Manage the SYSAUX tablespace
by Nirbhay Kumar Singh, Oracle Database Administrator, Rackspace Technology
Introduction
Oracle 10g introduced a new mandatory tablespace called SYSAUX with mandatory attributes like PERMANENT, READ WRITE, EXTENT MANAGEMENT LOCAL, and SEGMENT SPACE MANAGEMENT AUTO. This post explores how to manage this tablespace as it grows.
You can use the SYSAUX tablespace to perform the following tasks:
- Avoid SYSTEM tablespace fragmentation caused by install and deinstall options.
- Avoid risking SYSTEM tablespace corruption and out-of-space situations.
- Reduce maintenance for the database administrator.
- Store and reduce tablespaces for all auxiliary database metadata related to
- Oracle options and features, such as the following items:
SYSAUX occupants
There are 26 SYSAUX occupants that you can query, as shown in the following example:
SQL> select OCCUPANT_NAME,OCCUPANT_DESC
from V$SYSAUX_OCCUPANTS
order by SPACE_USAGE_KBYTES desc
Later versions of Oracle Database added even more features, as shown in the following image:
The following image shows the components added or deprecated in each Oracle Database version:
Proactive SYSAUX tablespace management
To proactively monitor and manage SYSAUX tablespace proactively, perform the following actions:
1. Set the SYSAUX tablespace to AUTOEXTEND off.
2. Confirm the STATISTICS_LEVEL value.
- The ALL value is sometimes resource-intensive.
- The Basic and Typical values tend to consume fewer resources.
3. Check how you use advisors, baselines, and SQL tuning sets. Advisors need to retain information in snapshots even if you plan to drop the snapshot range.
4. Run a query to determine which sysaux_occupant consumes the most space in the SYSAUX tablespace.
Remedial SYSAUX tablespace management
The SYSAUX tablespace grows for the following reasons:
- You set an excessive retention period.
- The segment advisor grows too large.
- The active session history (ASH) grows too large.
The following sections offer some remedial actions
Check the AWR retention period
To manage this, check the retention period of the Automatic Workload Repository (AWR) snapshot. The AWR collects the performance statistics for problem detection and performance tuning and stores the details in memory and database tables. The system deletes stored data based on the retention period. If the retention period is too high, this data consumes more space in the SYSAUX tablespace. So, you should set an appropriate retention period. Check the retention period by using the following query:
SQL> SELECT retention FROM dba_hist_wr_control;
You can alter the retention manually by using the
DBMS_WORKLOAD_REPOSITORY.MODIFY_SNAPSHOT_SETTINGS procedure. The following example sets the retention period to 5760 minutes (or four days:
4 days \* 24 hours per day \* 60 minutes per hour = 5760 minutes):
SQL> execute DBMS_WORKLOAD_REPOSITORY.MODIFY_SNAPSHOT(RETENTION=>5760);
Check the largest object in the SYSAUX tablespace
Run the following query to identify the largest occupant in the tablespace:
SQL> select OCCUPANT_NAME, SCHEMA_NAME, MOVE_PROCEDURE, SPACE_USAGE_KBYTES
from v$sysaux_occupants;
You use the move procedure command to move the occupant to another tablespace unless the move procedure is null.
The following example moves WKSYS occupant in XYZ tablespace:
SQL> execute WKSYS.MOVE_WK(‘XYZ’);
Check the Active Session History (ASH)
See if the ASH consumes more space in AWR data by using the following script:
SQL> @$ORACLE_HOME/rdbms/admin/awrinfo.sql
ASH usage is acceptable at 1.1 percent. If it is high, delete orphaned ASH rows. Check for orphaned rows by using the following query:
SQL> SELECT COUNT(1) Orphaned_ASH_Rows FROM wrh$_active_session_history a
WHERE NOT EXISTS (SELECT 1 FROM wrm$_snapshot WHERE snap_id = a.snap_id
AND dbid= a.dbid AND instance_number = a.instance_number);
Because the value is larger than zero, delete. View the orphans by using the following query:
SQL> DELETE FROM wrh$_active_session_history a WHERE NOT EXISTS
(SELECT 1 FROM wrm$_snapshot WHERE snap_id = a.snap_id
AND dbid = a.dbid AND instance_number = a.instance_number);
Then, run the following query to shrink the WRH$_ACTIVE_SESSION_HISTORY
table to reclaim the freed space:
SQL> alter table WRH$_ACTIVE_SESSION_HISTORY shrink space;
Conclusion
This post introduces the SYSAUX tablespace. It also provides suggestions on monitoring and managing space growth in the SYSAUX tablespace default occupants and identifying any non-default objects wrongly stored in the tablespace.
Recent Posts
Padrões de rede híbrida do Google Cloud - Parte 2
Outubro 16th, 2024
Padrões de rede híbrida do Google Cloud - Parte 2
Outubro 15th, 2024
How Rackspace Leverages AWS Systems Manager
Outubro 9th, 2024
O Windows Server impede a sincronização da hora com o Rackspace NTP
Outubro 3rd, 2024