by tprzybylak
I have several servers that I allow to auto-reboot in the middle of the night for updates, etc.
Some of them take +/- 10 minutes to reboot. It is NOT always at the same time.
During that time, 24x7 throws me an alert every time for every server
Is there a way to increase the allowed amount of downtime so that if a server is down for 10 minutes its ok, but if its down for over 15 I want the alert? Is this just changing polling frequency?
Like (1)
Reply
Replies (6)
Hello,
We believe you are looking for something like setting up a poll frequency only after alerts are to be triggered.
You can create escalation alerts to specific user groups. Please go ahead and create specific user groups and map one group for being alerted and the other for escalations alone. However, please note that this will apply for all the monitors.
(Admin > Configuration Profiles > Add Notification Profile > Contact Group for Status Escalation > Select group and time interval)
You can also set up persistent notifications wherein you have an option to be alerted after three or four continuous failures which is again an alternative that you can configure and see for yourself.
(Admin > Configuration Profiles > Add Notification Profile > Persistent Notification)
Cheers!
Sushma.
Like (0)
Reply
by tprzybylak
Hi
I think what I am looking for is for the server to not alert until it is down for a longer time.
It seems like if a server is down for 2 minutes - it sends an alert.
How would I change that alert to 15 or 20 minutes?
Like (0)
Reply
Hi,
We get your requirement. We do see this as a feature request. We are yet to bring in this enhancement at our end.
Will keep you posted once we initiate working on the same.
Cheers!
Sushma.
Like (0)
Reply
Has anything happened with this feature request? We too would like to have control over when Site24x7 alerts us for different things to avoid false positives.
Like (0)
Reply
We have provision to configure Maintenance for a monitor. This can be used to suppress alerts when such maintenance activity happens.
You can use our API's to create a maintenance for like 10 mins for the server, and suppress alert for 10 mins. If the server is down even after the maintenance period you will receive down alert for the server.
API can be referred in the following link:
Let me know if you need any further assistance regarding this.
Raghavan
Like (0)
Reply
This is fair enough for simple UP/DOWN alerting but for things like CPU and memory usage we want control over when it alerts and that can't be via an API. If we supress alerts for an hour for example and in that time the CPU has been fluttering over and under the threshold but the window exoires and at that moment the CPU has peaked over the threshold, we still will get alerted without needing to be.
What we want is the ability to say, if the server is in this state (Down, CPU over x amount/Memory over Y amount for Z amount of time), then alert.
Like (0)
Reply