Now accepting guest editorial from SAP, Oracle, Workday and ServiceNow practitioners.Pitch a guide
The Sign-Off Table

Lesson 4 of 5

Health, business process monitoring and alerts

10 min

Listen to this lesson

0:00 / 6:26

Interfaces and jobs are where most hypercare problems first appear, but two more views complete the picture. Health Monitoring tells Sam whether the system itself is fine. Business Process Monitoring tells Dan and Maria whether the business is flowing. Behind all the apps sits one alerting layer that decides who hears about a problem and how.

Health Monitoring

SAP describes Health Monitoring as collecting technical metrics from connected cloud services and systems on a regular basis. The overview shows the current status of every connected object, rated by how many metrics are in warning or critical status. You can drill into one object to see its components and metrics, each with current value and rating, and use the analytics view to look at history, which helps spot trends and find when a resource shortage began. Which metrics are collected depends on the service type.

Business Process Monitoring

Business Process Monitoring gives line-of-business and IT users a view of end-to-end process health using predefined KPIs. SAP's portal describes KPIs that are discovered automatically for supported processes, such as lead to cash, source to pay, recruit to retire and design to operate, plus custom KPIs for ABAP components and BTP applications. A KPI detail view lets you look at the business documents behind a number, and threshold violations raise alerts.

Health MonitoringBusiness Process Monitoring
Question it answersIs the system healthy?Is the business process flowing?
Typical signalA resource metric in critical statusA rising backlog of orders without deliveries
Main audienceSam and the Basis teamDan, Maria and the process experts
Hypercare useSpot capacity or availability problems earlySpot stuck documents before customers call
Two complementary views of the same production system.
The Sign-Off Table · thesignofftable.com/academy/sap-alm-205

How alerts work

Every monitoring app sends events to Intelligent Event Processing, the central component that routes them. SAP distinguishes event situations, which open when a problem starts, can be updated, and close when the condition clears (a database becoming unavailable, then recovering), from stateless events, which open and close at once (one application log message). Rules decide which actions run. Most rules are generated from each app's event settings and are changed in that app, not in Intelligent Event Processing.

An event situation stays open and can be updated until the condition clears; a stateless event opens and closes at once.
The Sign-Off Table · thesignofftable.com/academy/sap-alm-205
  1. Event configuredIn the monitoring app: what is critical, for which object, with which threshold
  2. Event raisedA situation opens or a stateless event fires
  3. Rules appliedIntelligent Event Processing decides which actions to run
  4. ActionsCreate alert, send email, create ticket, start operation flow, send chat message
  5. Alert inboxAssign a processor, comment, postpone or confirm
  6. CloseConfirming the alert closes the situation; a recurrence opens a new one
From a monitored condition to a handled alert.
The Sign-Off Table · thesignofftable.com/academy/sap-alm-205

SAP lists five action types: create an alert in the app's alert inbox, send email, create or update a ticket in an external IT service management system, start an operation flow in an automation engine, and send a chat message. Support varies by use case. For email, SAP's job monitoring guide says recipients are registered in Notification Management and then selected in the event configuration.

One aborted MRP job, two actions: an alert in the inbox for the team to work, and an email to a recipient registered in Notification Management.
The Sign-Off Table · thesignofftable.com/academy/sap-alm-205

Alerts that people act on

  • Start narrow. Alert on errors in the scenarios and jobs that matter for the go-live, not on everything that can be collected.
  • Choose the event action deliberately. SAP's job monitoring default is to raise at new occurrence; every occurrence is useful for a critical cutover job but noisy for a job that runs every five minutes.
  • Route to a named team. Each alert type has an owner: Sam for health and jobs, the integration specialist for messages, Maria and Dan for process KPIs.
  • Use maintenance windows. SAP notes that job events are not raised while a service is in maintenance or disruption in Business Service Management, which avoids false alarms during planned downtime.

What would you do?

In the first week of hypercare, Contoso's alert inbox receives 400 alerts a day, mostly warnings from a frequent interface job. The team starts ignoring the inbox. What should Sam change?

Key points
  • Health Monitoring rates technical metrics of systems and services and shows history for trend analysis.
  • Business Process Monitoring uses predefined and custom KPIs, with drill-down to business documents.
  • Intelligent Event Processing routes events to alerts, email, tickets, operation flows and chat.
  • Good hypercare alerts are few, owned, routed and set to the right event action.

Check your understanding · not graded

Lesson check

  1. 1. What is an event situation in Intelligent Event Processing?

  2. 2. Where are email recipients for job alerts registered, per SAP's guide?

  3. 3. Which question does Business Process Monitoring answer?

  4. 4. Where are most Intelligent Event Processing rules changed?