2083031 - Database recovery prematurely aborts as one of the services crashed with signal 9

HANA Database recovery prematurely aborts due to the below error in the backup.log of the master node:

2014-10-15T18:18:29+02:00 P033272 1491455b2b4 ERROR   RECOVERY RECOVER DATA finished with error: [448] recovery could not be completed, volume 7, reached log position 0, [110044] The service 'indexserver' at hdbnode2:30803' responsible for the volume '7' does crash

The indexserver trace file does not contain a crash stack or any other indication of the root cause.

In the next step, check if the daemon trace on the affected node contains the below lines:

[28579]{-1}[-1/-1] 2014-10-15 18:18:29.912580 i Daemon TrexDaemon.cpp(12316) : process hdbindexserver with pid 28647 exited because it caught signal 9

As the service was killed with SIGNAL 9, also check the operating system SYSLOG (/var/log/messages) for further details:

Oct 15 18:18:22 hdbnode2 kernel: [23108.222905] Out of memory: Kill process 28647 (hdbindexserver) score 856 or sacrifice child

SAP HANA Database

SIGKILL 9; OOM Killer; restore; , KBA , HAN-DB-BAC , SAP HANA Backup & Recovery , HAN-DB , SAP HANA Database , How To