records as well as the records that were deleted
from the ServiceNow® table.
Note the following:
The names of the columns in these views are in uppercase. You cannot use lowercase names to access these columns.
Columns with time and timestamps are always saved using the UTC timezone, regardless of the timezone set in the ServiceNow instance. As a result, depending on the ServiceNow instance configuration, their displayed values may differ from the values displayed in the ServiceNow instance. The difference relates only to displayed values, timestamps both in ServiceNow and Snowflake are referring to the same point in time.
There are no views for empty tables. After data appears in the table in ServiceNow®, the view is created.
Although the connector handles changes to the schema, the connector does not reload the data.
As a result, in the case of schema changes, records from the old schema are not updated.
The following is an example of the output for a SELECT statement that retrieves the data from the
dest_db.dest_schema.incident_view
view. In this example, the incident
table in ServiceNow® has columns
named ACTIVE
, APPROVAL
, CATEGORY
, and ESCALATION
.
SELECT ACTIVE, APPROVAL, CATEGORY, ESCALATION FROM DEST_DB.DEST_SCHEMA.INCIDENT__VIEW LIMIT 5; +--------+----------------+------------------+------------+ | ACTIVE | APPROVAL | CATEGORY | ESCALATION | +--------+----------------+------------------+------------+ | TRUE | not requested | software | 0 | | TRUE | not requested | Cloud Management | 0 | | TRUE | not requested | software | 0 | | TRUE | not requested | network | 0 | | TRUE | not requested | database | 0 | +--------+----------------+------------------+------------+
Viewing the Event Logs for a Table¶
The Snowflake Connector for ServiceNow®V2 can track the changes made to records in ServiceNow®. This tracking information is stored in tables called event logs.
For every ServiceNow® table enabled for synchronization, the connector creates an event log table within
Snowflake named <destination_db>.<destination_schema>.<table_name>__event_log
.
Each event log table has the following columns:
Column |
Data Type |
Description |
---|---|---|
|
VARCHAR |
The value of the |
|
VARCHAR |
The date the record was last updated in ServiceNow®.
If there is no |
|
TIMESTAMP_NTZ |
The date the event was inserted in the event log. Note that the displayed timestamp is provided in the UTC timezone with no offset, which may differ from the timezone of the dates displayed in the ServiceNow instance. |
|
VARIANT |
The current data of the record event. For DELETE events, this is the data of the record at the time of deletion. |
|
VARCHAR |
Specifies if the record was inserted, updated, or deleted from ServiceNow®. |
The event log reflects the history of data changes in the corresponding ServiceNow® table. For example, if a
new record is inserted into the u_ip_port
table in ServiceNow®, a record with event_type
set to
INSERT
event type is added to the dest_db.dest_schema.u_ip_port__event_log
table in Snowflake.
Similarly, if a record is updated or deleted in a table in ServiceNow®, a record with event_type
set to
UPDATE
or DELETE
is added to the dest_db.dest_schema.u_ip_port__event_log
table.
The tables in Snowflake that contain the raw data (dest_db.dest_schema.table_name
) are derived
from the corresponding event log tables (dest_db.dest_schema.table_name__event_log
). For example:
If a record for an
INSERT
event is added totable_name__event_log
, the connector adds a corresponding record to thetable_name
table.If an
UPDATE
event for the givensys_id
is added to the event log table, the connector updates the corresponding record with thesys_id
in thetable_name
table with new data.If a
DELETE
event occurs, theis_deleted
flag of the corresponding record intable_name
is set totrue
.
Getting the Display Value of a Reference Field¶
In ServiceNow® tables, some fields are reference fields, which contain references to records in other tables.
In the example below, the field opened_by
in the incident
table is a reference field that
contains a reference to the record with the sys_id
<sys_id>
in another
table (sys_user
):
{ "link": "https://myinstance.service-now.com/api/now/table/sys_user/<sys_id>", "value": "<sys_id>" }
To show the reference fields in the table, call the SHOW_REFERENCES_OF_TABLE
stored procedure with the following
argument:
CALL SHOW_REFERENCES_OF_TABLE('<table_name>');
Where:
table_name
Specifies the name of the table you want to show the reference fields for.
This stored procedure inspects the schema of the table and returns a JSON list of objects containing the following properties:
Property |
Description |
---|---|
|
Name of the reference field. |
|
Name of the field that the reference points to. |
|
Name of the referenced table. |
Enabling Data Synchronization for Referred Tables¶
If a table contains references to other tables, you can enable data synchronization of the referred tables.
To synchronize data for referred tables, call the ENABLE_REFERENCED_TABLES
stored procedure with the following argument:
CALL ENABLE_REFERENCED_TABLES('<table_name>');
Where:
table_name
Specifies the name of the table (with the table reference fields) for which you want to enable data synchronization.
Creating a View Containing Reference Fields¶
If the table containing the reference fields and the tables referenced by the those fields have been processed, you can create a view that replaces the references with display values.
To create this view, call the CREATE_VIEW_WITH_DISPLAY_VALUES
stored procedure.
CALL CREATE_VIEW_WITH_DISPLAY_VALUES('<table_name>');
Where:
table_name
Specifies the name of the table containing the table reference fields for which you want to create a view with display value.
Note
Only reference fields with the sys_id
as reference key are supported.
Important
This procedure is only run manually, therefore each time the table schema is changed the view must be recreated manually to reflect the schema change.
After the view is created successfully, the stored procedure returns the name of the newly created view.
The view name is the table name with the __view_with_references
suffix added.
For example, for a ServiceNow® table named incident
, the stored procedure creates the view incident__view_with_references
.
Reference fields are replaced with display values and a new metadata column is added for each reference field.
The display value column has the same name as the reference column being replaced and may be null when if the display value is null or
the reference is not resolved. The metadata column name is the name of the reference column with the __metadata
suffix.
For example, for a reference column named user
, the procedure creates a column named user__metadata
.
The content of this column is a JSON object with a field named reference_field
with the following properties:
Property |
Description |
---|---|
|
|
|
Name of the referenced table. If the reference is not resolved this property is null. |
|
ServiceNow® link to the referred row.
If the reference column or reference column field |
|
Display value. If the reference is not resolved this property is null. |
|
|
|
Reason the reference failed to resolve. For example |
The following example shows how a pair of display value and metadata columns in a view created by the stored procedure
CREATE_VIEW_WITH_DISPLAY_VALUES
looks like.
The example table incident
has opened_by
column which references (by sys_id
as reference key) to the sys_user
table.
The incident__view_with_references
view created by the stored procedure resolves the reference, so the displayed values can be obtained with a simple SELECT
.
SELECT OPENED_BY, OPENED_BY__METADATA
FROM DEST_DB.DEST_SCHEMA.INCIDENT__VIEW_WITH_REFERENCES;
This command displays information in the following format:
+-----------+------------------------------------+
| OPENED_BY | OPENED_BY__METADATA |
+-----------+------------------------------------+
| "JOHN" | { |
| | "reference_field": { |
| | "display_value": "JOHN", |
| | "key": "b177...", |
| | "link": "https://...", |
| | "reference_table": "sys_user", |
| | "resolved": true |
| | } |
| | } |
+-----------+------------------------------------+