Monitor Metrics for VM Cluster Resources

You can monitor the health, capacity, and performance of your VM clusters and databases with metrics, alarms, and notifications. You can use Oracle Cloud Infrastructure Console, Monitoring APIs, or Database Management APIs to view metrics.

Note

To view metrics you must have the required access as specified in an Oracle Cloud Infrastructure policy (whether you're using the Console, the REST API, or another tool). For more information on policies, see Getting Started with Policies.

Prerequisites for Using Metrics

The following prerequisites must be met for the metrics to flow out of the VM Cluster.

  • Metrics on the VM Clusters depend on the Oracle Trace File Analyzer (TFA) agent. Ensure that this component is up and running. AHF version 23.2.0 or higher is required for capturing metrics from the VM Clusters. To start, stop, or check the status of TFA, see Manage Oracle Trace File Analyzer.
  • To view the metrics on the Oracle Cloud Infrastructure Console, the TFA flag defaultocimonitoring must be set to ON. This flag is set to ON by default and you need not perform any action to set this. If you are not seeing metrics on the Console, then as root user on the guest VM, check if the flag is set to ON.
    tfactl get defaultocimonitoring
    .---------------------------------------------------------------------.
    |                             <host name>                             |
    +-------------------------------------------------------------+-------+
    | Configuration Parameter                                     | Value |
    +-------------------------------------------------------------+-------+
    | Send CEF metrics to OCI Monitoring ( defaultOciMonitoring ) | ON    |
    '-------------------------------------------------------------+-------'
    If the defaultocimonitoring flag is set to OFF, then run the tfactl set defaultocimonitoring=on or tfactl set defaultocimonitoring=ON command to turn it on:
    tfactl set defaultocimonitoring=on
    Successfully set defaultOciMonitoring=ON
    .---------------------------------------------------------------------.
    |                             <host name>                             |
    +-------------------------------------------------------------+-------+
    | Configuration Parameter                                     | Value |
    +-------------------------------------------------------------+-------+
    | Send CEF metrics to OCI Monitoring ( defaultOciMonitoring ) | ON    |
    '-------------------------------------------------------------+-------'
  • The network communication uses the existing path that the Control Plane Server uses to reach the OCI Services. For more information, see Network Requirements for Oracle Exadata Database Service on Cloud@Customer.

View Metrics for a VM Cluster

To view the metrics for Guest VMs using the console, use this procedure.

Note

When there is a network problem and Oracle Trace File Analyzer (TFA) is unable to post metrics, TFA will wait for one hour before attempting to retry posting the metrics. This is required to avoid creating a backlog of metrics processing on TFA.

Potentially one hour of metrics will be lost between network restore and the first metric posted.

  1. Open the navigation menu. Click Oracle Database, then click Exadata Database Service on Cloud@Customer.
  2. Choose your Compartment.

    A list of VM Clusters is displayed.

  3. In the list of VM Clusters, click the VM Cluster for which you want to view the metrics.

    Details of the VM Cluster you selected are displayed.

  4. Under Resources, click Metrics.

    A chart for each metric is displayed. By default, the metrics for the last hour are displayed.

    You can only select the oci_database_cluster namespace from the Metric namespace drop-down.

  5. If you want to change the interval, select the required start time and end time.

    Alternatively, you can select the interval from the Quick Selects drop-down menu. The metrics are refreshed immediately for the selected interval.

  6. For each metric, you can choose the interval and statistic independently.
    • Interval: The time period for which the metric is calculated.
    • Statistic: The mathematical method by which the metric is calculated.
  7. For each metric, you can choose the following options from the Options drop-down menu.
    • View Query in Metrics Explorer
    • Copy Chart URL
    • Copy Query (MQL)
    • Create an Alarm on this Query
    • Table View

    For Detailed information on various options for viewing the metrics chart, see Viewing Default Metric Charts.

View Metrics for a Database

To view the metrics for a database using the console, use this procedure.

Note

When there is a network problem and Oracle Trace File Analyzer (TFA) is unable to post metrics, TFA will wait for one hour before attempting to retry posting the metrics. This is required to avoid creating a backlog of metrics processing on TFA.

Potentially one hour of metrics will be lost between network restore and the first metric posted.

  1. Open the navigation menu. Click Oracle Database, then click Exadata Database Service on Cloud@Customer.
  2. Choose your Compartment.

    A list of VM Clusters is displayed.

  3. In the list of VM Clusters, click the VM Cluster for which you want to view the metrics.

    Details of the VM Cluster you selected are displayed.

  4. In the list of databases, click the database for which you want to view the metrics.
  5. Under Resources, click Metrics.

    A chart for each metric is displayed. By default, the metrics for the last hour are displayed.

    You can only select the oci_database namespace from the Metric namespace drop-down.

  6. If you want to change the interval, select the required start time and end time.

    Alternatively, you can select the interval from the Quick Selects drop-down menu. The metrics are refreshed immediately for the selected interval.

  7. For each metric, you can choose the interval and statistic independently.
    • Interval: The time period for which the metric is calculated.
    • Statistic: The mathematical method by which the metric is calculated.
  8. For each metric, you can choose the following options from the Options drop-down menu.
    • View Query in Metrics Explorer
    • Copy Chart URL
    • Copy Query (MQL)
    • Create an Alarm on this Query
    • Table View

    For Detailed information on various options for viewing the metrics chart, see Viewing Default Metric Charts.

View Metrics for VM Clusters in a Compartment

To view the metrics for databases in a compartment using the console, use this procedure.

Note

When there is a network problem and Oracle Trace File Analyzer (TFA) is unable to post metrics, TFA will wait for one hour before attempting to retry posting the metrics. This is required to avoid creating a backlog of metrics processing on TFA.

Potentially one hour of metrics will be lost between network restore and the first metric posted.

  1. Open the Oracle Cloud Infrastructure Console by clicking the menu icon next to Oracle Cloud.
  2. From the left navigation list click Observability & Management.
  3. Under Monitoring, click Service Metrics.
  4. On the Service Metrics page, under Compartment, select your compartment.
  5. On the Service Metrics page, under Metric Namespace, select oci_database_cluster.
  6. If there are multiple VM Clusters in the compartment, then you can show metrics aggregated across the clusters by selecting Aggregate Metric Streams.
  7. If you want to limit the metrics you see, next to Dimensions, click Add (click Edit if you have already added dimensions).
  8. In the Dimension Name field, select a dimension.
  9. In the Dimension Value field, select a value.
  10. Click Done.
  11. In the Edit dimensions dialog click +Additional Dimension to add an additional dimension. Click X to remove a dimension.
  12. To create an alarm on a specific metric, click Options and select Create an Alarm on this Query. See Managing Alarms for information on setting and using alarms.
    Note

    If you don't see any metrics, check the network settings and AHF version listed in the prerequisites section.

View Metrics for Databases in a Compartment

To view the metrics for databases in a compartment using the console, use this procedure.

Note

When there is a network problem and Oracle Trace File Analyzer (TFA) is unable to post metrics, TFA will wait for one hour before attempting to retry posting the metrics. This is required to avoid creating a backlog of metrics processing on TFA.

Potentially one hour of metrics will be lost between network restore and the first metric posted.

  1. Open the Oracle Cloud Infrastructure Console by clicking the menu icon next to Oracle Cloud.
  2. From the left navigation list click Observability & Management.
  3. Under Monitoring, click Service Metrics.
  4. On the Service Metrics page, under Compartment, select your compartment.
  5. On the Service Metrics page, under Metric Namespace, select oci_database.
  6. If there are multiple databases in the compartment, then you can show metrics aggregated across the databases by selecting Aggregate Metric Streams.
  7. If you want to limit the metrics you see, next to Dimensions, click Add (click Edit if you have already added dimensions).
  8. In the Dimension Name field, select a dimension.
  9. In the Dimension Value field, select a value.
  10. Click Done.
  11. In the Edit dimensions dialog click +Additional Dimension to add an additional dimension. Click X to remove a dimension.
  12. To create an alarm on a specific metric, click Options and select Create an Alarm on this Query. See Managing Alarms for information on setting and using alarms.
    Note

    If you don't see any metrics, check the network settings and AHF version listed in the prerequisites section.

Manage Oracle Trace File Analyzer

The deployment of the cloud-certified Autonomous Health Framework (AHF), which includes Oracle Trace File Analyzer, is managed by Oracle. You shouldn’t install this manually on the guest VMs.

  • To check the run status of Oracle Trace File Analyzer, run the tfactl status command as root or a non-root user:
    # tfactl status 
    .-------------------------------------------------------------------------------------------------.
    | Host           | Status of TFA | PID    | Port | Version    | Build ID             | Inventory Status|
    +----------------+---------------+--------+------+------------+----------------------+------------+
    | node1          | RUNNING       |  41312 | 5000 | 22.1.0.0.0 | 22100020220310214615| COMPLETE    |
    | node2          | RUNNING       | 272300 | 5000 | 22.1.0.0.0 | 22100020220310214615| COMPLETE    |
    '----------------+---------------+--------+------+------------+----------------------+------------'
  • To start the Oracle Trace File Analyzer daemon on the local node, run the tfactl start command as root:
    # tfactl start
    Starting TFA..
    Waiting up to 100 seconds for TFA to be started..
    . . . . .
    . . . . .
    . . . . .
    . . . . .
    . . . . .
    . . . . .
    . . . . .
    . . . . .
    Successfully started TFA Process..
    . . . . .
    TFA Started and listening for commands
  • To stop the Oracle Trace File Analyzer daemon on the local node, run the tfactl stop command as root:
    # tfactl stop
    Stopping TFA from the Command Line
    Nothing to do !
    Please wait while TFA stops
    Please wait while TFA stops
    TFA-00002 Oracle Trace File Analyzer (TFA) is not running
    TFA Stopped Successfully
    Successfully stopped TFA..