Using our new Intelligent Monitoring and Alerts to create custom Collector Down Alerts


Intelligent monitoring can be used to create an alert to monitor the status of your Collector and thus send you email alerts when the Vyopta Data Collector is Down.

If you have an on-premise data collector to monitor you on-premise infrastructure  (Cisco, Polycom, Pexip etc),  you will need to monitor the collector,  to make sure the collector stays online so we are capturing all your valuable meeting and call data.  One of the most common problems customers run into,  is when the data collector is taken down,  network connectivity causes connectivity issues to our cloud, customer deployed environment change,  which all prevent us from collecting your systems data.

With our new release of our Intelligent Monitoring and Alerts,  we have created the ability to monitor the state of your on-premise collector.  What we do is we have the ability to track the amount of realtime calls (streams) that are currently being collected from your on premise collector,  and if we see a drop of your active calls,  we can send you an alert to notify you when this happens.   Not only will we send a an alert,  we can also show you history of the streams to show when potential collector problems started.

Finally you can direct the alert to go to your team (or members of your team),  with instructions on how to take action  (which you can change anytime)   Gone is the days when an alert comes in about a collector issue,  and nothing happens since the recipient is on vacation.

Here is how you can set up your Critical Alert for your environment today.

If you are using Vyopta to monitor only UCaaS systems like Zoom,  Webex, and Microsoft Teams only,  you wont need to set up a collector alert,  since all UCaaS systems are collected from the Vyopta Cloud.

The Intelligent monitoring rules can be found by navigating to CPM Monitoring -> Monitors -> Rules Tab to see all the Rules created for Intelligent monitoring. You can see the list of all monitors at the organisation including the current state of each monitor


Current State: The live reading of the current state 

Enabled: The state of the monitor, whether or not it’s on (options include “Enabled” or “Disabled). Clicking the toggle in the table will change its “Enabled” value. 

Name: The user-inputted name for the monitor (note, this is not necessarily the same as the email subject!). It is hyperlinked to the read-only view.

Category: The user-selected metric category of the monitor (options include Bridges, Call Controls, Calls, Endpoints, Meetings, Trunks).

Metric: The user-selected metric of the monitor.

Filtered by: The list of the filters applied to the metric defined in the monitor.

Created by: The Vyopta user who created the monitor.

Created on: The date which the Vyopta user who created the monitor hit “save” first. 




By clicking on the ‘Name’ you can edit the parameters of an existing Intelligent monitoring.


Creating a monitoring alert consists of the following steps:

  1. Adding the monitor
  2. Defining the metrics
  3. Setting the alert conditions and
  4. Notification details.

In order to create an Intelligent monitoring alert, click on ‘Add monitor’ and provide a name for the alert that best describes your Alert. In this example we will be creating an alert that will help you monitor the status of the Vyopta Data Collector by checking whether Vyopta is receiving any call data – the absence of all call data would indicate a possible problem with your collector if this takes place during expected working hours.


Name: “Critical - Collector Down Alert”



The first Section: Define the Metric

  • Category: Select the Category for which you want to create the monitoring. In this scenario select ‘Calls’ ( see note at the end of this document )


  • Metric: Based on the Category selected, the values in the Metric dropdown field will be populated with the relevant options.  In this scenario,  select 'Call Streams' which will supply a count of real-time calls being captured by your on-premise collector.


Once the Category and metric has been selected, you can add any additional filter as needed.

To create a Collector down alert, we will be adding the following filter:

“Data Source does not contain WEBEX and Data Source does not contain GMEET  and Data Source does not contain TEAMS and Data Source does not contain BJN and Data Source does not contain ZOOM"


The above UCaaS data sources are excluded since this data is not gathered by your on-premise  data collector, instead being collected on your behalf by collectors in the Vyopta Cloud.


Once the metrics have been defined, Set the alert conditions for which you want the alert to be triggered.

For Creating a Collector down alert, Select the following values.

This alert conditions means – When the Call streams is equal to 0 for consecutive 30 minutes, the Vyopta Data collector is down.



Operator: The metric is compared to the thresholds with the user-inputted operator as the comparison. There are 6 options - “Greater than or Equals”, “Greater than”, “Less than”, “Less than or Equals”, “Equals”, “Not Equals”. This is a mandatory field. 

Thresholds: User inputted field(s) the metric is compared to with the operator. Critical is meant to signify the most severe, then warn, then info as least severe. All thresholds are optional individually, but each monitor must have at least one threshold filled out. The Notify section is only enabled for each threshold level once a value is inputted. 

Duration: The user inputted value (in minutes) that the metric must satisfy the comparison of the operator vs threshold in order to count trigger a change of monitor state. This is a mandatory field. 


Once the Alert conditions are set, Select the details that need to be sent out to your team in the notification. You can create different notification to be sent for ‘Critical, Warn and Info’ status.


Enable Email Per Threshold: Checking this box signifies an email will send if that threshold’s state is reached. Unchecking the box will not send an email. Note, unchecking does not invalidate that state from the “current state” live monitoring - you’ll still see it, just no email.

Subject: This is the subject of the email being sent. Note that it can be the same or different from the monitor name. We will automatically append the threshold state before the subject (i.e. “[Critical] This is the email subject” in the above example).

To (Recipient) :  This is going to contain the emails that the alert would be sent to.  When selecting the recipient,  you will want someone on the list that will be able to take action on the alert once they receive it

Send As: Signifies the type of email being delivered. “HTML”, “HTML - Custom”, and “Plain Text” are the three options. “HTML - Custom” allows for an additional input to customize the text included in the body of the email. 


An example of the email that gets sent out is shown below:

We haven't received any active calls with the Vyopta Data Collector located on the server "ABC" over the last 30 minutes. If you have received this alert, please check the VyoptaCPMDataCollector service is up and running.

For a more detailed instructions, please follow the knowledge article supplied by Vyopta -

Note: This knowledge base article will guide you through the steps that is needed to restart the collector.

Frequency: Currently this is under evaluation to include/remove/update given it is illogical in the “email send on change of state” guidance we are operating under.

Days & Active Schedule: These are the user inputted values for when emails should fire. In this example the alerts would only send an email Monday through Friday 12am to 12 pm PST. All state changes outside of that schedule would still be tracked in the UI, but no email would send. 

You should be aware, however, that when an alert becomes 'active' by a schedule, the criteria are applied immediately:- therefore, should an alert be scheduled to become active at 08:00, and examines a 30-minute window (duration) data will be used for the preceding 30 minutes - thus if the alert would have been triggered with the data from 07:30 - 08:00 then an alert will be sent. 

To avoid this, set the scheduled activation time to be one duration interval after the time that measurement should start - in this case this would be 08:00 plus 30 minutes, or 08:30


Note: In this example we have used a calls metric to alert us when the Vyopta Data Collector may be down. Depending upon your environment and expected working hours it may be appropriate to use a different metric such as monitoring Call controls or other infrastructure status.


Was this article helpful?
0 out of 0 found this helpful



Please sign in to leave a comment.