The number of active agents in the repository. If this number is 0, then Enterprise Manager is not monitoring any external targets. May be a problem if unexpected.

Data Source

The number of agents whose status is up in the mgmt_current_availability table.

User Action

If no agents are running, determine the reasons they are down, correct if needed and restart. Log files in the agent's $ORACLE_HOME/sysman/log directory can provide information about possible causes of agent shutdown.

This is the number of rows processed.

Data Source

The mgmt_system_performance_log table in the Management Repository.

User Action

If this number continues to rise over time, then the user may want to consider adding another Management Service or increasing the number of loader threads for this Management Service. To increase the number of loader threads, add or change the em.loader.threadPoolSize entry in the emoms.properties file. The default number of threads is 2. Values between 2 and 10 are common.

This is the amount of time in seconds that the loader thread has been running in the past hour.

Data Source

The mgmt_system_performance_log table in the Management Repository.

User Action

If this number is steadily increasing along with the Loader Throughput (rows per hour) metric, then perform the actions described in the User Action section of the help topic for the Loader Throughput (rows per hour) metric. If this number increases but the loader throughput does not, check for resource constraints, such as high CPU utilization by some process, deadlocks in the Management Repository database, or processor memory problems.

The total number of notifications delivered by the Management Service over the previous 10 minutes. The metric is collected every 10 mins and no alerts will be generated.

Data Source

The mgmt_system_performance_log table in the Management Repository.

User Action

If the number of notifications processed is continually increasing over several days, then you may want to consider adding another Management Service.

This indicates average number of EM console accesses in a minute. The metric is collected every 10 mins and alerts will not be generated.

Data Source

This metric is obtained using the following query of the mgmt_oms_parameters table in the Management Repository.

SELECT value FROM mgmt_oms_parameters 
where name='loaderOldestFile'

User Action

None.

The number of times the agent has been restarted in the past 24 hrs.

Data Source

Derived by:

(SELECT t.target_name, COUNT(*) down_count
  FROM mgmt_availability a, mgmt_targets t
  WHERE a.start_collection_timestamp = a.end_collection_timestamp
    AND a.target_guid = t.target_guid
    AND t.target_type = MGMT_GLOBAL.G_AGENT_TARGET_TYPE
    AND a.start_collection_timestamp > SYSDATE-1
  GROUP BY t.target_name)

User Action

If this number is high, check the agent logs to see if a system condition exists causing the system to bounce. If an agent is constantly restarting, the Targets Not Uploading Data metric may also be set for targets on the agents with restart problems. Restart problems may be due to system resource constraints or configuration problems.

The number of administrators defined for Enterprise Manager.

Data Source

The mgmt_created_users table in the Management Repository.

The number of groups defined for Enterprise Manager.

Data Source

The mgmt_targets table in the Management Repository.

User Action

If you have a problem viewing the All Targets page, you may want to check the number of roles and groups.

The number of roles defined for Enterprise Manager.

Data Source

The mgmt_roles table in the Management Repository.

User Action

If you have a problem viewing the All Targets page, you may want to check the number of roles and groups.

The number of targets defined for Enterprise Manager.

Data Source

The mgmt_targets table in the Management Repository.

User Action

This metric is informational only

This is the total number of MB that the Management Repository tablespaces are currently using.

Data Source

The dba_data_files table in the Management Repository.

User Action

This metric is informational only.

The rate at which targets are being created. The target addition rate should be greatest shortly after EM is installed and then should increase briefly whenever a new agent is added. If the rate is increasing abnormally, you should check for abnormal agent or administrator activity and verify that the targets are useful. Check to see that group creation is not being over utilized.

Data Source

The metric is derived from the mgmt_target table, the current target count - target count at last sampling.

User Action

This metric is informational only.

The total MB allocated to the Management Repository tablespaces. This will always be greater than or equal to the space used.

Data Source

The dba_free_space table in the Management Repository.

User Action

This metric is informational only.

The rate at which users are being created. The target addition rate should be low. If the rate is increasing abnormally, you should check for abnormal administrator activity.

Data Source

The metric is derived from the mgmt_created_users table, the current user count - user count at last sampling.

User Action

This metric is informational only.

This metric flags a DBMS job whose schedule is invalid. A schedule is marked 'Invalid' if it is scheduled for more than one hour in the past, or more than one year in the future. An invalid schedule means that the job is in serious trouble.

Data Source

The user_schedule_jobs table in the Management Repository.

User Action

None.

The percentage of the past hour the job has been running.

Data Source

The mgmt_system_performance_log table in the Management Repository.

User Action

If the value of this metric is greater than 50%, then there may be a problem with the job. Check the System Errors page for errors reported by the job. Check the Alerts log for any alerts related to the job.

The down condition equates to the dbms_job "broken" state. The Up Arrow means not broken.

Data Source

The broken column is from the all_users table in the Management Repository.

User Action

Determine the reason for the dbms job failure. Once the reason for the failure has been determined and corrected, the job can be restarted through the dbms_job.run command.

To determine the reason the dbms job failed, take the following steps (replacing myjob with the displayed name of the down job):

1. Copy down the DBMS Job Name that is down from the row in the table. This DBMS Job Name is 'yourDBMSjobname' in the following example.

2. Log onto the database as the repository owner.

3. Issue the following SQL statement:

   select dbms_jobname 
     from mgmt_performance_names 
     where display_name='yourDBMSjobname';
   

4. If the dbms_jobname is 'myjob', then issue the following SQL statement:

   select job
     from all_jobs
     where what='myjob';
   

5. Using the job id returned, look for ORA-12012 messages for this jobid in the alerts log and trace files and try to determine and correct the problem.

The job can be manually restarted through the following database command:

execute dbms_job.run (jobid);

The number of notifications delivered per second, averaged over the past hour.

Data Source

The mgmt_system_performance_log table in the Management Repository.

User Action

This metric is informational only.

This metric displays the average time taken to dequeue a message from the event queues.

Data Source

The data for this metric comes from entries in mgmt_system_performance_log where name= 'DequeueTime'

User Action

If the average event dequeue time is consistently higher for more than an hour, then it might indicate that the database queue table (em_event_bus_table) requires maintenance.

This metric displays the average time taken to apply the incident rules to an event or an incident or a problem.

Data Source

The data for this metric comes from entries in mgmt_system_performance_log where name= 'ProcessingTime'

User Action

If the average event processing time is continually increasing for more than an hour, then remove any unnecessary or out-of-date incident rules. Fix any operational issues present in the Management Repository database. You might have to add an additional Management Service to increase the event system capacity.

This metric displays the average time taken to query the event bus (aq$em_event_bus_table).

Data Source

The data for this metric comes from entries in mgmt_system_performance_log where name= 'QueryTime'.

User Action

If the average event queue query time is consistently higher for more than an hour, then it might indicate that the database queue table (em_event_bus_table) requires maintenance.

This metric displays the average time an event or an incident or a problem waits in the event queue before it is picked up for applying incident rules.

Data Source

The data for this metric comes from entries in mgmt_system_performance_log where name= 'Latency'

User Action

If the average latency is continually increasing, then remove any unnecessary or out-of-date incident rules. Fix any operational issues present in the Management Repository database. You might have to add an additional Management Service to increase the event system capacity.

This metric displays the status of the event system. If there are events waiting to be processed in the event queues and no events were processed for more than 5 minutes, then the event system is DOWN. Otherwise, the event system is UP.

Data Source

The data for this metric comes from entries in the aq$em_event_bus_table view and the em_event_bus_queues table.

User Action

Restart the Management Service. This might resolve the issue.

If any of the event queues is assigned a nonexisting Management Service, then this metric can have a positive count. Otherwise, it is 0.

Data Source

The data for this metric comes from entries in em_event_bus_queues, em_event_coordinators and mgmt_failover_table.

User Action

If the value stays positive continually for more than half an hour, then restart Oracle Management Service, which might resolve the issue.

If any event queue is not assigned to a Management Service, this metric has a positive value. Otherwise, it is 0.

Data Source

The data for this metric comes from entries in em_event_bus_queues, em_event_coordinators and mgmt_failover_table.

User Action

If the value stays positive continually for at least half an hour, then restart Oracle Management Service, which might resolve the issue.

This metric displays the number of events waiting to be processed in event queues.

Data Source

The data for this metric comes from entries in aq$em_event_bus_table, where consumer_name does not start with 'ADM' and msg_state is 'READY'.

User Action

If the value stays high continually for at least half an hour, you might be experiencing an event flood and the issue could be temporary. Fix any operational issues present in the Management Repository database. You might need to add an additional Oracle Management Service to increase the event system capacity.

This metric is a total number of events currently waiting to be processed in event queues.

Data Source

The data for this metric comes from entries in aq$em_event_bus_table, where consumer_name does not start with 'ADM' and msg_state is 'READY'.

User Action

If the value stays high continually, you might experience an event flood. You might have to add an additional Oracle Management Service to increase the event system capacity.

This metric measures the percentage of CPU time elapsed while processing the load from the queues of a specific queue type.

Data Source

The data for this metric comes from entries in mgmt_system_performance_log where name= 'ProcessingTime'.

User Action

If the value stays high continually, you might experience an event flood. You might have to add an additional Oracle Management Service to increase the event system capacity.

This metric displays the number of events currently waiting to be processed in event system queues of a specific queue type.

Data Source

The data for this metric comes from entries in the mgmt_system_performance_log where name= 'ProcessingTime'.

User Action

If the value stays high continually, you might experience an event flood. You might have to add an additional Oracle Management Service to increase the event system capacity.

This metric displays the number of events currently waiting to be processed in event system queues of a specific queue type.

Data Source

The data for this metric comes from entries in aq$em_event_bus_table, where consumer_name starts with a specific queue type value and msg_state is 'READY'.

User Action

If the value stays high continually, you might experience an event flood. You might have to add an additional Oracle Management Service to increase the event system capacity.

This metric displays the name of the blackout

This metric displays the URL of the Management Agent that is installed on the host.

This metric displays the number of targets affected by the expired blackout.

The alert log error trace file is the name of an associated server trace file generated when the problem generating this incident occurred. If no additional trace file was generated, this field will be blank.

The following table shows how often the metric's value is collected.

Data Source

The alert log error trace file name is extracted from the database alert log.

User Action

The alert log error trace file name is provided so that the user can look in this file for more information about the problem that occurred.

The fully specified (includes directory path) name of the current XML alert log file.

The following table shows how often the metric's value is collected.

Data Source

This name is retrieved by searching the OMS ADR_HOME/alert directory for the most recent (current) log file.

User Action

The alert log file name is provided so that the user can look in this file for more information about the problem that occurred.

A diagnostic incident is a single occurrence of a problem (critical error) that occurred in the OMS process while using Enterprise Manager.

The following table shows how often the metric's value is collected.

Data Source

Text describing a diagnostic incident is extracted from the database alert log, which is an XML file stored in the Automatic Diagnostic Repository (ADR) that stores a chronological list of database messages and errors.

User Action

Diagnostic incidents usually indicate software errors and should be reported to Oracle using the Enterprise Manager Support Workbench.

The Execution Context ID (ECID) tracks requests as they move through the application server. This information is useful for diagnostic purposes because it can be used to correlate related problems encountered by a single user attempting to accomplish a single task.

The following table shows how often the metric's value is collected.

Data Source

The ECID is extracted from the database alert log.

User Action

Diagnostic incidents usually indicate software errors and should be reported to Oracle using the Enterprise Manager Support Workbench. When packaging problems using Support Workbench, the ECID will be used by Support Workbench to correlate and include any additional problems in the package.

An optional field (may be empty) assessing the impact of the problem that occurred.

The following table shows how often the metric's value is collected.

Data Source

The impact is extracted from the database alert log.

User Action

This field is purely informational. Diagnostic incidents usually indicate software errors and should be reported to Oracle using the Enterprise Manager Support Workbench.

The Incident ID is a number that uniquely identifies a diagnostic incident (single occurrence of a problem).

The following table shows how often the metric's value is collected.

Data Source

The incident ID is extracted from the database alert log.

User Action

Diagnostic incidents usually indicate software errors and should be reported to Oracle using the Enterprise Manager Support Workbench. Problems are one or more occurrences of the same incident. Using Support Workbench, the incident ID can be used to select the correct Problem to package and send to Oracle. Using the command line tool ADRCI, the incident ID can also be used with the show incident command to get details about the incident.

The job dispatcher is responsible for scheduling jobs as required. It starts up periodically and checks if jobs need to be run. If job dispatcher is running more than the threshold levels, then it is having problems handling the job load.

Data Source

This is the sum of the amount of time the job has run over the last hour from the mgmt_system_performance_log table in the Management Repository divided by one hour, multiplied by 100 to arrive at the percent.

User Action

This metric is informational only.

The number of job steps processed per second by the job dispatcher, averaged over the past hour and sampled every 10 minutes.

Data Source

The mgmt_job_execution table in the Management Repository.

User Action

This metric is informational only.

This metric displays number of metric collection errors cleared for the day.

This metric displays the number of metric collection errors reported each day.

This metric displays the average time taken in seconds to deliver the notification from the time this issue was published to Enterprise Manager.

Data Source

• The data for this metric comes from the em_notify_requests and em_notify_deliveries tables (Release 12c (12.1.0.4 and later))

• The data for this metric comes from entries in mgmt_system_performance_log where name=method_name||_OMS_SECONDS

User Action

If the average notification time is steadily increasing, verify that the notification methods specified are performing as expected. Remove any unnecessary or out-of-date incident rules.

This metric displays the percentage of CPU (elapsed time) that the notification system was active in sending notifications.

User Action

• The data for this metric comes from the em_notify_requests and em_notify_deliveries tables (Release 12c (12.1.0.4 and later)

• The data for this metric comes from entries in mgmt_system_performance_log where name=method_name||DELIVERY_MILLIS (All releases earlier than 12c (12.1.0.4))

User Action

If the notification processing time is steadily increasing, then verify that the notification methods are performing as expected. Remove any unnecessary or out-of-date incident rules.

This metric displays the total number of notifications delivered by the Management Service over the past hour.

Data Source

• The data for this metric comes from the em_notify_requests and em_notify_deliveries tables (Release 12c (12.1.0.4 and later)

• The data for this metric comes from entries in mgmt_system_performance_log where name=method_name||DELIVERY_MILLIS (All releases earlier than 12c (12.1.0.4))

User Action

If the number of notifications processed is continually increasing over several days, then you might want to consider adding another Management Service.

This metric displays the total number of notifications pending delivery.

Data Source

• The data for this metric comes from the em_notify_requests and em_notify_deliveries tables (Release 12c (12.1.0.4 and later)

• The data for this metric comes from aq$em_notify_qtable where consumer_name like 'method_name%' (All releases earlier than 12c (12.1.0.4))

User Action

If the number of notifications pending is continually increasing, then fix any operational issues. You might have to add an additional Management Service to increase the notification system capacity.

If one of the following conditions exist, then the metric's value is DOWN. Otherwise, it is UP.

• If more than 500 notifications are waiting to be completed for more than 1 hour

• If an advanced notification device has not processed notifications for more than one hour and it has requested for notifications to be retried later

Data Source

• If more than 500 notifications are waiting to be completed for more than one hour, then data is obtained from aq$em_notify_qtable:

  SELECT count(1)
   FROM aq$em_notify_qtable
  WHERE msg_state='READY'
   AND consumer_name not like 'ADM%'
   AND consumer_name not like 'RCA1'
   AND enq_time <= sysdate - (1/24);
  

• If an advanced notification device has not processed notifications for more than one hour and it has requested for notifications to be retried later, then data is obtained from the em_notify_requeue table:

    SELECT count(1)
          FROM (SELECT device_id, min(insertion_timestamp) as first_ts
              FROM EM_NOTIFY_REQUEUE
             GROUP BY device_id) a
  WHERE a.first_ts > sysdate - (1/24);
  

User Action

• If the pending notifications count is high, then check if notification methods are performing as expected. Remove any unnecessary or out-of-date incident rules. You might have to add an additional Management Service to increase the notification system capacity.

• If an advanced notification method is requesting for notifications to be retried later, then fix the underlying issue so that it stops requesting retries.

This metric displays the number of backoff requests in the last 10 minutes.

This metric displays the number of files in the upload backlog.

This metric displays the number of rows processed by the data loader in the last hour.

This metric displays the size (in MB) of the upload backlog.

This metric displays the percentage of targets that are not monitored.

This metric displays the rate of data upload to the Management Repository.

This metric displays the number of pending templates that will apply to your target.

For more information about Monitoring Templates, see Oracle Enterprise Manager Administrator's Guide.

This metric displays the number of metric extensions pending deployment.

For more information about Metric Extensions, see Oracle Enterprise Manager Administrator's Guide

The total amount of time in seconds the collection workers were running in last 10 minutes. This is an indicator of the load on the repository collection subsystem. This could be due to two reasons, the number of collections have increased or some of the metrics are taking a long time to complete. This needs to be related with collections processed metric to find out if number of collections have increased or metrics are taking a long time.

Data Source

The data for this metrics come from entries in mgmt_system_performance log where job_name=MGMT_COLLECTION.Collection Subsystem.

The total number of collections that were processed in the last 10 minutes.

Data Source

The data for this metrics come from entries in mgmt_system_performance log where job_name=MGMT_COLLECTION.Collection Subsystem

The total number of collections that were waiting to run at the point this metric was collected. An increasing value would mean the collection workers are falling behind and would need to be increased. The collections waiting to run could be high initially on system startup and should ideally go down towards zero.

Data Source

The data for this metrics come from entries in mgmt_collection_tasks table which holds all the list of collections.

User Action

This metric is informational only.

The total number of workers that were processing the collections.

Data Source

The data for this metric come from entries in mgmt_collection_workers table.

User Action

This metric is informational only.

The total number of collections per second processed by all the collection workers.

Data Source

The data for this metrics come from entries in mgmt_system_performance log where job_name=MGMT_COLLECTION.Collection Subsystem.

User Action

This metric is informational only.

This metric displays the run duration (in seconds) for the collection task.

The number of job steps that were ready to be scheduled but could not be because all the dispatchers were busy.

When this number grows steadily, it means the job scheduler is not able to keep up with the workload.

User Action

This is the sum of job steps whose next scheduled time is in the past - job steps eligible to run but not yet running. If the graph of this number increases steadily over time, the user should take one of the following actions:

• Increase the em.jobs.shortPoolSize, em.jobs.longPoolSize and em.jobs.systemPoolSize properties in the web.xml file. The web.xml file specifies the number of threads allocated to process different types of job steps. The short pool size should be larger than the long pool size.

  Property	Default Value	Recommended Value	Description
  em.jobs.shortPoolSize	10	10 50	Steps taking less than 15 minutes
  em.jobs.longPoolSize	8	8 - 30	Stars taking more than 15 minutes
  em.jobs.systemPoolSize	8	8 - 20	Internal jobs (e.g. agent ping)

• Add another Management Service on a different host.

• Check the job step contents to see if they can be made more efficient.

This metric displays the status of the Repository Job Scheduler.

This metric displays the status of the long running metric collections.

This metric displays the status of the short running metric collections.

A count of the number of sessions between the Management Service and Management Repository database.

Data Source

The gv$session system view.

User Action

This metric is informational only.

This metric indicates whether Enterprise Manager is up or down. If you have configured the agent monitoring the oracle_emrep target with a valid email address, you will receive an email notification when Enterprise Manager is down.

The following table shows how often the metric's value is collected and compared against the default thresholds.

Data Source

sysman/admin/scripts/emrepresp.pl

User Action

This metric checks for the following:

• Is the Management Repository database up and accessible?

  If the Management Repository database is down, start it. If 'Invalid Username or Password' error is displayed, verify that the name and password for the oracle_emrep target is the same as the repository owner's name and password.

• Is at least one Management Service running?

  If a Management Service is not running, start one.

• Is the Repository Metrics dbms job running?

  If the DBMS job is down or has an invalid schedule, it should be restarted by following the instructions in the User Action section of the help topic for the DBMS Job Bad Schedule metric.

This metric is generated if any of the OMS services (such as Loader, Notification, or PingRecorder) failed to get initialized during the OMS startup. At present this metric is used only by Loader service.

This metric has two key columns and one non-key columns:

• The key columns are Management Service and Service Name. The key values uniquely identify the Service instance that has initialization errors.

• The non-key column is Service Status. This column indicates whether the Service is running fine or encountered an error during OMS startup.