One of our clients have a master, a slave, and a standby HANA database. They have some issue where their slave node got moved over to the standby node. The system performance was not good and very bad, so they thought to shutdown down and return the nodes all their original locations. but servers are no longer able to start the HANA database back up.
Raised SAP message sap suggested below:
Thanks for the update.
The root cause of the DB start up fail is because root user execute some hdbnsutil command(which should run using sidadm user) and create file .SHM_SAP_IPMM_00XX_SID in /var/lib/hdb/SID/shmgrp
The file .SHM_SAP_IPMM_00XX_SID is uesed for HANA made inter process memory management. Since the file is created by root user, sidadm user could not control the HANA inter process memory, and it leads to DB start up fail.
Please make sure no HANA service or hdbnsutil command is executed by root user to avoid this situation.