Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

On this page the bottom and most important half of the Ymonitor application is described. Here you can find your data displayed in different ways. To select what you want to see, there are several buttons at the top of the website. These buttons show the different views that are possible to select. Each shows different data or the same data in a different way. The currently selected view will have a black background.

Table of Contents
minLevel1
maxLevel7

General information

Exporting data

...

If you click this icon you get a menu with several with several export options. This could be an export to the clipboard, a png file , the clipboard or a CSV document in case of table, and an export to a png file in case of graph.

Working with graphs

At any place where there is a graph and not a table it is possible to filter what the graph shows. If you click on an option in the legend below the graph in the legend on an option , it will remove that option from the graph. This could be useful to rescale the graph for more details.
This goes for any graph so not only the line graphs but also bar or pie charts. It doesn’t doesn't work in the several tables that are shown.

Business hours

...

This view will be the one you see when you log into Ymonitor. In the view are two tables. The top one shows information from all sentinels combined and the bottom table shows data split up by the different sentinels.
It is important to note here that the data shown is only from the defined business hours. The current set business hours can be seen in the text above the table. If this needs to be changed please contact your Ketenbewaking team.

In the table you can see several fields all the fields are about the currently selected time period within the business hours:

...

Name of the field

...

What does it show?

...

Perf.

...

The average performance (in seconds)

...

Ava.

...

The percentage of transactions that have successfully been executed.

...

No.

...

The amount of transaction executed.

...

Min

...

The lowest measured responsetime.

...

Max

...

The highest measured responsetime

...

Failed

...

The amount of transactions that have been unavailable.

...

Timeout

...

The amount of transactions that reached their timeout value. (see thresholds)

Availability

...

Availability

...

On this page you can see the availability of your application. First you will see a graph with the transactions. Then , then a table with the transactions. Under that is the same but showing sentinels instead of transactions.
In the case you see a grey block here it means that there has been no measurement on at that time. This could have several reasons but one of them is that the script came across an error. In most cases Ymor Sentia will build a script so that it will stop after failing one transaction. The reason for this is that it is impossible for the script to know if the other transactions can be performed. For instance, when the login transaction fails a lot of the other transactions after it won’t won't be possible to do (since the user is not logged in). So in most cases you will see grey blocks after a failed transaction (red block).

Performance

...

In here you can see the performance of your application in seconds. As with the availability view, you will first see a graph and table depicting the transactions with under that the same as for the sentinels.
The  The colouring of the blocks is done using the set thresholds. To see the settings you can click on the threshold button in the top right.
In  In the case that a measurement is unavailable it will show as a grey block within this view. This is because we can’t can't tell the performance of a failed transaction. It could be quicker because the server gives an error message or it could be longer than it should be because of timeouts.
Note: There are cases where this view doesn’t doesn't actually show the performance but is measuring something else. This is the case with services that for example measure the amount of currently active Citrix sessions. If a script is measuring this, it will use the performance view as a way of portraying the number.

Percentile

...

The data shown in the graph and matrix on this view can help with filtering outliers from your information. The data that is being calculated can be summed up in the following sentence:

Info

The percentile shows you how many percent of the measurements has a certain responsetime or a faster responsetime.

If for example the 95th percentile has a value of 5 (seconds) it means that of all the measurements done 95% of the measurements have finished in 5 seconds or less. Therefore 5% of the measurements took 5.1 seconds or higher.

Network

...

In URL monitoring it is possible to see some detailed information about a certain URL. Information like how long did it take to resolve the DNS, how long was the wait time on the server, etc.
If there is URL monitoring set up by the ketenbewaking team the data will be shown here. It is possible that this hasn’t been set up.

In the case that it is configured at the top of the page you will be able to see a select box showing the different URL monitors that are attached to this service.

...

By selecting a different URL the data will change to show the information of that particular URL

At the top panel you can see how the network has been over time this is from all sentinels combined. The graph at the bottom shows it split for each sentinel or group.
In the graph there are several definitions used you can find the explanation in the following table:

...

Name of the field

...

What does it show?

...

DNS

...

How long did it take to convert the URL into an IP address.

...

Connection

...

How long did it take for the sentinel to set up a TCP connection with the server

...

Network

...

How long did it take between sending the request and receiving the response

...

Download

...

How long it took to download the response

Webtrace

...

A web trace shows the route that a request to a website has taken. In this case the route is the different network components that make up the internet.
If there is a web trace set up by the ketenbewaking team the data will be shown here. It is possible that this hasn’t been set up.

If a webtrace is set up you will be able to make a selection at the top about both the destination and which sentinel to show.

...

Unlike the URL monitoring it is important to have the Sentinel selection here. This is because with a traceroute your starting point (sentinel) is very important.

Below the selection there are three graphs.First off there is a line graph showing you how long each hop took over time. Using this its also possible to see when hops are suddenly added or removed.

The lower left shows a pie chart showing how long each hop took relative to the other hops. This data is over the whole period that is selected in Ymonitor.

The lower right is a graph that doesn’t change when you select a different sentinel. This pie chart shows how long the full traceroute takes compared to the other sentinels.

Note: Traceroute information can be very useful but it can also be misleading. For example if the traceroute will take 100ms longer to reach the server it doesn’t have to mean that your website will load 100ms slower. The most useful information you can get from traceroute is which network components are hit to make the request along with possible changes in the path the requests take.
Please contact your ketenbewaking team if you are unsure about what the data is saying.

Alerts

...

Alerts

...

In this view you can see the amount of triggered alerts. The Alerts have been set up by Ketenbewaking and if there need needs to be more alerting or less please contact them.
If you click on one of the bars you get a list of all the alerts in that timeframe along with the content of those alerts.

...

In this view you can see the amount of errors (failed transactions) there have been over the selected period.
If you click on one of the bars you will get a list of all the errors in that timeframe. It will also show wich Sentinel which sentinel generated the error along with the message of the error. On the right side, if you click on the download icon you can download screenshots showing what our tooling saw at the moment of the error. It is possible that screenshots aren’t aren't available. This is dependent on application being monitored and the tooling used to do so. 

There is another button available next to the download button which is 'compare screenshots'. More detailed information can be found on Screenshot compare page.

Apdex

...

On this page you can see the apdex value of your application. The view for the apdex value is similar to the availability and performance views. The first graph and table shows the apdex value of the transaction and the graph and table below them are same as for the sentinels. The colouring of the blocks is done using the set thresholds. To see the settings you can click on the threshold button in the top right. To have more detailed information on Apdex value, please visit Report page.

Percentile

...

The data shown in the graph and matrix on this view can help with filtering outliers from your information. The data that is being calculated can be summed up in the following sentence:

Info

The percentile shows you how many percent of the measurements has a certain response time or a faster response time.


If for example the 95th percentile has a value of 5 (seconds) it means that of all the measurements done 95% of the measurements have finished in 5 seconds or less. Therefore 5% of the measurements took 5.1 seconds or higher.

Tresholds

...

By clicking on 'Tresholds' button, the availability, performance and APDEX thresholds of the currently selected service will be shown. The threshold is the value a measurement has to reach before the colour in Ymonitor will change. Below image shows the values that are set for 3 situations.

...

If the measurement of the transaction is completed in less than 1.258 seconds, that will be shown in green in Ymonitor.

If the measurement of the transaction takes more than 1.585 seconds, that will be shown in red in Ymonitor.

If the measurement of the transaction lasts between 1.258 and1.585 seconds, that will be shown in orange in Ymonitor.