I'd like the ability to alert off of a group's status. You are able to set "Monitor count threshold to decide the monitor group's status" but you can't then alert off of this.
Our use case is that we group monitors by datacenter or zone and we'd like to create an escalated incident (via pagerduty) when the group hits a specific threshold. This allows us to page an "all hands on deck" type scenario.
Like (4)
Reply
Replies (4)
Hi Damian,
Thanks for bringing up this request.
Few questions regarding this:
- Are you using Group Status threshold feature in Monitor Groups ?
- Do you also need EMail & SMS alerts for Monitor Group health ?
- Do you also need performance threshold alerts for Monitor Groups ?
- How did you grouped your monitors ? Is it based on monitor type / functionality / datacenter ?
- Do you need Escalation configurations for Monitor Group ?
- Do you need specific mail templates for notification ?
These questions mainly concentrate to get an idea whether Monitor Group should be given all functionalities of Monitors.
Kindly provide your response on this.
Like (0)
Reply
I'm sure that other customers have use cases outside of ours that may modify the importance of each of these feature but here's my thoughts/answers.
- Are you using Group Status threshold feature in Monitor Groups ?
Yes. We are using basic HTTP checks against our customer applications. If a specific number of these are down we would consider it a more catastrophic failure that we would want to alert a broader group of people.
- Do you also need Email & SMS alerts for Monitor Group health ?
Yes and preferably the ability to alert pagerduty to a different pagerduty "service" but this is a larger product requirement. Our situation, as the product stands today, would be that we'd send a Monitor Group Down alert email notification to a user we setup with a pagerduty email address.
- Do you also need performance threshold alerts for Monitor Groups ?
Not in our use case since we would use the Group Status threshold. I think this would probably be sufficient for all customers since the individual monitors would have the performance thresholds set to them.
- How did you grouped your monitors ? Is it based on monitor type / functionality / datacenter ?
Today we are grouping by datacenter. I think this would be customer specific as they may have a group of VMs on a single host that they are monitoring an application on and if all x applications are down (in the group) then they would see it as a larger problem. Probably better examples for this that other customers may be able to comment on based on their use case.
- Do you need Escalation configurations for Monitor Group ?
We aren't currently using this feature as our escalation would be done via pagerduty.
- Do you need specific mail templates for notification ?
This would be a "nice to have."
Like (0)
Reply
We want exactly this feature as well. We have grouped our monitors for
specific areas and want to be able to get an alert, not just a color
change of the group when a threshold of X number of monitors in that
group are in a warning or down status.
Previously we have been using sitescope and they call this
functionality a composite monitor, basically a grouping of monitors
that has its own ability to alert.
Like (0)
Reply