opManager: Alert Notification and Escalation
I have found the following document recommending some steps towards 'Fail-safe alerting' in opManager:
http://manageengine.adventnet.com/products/opmanager/fail-proof-alerting.html
This document mentions alerting in different ways depending on the clock. But I am confused. From what I can tell there is no method to schedule alerts and escalation against the clock, or to a roster. There is no method to reliably escalate alerts at all in fact.
Is there a method, or plugin that enables sensible alerting and escalation? For example:
A server dies during business hours, an alert is generated, an email sent to 'support@blah' and an SMS paged out to the rostered individual for that day. Should the alert not be acknowledged within 15min this happens again and maybe a manager get paged.
But, should a server fail outside of business hours then an email is sent and the afterhours support staff paged. Again, if the alert is not acknowledged then alternate escalations occur.
I am not so concerned about querying a roster each time an alert is generated, but more so that should an alert be missed, it get's escalated within a reasonable timeframe. And finally that different notification destinations/methods be used depending on the time of day.
I am assuming I am missing something obvious, as monitoring anything would surely go hand-in-hand with reliable notification policies?
http://manageengine.adventnet.com/products/opmanager/fail-proof-alerting.html
This document mentions alerting in different ways depending on the clock. But I am confused. From what I can tell there is no method to schedule alerts and escalation against the clock, or to a roster. There is no method to reliably escalate alerts at all in fact.
Is there a method, or plugin that enables sensible alerting and escalation? For example:
A server dies during business hours, an alert is generated, an email sent to 'support@blah' and an SMS paged out to the rostered individual for that day. Should the alert not be acknowledged within 15min this happens again and maybe a manager get paged.
But, should a server fail outside of business hours then an email is sent and the afterhours support staff paged. Again, if the alert is not acknowledged then alternate escalations occur.
I am not so concerned about querying a roster each time an alert is generated, but more so that should an alert be missed, it get's escalated within a reasonable timeframe. And finally that different notification destinations/methods be used depending on the time of day.
I am assuming I am missing something obvious, as monitoring anything would surely go hand-in-hand with reliable notification policies?