At present we have global/default alarms set for CPU Utilization, Memory Utilization and FilesystemCapacity.
I want to find out how I can set individual alarms for an individual monitored host.
For example, currently we get a warning alarm if CPU Utilization goes above 80% and/or if Memory Utilization goes above 80% and/or if a FileSystem Capacity goes above 70% capacity. We have some monitored hosts which these global alarms are not satisfactory for. For example, I may have host1 which I would like the CPU alarm set to 95% because it always has high utilization so 80% is too low. I may have host2 which has a C: drive which is usually between 70%-80% capacity so I would like the Filesystem alarm for it set to 85%. And I may have host3 which consistently runs at 80%-90% CPU utilization so I would like the CPU alarm set to 95%. As well as setting different alarm rule conditions on an individual monitored host basis we also have the need for a group of servers to have the same alarm rule conditions.
For example, I may have host10, host11, host12, host13, host14 and host15 which all are webservers and usually perform around the same as each other. In this case Id like if possible to see if I need to setup individual alarm rule conditions for each monitored host as I was asking about in the previous paragraph or is it possible to assign an alarm threshold for these group of 5 monitored hosts.
In both cases above, individual monitored host alarms or group monitored host alarms, I also need to find out how I can have the alarms send an email to a specific list of people. For example, I know how to set a rule such as Capacity to send to a different email address list than say the rule for CPU Utilization. But how do I set a different email address list for a specific monitored host?
For example all Capacity Rule alarms currently go to me, so how can I set exceptions so that if the Capacity Rule alarm being triggered is for hos
The best way to accomplish what you wish is through the use of the registry variables the CPU Utilization, Memory Shortage and Capacity rules use in their evaluations.
For CPU Utilization, the registry variables that will need to be modified are:
CPUCritical, CPUFatal, and CPUWarning
For Memory Shortage, the registry variables that will need to be modified are:
MemoryCritical, MemoryFatal, and MemoryWarning
For Capacity, the registry variables that will need to be modified are:
CapacityCritical, CapacityFatal, and CapacityWarning
Unfortunately, there is no way to group hosts, so you will have to do this manually.
To get started, well use the CPUCritical registry variable as an example. The same ideas apply to all the registry variables you wish to modify.
1) Go to Administration | Rules & Notifications | Manage Registry Variables and find CPUCritical and click on it to modify.
2) You should see a section called Registry Values at the bottom of this page. Click on Add Registry Value.
3) This will take you to a dropdown page. Under Topology Type, selected Host
4) This should refresh the Topology Object dropdown so that all the hosts in your environment are now in this dropdown list. Select a host that you wish to have a different value of CPUCritical (so instead of 80%, a host that only fires at a higher number like host1)
5) Enter a value under Default Value. For example, if you wish host1 to fire only on 95% then enter 95 and click Add.
You will now do that for all the hosts you wish to have different values for and for each registry variable you wish to modify.
To send out emails to different hosts, you would do something very similar to the above 5 steps. This time, you would modify the SYSADMIN registry variable (assuming that your email actions for CPU Utilization, Memory Shortage and Capacity rules use SYSADMIN. If they dont, and are using a different registry variable then use that variable instead).
1) Go to Administration | Rules & Notifications | Manage Registry Variables and find SYSADMIN and click on it to modify.
2) You should see a section called Registry Values at the bottom of this page. Click on Add Registry Value.
3) This will take you to a dropdown page. Under Topology Type, selected Host
4) This should refresh the Topology Object dropdown so that all the hosts in your environment are now in this dropdown list. Select a host that you wish to send emails for.
5) Enter a comma separated list of emails you wish to receive emails to when this host fires an alarm.
It is possible to have separate email addresses go to different alarms. This can be done by creating new registry variables based off the SYSADMIN variable, and modifying the email addresses for those hosts you wish to receive alerts for and associating those new variables to their rules.
For example, you would create a copy of SYSADMIN variable and name it something appropriate, such as CAPACITYADMIN. In it, you would put host1 as the topology object and put the email addresses of those users. Then in your rules for CPU, you would add the email action to use CAPACITYADMIN instead of using SYSADMIN. You would repeat this for as many different email addresses and alerts as you wish
