PGXC_WLM_WORKLOAD_RECORDS
PGXC_WLM_WORKLOAD_RECORDS displays the status of job executed by the current user on CNs. Only the system administrator or the preset role gs_role_read_all_stats can access this view. This view is available only when enable_dynamic_workload is set to on.
Column |
Type |
Description |
---|---|---|
node_name |
Text |
Name of the CN where the job is executed. |
thread_id |
Bigint |
ID of the backend thread. |
processid |
Integer |
lwpid of the thread. |
Timestamp |
Bigint |
Start time of statement execution. |
username |
Name |
Username logged in to the backend. |
memory |
Integer |
Memory required for the statement. |
active_points |
Integer |
Number of resources consumed by the statement on the resource pool. |
max_points |
Integer |
Maximum number of resources in the resource pool. |
priority |
Integer |
Priority of a job. |
resource_pool |
Text |
Resource pool where a job is. |
status |
Text |
Job execution status. The options are: pending running finished aborted unknown |
control_group |
Text |
Cgroups used by a job. |
enqueue |
Text |
Queue for the job, including: GLOBAL: global queue. RESPOOL: resource pool queue. ACTIVE: not queued. |
query |
Text |
Statement currently being executed. |
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.