Postgres

Use the following procedure for monitoring the Postgres database.

1. Navigate to Database monitoring on the left pane.
2. Select Postgres from the Select database drop-down list.
3. Select a service from the Select service drop-down list.
4. Select a database from the Select database drop-down list.

The following screen appears along with the various available tabs.

• Hardware
• Query Performance
• Connections
• Metrics

Hardware

The Hardware tab is selected by default. The following widgets are displayed.

Widget Name	Description
CPU usage	CPU utilization of host
Memory usage	Memory utilization of host
Netowork I/O	Network input/output data in MB
Connections & sessions	database number of connections & sessions count

Query Performance

Clicking the Query Performance tab and selecting the threshold from the Query threshold drop-down list displays the following screen.

note

• Click the   to download the report in csv or pdf format.

• Once clicked on download icon, below columns selection will be dispalyed to select which columns to download in the report.

  

• Once selected hit Download as CSV or Download as PDF to download the report.

• Only the top 100 entries are downloaded if there are more than 100 entries for downloading.

Query Execution Plan

1. Clicking on ID/Query it will Query details tab will be opened with following screen.

2. The Query Details tab is selected by default. The following widgets are displayed.

Widget Name	Description
Query	Query executed in Database
Wait States	Query wait conditions (IO/CPU/Lock etc.)
Resource consumption over Time	CPU, memory, and I/O usage trends during execution
Disk and Buffer usage	Physical disk reads/writes and shared buffer utilization
Session ID	Unique identifier for database session
Elapsed Time	Total time since query started execution
Execution Time %	Percentage of time spent actively executing
Users	Database role/user executing the query
Query Active in Schema	Schema where query is executing
Query ID/SQL Handle	Unique identifier for normalized SQL statement
Procedures	Stored procedures/functions involved in execution

3. Clicking on Query Execution Plan tab following screens will be displayed.

4. The Query Execution Plan tab is selected by default. The following widgets are displayed.

Widget Name	Description
Query	Query executed in Database
Cached Execution Plan	Indicates whether the query execution plan is reused from PostgreSQL plan cache instead of being recompiled
Last Used	Timestamp of the last execution using this cached plan
Executions	Total number of times the query plan has been executed
Average Elapsed Time (ms)	Average time taken per execution in milliseconds
Plan Hash Value	Unique hash identifying the execution plan structure
Plan Details	Full breakdown of the execution plan generated by EXPLAIN/EXPLAIN ANALYZE
Rows	Number of rows processed or returned at each plan step
Data Processed	Total volume of data scanned/processed during execution
Estimated Cost (CPU%)	Optimizer-estimated cost of executing the query (CPU + I/O cost model)
Number of Steps	Total number of nodes/operations in the execution plan tree
Explained for schema	Schema context for which the query plan was generated
Referenced Objects	Tables, indexes, views, or functions used in the execution plan
Tree Node	Individual node in the execution plan hierarchy (scan, join, sort, etc.)
Operation	Type of operation performed (Seq Scan, Index Scan, Hash Join, Sort, etc.)
Object Type	Type of database object involved (table, index, view, etc.)
Object Name	Name of the referenced database object used in execution

Connections

Clicking the Connections tab displays the following widgets as in the following screen.

Widget Name	Description
No. of connections	Database number of connections count
No. of sessions	Database number of sessions count
Session details	Session details having users and queries count with time

Metrics

Clicking the Metrics tab displays the following widgets.

Widget Name	Description
Bulk operations	Buffer and block hits count
Confis operations	Queries cancelled due to deadlock, pinned buffer, timeout, old snapshot, and dropped tablespaces
Tup operations	number of operations count like fetch, delete, insert, returned, and update
Xact operations	Commit and rolled back transactions count

Multi-column Filter

Query performance

1. Multi-column filter can be applied on numeric columns within the Query performance Table to refine the displayed data based on specific conditions.
2. When the user clicks on the “Filter by” option, user can apply filter below numeric columns

Column Name	Description
Total execution	Total number of query executions recorded.
Maximum execution time (s)	The longest query execution time recorded in seconds.
Maximum execution timestamp	Date and time when the longest-running query was recorded.

2. Each filter condition is applied using the below operators and value ( user input based on the unit)

Operator
Greater than
Less than
Equals
Greater than or equals
Less than or equals

3. Once filters & conditions are applied save button is enabled & click "Save" to apply the filter.
4. Once filters & conditions are applied it displays no. of filters are applied.

note

• Filters are applied using AND logic — all specified conditions must be met for a record to be displayed.
• Only numeric columns support filtering; non-numeric fields (e.g., text or categorical values) are excluded from the filter options.
• When the Maximum Execution Timestamp (s) filter is applied, the system reviews each query in the table. If a query was executed within the time frame selected in the top-right corner, it displays that query’s maximum execution timestamp and maximum execution time.
• If no records match the selected filters, the table will display a message such as “No data".
• The multi-column filter state persists until the user clears filters by clicking on "Clear" button when Filter by button is clicked.
• A "Reset" button is available to clear all applied filters and restore the table to its original, unfiltered state. This button is enabled only when one or more filters are currently applied.

Connections --> Session details

1. Multi-column filter can be applied on numeric columns within the Connections --> Session details Table to refine the displayed data based on specific conditions.
2. When the user clicks on the “Filter by” option, user can apply filter below numeric columns

Column Name	Description
Session count	Total number of active sessions recorded in the PostgreSQL database.
Timestamp	Date and time when the session count was captured.

2. Each filter condition is applied using the below operators and value ( user input based on the unit)

Operator
Greater than
Less than
Equals
Greater than or equals
Less than or equals

3. Once filters & conditions are applied save button is enabled & click "Save" to apply the filter.
4. Once filters & conditions are applied it displays no. of filters are applied.

note

• Filters are applied using AND logic — all specified conditions must be met for a record to be displayed.
• Only numeric columns support filtering; non-numeric fields (e.g., text or categorical values) are excluded from the filter options.
• If no records match the selected filters, the table will display a message such as “No data".
• The multi-column filter state persists until the user clears filters by clicking on "Clear" button when Filter by button is clicked.
• A "Reset" button is available to clear all applied filters and restore the table to its original, unfiltered state. This button is enabled only when one or more filters are currently applied.