Restrictions

When using the baseline O&M feature, pay attention to the following requirements to ensure that the task priority takes effect:

GaussDB(DWS) restrictions:

To enable priority for DWS SQL nodes of DataArts Studio, run the following SQL statements as an administrator to set priority parameters:

SELECT * FROM gs_wlm_set_queryband_action('Priority=1','priority=low');
DWS priority 1 corresponds to priority 1 of DataArts Factory baseline O&M jobs.
SELECT * FROM gs_wlm_set_queryband_action('Priority=2','priority=medium');
DWS priority 2 corresponds to priorities 2, 3, and 4 of DataArts Factory baseline O&M jobs.
SELECT * FROM gs_wlm_set_queryband_action('Priority=3','priority=high');
DWS priority 3 corresponds to priority 5 of DataArts Factory baseline O&M jobs.

MRS restrictions:

For MRS clusters, priority configuration is available for data connections in MRS API mode and unavailable for data connections in proxy mode.
To enable priority for MRS nodes of DataArts Studio (if the MRS cluster is in security mode, that is, Kerberos authentication is enabled), add the whitelist configured for MRS Hive, configure the following parameters, and click Save. The configuration takes effect after a rolling restart of the parameter. Set this parameter in the basic configuration on the cluster O&M management page of MRS Hive. The operations are as follows:
1. Log in to FusionInsight Manager and choose Cluster > Services > Hive. Click Configurations then All Configurations.
2. In the navigation pane on the left, choose Hive > HiveServer. Add mapreduce.job.priority to the value of hive.security.authorization.sqlstd.confwhitelist.
  Figure 1 Configuring hive.security.authorization.sqlstd.confwhitelist
3. Save the change and restart Hive.
  
  Priority configuration is available for the following MRS nodes: MRS Spark SQL, MRS Hive SQL, MRS Spark, MRS Flink Job, and MRS MapReduce.
  
  To make the Hive priority take effect, contact O&M engineers to enable the MRS Hive priority configuration item.
Before using baseline O&M, you must create a topic in MRS. For details, see Creating a Topic on Kafka UI.

DLI restrictions:

DLI allows you to set job priorities. When resources are insufficient, computing resources are preferentially provided for jobs with higher priorities. DLI priority configuration is available for DLI Flink Job, DLI SQL, and DLI Spark job operators.