In the new NetEye version 3.5 that will be shortly released, it has been implemented the Shut Down Management module that allows to configure automatic shutdown procedures in a data center.
I’ll try now to provide a simple example to let you understand the potentiality and the necessity of this feature. For example if there are problems with the power supply in the data center, the UPS usually will start with an half an hour autonomy. Therefore, it is necessary to shut down all the servers before the power will be definitely interrupted. In this case a “Business process” can be configured in NetEye to execute the desired logic for the checks (i.e. it will check that the UPS is started). In the Shut Down Management module the user can configure that in case the check fails, and a determined tolerance time (30 minutes ) has been elapsed, the automatic shutdown procedure will be started to stop on time all the servers with a certain order and logic. The tolerance time has to be calculated in terms of total UPS battery autonomy time minus the total time required to complete the shutdown procedure of all hosts.
Let’s see how to set up a shutdown management procedure.
When configuring a new Shutdown Procedure definition it is possible to choose the following principal settings:Require Explicit User Confirm:
Finally a shutdown definition consists of a list of hosts to be shut down and a clear shutdown sequence. To helps to ensure to shut down the hosts in the right sequence in order to avoid data corruption and system inconsistencies. Already build-in shutdown Commands are available and can be attributed to the hosts. These commands can interact directly with the system or invoke the shutdown process via remote NetEye agent.
Once the Nagios monitoring status becomes a Critical with a HARD status ( confirmed critical ) the shutdown management checks the status and makes sure: