About Integrated Diagnostics
The Integrated Diagnostics REST API helps obtain the performance data to be able to diagnose performance issues related to integrated Extract, integrated Replicat, and parallel Replicat integrated Replicat. This is a replacement over the previously used Replication Performance Advisory Utility (UTLRPDAV).
Integrated Diagnostics feature is responsible for collecting diagnostic and performance data for each of the Oracle GoldenGate Extract and Replicat processes. This process preserves interval data in memory and computes all statistical values related to CPU usage, session wait times, various counts and status.
-
Collecting data
-
Displaying data
There can be multiple instances of Integrated Diagnostics running, which
collect data for a different replication process. The Diagnostic Collector will write
the collected diagnostic data in JSON format to the
{deployment}/var/lib/reports
directory. The filename is in the
format: {ProcessName}.diagnostics.YYYY-MM-DD-HH-MM-SSZ
A maximum of 10 collection files for each process will exist. When a new collection starts, if there are already 10 files, the least current collection will be purged. Collected diagnostics can be accessed while the collector is running as it saves the data after every sample is collected. You can download the existing reports at any time.
The Diagnostic Collector will only allow one instance to be running for a specific Extract or Replicat at a time. The Administrative Service retrieves the collected diagnostic and performance data and provide listings of previously collected data.
How to Configure Integrated Diagnostics
-
From the Administration Service, left-navigation pane, select the Extract or Replicat process for which you want to configure performance monitoring with Integrated Diagnostics. The process menu expands in the left pane.
-
Click the Integrated Diagnostics option to open the configuration page for Integrated Diagnostics.
-
Click the plus (+) sign to open the Create Integrated Diagnostics Collection dialog box.
-
Specify the following values in this dialog box:
-
Collection Frequency (in seconds): Sets the frequency time interval for collecting performance statistics for the selected Extract/Replicat process. Minimum value is 5 seconds.
-
Collection Duration (in minutes): Set the duration for which the data is collected.
For example, if you collect data in a 10 second interval for 5 minutes (300 seconds), the collection contains 30 measurement points.
-
-
Click Submit to create a query. This query will run for the specified Collection Duration. Notice that the query shows in Running state till the collection duration is over. The status changes to Completed after the data collection duration is over, as shown in the following image.
-
Click the Details icon to view the detailed results of the query. A dialog box is displayed that shows the monitoring data for the selected process.
As shown in the image, the information is depicted using graphical charts and a table as well. You can download this table as a .csv file also.
The information displayed includes measuring(Display Properties). idle time, top-wait event, CPU usage, System I/O file sequential read values for Extract, Replicat processes. These values are measured for Lag, messages Captured/sent, LogMiner reader, LogMiner Preparer, LogMiner Merger, LogMiner Builder, and the Capture or Replicat process.
-
Use the Download icon, to download the results of the query.