Service provider - monitoring questions



  • Those of you that provide services, either in-house as IT guys or as an MSP, how much of your work originates from tickets and how much originates from your own monitoring?

    Tickets are pretty straight forward. Someone has a problem which in turn triggers your work which is a sequence of responses, trouble shooting and fixes. I assume tickets are mostly related to client devices and not servers/services?

    But on the monitoring side, I assume you monitor servers and VMs but does your monitoring also goes down to application / services level? With that I mean that not only do you monitor that a server/VM is healthy but also that, for instance, the http server actually is running and have response time that is reasonable and any database connections or other connections that the http server depends on are also actually working. Or do you depend on someone noticing the problem and writing a ticket?

    Also do you respond to monitoring triggers/tickets 24/7 or just during business hours? I mean if something goes down on Friday evening, will it stay down until Monday morning? Or is this more of an issue in enterprises and not really in the SMB space?



  • @Pete-S We do not monitor anything in detail, because clients have not chosen to pay us to setup that level of monitoring.

    That said, various tools provide constant email for things (Veeam), and if it has an error, I check it.

    Other things like that.

    I never put off things until "Business Hours"

    But then we only have a handle full of clients that we do full time IT for. It is not what we are after.



  • @Pete-S said in Service provider - monitoring questions:

    But on the monitoring side, I assume you monitor servers and VMs but does your monitoring also goes down to application / services level? With that I mean that not only do you monitor that a server/VM is healthy but also that, for instance, the http server actually is running and have response time that is reasonable and any database connections or other connections that the http server depends on are also actually working. Or do you depend on someone noticing the problem and writing a ticket?

    Services that are external facing (like a web page for example) gets monitored via uptimerobot.com.



  • @Pete-S said in Service provider - monitoring questions:

    Also do you respond to monitoring triggers/tickets 24/7 or just during business hours? I mean if something goes down on Friday evening, will it stay down until Monday morning? Or is this more of an issue in enterprises and not really in the SMB space?

    The severity of the issue will dictate whether it needs to be responded to immediately or if it can wait. We have agreements with clients as to what we can action without their prior approval/notification for those critical services so we aren't hung-up waiting for a reply.



  • @JaredBusch same here, we do basic monitoring, but monitoring is complicated.

    1. It uses resources both technical and technician. If people don't pay for it, they don't get automatic work from it. Some do, but most do not.
    2. What do you monitor? Monitoring in an enterprise with thousands of critical apps is more or less obvious. In an SMB, it can be hard. 80% of people we call due to monitoring firing off have just lost power, lost Internet, rebooted the machine themselves, turned off for the night, etc. The "uptime" of services concept isn't critical to many, and so even calling them to see if things are okay wastes their time.
    3. Monitoring end to end is hard. The server might be up. The app might be up. People might be using it. But it is "slow" or "doesn't do what is expected" and monitoring for those things often doesn't work.


  • Thanks guys for your replies!

    I had a feeling that monitoring was "underutilized" compared to what is technically possible. But as always, it's the business needs and the effort (cost) that determines the service level.



  • @Pete-S said in Service provider - monitoring questions:

    Thanks guys for your replies!

    I had a feeling that monitoring was "underutilized" compared to what is technically possible. But as always, it's the business needs and the effort (cost) that determines the service level.

    totally underutilized compared to what is possible. you are 100% spot on with that.


Log in to reply