Lesson 6 of 6
Monitoring and reprocessing
9 min
Listen to this lesson
Replication errors are normal in production: HR enters a value that has no mapping, a position replicates after the employee, a business partner sync hits an existing supplier. The difference between a healthy integration and a broken one is how quickly errors are seen, who owns them and how they are reprocessed without creating new problems.
Where to look
| Monitor | What it shows | Who uses it |
|---|---|---|
| Data Replication Monitor (EC) | Status per employee and run: success, warning, error with message; manual or automatic rescheduling | Kofi's HRIS team daily |
| Cloud Integration message monitor | Each integration flow run, payloads and errors | Theo's team |
| S/4HANA application log (SLG1) | Inbound processing and business partner sync messages | Theo's team with Kenji's developers |
| Query status (SFIOM_QRY_ADM) and job logs | Last query time per object type; job success | Theo's team |
| IDoc monitoring for cost centers | Outbound IDoc status for cost center replication | Theo's team |
The monitors along the chain and their owners at Contoso.
Reprocessing
Most employee errors are reprocessed by fixing the cause and rescheduling the employee in the Data Replication Monitor, which triggers replication again for that person; SAP Press notes the monitor supports manual and automatic rescheduling. Org object errors are reprocessed by the next org object run once the data is fixed. Business partner sync errors are rerun for the single employee with /SHCM/RH_SYNC_BUPA_EMPL_SINGLE after the cause (for example an existing supplier business partner with the wrong category) is fixed. Never fix a replicated record directly in S/4HANA infotypes: the next replication overwrites it, and the source and target disagree in between.
Operating rhythm
- 06:30 dailyMorning checkKofi reviews the Data Replication Monitor; Theo checks job logs and SLG1
- 09:00 dailyTriageErrors routed: EC data to HRIS, mapping to integration, BP sync to developers
- Same dayReprocessFixed employees rescheduled; results verified
- WeeklyReconciliationHeadcount and cost center assignments compared between EC and S/4HANA
- MonthlyFinance checkDan compares payroll postings by cost center with EC assignments
Reconciliation as a control
Monitors show what failed; reconciliation shows what is missing. Contoso runs a weekly comparison of active employees with company, cost center and manager between EC and S/4HANA, using an EC report and an S/4HANA query on infotype 0001 for the same key date. Differences are listed per employee and cleared within a week. The monthly finance check compares payroll postings by cost center with EC assignments; a difference there is often the first sign of a replication gap that the monitors did not show.
Common errors and their owners
| Error pattern | Typical cause | Owner | Fix |
|---|---|---|---|
| No value mapping for an EC code | New event reason, location or pay group in EC | Theo | Add mapping, transport, reschedule |
| Position or org unit not found | Org object not yet replicated | Theo with Kofi | Let org replication run or fix org data, reschedule |
| Cost center not valid on date | Cost center expired or not yet valid in S/4HANA | Dan with Kofi | Correct date or assignment, reschedule |
| BP sync: business partner already exists | Existing supplier BP for the person | Kenji's team | Clean up the existing BP, rerun single sync |
| No runs at all | Query job missing after refresh | Theo with Sam | Restore job schedule, check last run time |
Error patterns Contoso met in testing and hypercare.
What would you do?
After a reorganization, 23 of 180 transfers fail with "position not found". The other 157 succeeded. What should Theo and Kofi do?
- Five monitors: Data Replication Monitor, Cloud Integration, SLG1, query status and job logs, cost center IDocs.
- Fix the cause, then reschedule in the Data Replication Monitor; never fix replicated records in S/4HANA.
- Rerun business partner sync per employee after fixing the cause.
- Reconcile weekly and monthly: monitors show failures, reconciliation shows gaps.
Check your understanding · not graded
Lesson check
1. How is a failed employee replication usually reprocessed?
2. Which report reruns business partner sync for one employee?
3. What does reconciliation find that monitors miss?
4. Why did bulk rescheduling of 23 failed transfers fail again?