Configure job policies

Daria Barysheva

The Configuration → Job policy page in the ADB Control web interface allows you to configure schedules for different system jobs. This page includes two tabs, each of which is described in detail below.

Data cleanup

On the Configuration → Job policy → Data cleanup tab, you can configure a schedule for removing old metrics from ADB Control.

adbc config jobs
The "Configuration → Job policy → Data cleanup" tab

To configure the job, follow the steps:

  1. Activate the Enable switcher.

  2. Fill in the following parameters in the Main parameters section.

    Field Description Default value

    Batch size

    A maximum number of rows in the data batch

    500

    Expire duration

    A maximum time period that is used to store metrics in ADB Control. If this period expires, the metrics will be removed during the next job launch. Use the following format — <value><unit> (without spaces), where <unit> can have the following values:

    • ms — milliseconds;

    • sec — seconds;

    • min — minutes;

    • hr — hours;

    • d — days;

    • w — weeks.

    Examples: 5min, 1hr, 1d. The minimal value is 100ms; the maximum is 1w

    5d

    Data cleanup schedule

    A schedule for automatic data cleanup

    0 0 23 * * ? *

  3. Click Apply. The Revert button is designed to undo the changes that have not been yet saved by clicking Apply.

If all steps are completed successfully, the next launch time of the Data cleanup job is updated on the Jobs → ADCC → Schedule page.

CAUTION

  • The data cleanup job is enabled in ADB Control by default. It is not recommended to disable the job, since it can cause a lack of disk space.

  • For long-term storage of historical data, you can configure loading metrics to an external database.

Metrics offload

On the Configuration → Job policy → Metrics offload tab, you can configure a schedule for uploading ADB Control metrics and events to an external database. To configure the job, follow the steps:

  1. On the target host, ensure you have a database to store the uploaded metrics. Its name will be used when configuring the JDBC connection later.

  2. Provide the ADB Control access to the selected external database. For example, PostgreSQL requires the following record in the pga_hba.conf file:

    host    <database_name>  <user_name>       <adbc_address>      trust

    where:

    • <database_name> — a database name in PostgreSQL. For example, adbc_metrics.

    • <user_name> — a user name in PostgreSQL. For example, postgres.

    • <adbc_address> — an IP address of the host where ADB Control is deployed (with a subnet number). For example, 10.92.18.1/32.

  3. Go to the Configuration → Job policy → Metrics offload tab in the ADB Control interface.

    adbc config jobs2
    The "Configuration → Job policy → Metrics offload" tab

  4. Activate the Service enabled switcher.

  5. Click Edit connection in the Analytic database connection section.

  6. In the window that opens, fill in the following fields:

    • JDBC URL — a URL of the JDBC connection to the host where the target database is located. For PostgreSQL, use the following format:

      jdbc:postgresql://<external_ip>:5432/<db_name>

      where:

      • <external_ip> — an IP address of the host where the target database is located. For example, 10.92.6.225.

      • <db_name> — a target database name. For example, adbc_metrics.

    • User — a user name. In the following example, the default user postgres is used.

    • Password — a user password.

      adbc config jobs3
      JDBC connection parameters

  7. Click Save.

  8. Fill in the following fields in the Main parameters section of the previous form.

    Field Description Default value

    Batch size

    A maximum number of rows in the data batch that is sent to an external database

    1000

    Offload metrics job schedule

    A schedule for the automatic data export to an external database

    0 0 * * * ? *

  9. Click Apply. The Revert button is designed to undo the changes that have not been yet saved by clicking Apply.

    adbc config jobs4
    Data is filled

  10. Get the following message on the successful result.

    adbc config jobs5
    The successful result

  11. As a result, the adcc schema should be created in the external database. The tables in this schema will store metrics and events collected from ADB Control. To check whether all necessary objects are created, you can run the following query in PostgreSQL:

    SELECT table_name
FROM information_schema.tables
WHERE table_schema = 'adcc' ORDER BY  table_name;

    Result:

             table_name
-----------------------------
 audit_auth
 audit_auth_adb
 audit_operation
 cluster
 cluster_query_metric
 query
 query_error
 resgroup_config_audit
 resgroup_status
 resgroup_status_per_segment
 transaction
(11 rows)

If all steps are completed successfully, the new scheduled job Export job should be added to the Jobs → ADCC → Schedule page.

IMPORTANT

In addition to the job schedule, you should specify the clusters and databases which metrics should be exported to the external database. For more information, see Cluster management.
Found a mistake? Seleсt text and press Ctrl+Enter to report it