PRIMECLUSTER Wizard for Oracle Configuration and Administration Guide 4.2 - Linux for Itanium -

A.1.3 Warning state

An Oracle instance resource has entered Warning.

[Case 1] (Standby, Oracle RAC Scalable)

An Oracle instance resource enters Warning when

• The Oracle system user password is not registered in PRIMECLUSTER Wizard for Oracle,
• The Oracle system user password is registered in PRIMECLUSTER Wizard for Oracle, but it is incorrect, or
• The correct Oracle system user password is registered in PRIMECLUSTER Wizard for Oracle, but it is disabled.

If ORA-01017 is output to syslog, you can determine the cause of this problem.

Register the correct system user password again. Refer to "3.3 Oracle System User Password Change".

[Case 2] (Standby, Oracle RAC Scalable)

An Oracle instance resource enters Warning when the Oracle system user account is locked, so you are not allowed to connect to Oracle.

If ORA-28000 is output to syslog, you can determine the cause of this problem.

You can also check the system user account by referring to the following SQL.

If the "account_status" column indicates "LOCKED (TIMED)" or "LOCKED", you can determine the cause of this problem.

# su - <Oracle user>
$ sqlplus /nolog
SQL> connect / as sysdba
SQL> select username, account_status from dba_users;

Access Oracle using sysdba privileges then release account locking of the system user.

# su - < Oracle user >
$ sqlplus /nolog
SQL> connect / as sysdba
SQL> alter user system account unlock;

If ORA-01017 is output to syslog as well as ORA-28000, an incorrect password might have been registered. If this is the case, correct the system user password. Refer to "3.3 Oracle System User Password Change". Then, release account locking using the procedure above.

[Case 3] (Standby, Oracle RAC Scalable)

An Oracle instance resource enters Warning when instance monitoring is discontinued.
Check the monitoring status by executing the "clorainfo -m" command. Refer to "4.6 clorainfo - Display Resource Configuration and Monitoring Status".

If Oracle instance monitoring has been discontinued, check if the Oracle instance has been activated with the OPEN status then restart monitoring. Refer to "4.2 hvoraenable - Restart Monitoring Resources".

[Case 4] (Standby, Oracle RAC Scalable)

In Oracle instance monitoring, an Oracle instance resource enters Warning if there is no reply from SQL (e.g. insert, update, etc.) within a specified time, but you can access Oracle, and the instance indicates OPEN.
Take corrective action on Oracle.

In Oracle instance monitoring of PRIMECLUSTER Wizard for Oracle, if there is no reply from Oracle within a specified time, monitoring timeout will be considered. At the first monitoring timeout, the resource will only enter Warning, however, if it occurs twice in a row, a resource failure will be determined.

[Case 5] (Standby, Oracle RAC Scalable)

In Oracle instance monitoring, an Oracle instance resource enters Warning if you cannot access Oracle because of a max session error or max process error.

If ORA-00018 or ORA-00020 is output to syslog, you can determine the cause of this problem.

For the maximum number of sessions that are set in the Oracle SESSIONS parameter, or the maximum number of processes that are set in the PROCESSES parameter, estimate enough values considering monitoring sessions for PRIMECLUSTER Wizard for Oracle.

For details, refer to the Oracle manual.

[Case 6] (Oracle RAC Scalable)

The monitoring of Oracle RAC Instance resource becomes failure when setting of WFTime is the followings, the resource state enter Warning until it is recovered by Oracle Clusterware.

• [PersistentWarning]
• [<second>]

It is possible to check setting of WFTime by the "clorainfo -c" command. (For detail of "WFTime", refer to step "12" of"2.3.5 userApplication with Oracle RAC Instances and Listeners").
Check whether the cause of resource failure is reported referring to the Oracle alert log or log files.