SAP Hana Scale-Up cluster maintenance procedure for the pacemaker 2.x on SLES for SAP

Unlock the power of Pacemaker 2.x for efficient SAP HANA cluster maintenance. This guide will show you how to move beyond full-cluster maintenance and focus on targeted tasks. By utilizing SAP HANA Multi-State resource maintenance, you can perform administrative actions on cluster nodes and the database level, all while preserving the integrity of your primary and secondary HANA DB roles and preventing accidental failovers.

• Adhere to Linux maintenance procedures, ensuring HANA runs on the same nodes as before.
• Verify each step and test the procedure before production implementation.

Steps:

These steps need to be applied on pacemaker version 2 and above cluster setups.

1. Check the status of the Linux cluster, HANA, and replication. Ensure both HANA DBs are running, replication is active, and no failed resources or warnings are present. Clear any location constraints and failed resource warnings from the cluster.

# crm status

# SAPHanaSR-showAttr

# crm configure show | grep cli-

# su – <sidadm>

# HDBSettings.sh systemReplicationStatus.py

2. Put the HANA multistate resource into maintenance mode.

# crm resource maintenance mst_SAPHana_<SID>_HDB<INSTANCE_NUMBER>

3. Put the Linux cluster into maintenance mode on one node.

# crm maintenance on

4. Stop the Linux cluster on all nodes.

# crm cluster run “crm cluster stop”

      5. Perform Linux maintenance, ensuring both HANA DBs are running and replication is active.

6. Start the Linux cluster on all nodes.

# crm cluster run “crm cluster start”

7. Set the cluster ready for operations on one node.

# crm maintenance off

8. Allow the Linux cluster to detect the HANA multistate resource status on one node.

# crm resource refresh mst_SAPHana_<SID>_HDB<INSTANCE_NUMBER>

9. Get the current HANA DB nodes’ attribute status.

# SAPHanaSR-showAttr 

10. Set the HANA multistate resource ready for operations on one node.

# crm resource maintenance mst_SAPHana_<SID>_HDB<INSTANCE_NUMBER> off

11. Check status of Linux cluster and HANA DB as well as replication, see above point #1.

By following above steps, our Primary and Secondary Hana DBs role will remain same.

Additional Information

SLES for SAP HANA Maintenance Procedures – Part -2 (Manual Administrative Tasks, OS reboots and Updation of OS and HANA)

https://documentation.suse.com/sbp/sap-15/html/SLES4SAP-hana-sr-guide-PerfOpt-15/index.html#id-maintenance

man SAPHanaSR_maintenance_examples