Application details - Serverless App Engine - Alibaba Cloud Documentation Center

The application details page provides an application overview, topology, JVM monitoring, thread and connection pool monitoring, SQL call analysis, exception analysis, and information about upstream and downstream applications.

Access the feature

On the SAE Applications page, select the destination region and namespace from the top navigation bar, and then click the Application ID to open the application details page.
In the navigation pane on the left, choose Application Monitoring > Application Details and view the application details.

Overview

On the Overview tab, you can view the following information.

djAnhMwfuM

Request Count: Shows the time-series curve of the application's request count over a specified period.
Response Time: This area shows a time-series curve of the application's response time over a specified period.
Slow Call Count: This area displays a time series curve of the application's slow calls over a specified period.
HTTP Status Code: This area displays the time-series curve of HTTP status codes for the application over a specified period.

Optional: In each area, you can perform the following operations:

Hover over the chart to view statistics.
You can select a time range on the chart to perform the following operations.
- Click Enter Selected Time Period to view monitoring information for the selected period.
- Click View Logs For Selected Time Period to view the corresponding logs. You must first enable application log collection.
Click the icon to view statistics for the metric over a specific period or compare statistics for the same period on different dates.
Click the icon to view the API details for the metric.
Click the icon to zoom in.

Topology view

On the Topology tab, the topology graph visualizes the call relationships between the application's upstream and downstream components. This helps you quickly identify application bottlenecks.

CSenOqiJg8

JVM monitoring

View JVM monitoring metrics

The JVM Monitoring tab displays time-series curves for Instantaneous GC Count, Instantaneous GC Time, Heap Memory Details, Metaspace Details, Non-Heap Memory, Direct Buffer, and JVM Thread Count.

Note

For garbage collection (GC) types other than ZGC and Shenandoah, the time consumed refers to the Stop-the-World (STW) duration, during which all Java application threads are paused. For ZGC and Shenandoah, Pauses correspond to the STW duration, and Cycles represent the total time for one GC.

In the upper-right corner of the Instantaneous GC Count and Instantaneous GC Time areas, click the Instantaneous Value and Accumulated Value buttons to switch between the time series curves.
Click a metric name, such as FullGC Count, on any monitoring panel to show or hide its curve on the chart.
Note
Each chart must have at least one visible metric. This means you cannot hide a metric if it is the only one displayed on the chart.
Click the icon on any monitoring panel to view statistics for the metric over a specific period or compare statistics for the same period on different dates.
Click the icon on any monitoring panel to zoom in.
Click the icon in the upper-right corner of Heap Memory Details/min, Metaspace Details/min, Non-Heap Memory/min, Direct Buffer/min, or JVM Threads/min to view the API details for the corresponding monitoring metric.

d754b20c961b43a17974c3f7a0fe47e2

Function introduction

The JVM monitoring feature monitors the following metrics:

Metric	Category
Instantaneous and accumulated GC details	Number of full GCs Number of young generation GCs Time consumed for full GCs Time consumed for young generation GCs
Heap memory details	Total heap memory Amount of heap memory in the old generation in bytes Amount of heap memory in the young generation (survivor space) in bytes Amount of heap memory in the young generation (eden space) in bytes Amount of committed heap memory in bytes
Metaspace	Size of metaspace in bytes
Non-heap memory	Amount of committed non-heap memory in bytes Initial amount of non-heap memory in bytes Maximum amount of non-heap memory in bytes
Direct buffer	Total size of DirectBuffer in bytes Used size of DirectBuffer in bytes
JVM thread count	Total number of threads Number of deadlocked threads Number of new threads Number of blocked threads Number of runnable threads Number of terminated threads Number of threads in the timed waiting state Number of waiting threads

References

The data collected by ARMS Application Monitoring comes from Java Management Extensions (JMX). The memory regions included in non-heap memory are fewer than the actual non-heap memory regions in a Java process. Therefore, the sum of heap and non-heap memory in monitoring may differ from the resident set size (RES) shown by the top command. For more information, see JVM monitoring memory details.

Thread pool and connection pool monitoring

On the Thread Pool Monitoring and Connection Pool Monitoring tabs, you can view metrics for the application's thread and connection pools, including Core Thread Count, Maximum Thread Count, Active Thread Count, Current Thread Count, Historical Maximum Thread Count, Completed Task Count, Scheduled Task Count, and Queue Size.

For information about the frameworks and metrics supported for thread pool and connection pool monitoring, see Thread pool and connection pool monitoring.
For information about specific agent versions, see Agent management.

Note

To upgrade the agent version, contact the technical staff in the DingTalk group (ID: 32874633) for assistance.

SQL call analysis

Note

The default threshold for a slow SQL query is 500 ms. To adjust this threshold, see Threshold settings.

The SQL Call Statistics area displays a time series curve of the application's SQL calls over a specified period.

Optional: On the SQL Call Statistics tab, you can perform the following operations:

Hover over the chart to view statistics.
Click the icon to view statistics for the metric over a specific period or compare statistics for the same period on different dates.
Click the icon to view the API details for the metric.
Click the icon to zoom in.

Exception analysis

What is an exception?

In the ARMS exception analysis feature, an exception corresponds to a Java `Exception`. Because of the `try-catch` mechanism, an exception in an API call may not be perceived by the requester. Multiple `try-catch` blocks can cause multiple exceptions for a single API call. If an uncaught exception affects the return result of an API call, it constitutes an error.

The Exception Statistics area displays a stacked column chart and a list of application exceptions during a specified period.

Optional: In the Exception Statistics area, you can perform the following operations:

Hover over the chart to view statistics.
You can select a time range on the chart to view statistics for that period.
Click the icon to view statistics for the metric over a specific period or compare statistics for the same period on different dates.
Click the icon to view the API details for the metric.
Click the icon to zoom in.

Upstream and downstream applications

Note

Upstream application: An application that calls the current application. It actively sends requests to the current application to trigger business logic.
Downstream application: An application that is called by the current application. It is a service that the current application depends on to execute its business logic.

On the Upstream Applications and Downstream Applications tabs, you can view the Response Time, Request Count, and Error Count.

Error analysis and trace query

The SAE console does not provide direct access to error analysis or trace queries, but you can click the Error Analysis and Trace Query tabs to view them in the ARMS console.

Related documents: