The HugePages was configured for newly built Linix servers, and we did not change any configurations of GIMR database “-MGMTDB”. Just found MGMTDB is using HugePages automatically if HugePages is configured according to the alert log.
********************************************************************** Mon Sep 19 12:52:15 2016 Dump of system resources acquired for SHARED GLOBAL AREA (SGA) Mon Sep 19 12:52:15 2016 Per process system memlock (soft) limit = UNLIMITED Mon Sep 19 12:52:15 2016 Expected per process system memlock (soft) limit to lock SHARED GLOBAL AREA (SGA) into memory: 754M Mon Sep 19 12:52:15 2016 Available system pagesizes: 4K, 2048K Mon Sep 19 12:52:15 2016 Supported system pagesize(s): Mon Sep 19 12:52:15 2016 PAGESIZE AVAILABLE_PAGES EXPECTED_PAGES ALLOCATED_PAGES ERROR(s) Mon Sep 19 12:52:15 2016 4K Configured 4 4 NONE Mon Sep 19 12:52:15 2016 2048K 59398 377 377 NONE Mon Sep 19 12:52:15 2016 ********************************************************************** System parameters with non-default values: processes = 300 cpu_count = 2 sga_target = 752M control_files = "+OCR_VOTE/_MGMTDB/CONTROLFILE/current.260.922884271" db_block_size = 8192 compatible = "12.1.0.2.0" cluster_database = FALSE db_create_file_dest = "+OCR_VOTE" _catalog_foreign_restore = FALSE undo_tablespace = "UNDOTBS1" _disable_txn_alert = 3 _partition_large_extents = "false" remote_login_passwordfile= "EXCLUSIVE" db_domain = "" dispatchers = "(PROTOCOL=TCP) (SERVICE=-MGMTDBXDB)" job_queue_processes = 0 _kolfuseslf = TRUE parallel_min_servers = 0 audit_trail = "NONE" db_name = "_mgmtdb" open_cursors = 100 pga_aggregate_target = 325M statistics_level = "TYPICAL" diagnostic_dest = "/u01/app/grid" enable_pluggable_database= TRUE
Make DBA Life Easy 