Send Dynatrace notifications to ServiceNow

For ServiceNow, Dynatrace offers:

Incident integration
Configuration Management Database (CMDB) integration via ServiceNow-provided application
Event integration

To connect your Dynatrace monitoring environment to your ServiceNow instance, configuration is required both on the ServiceNow instance and in the Dynatrace web UI.

Once you configure Dynatrace Incident Integration, Dynatrace automatically creates an incident within your ServiceNow instance for every auto-discovered problem.

How it works

Dynatrace Incident Integration pushes incidents from your Dynatrace environment to your ServiceNow instance.

Configuration in ServiceNow

Go to Dynatrace Incident Integration page in the ServiceNow Store and select Get to install Dynatrace Incident Integration in your ServiceNow instance.
Follow the Guided Setup for an initial configuration of the application.
After the application is installed, go to the new Dynatrace Incident Integration menu on your ServiceNow instance to further configure (Setup, Settings) and explore (Problems, ServiceNow Incidents) the application.
Use the Transform Map module to transform the incoming Dynatrace data in the Problems import set table into an Incident table. The Problems import set table automatically transforms any incoming Dynatrace detected problem into an incident within the ServiceNow incident table.

Configuration in Dynatrace

After installing and configuring Dynatrace integration on the ServiceNow side, switch over to your Dynatrace environment.

In the Dynatrace menu, go to Settings and select Integration > Problem notifications.
Select Add notification.
Select ServiceNow from the available notification types.
Enter the requested information as described on the Problem notifications page.

The Available placeholders section of the configuration page lists placeholders you can use in the Description field to customize the text message associated with the problem notification. Placeholders are automatically replaced with problem-related information.
Turn on Send incidents into ServiceNow ITSM.
Select Send test notification to make sure your ServiceNow integration is working.
Save changes.

Workflow

After configuring the application on both sides, problem notifications are pushed from Dynatrace to your ServiceNow instance.
The Scripted REST APIs module is the entry point of the Dynatrace information into ServiceNow. It pushes the data into the Problems import set table.
The Problems import set table automatically transforms any incoming Dynatrace-detected problem into an incident within the ServiceNow incident table.
When a problem is closed in Dynatrace, the incident is marked as Resolved in ServiceNow.
Incident Integration flow

Troubleshooting

To allow Dynatrace to report discovered CIs in your CMDB, verify that the discovery data source SG-Dynatrace exists. To do this:

Go to System Definition > Choice Lists.
In the Value field, enter =SG-DYNATRACE and run a search.
If you get no results, select New.
Create a new entry with the following details.
- Table: Configuration Item [cmdb_ci]
- Element: discovery_source
- Label: SG-DYNATRACE
- Value: SG-DYNATRACE

If the hosts, process groups, services, and applications still don't show up in the CMDB, check the ServiceNow system log for error messages within the identification engine. It might be that a non-standard CMDB-type attribute has been defined as a mandatory field.

When Dynatrace-monitored servers aren't showing up in the CMDB, verify that:

Host synchronization has been enabled in the ServiceNow settings.
The Dynatrace URL and API token have been correctly entered in the environments table of ServiceNow.
The self-check has been run on the Support page in the ServiceGraph Connector for Dynatrace.
SG-DYNATRACE has been defined as a discovery source.
No errors have been logged in the system by the ServiceNow identification engine.
The identification log doesn't state anything about a customer-defined mandatory field that might be missing during the Dynatrace CMDB import.

Upgrading from a previous version of the integration application doesn't automatically delete the CIs from your CI table. If you want to start from scratch, you need to do it exclusively via a background script or filter CIs by Discovery Source: SG-DYNATRACE and delete it from the CI table manually. We don’t recommend that you do this in your ServiceNow PROD instance because historical incidents and events might be referencing it, and data would be lost.

This can happen because of the following reasons.

The Send events into ServiceNow ITOM button hasn't been enabled in Dynatrace.
The user enabling the integration doesn't have the x_dynat_ruxit_integration role.

Go to Service Graph Connector for Dynatrace > Settings and check if the synchronization of hosts, processes, services and applications is enabled.
In ServiceNow, enable Debug in Settings and check System Logs > Application logs and System Logs > Errors.

Add the following IP addresses and the ones listed in the error to your ServiceNow allow list.

52.26.178.159
52.0.97.215
54.154.238.149
52.74.69.201
54.66.240.72
54.147.40.70
3.231.89.157

For information about allowing specific IP addresses in ServiceNow, see Define IP Address access control.

Verify that the Dynatrace Incident Integration is correctly set up in Dynatrace.
Trigger a push from Dynatrace and check if the test entry is created under Dynatrace > Problems in ServiceNow.
If the problem is registered in the Problems import set table, but a corresponding incident entry isn't created, check for Problem to Incident Transformation Map and other errors.
Check for mandatory fields left empty during the transformation between problem entry and incident creation.

If there's no server with the same name, search for a server that has the Dynatrace Entity ID (for example, HOST-12121212) set within the CMDB CI attribute correlation_id.
Check if the Service Graph connector is downloaded, and the setup for CMDB Integration.
Check if the Dynatrace ID sync field in ServiceNow is correlation_id.

Check within Dynatrace if the toggle Use text format for problem details instead of HTML was enabled.
If problem details are sent in plaintext format, you need to also change the ServiceNow table field from type HTML to type text.
Find the table x_dynat_ruxit_problems and change the type of the problem details field x_dynat_ruxit_problems.details from HTML to type string.

Frequently asked questions

For your Dynatrace instance to communicate with ServiceNow, you need to provide ServiceNow with a list of Dynatrace IP addresses that will be allowed to send information to ServiceNow. Dynatrace servers are distributed throughout various regions. The best way to ensure that you get the right IP addresses for your region is to look up the IP addresses by running the nslookup command. You must first create your environment in Dynatrace and then execute nslookup.

Example:

1C:\>nslookup abc.live.dynatrace.com

Optionally, you can review the OneAgent download page to see the IP addresses for your region.

Dynatrace sends all detected problems into the ServiceNow incident table (incident).
All single events that are correlated with a Dynatrace-detected problem are sent to the ITOM event table (em_event).
The relationship between a CI and a Dynatrace-detected problem is sent into the CIs Affected table (task_ci).
If enabled, Dynatrace synchronizes all auto-discovered web services into the CMDB application services table (cmdb_ci_services_discovered).
All hosts and process groups running on those hosts are synchronized into the CMDB server table and derived Linux and Windows server tables (cmdb_ci_server), and into the process groups related tables in ServiceNow (cmdb_ci_appl).

Dynatrace introduces one scheduled job that pulls a subset of the topology information from Dynatrace Smartscape at 10-minute intervals. The SG-Dynatrace scheduled job pulls all web application information along with their relation to software services, all host and process group information, and all software services and relationships. The information pull is performed using the official Dynatrace REST API.

Incidents are imported through an import set table (x_dynat_ruxit_problems). You can reconfigure the default transformation map Problem to Incident Transformation Map to route the information to different tables or fields.

The Dynatrace ServiceNow application is built entirely on the public Dynatrace API and uses a subset of this information by default. We recommend using the ServiceNow Dynatrace application as a starting point and customizing information retrieval based on your requirements. You can also fetch detailed near real-time metrics for each of your Dynatrace monitored configuration items using the Dynatrace API.

Deduplication is done through ServiceNow CMDB Identification and Reconciliation. Custom CI identification rules are used to merge identical CIs based on a given attribute (for example, host names).

Multiple Dynatrace environments can be configured within the Dynatrace environments table in ServiceNow.