
Copyright © 2001, 2015, Oracle and/or its affiliates. All rights reserved.
To use these services, users need to configure new connect strings in the corresponding
glog.properties file
5
for each application server like the following:
otm_oltp
glog.database.dbaOnly.t2client.connectionURL=jdbc:oracle:thin:@(DESCRIP
TION=(ADDRESS=(PROTOCOL=TCP)(HOST=otm-rac01-
vip)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=otm-rac02-
vip)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=otm-rac03-
vip)(PORT=1521))(LOAD_BALANCE=yes)(CONNECT_DATA=(SERVER=DEDICATED)(SERV
ICE_NAME=otm_oltp)))\
otm_batch
glog.database.dbaOnly.t2client.connectionURL=jdbc:oracle:thin:@(DESCRIP
TION=(ADDRESS=(PROTOCOL=TCP)(HOST=otm-rac01-
vip)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=otm-rac02-
vip)(PORT=1521))(ADDRESS=(PROTOCOL=TCP)(HOST=otm-rac03-
vip)(PORT=1521))(LOAD_BALANCE=yes)(CONNECT_DATA=(SERVER=DEDICATED)(SERV
ICE_NAME=otm_batch)))
Hash Partitioning Indexes on RAC Databases
Oracle Transportation Management installations on the RAC database may experience high cluster
waits such as “gc buffer busy” on the indexes for the columns whose data is populated by database
sequence generated number. Converting such indexes into hash partitioned index reduces cluster
waits. Perform following steps to convert about twenty eight indexes into hash partition. Oracle
Transportation Management application and web servers should be off line during this procedure.
Note: Converting regular indexes into hash partitioned index may affect range scan on the
corresponding column. If user community is heavily searching column data by range then
you may want to keep index same as it is.
1. Make sure all constraints are valid by running following query on the database.
SELECT owner, status, constraint_type, COUNT(*) FROM
dba_constraints WHERE owner IN ('GLOGOWNER','REPORTOWNER') GROUP BY
owner, status, constraint_type;
2. Shut down all web and application servers.
3. Go to the Application servers <otm home>/glog/oracle/script8 directory. From the script8
directory log into the database as “glogower” using SQLPlus and run the following SQL
command:
@gen_hash_indexes.sql
This command will create script create_hash_indexes.sql in the same directory
4. Review create_hash_indexes.sql. This file will have about 28 indexes for hash partitioning. You
may remove indexes which are not good candidate for hash partitioning.
5. Using SQL PLUS, log in as GLOGOWNER and run
@create_hash_indexes.sql
6. Review generated log file create_hash_indexes.log and make sure that there are no errors.
5
Note these properties are reserved and cannot be added via property sets.
Comentarios a estos manuales