ServiceNow

Letzte Aktualisierung: 20.10.2026

ServiceNow is a cloud platform for IT service management (ITSM), IT operations management, HR service delivery, security operations, and customer service workflows. You can use ServiceNow as a data source when landing or replicating data.

ServiceNow integration replicates data using the ServiceNow Table API at the /api/now/table endpoint.

Preparing for authentication

To access your data, you need to authenticate the connection with your account credentials.

Grant the integration user read access through ACLs to the tables you intend to replicate and to the sys_db_object and sys_dictionary system tables, which the connector uses to discover available tables and their column definitions.

You need administrator permissions in your ServiceNow account to create the integration user.

To create a ServiceNow integration user:

1. Log into your ServiceNow instance, for example https://instance.service-now.com.
2. Navigate to All > User Administration > Users and click New.
3. Enter a User ID and set a password.
4. Clear Password needs reset to avoid requiring a password change at the user's next login.
5. Click Save.
6. Assign a role that grants read access to the tables you intend to replicate. A read-only role scoped to the required tables is preferred over a broadly privileged role such as admin.

Creating the connection

Weitere Informationen finden Sie unter Herstellung einer Verbindung zu SaaS-Anwendungen.

1. Füllen Sie die erforderlichen Verbindungseigenschaften aus.

2. Geben Sie einen Namen für die Verbindung in Verbindungsname ein.

3. Wählen Sie Verbindungsmetadaten öffnen aus, um nach der Erstellung Metadaten für die Verbindung zu definieren.

4. Klicken Sie auf Erstellen.

Schema

The set of available tables is instance-specific and determined at discovery time. The connector enumerates all tables from sys_db_object, resolves each table's inherited columns by walking its super_class chain, and reads column data types from sys_dictionary. You then select which discovered tables and fields to replicate. Commonly replicated tables include incident, problem, change_request, task, sys_user, and cmdb_ci, but the exact catalog depends on your instance and the integration user's permissions.

Every table uses sys_id as its primary key.

The replication method used for each table depends on whether the table has a sys_updated_on column:

• Incremental replication is used when a table has a sys_updated_on column. New and changed records are picked up on each sync using sys_updated_on as the replication key.
• Full-table replication is used when a table has no sys_updated_on column.

Tables excluded by default

To protect sync performance, the following high-volume system and log tables are excluded from discovery by default: sys_audit, sys_audit_delete, syslog, syslog_transaction, sys_email_log, sys_history_line, sys_history_set, ha_log, sys_cache_flush, sys_cluster_state. These exclusions cannot be changed from the connection settings in this release.

Limitations and considerations

• HTTP Basic authentication only. This connector authenticates with a username and password. OAuth 2.0 is not supported in this release.
• Discovery scope follows the account's permissions. Tables the account cannot read are skipped during discovery and reported in the run log. If the account cannot read any table, discovery fails with a permissions error.
• Large catalogs. A broadly privileged account can surface thousands of tables. Scope the integration user's permissions and select only the tables you need.
• Incremental replication re-reads the boundary record. The most recent record from the previous sync may be re-read on the next sync. Records upsert on sys_id, so duplicate reads are safe.
• Full-table replication re-reads the entire table on every sync. Tables without sys_updated_on are re-read in full on each sync. Sync time does not decrease as data accumulates.
• Start Date applies to the initial sync only. Once a table has a stored bookmark, later syncs resume from it. Changing Start Date does not re-replicate historical data.
• Empty records are skipped. Empty ({}) rows returned by ServiceNow are skipped and counted in the run log.
• Retries. Rate-limit responses (HTTP 429) and transient connection errors are retried automatically with exponential backoff (up to 5 attempts), honoring ServiceNow's Retry-After header.
• Pagination. The connector uses keyset pagination based on sys_id with a page size of 1,000 records and a default request timeout of 300 seconds.