Oracle Cluster Health Monitor – changes in 12.1.0.2

From Oracle 12.1.0.2, the Oracle Cluster Health Monitor becomes an Oracle database by default [replacing the old Berkley DB], and it’s called “-MGMTDB” (note the leading “-“)

cat /etc/oratab

+ASM1:/u01/app/grid:N # line added by Agent
-MGMTDB:/u01/app/grid:N # line added by Agent

It lives on one of the nodes on your RAC cluster and occupied space on Disk Group provisioned during install.

The DB will take about 750MB RAM [and 1GB on disk to start with], so even more memory taken on top of the +ASM database memory. This makes a good case for running the -MGMTDB on one node but using Flex ASM on the other nodes if you have mode than a 2 node cluster. Given 80% of all the RAC installs globally are 2 node clusters… that’s not going to help so you’re just losing memory.

It’s scheduled to regularly minutes and gather O/S stats and store them in the DB.

Check out the settings using:

srvctl config mgmtdb

Database unique name: _mgmtdb
Database name:
Oracle home: 
Oracle user: oracle
Spfile: +OCR_DG/_MGMTDB/PARAMETERFILE/spfile.268.884776109
Password file:
Domain:
Start options: open
Stop options: immediate
Database role: PRIMARY
Management policy: AUTOMATIC
Type: Management
PDB name: rac12c_cluster
PDB service: rac12c_cluster
Cluster name: rac12c-cluster
Database instance: -MGMTDB

You can switch the CHM process on and off using:

crsctl stop res ora.crf -init
crsctl start res ora.crf -init