SAP HANA Administration
How to Start and Stop SAP HANA Safely
Learn the correct commands and operational sequence to start and stop SAP HANA on-premise, including HDB, sapcontrol, SAP HANA cockpit, and tenant database checks.
Starting and stopping SAP HANA on-premise requires the right operating-system user, command, and system layout. This guide covers the local HDB script, sapcontrol for distributed systems, SAP HANA cockpit, and verification steps for system and tenant databases.
Understand the SAP HANA startup sequence
SAP HANA 2.0 runs as a multitenant system with a system database and tenant databases. The startup sequence begins with the SAP HANA services on the host, followed by the system database and the tenant databases according to their configured startup behavior.
A successful operating-system start does not replace database-level verification. After starting the system, check service status, connect to the system database, and confirm that required tenant databases are available to applications. SAP HANA cockpit and SAP HANA database explorer provide practical administration views for these checks.
Start SAP HANA with HDB
The local HDB script starts and stops SAP HANA on the current host. The script is located under /usr/sap/<SID>/HDB<instance number>/HDB, and Linux is case sensitive.
Run the command as the <sid>adm user:
cd /usr/sap/<SID>/HDB<instance number>
./HDB start
The stop command uses the same script:
cd /usr/sap/<SID>/HDB<instance number>
./HDB stop
HDB acts on the local host. It is suitable for a single-host operation, while distributed systems require coordinated control across hosts. Use sapcontrol for that distributed-system workflow.
Use sapcontrol for distributed systems
For a distributed SAP HANA system, run sapcontrol as the <sid>adm user with the instance number:
sapcontrol -nr <instance number> -function StartSystem
To stop the system, use:
sapcontrol -nr <instance number> -function StopSystem
These commands provide the documented system-level control path for distributed installations. After either operation, inspect the returned status and verify all relevant hosts and services before releasing the system to users.
For command-line database work, the SAP HANA hdbsql command-line guide explains connection and execution practices that complement system-level lifecycle commands.
Verify the system after startup
Use SAP HANA cockpit to review host and service status, database availability, alerts, and recent operational events. SAP HANA database explorer is useful for connecting to the system database or a tenant database and confirming that SQL access is available.
Check the following items:
- The system database accepts an administrative connection.
- Required tenant databases are running and reachable.
- Expected services report a healthy state.
- Application connections succeed after database startup.
- Recent alerts do not indicate resource, storage, or service failures.
Memory pressure can delay startup or cause services to become unstable. The SAP HANA memory usage guide describes the host and service memory views used during investigation. For ongoing operational monitoring, the SAP HANA cockpit alerts guide provides related alert-handling guidance.
Stop SAP HANA with operational control
Plan a controlled stop when applications, jobs, interfaces, and backups are not using the system. Notify dependent teams, stop application activity according to the local runbook, and then use HDB stop for a local host or sapcontrol -nr <instance number> -function StopSystem for a distributed system.
After the stop completes, confirm that the expected services have stopped and that no dependent process is still writing to the database. Record the reason, time, command path, and verification result in the operations log.
A stop can affect backup and recovery workflows. Review the SAP HANA backup and recovery guide when the operation is part of maintenance, recovery preparation, or a planned outage.
Troubleshoot startup and shutdown problems
Begin with the command output and the service state reported by SAP HANA cockpit or SAP HANA database explorer. Then review the relevant trace files, operating-system messages, storage availability, and network reachability between hosts.
Common investigation areas include:
- Confirming that the command ran as
<sid>adm. - Checking the instance number and SAP system identifier.
- Verifying that required file systems are mounted and writable.
- Reviewing service and host status across every distributed-system host.
- Checking tenant database availability after the system database starts.
- Reviewing recent traces for the service that failed to start or stop.
The SAP HANA trace files guide explains where to focus when service diagnostics are required. Preserve relevant evidence before restarting repeatedly, especially when the issue affects a production system.
Build a safe start and stop runbook
A reliable runbook identifies the system, instance number, hosts, <sid>adm account, preferred control method, application dependencies, and verification checks. It also records who authorized the operation and how the result was confirmed.
For routine work, use this sequence:
- Confirm the maintenance window and application impact.
- Identify whether the system is local-host or distributed.
- Select
HDBorsapcontrolaccordingly. - Run the command as
<sid>adm. - Verify system, service, and tenant database status.
- Check alerts, connections, and dependent applications.
- Record the outcome and any follow-up action.
This approach separates the operating-system lifecycle action from database availability checks, which helps operators detect partial startup and incomplete shutdown conditions early.