The fine art of setting up alerts. Its important to get this right. Not enough alerting could mean, you are missing important notifications. Too many, means you will end up ignoring them. Therefore I always run a test for a couple of weeks to my own email address, just for testing purposes before I start sending alerts to distribution lists. Just a thought.
Pre-requisite: add all interfaces on a node that you would like to monitor and alert on, to Solarwinds NPM.
On your Orion box, got to Programs>Solarwinds Orion>Alerting, Reporting and Mapping>Advanced Alert manager.
This will open the pane as shown below in figure 1. It will provide you with the option to create a new alert, or edit existing ones. IN this post I will show an alert i created for ISDN outage called "E1_down" (red arrow in fig. 1).
|Fig.1 new and existing alerts|
If you decide to add a new alert, follow the TABS sequentially, as per Fig.2.
Follow the Tabs as per fig.2
|Fig. 2 - Enable alert, naming alert and polling interval|
Set up the general parameters, in the example above in fig. 2, the alert will only be checked every 5 minutes. Be smart about this,and set this up with your response times and SLA in the back of your mind. I mean no point polling something every minute, if you only want to respond after 1 hour. Also keep in mind that the more frequent you poll, the more traffic it will generate.
|Fig.3 - Setting trigger conditions|
Fig. 3 shows the trigger conditions, in other words, what conditions will need to be met for this trigger to occur. So let us say we are interested in alerts on Serial/ISDN interfaces. in Figure 3, I added the condition "field interface status is equal to value Down" . But I also added "field node Name contains voip", because I am only interested in voice gateways. Of course in order for this to work, you will need to differentiate your voice gateway nodes from all the others, by adding the word "voip" in their name. Well at least, that is the way I resolved it. Now note that in Fig. 3, I configured it so that both criteria will need to be met in order for the trigger to be invoked.
|Fig.4 Reset Condition|
OK, by "Reset" (Fig.4), Solarwinds means "recovery" of the alert. In my example; when the serial/ISDN interface is no longer down. This is why I selected the corresponding radio button in Fig.4. I will explain later what you can do when that condition occurs.
|Fig. 5 Alert Suppression|
Now in my example, we are monitoring the status of a serial/ISDN interface. I do not want these alerts to be generated when the whole node is down, for instance because of power outages. Therefore I use the alert suppression tab in fig. 5 to set that as a criterium to suppress the alert.
|Fig. 6 Trigger Action|
The action defines what happens when the alert conditions are met. I just want an email to a predefined email address, with a customized message saying, what interface on what node is down. You can also set things like SMS triggers actions, syslog, or sending webpages and many more.