4 Replies Latest reply on Jan 23, 2018 7:23 AM by anitazuri

    Alerts and too many Emails

    anitazuri

      Hi all,

       

      I have created an alert to trigger when specific services (33 of them) go down and come back up, it works great however my manager is not happy that we receive an email for each service.

       

      Is there a way of having one email that reports all the services that have gone down.

       

      I have created a template in SAM for all of the services that need to be monitored then created an alert that goes like this:-

       

      When the Node is equal to 'Shared Services Server'

      and

      the Node is Not equal to Down

      and

      the Component is equal to Down

      send an email

       

      You can imagine how many emails we get if something goes wrong!

       

      Surely there's got to be a better way

       

      Thanks

      Anita

        • Re: Alerts and too many Emails
          yaquaholic

          Hi Anita,

           

          It will be down to the alert itself, I am guessing that the alert is Component based and not Application?

          Try creating a new alert, but this time based of [the] Application and use the overall application status (warning, critical, etc) as the trigger.

          There are some nice variables that you can insert into your emails, that will then detail the components at fault, within your application template. 

           

          Which will take your one email for each component to one email for them all.

          1 of 1 people found this helpful
          • Re: Alerts and too many Emails
            Geoff Smith

            You could change it from a Component alert (when does the service go down) to an Application alert (when does the Application go down) and then use one of the "Components with status..." items in the E-Mail to list each component and whether its currently up or down or whatever.

             

            If they're all running on multiple servers, throw all the components into a group and use a group alert and use the "Group Status Root Cause" item in the alert.

            1 of 1 people found this helpful