SM37 — Background Job Monitor
SM37 is used to monitor scheduled background jobs, job steps, variants, logs, spool output and cancellation status. It is most useful when a scheduled report did not run, ran late, cancelled, or produced unexpected output. For reliable support work, start with the exact system, client, user and business context, then use the transaction's own logs or status information before changing configuration or data.
This practitioner page covers SM37, Background Job Monitor. It focuses on the real operational purpose of the transaction, the evidence to collect before changing anything, the objects and status information that matter, and the failure patterns consultants repeatedly see in production support and project testing.
Published 19 Sept 2026· 766 words
Esta página ainda não está disponível em português.
Purpose
Monitor scheduled background jobs, job steps, variants, logs, spool output and cancellation status. SM37 should be treated as a diagnostic or controlled business tool rather than simply a screen name: the value comes from understanding what evidence it exposes or what state it changes. In support, capture the exact system, client, user, timestamp and affected business object before drawing a conclusion, because the same transaction can show very different results across organizational and execution contexts.
When it is used
SM37 is typically used when a scheduled report did not run, ran late, cancelled, or produced unexpected output. Consultants also reach for it during test cycles and incident reproduction because it gives a direct view of the relevant SAP runtime or configuration state. In production, use the narrowest selection that reproduces the issue, and distinguish a display/analysis action from any action that changes or deletes system state.
How to use it in practice
- Filter tightly by job name, user and date to avoid reviewing unrelated runs.
- Open the job log before checking anything else; the first explicit error usually defines the next diagnostic step.
- Inspect the step program, variant and execution user because background context often differs from dialog testing.
- For Cancelled jobs, correlate the timestamp with ST22 and SM21.
- Confirm whether a restart is safe before rescheduling business-critical processing.
Key data objects
The following fields, logs or repository objects are the most useful anchors when working in SM37. They are the pieces of context to capture in screenshots, tickets and handovers so another consultant can reproduce the same finding rather than starting from a generic symptom.
- job name and user — verify the exact value and its relationship to the failing business or technical step.
- start condition — verify the exact value and its relationship to the failing business or technical step.
- job status — verify the exact value and its relationship to the failing business or technical step.
- step program and variant — verify the exact value and its relationship to the failing business or technical step.
- job log and spool — verify the exact value and its relationship to the failing business or technical step.
How to prove it in the data
Do not stop at the first visible error. Reproduce the issue with the same user, client and input, capture the key values above, and correlate them with the nearest application log, job/update/RFC record or repository object. A good proof shows the before-state, the exact failure or status, and the after-state following a controlled correction; that makes the diagnosis auditable and prevents a coincidental retry from being mistaken for a fix.
ECC vs S/4HANA
SM37 continues to be central in S/4HANA for ABAP background jobs, even where application scheduling is exposed through Fiori apps or framework-specific schedulers. The practical rule is to separate “still technically available” from “preferred design for new work.” During an S/4HANA program, keep the transaction as a support/reference tool where valid, but challenge legacy implementation patterns that conflict with released APIs, Fiori-first processes or clean-core principles.
Common pitfalls and how to diagnose them
- Assuming a Finished status means the business result is correct. Diagnose this by returning to the exact user, timestamp, object and log evidence before changing settings.
- Testing the report in dialog under your own user and overlooking the background user's authorizations. Diagnose this by returning to the exact user, timestamp, object and log evidence before changing settings.
- Restarting a cancelled job without checking for partial updates or duplicate output. Diagnose this by returning to the exact user, timestamp, object and log evidence before changing settings.
Whose problem this is
Primary ownership normally sits with the BASIS functional or technical team, with Basis, Security or development joining only when the evidence crosses into infrastructure, authorization or custom code. A strong escalation includes the transaction, exact selection/input, affected object, timestamp, expected result, actual result and the checks already completed.
Related SAP objects
Reviewed pages this object connects to in the ERPClimb knowledge graph.
Source: ERPClimb — https://erpclimb.com/sap-tcodes/sm37ERPClimb is an independent platform and is not affiliated with SAP SE. Reference pages are written and reviewed by SAP consultants for learning and troubleshooting.