Why does SAP Migration Cockpit show 'File import failed' with no error details in the UI?

This almost always means the Excel or CSV template version does not match the LTMC migration object version installed in your system. Download a fresh template directly from LTMC (select your Migration Object → Download Template), repopulate your source data into that template, and re-upload. Also verify the file is fully closed in Excel before uploading — LTMC cannot read a Windows-locked file and silently fails. If the problem persists after these steps, open transaction SMICM → Goto → Trace File to check for an HTTP 413 Request Entity Too Large response, which means you need to increase the profile parameter icm/HTTP/max_request_body_size and restart ICM.

How do I fix RFC_NO_AUTHORITY errors during SAP data migration?

RFC_NO_AUTHORITY means the technical user executing the migration lacks the S_RFC authorization object. Immediately after the error, run transaction SU53 as the migration user to capture the exact missing authorization check — it will show you the specific RFC function group name that was blocked. Take that information to your BASIS team and ask them to add S_RFC with Activity 16 (Execute) and RFC_NAME set to that function group in the migration role via PFCG. Never assign SAP_ALL in a production system. For system-to-system RFC migrations, also confirm in SM59 that the RFC destination user account has the same role assigned in the target system.

What causes TIME_OUT dumps specifically in LTMC but not in normal SAP transactions?

LTMC dialog-mode migration executes inside a single ABAP dialog work process. When the migration package size is large (default 1000 rows), the work process must complete the full loop of data validation, transformation, and database INSERT within rdisp/max_wprun_time (default 300 seconds). Normal user transactions involve lightweight operations that complete in seconds. Fix options in order of preference: (1) Switch LTMC execution to Background mode in Settings — this bypasses the dialog timeout entirely and no restart is required. (2) Reduce the package size to 200 rows so each batch finishes within 300 seconds. (3) Increase rdisp/max_wprun_time to 3600 in RZ10 and restart the instance. Background mode is strongly preferred because it also survives browser session disconnections.

How do I migrate more than 200,000 records through LTMC without timeouts or memory errors?

For datasets above 50,000 rows: first split the source file into chunks of 10,000 to 25,000 rows using Excel or a simple script. In LTMC Expert Settings, enable Parallel Processing with 3 to 4 workers and reduce Package Size to 500. Always run in Background mode and verify enough background work processes are available in SM50 before starting. Monitor progress in transaction LTMOM rather than LTMC — LTMOM provides real-time package-level status and error detail. For S/4HANA Cloud or very large datasets, switch to the Staging Tables approach instead of file upload, which bypasses browser upload size limits and allows direct database-level staging before the migration step executes.

Why does SAP data migration work in the test system but fail with connection or timeout errors in production?

Three causes account for nearly all test-passes-production-fails scenarios. First, firewall policy differences: production networks typically have stricter rules that drop idle RFC connections after 60–120 seconds, causing TIMEOUT in SM59 even when the gateway is running. Work with your network team to verify the same ports (33xx, 36xx) are open and that stateful firewall idle timeouts are raised for SAP service IPs. Second, profile parameter differences: production may have lower rdisp/max_wprun_time or fewer background work processes than test. Compare RZ10 profiles side by side between systems and align them. Third, authorization gaps: the migration user account often has a more permissive role in test (sometimes SAP_ALL) but a restricted custom role in production. Run SU53 in production immediately after the first failure to capture the exact missing authorization object.

SAP Data Migration Troubleshooting: Connection Refused, TIME_OUT Dumps & Slow Performance Fixes

Fix SAP data migration failures fast: resolve CPIC_CALL_BACK_NOT_REGISTERED, TIME_OUT dumps, and slow LTMC loads with profile tuning, background mode, and RFC a

Last updated: February 23, 2026

Last verified: February 23, 2026

2,673 words

Key Takeaways

Connection refused errors (CPIC_CALL_BACK_NOT_REGISTERED, NIECONN_REFUSED) are caused by a stopped SAP gateway or message server process, or firewall rules blocking ports 33xx/36xx/39xx — verify with ps and ss on the application server, then restart via sapcontrol.
TIME_OUT short dumps during LTMC migration occur because the dialog work process runtime limit (rdisp/max_wprun_time, default 300s) is shorter than the time needed for a large migration batch — switch LTMC execution to Background mode or raise the profile parameter to 3600 in RZ10.
Slow migration performance is caused by missing table statistics after bulk inserts, lock contention visible in SM12, or too few background work processes — run UPDATE STATISTICS in HANA, reduce LTMC package size to 200–500 rows, and increase rdisp/wp_no_btc to at least 6.
Quick diagnostic path: ST22 for dump type → SM21 for system log events → SM59 for RFC status → RZ10 for profile parameter review → apply targeted fix and restart only if required.

Fix Approaches Compared
Method	When to Use	Time	Risk
Switch LTMC to Background execution mode	TIME_OUT dumps during dialog-mode migration	2–5 min config, no restart	Low — no downtime, survives browser disconnect
Increase rdisp/max_wprun_time in RZ10	Persistent TIME_OUT even in background; long-running BAPIs	15 min + instance restart	Low — only extends runtime window
Restart SAP instance via sapcontrol	Connection refused / gateway or ICM crash	10–20 min downtime	Medium — interrupts active user sessions
Reduce LTMC batch/package size to 200–500	DBIF_RSQL_SQL_ERROR or lock contention mid-migration	5 min config, reruns needed	Low — may increase total migration clock time
Run UPDATE STATISTICS on target tables	Slow INSERT after bulk load on HANA	Minutes to hours depending on table size	Low — metadata-only read operation
Add S_RFC authorization object in PFCG	RFC_NO_AUTHORITY blocking source system access	30–60 min with BASIS team	Low — additive permission change only

Understanding SAP Data Migration Failures

SAP data migration spans multiple tools and layers: the SAP Migration Cockpit (LTMC/LTMOM), Legacy System Migration Workbench (LSMW), BAPI-based batch loaders, and RFC-connected ETL tools. Each layer introduces distinct failure modes. A connection refused at the RFC gateway looks entirely different from a TIME_OUT dump in the ABAP runtime, even though both prevent data from moving. Misidentifying the layer wastes hours of investigation.

The four most common failure categories are:

Network and gateway failures: CPIC_CALL_BACK_NOT_REGISTERED, NIECONN_REFUSED, RFC_NO_AUTHORITY
Runtime timeout dumps: TIME_OUT, ICM_HTTP_CONNECTION_BROKEN, SYSTEM_FAILURE
Database errors: DBIF_RSQL_SQL_ERROR, DBIF_REPO_SQL_ERROR, duplicate primary key violations
Data quality errors: mandatory field missing, type conversion failure, orphaned foreign key references

Step 1: Identify the Failure Layer

Check ST22 first — ABAP short dumps. This is the fastest starting point for any SAP crash or timeout.

Transaction ST22 -> Set date range -> Look for:
  TIME_OUT             — work process runtime exceeded rdisp/max_wprun_time
  DBIF_RSQL_SQL_ERROR  — database rejected an INSERT or UPDATE
  SNAP_NO_NEW_ENTRY    — SNAP table (TSP01) is full, call BASIS
  MEMORY_NO_MORE_PAGING — system swapping heavily, memory exhausted
  DYNPRO_SEND_IN_BACKGROUND — UI call attempted from background process

Check SM21 for infrastructure-level events:

Transaction SM21 -> Local Analysis -> Last 1 hour -> Look for:
  "Work process restarted"       — process crashed, note reason code
  "Enqueue: timeout waiting"     — lock contention blocking migration
  "RFC: no more sessions"        — RFC connection pool exhausted
  "DB error 54" or "DB error 60" — Oracle deadlock or lock timeout

Check SM59 for RFC connectivity when migrating between systems:

Transaction SM59 -> Select RFC destination -> Connection Test
Green: "Connection test was successful"
Red:   "CPIC_CALL_BACK_NOT_REGISTERED" -> gateway process not running
Red:   "RFC_NO_AUTHORITY"              -> missing S_RFC authorization object
Red:   "TIMEOUT"                       -> network latency or stateful firewall dropping idle RFC connections

Step 2: Resolve Connection Refused Errors

Symptoms: NIECONN_REFUSED, CPIC_CALL_BACK_NOT_REGISTERED, SM59 connection test fails with a red status.

Root cause: The SAP gateway (gwrd), message server (msServ), or ICM (icman) process has stopped, or a firewall rule is blocking ports required for RFC communication.

Step 2a — Verify running processes on the application server:

# Run as <sid>adm on the SAP application server
ps aux | grep -E 'disp\+work|gwrd|msServ|icman' | grep -v grep

# Expected: one line per process
# <sid>adm ... /usr/sap/<SID>/D<NR>/exe/disp+work
# <sid>adm ... /usr/sap/<SID>/D<NR>/exe/gwrd
# <sid>adm ... /usr/sap/<SID>/D<NR>/exe/icman

# If any process is missing, restart the instance:
sapcontrol -nr <INSTANCE_NR> -function RestartSystem IsRestarted 10 120

Step 2b — Confirm ports are listening (replace 00 with your instance number):

ss -tlnp | grep -E '3200|3300|3600|3900|8000'
# 32<NR> = SAP dispatcher (dialog)
# 33<NR> = RFC/CPIC gateway  
# 36<NR> = Gateway server
# 39<NR> = Message server
# 80<NR> or 8000 = HTTP/ICM

Step 2c — Fix gateway ACL if CPIC_CALL_BACK_NOT_REGISTERED persists after restart:

Edit /usr/sap/<SID>/SYS/profile/DEFAULT.PFL and ensure:

gw/acl_mode = 1
gw/sec_info = $(DIR_GLOBAL)/secinfo
gw/reg_info = $(DIR_GLOBAL)/reginfo

Verify the secinfo file contains a P (permit) entry for your source system hostname and IP. Restart the gateway process after any ACL change.

Step 3: Resolve Timeout Errors During Migration

Symptoms: ST22 shows TIME_OUT dump; LTMC browser session shows ICM_HTTP_CONNECTION_BROKEN; migration aborts mid-package with no partial commit.

Root cause: Dialog work process runtime limit (rdisp/max_wprun_time, default 300 seconds) is shorter than the time required to process a migration batch of 1000 rows with full validation and commit.

Step 3a — Switch to Background execution in LTMC (preferred — no restart):

LTMC → Open your migration project
Navigate to the Migrate Data step
Click Settings → change Execution Mode from Dialog to Background
Set Background Job Priority to A (highest)
Optionally verify the scheduled job in SM37 after launching

Step 3b — Extend profile parameters for persistent timeout issues (requires restart):

In RZ10 → Instance Profile → Extended Maintenance, add or modify:

rdisp/max_wprun_time = 3600
rdisp/scheduler/prio_high/max_wprun_time = 7200
icm/keep_alive_timeout = 600
icm/conn_timeout = 600
http/session_timeout = 3600

Then restart the instance:

sapcontrol -nr <NR> -function Stop
sapcontrol -nr <NR> -function Start
sapcontrol -nr <NR> -function GetProcessList

Step 4: Resolve Slow Migration Performance

Symptoms: Migration runs but takes 10x longer than expected; SM50 shows work processes locked in "Running" state for minutes at a time; RFC calls to the source system intermittently time out.

Step 4a — Diagnose lock contention via SM12 and SM50:

Transaction SM12 -> Display Lock Entries -> Sort by Lock Time
Entries older than 5 minutes during an active migration indicate a blocking transaction.
Note the locked table name and the owning user/session.
Go to SM50 and find the matching work process.
If the process is confirmed hung, use SM50 -> Process -> Cancel with Core (emergency only).

Step 4b — Identify expensive database statements on HANA via DBACOCKPIT:

-- Find the slowest statements in the last migration window
SELECT TOP 20
  SUBSTR(STATEMENT_STRING, 1, 120) AS SQL_SNIPPET,
  DURATION_MICROSECOND / 1000000  AS DURATION_SEC,
  EXECUTION_COUNT,
  USER_NAME
FROM M_EXPENSIVE_STATEMENTS
WHERE START_TIME > ADD_SECONDS(NOW(), -3600)
  AND DURATION_MICROSECOND > 10000000
ORDER BY DURATION_MICROSECOND DESC;

-- Force statistics refresh on the migration target table
UPDATE STATISTICS "SAPHANADB"."MARA";

Step 4c — Reduce LTMC batch size and enable parallel packages:

LTMC → Migration Project → Settings → Expert Settings:

Set Package Size from 1000 (default) to 200–500
Enable Parallel Processing: set parallel workers to 2–4 (must not exceed available background WPs minus 2)

Step 4d — Ensure sufficient background work processes in RZ10:

rdisp/wp_no_btc = 8
# Set to: (parallel LTMC workers) + 2 at minimum
# Check current allocation: Transaction SM50 -> Process Overview

Step 5: Resolve Data Quality Errors

Symptoms: LTMC error log shows rows in red; messages include Field LIFNR is mandatory, Duplicate entry for primary key MANDT-MATNR, or Foreign key constraint violation.

Step 5a — Export and triage the LTMC error log:

LTMC → Migration Project → Migrate Data step → Error Log button
Export as Excel/CSV using the download icon
Filter column Message Type = E
Sort by Message ID to group the most frequent errors

Step 5b — Check for duplicate key conflicts before re-import:

" Run in SE38 — adjust table and key field to your migration object
REPORT z_check_migration_duplicates.

SELECT matnr, COUNT(*) AS cnt
  FROM mara
  GROUP BY matnr
  HAVING COUNT(*) > 1
  INTO TABLE @DATA(lt_dups).

IF lt_dups IS INITIAL.
  WRITE: / 'No duplicate keys found — safe to re-import'.
ELSE.
  LOOP AT lt_dups INTO DATA(ls_dup).
    WRITE: / 'Duplicate:', ls_dup-matnr, 'Count:', ls_dup-cnt.
  ENDLOOP.
ENDIF.

Step 5c — Correct source data and re-run migration:

Fix the source spreadsheet based on error log findings
In LTMC → Migration Project → click Revert to File Upload to reset the step
Upload the corrected file, re-run validation, then execute migration

Baseline SAP Configuration Before Any Migration Project

Verify the following parameters are correctly set in RZ10 before starting a large migration. Document before-values and keep a rollback plan. Always validate in a non-production system first.

rdisp/max_wprun_time       = 3600
rdisp/wp_no_dia            = 10
rdisp/wp_no_btc            = 8
ztta/roll_area             = 3000000
ztta/roll_first            = 1
em/initial_size_MB         = 4096
abap/heap_area_dia         = 500000000
icm/server_port_0          = PROT=HTTP,PORT=8000,TIMEOUT=3600
icm/keep_alive_timeout     = 600
http/session_timeout       = 3600

Frequently Asked Questions

bash

#!/bin/bash
# SAP Data Migration Diagnostic Script
# Usage: ./sap_migration_diag.sh <SID> <INSTANCE_NR>
# Run as <sid>adm on the SAP application server

SID=${1:-SAP}
NR=${2:-00}

echo ================================================
echo  SAP Migration Diagnostic: SID=$SID Instance=$NR
echo  $(date)
echo ================================================

# 1. SAP process status via sapcontrol
echo
echo [1] SAP Process List
echo -------------------
sapcontrol -nr $NR -function GetProcessList 2>/dev/null \
  || echo ERROR: sapcontrol failed -- verify SID and instance number

# 2. Check critical OS-level processes
echo
echo [2] Critical SAP Process Check
echo --------------------------------
for PROC in disp+work gwrd msServ icman; do
  PID=$(pgrep -f $PROC 2>/dev/null | head -1)
  if [ -n "$PID" ]; then
    echo "  OK   $PROC  (PID $PID)"
  else
    echo "  FAIL $PROC  NOT RUNNING -- investigate immediately"
  fi
done

# 3. Check SAP ports are listening
echo
echo [3] SAP Port Availability
echo --------------------------
for PORT in 32${NR} 33${NR} 36${NR} 39${NR} 80${NR} 8000; do
  ss -tlnp 2>/dev/null | grep -q ":${PORT} " \
    && echo "  OPEN   port $PORT" \
    || echo "  CLOSED port $PORT -- check firewall and gateway config"
done

# 4. Check relevant profile parameters
echo
echo [4] Key Profile Parameters
echo ---------------------------
PROF=/usr/sap/${SID}/SYS/profile
for PARAM in rdisp/max_wprun_time rdisp/wp_no_dia rdisp/wp_no_btc \
             icm/keep_alive_timeout http/session_timeout em/initial_size_MB; do
  VAL=$(grep -rh "^${PARAM}" ${PROF}/ 2>/dev/null | tail -1)
  if [ -n "$VAL" ]; then
    echo "  SET     $VAL"
  else
    echo "  DEFAULT $PARAM (not explicitly set -- check SAP default)"
  fi
done

# 5. Memory and disk
echo
echo [5] System Resources
echo --------------------
echo -- Memory --
free -h 2>/dev/null
echo -- Disk --
df -h /usr/sap/${SID} /sapmnt 2>/dev/null || df -h /usr/sap 2>/dev/null

# 6. Recent core dumps in SAP work directory
echo
echo [6] Recent Core Dumps in SAP Work Dir (last 24h)
echo --------------------------------------------------
find /usr/sap/${SID}/D${NR}/work/ -name "core*" -mtime -1 \
  -exec ls -lh {} \; 2>/dev/null && echo "  (none found)" || true

# 7. HANA expensive statements (if hdbsql is available)
echo
echo [7] HANA Expensive Statements Check
echo -------------------------------------
if command -v hdbsql &>/dev/null; then
  hdbsql -u SYSTEM -p "${HANA_PASS}" -d SYSTEMDB 2>/dev/null <<'HDBEOF'
SELECT TOP 10
  SUBSTR(STATEMENT_STRING,1,80) AS SQL,
  DURATION_MICROSECOND/1000000 AS SEC,
  USER_NAME
FROM M_EXPENSIVE_STATEMENTS
WHERE DURATION_MICROSECOND > 10000000
ORDER BY DURATION_MICROSECOND DESC;
HDBEOF
else
  echo "  hdbsql not found on PATH -- run query via DBACOCKPIT in SAP GUI"
fi

# 8. SAP GUI transaction checklist
echo
echo [8] SAP GUI Follow-Up Checklist
echo ---------------------------------
echo "  ST22  -- Short dumps today: filter TIME_OUT, DBIF_RSQL_SQL_ERROR"
echo "  SM21  -- System log last 1h: RFC errors, work process restarts"
echo "  SM59  -- RFC destinations: run Connection Test for migration source"
echo "  SM12  -- Lock entries: identify blocking locks during migration"
echo "  SM50  -- Work process overview: find processes stuck in Running state"
echo "  LTMOM -- Migration Cockpit monitor: per-package status and error log"
echo "  RZ10  -- Instance profile: review and tune migration parameters"
echo
echo Diagnostic complete.

Error Medic Editorial

The Error Medic Editorial team comprises senior DevOps engineers, SAP BASIS consultants, and SRE practitioners with 10+ years of hands-on experience managing enterprise ERP infrastructure. We specialize in SAP S/4HANA system administration, large-scale data migration projects (LTMC, LSMW, SLT), ABAP runtime tuning, and production incident response across cloud and on-premise deployments.

Sources

Explore More Enterprise Software Guides

Confluence Troubleshooting Guide: Fixing 'Connection Refused', Timeouts, and Slow Performance

Comprehensive guide to resolving Confluence configuration issues, 'Connection Refused' errors, timeouts, and slow performance. Learn to diagnose and apply fixes

Fixing Microsoft Dynamics 365 Data Migration Error 0x80040216 and API Throttling (429)

Resolve Dynamics 365 data migration timeouts, Error 0x80040216, and API throttling (429). Learn optimal Dataverse configuration tweaks to stabilize bulk ingesti

How to Fix REQUEST_LIMIT_EXCEEDED and API Limits in Salesforce

Resolve the Salesforce REQUEST_LIMIT_EXCEEDED error. Learn to diagnose total API limits, fix concurrent request bottlenecks, and optimize bulk batch items.