Updated on 2022-09-23 GMT+08:00

Monitoring a Real-Time Job

In the real-time processing mode, data is processed in real time, which is used in scenarios with high real-time performance. This type of job is a pipeline that consists of one or more nodes. You can configure scheduling policies for each node, and the tasks started by operators can keep running for an unlimited period of time. In this type of job, lines with arrows represent only service relationships, rather than task execution processes or data flows.

You can choose Monitor Job and click the Real-Time Job Monitoring tab to view the job status, start time, and end time, and perform the operations listed in Table 1.

Figure 1 Real-time job monitoring page
Table 1 Operations supported by real-time job monitoring

No.

Operation

Description

1

Searching for a job based on the job name or owner

-

2

Filtering jobs based on the job status or job label

-

3

Perform operations on jobs in a batch

Select multiple jobs and perform operations on them.

4

Viewing job instance status

Click job in front of the name. The Last Instance page is displayed. You can view information about the last instance of the job.

5

Job status-related operations

In the Operation column of the job, you can start, pause, recover, and stop job scheduling.

6

Adding a job label

In the Operation column of a job, choose More > Add Job Label.

7

Viewing node information of a job

Click a job name. On the displayed page, click a node to view its associated job/scripts and monitoring information.

NOTE:

If event-driven scheduling is configured for a node in the job, the subjob monitoring page is displayed when you click the node.

8

Disabling and restoring a node

Click a job name. On the displayed page, right-click a node and select Disable. After the node is disabled, you can right-click it and select Restore to restore it on another location. For details, see Real-Time Job Monitoring: Disabling and Restoring a Node.

9

Viewing the boot log

Click a job name. On the displayed page, right-click a node and select View Run Log to view logs of the node.

10

Configuring scheduling

Click a job name. On the displayed page, right-click the node where event-driven scheduling is configured and select Configure Scheduling to view and modify the scheduling information about the node. For details, see Real-Time Job Monitoring: Configuring Scheduling for a Node Where Event-driven Scheduling Is Configured.

11

Monitoring subjobs

Click a job name. On the displayed page, click the node where event-driven scheduling is configured to go to the subjob monitoring page. For details, see Real-Time Job Monitoring: Monitoring Subjobs.

12

Clearing stream messages

Click a job name. On the displayed page, right-click the node where event-driven scheduling is configured and select Clear Stream Message.

Real-Time Job Monitoring: Disabling and Restoring a Node

You can disable a node in a real-time job and restore it in another location.

  1. Log in to the DataArts Studio console. Locate an instance and click Access. On the displayed page, locate a workspace and click DataArts Factory.
  2. In the left navigation pane of DataArts Factory, choose Monitoring > Monitor Job.
  3. On the Real-Time Job Monitoring tab page, click a job name.
  4. On the displayed page, right-click the node and select Disable.
  5. Right-click the node and choose Resume from the shortcut menu. The Resume Node Running dialog box is displayed, as shown in Table 2.
    Table 2 Resumption parameters

    Parameter

    Description

    Last Paused

    Start time when a node is suspended.

    Tasks Not Run

    Number of tasks that are not running during node suspension.

    Run From

    Parameters for performing the tasks generated during the pause period.

    Position from which running restarts.

    • Paused node
    • The first node of the subjob

    Concurrent Tasks

    Parameters for performing the tasks generated during the pause period.

    Number of tasks to be processed.

    Task Name

    Parameters for performing the tasks generated during the pause period.

    Task to be resumed.

Real-Time Job Monitoring: Configuring Scheduling for a Node Where Event-driven Scheduling Is Configured

If event-driven scheduling is configured for a node in a real-time job, right-click the node on the job monitoring details page and choose Configure Scheduling from the shortcut menu to view and modify the scheduling information about the node.

  1. Log in to the DataArts Studio console. Locate an instance and click Access. On the displayed page, locate a workspace and click DataArts Factory.
  2. In the left navigation pane of DataArts Factory, choose Monitoring > Monitor Job.
  3. On the Real-Time Job Monitoring tab page, click a job name.
  4. On the displayed page, right-click the node where event-driven scheduling is configured, select Configure Scheduling, and configure the parameters shown in Table 3.
    Figure 2 Configuring scheduling
    Table 3 Policy parameters

    Parameter

    Description

    Concurrent Events

    Number of jobs that can be concurrently processed. The maximum number of concurrent events is 10.

    Event Detection Interval

    Interval for event detection. The unit of the interval can be Second or Minute.

    Failure Policy

    Select a policy to be performed after scheduling fails.

    • Stop scheduling
    • Ignore failure and proceed

Real-Time Job Monitoring: Monitoring Subjobs

When event-based scheduling is configured for a node in a job, you can click this node to query monitoring information of subjobs. On the Subjob page, you can stop, rerun, continue, and succeed subjobs as well as view subjob events.

  1. Log in to the DataArts Studio console. Locate an instance and click Access. On the displayed page, locate a workspace and click DataArts Factory.
  2. In the left navigation pane of DataArts Factory, choose Monitoring > Monitor Job.
  3. On the Real-Time Job Monitoring tab page, click a job name.
  4. Click a node with event-based scheduling configured, as shown in Figure 3.
    Figure 3 Subjob monitoring page

    Table 4 describes the actions listed in the Operation column of each subjob.

    Table 4 Subjob monitoring operations

    Operation

    Description

    Stop

    Stops a subjob instance that is in the Running state.

    Rerun

    Reruns a subjob instance that is in the Succeed or Failed state.

    Continue

    If a subjob instance is in the Abnormal state, you can click Continue to begin running the subsequent nodes in the subjob instance.

    NOTE:

    This operation can be performed only when Failure Policy is set to Suspend the current job execution plan. To view the current failure policy, click a node and then click Advanced Settings on the Node Properties page.

    Forcibly Succeed

    Forcibly changes the status of a subjob instance from Failed to Succeed.

    View Event

    Displays the event content of a subjob.

  5. Click in the Status column. The running records of the subjob node are displayed.
    Table 5 describes the operations that can be performed on the node.
    Table 5 Operations (node)

    Operation

    Description

    View Log

    View the log information of a node.

    More > Manual Retry

    To run a node again after it fails, click Retry.

    NOTE:

    This operation can be performed only when Failure Policy is set to Suspend the current job execution plan. To view the current failure policy, click a node and then click Advanced Settings on the Node Properties page.

    More > Succeed

    Change the status of a node from Failed to Succeed.

    NOTE:

    This operation can be performed only when Failure Policy is set to Suspend the current job execution plan. To view the current failure policy, click a node and then click Advanced Settings on the Node Properties page.

    More > Skip

    To skip a node that is to be run or that has been paused, click Skip.

    More > Pause

    To pause a node that is to be run, click Pause. Nodes queued after the paused node will be blocked.

    More > Resume

    To resume a paused node, click Resume.