Writing this post because I was working on customizing alerts that can be generated by Cisco's Real Time Monitoring Tool. RTMT is by far the easiest tool for for performance monitoring and alert generation, on either CUCM, CUC as well as CUPS. So Lets get cracking.
The start off point for most people, when using RTMT, is the alert central (See below)
|Fig. 1 - RTMT Alert Central|
In alert Central, CUCM (middle TAB of the ones that I encircled in Fig.1) has a number of precanned alerts, such as SDLLinkOutOfService, Route List Exhausted etc etc. Without any modification, the precanned alerts do nothing but show up, only when an administrator fires up RTMT and goes into Alert Central and checks the list. This is not a very good way of monitoring your system, in fact it is useless.
So let's add an email notification to a certain pre-canned alert. Let's have a look at a MaliciousCall Trace.
|Fig. 2 - Alert Properties|
Right click the alert and go to Properties, here you can set the level (Critical, Warning etc.) and also the alert Frequency and schedule. Frequency might be particularly important to tweak properly. This because if you set the alert to notify every time it polls, you might end up with a substantial amount of emails for the same alert, over and over, and over again. After this you will get to the email notification window (see below in Fig. 3)
|Fig.3 -Email notification configuration|
This window is where you configure the recipient of the alert, most likely a shared mailbox of some sort (in this example some dude called firstname.lastname@example.org ). These steps result in an email being sent out to email@example.com everytime someone hits the MCID softkey, or should I say, when a malicious call is being perceived.
As I mentioned these pre-canned alerts are good if you want to be notified of critical stuff like, server down, DB replications issues, critical services going down etc. ect. In fact, I think the pre-canned objects are a very good starting point of monitoring your CUC, CUCM and CUPS machines.
Personally, I dont like pre-canned and non-customisable (that is probably why I dont like Apple products either). I like being able to monitor stuff as part of problem diagnostics for instance; get an alert when a certain event is triggered so I can jump on the gateway and do troubleshooting.
I have used the term "customisable alert", so far. What it really is, is a performance object that is being monitors and when it gets to a certain value, it spits out an email notification. So let's go back to the main window of RTMT (Figure 1) and go to "Performance". Now here is the catch, and I hope someone can prove me wrong, there is no "official" Cisco documentation of the counters that are available for monitoring. This means that if you are looking for anything specific, you will need to go through all the categories.
Performance counters, yes. using RTMT you will see a view of your cluster and its IP addresses:
|Fig. 4 Performance Monitoring|
Expand the CUCM server you are interested in. This will give you an overview the available categories, in which the various counters are grouped. In Fig.5 I have expanded on the PRI Active Channels on an MGCP Gateway. The instances are the actual gateways that can be monitored, so pick one of the instances/gateways (in this example the active B channels on an E1).
which will provide a continuous graph of that particular counter (Fig. 6)
|Fig. 6 Counter Graphing|
|Fig. 7 Alert properties|
In terms of performance monitoring another good feature that RTMT offers is the performance log viewer. This isn't so much related to alerts, but good to mention anyway. It will provides a portal to access historical counters.If you go to the Performance Log Viewer (Figure 4). Anyway, I will let you play with that, as I want to publish this post
Namaste! und Viel Glueck!