In many other indicators evaluating website usability it is certainly the most important. A site must first be used to evaluate other aspects. Web site is not available or can not find one of us rely on customer complaints, etc., need to be equipped with active means of monitoring. So today we want to talk and everyone is Azure Monitor Application Insights one of the following two services. Application Insights we can use in multiple areas, all the time to visit the site launch, when the poor availability immediately alarm to alert people that investigation. With more than a boss keen alert, since you can sleep at ease at night.
Azure Monitor Application Insights is one of the two services, he can run near real-time in the private cloud, hybrid cloud and any public cloud-hosted applications for various platforms (.Net / Nodejs / Java / Python, etc.) monitor. It will automatically detect anomalies applications and to help diagnose the problem by its own powerful analytical tools, so as to enhance the performance and availability of applications.
Speaking of which may have friends will ask: Application Insights can provide us with those types of monitoring it? In fact, the type of monitoring Application Insights provides as follows:
- Request, response time, and failure: Learn most popular pages, time and location of the user. See which pages best. When there are more requests, if the response time is long and the failure rate is high, there may be a resource problem.
- Dependency rate, response time and failure rate: find out if the external service is slow down the speed.
- Abnormal: Statistical analysis of the aggregated information, or select a specific example of the stack and drill track and related requests. Reporting server and browser abnormal.
- Page views and load performance: report by the user's browser.
- AJAX call (from the website): speed, response time and failure rate.
- And a user session count.
- Windows or Linux computer server performance counters such as CPU, memory and network usage.
- Host diagnosis of Docker or Azure.
- Application of diagnostic trace log: You can track events associated with the request.
- The client or server code to write your own custom events and indicators used to track business events.
Learn the monitor type in the future, we look at how Aplication Insight to monitor the application:
To use the Application Insights monitoring application, we need to now create Application Insights resources Azure, and then install a small package of detection in the application. Testing kits can detect not only Web services applications, you can also detect all background components and the performance of the host Web pages and running JavaScript applications. And sends the data to the application insights and resources associated with it inherited azure monitor in
addition to Application Insights can also be set to periodically send a request to the Web service integrated Web test.
Having said that, I believe you for Application Insights also have a certain understanding, here we look at how you can use the site to monitor the availability of Application Insights.
Creating Application Insights resources
click on the "New Resource" - "IT & Management Tools" - "Application Insights": 
Enter the "Resource Group", "name" and "area" after clicking "review and create": 
after confirmation correct and click "Create ": 
is created, as shown below:
Set ULR monitoring (site availability monitoring)
In the Application Insights we just created, find the "Availability" tab, then click the "Add Test": 
Next we need the following settings:
- Test Name - the name of this test we will use it when we experience a subsequent query data.
- Test Type - divided into the URL test and multi-step test. The URL should be noted that the test is completely free. Multi-step test would incur costs.
- We need to monitor URL- URL test site
- Test frequency - Here I choose five minutes
- Test sites - because my site deployment in Southeast Asia, but users spread across multiple regions, so I chose the five test position. Of course, we can be configured according to the actual situation in a real production environment.
- Testing Success Criteria - we can set the timeout for state testing and code information
- Alert - In this setting we can generate alarm information upon detection of the site is unavailable
Once created click the "context menu" - "Edit Alert":
Set the alarm for the rule in the rule management page, and click Save:
Wait a few minutes, we can see the corresponding data in the "Availability" widget "Overview" page:
Click the widget to open a more detailed monitoring dashboard. Choose Details can provide us with additional information, such as test duration, and the percentage of the number of available tests performed. We can also test the bottom right edit or pause.
Here we can see that I was a few outliers, hover over the corresponding outliers can be seen at the duration and location of the test:
You can see the detailed monitoring information click into the future:
Journal
Because Application Insights and Log Analytics using the same query language, so we can click on the "Log" in the "Overview" page:
Because the "URL test" is stored in the "AvailabilityResults" table, so we can correspond query using the following command:
availabilityResults
| project-away performanceBucket, operation_Name We can see the name in the Name column, if we have the availability of testing multiple sites you can make appropriate screening:
Using these monitoring data, we can generate different visual icon or monitor alerts. For example - this scattergram, is divided according to the duration of the average value of the position
This is how Application Insights monitoring website availability we can use.