Dynamic monitoring - Alarm messages seems to have some form of corruption at the end of the alarm message in IM alarm subconsole.
OC shows the same strange characters
That same alarm in Netcool ticketing application shows '??' instead of the 'l with the umlaut' and the italic 'f'
XXX-Unix Servers:P2:Average total CPU is 47.04% from source <xxxxx.xxxx.xxxxxxx.com> targeting xxxxxxxxx has crossed the level 5 threshold of +3.0 σ
Using ${unit} instead of ${threshold_symbol} worked to display the percent symbol properly and this avoids the custom alarm message corruption.
Here is the out of the box example provided when you click on the ? icon when using the Admin Console, and configuring ‘custom’ alarm messages:
${qos_name} = ${value}${unit} from source ${source} targeting ${target} has crossed the level ${level} threshold of ${threshold_sign}${threshold}${threshold_symbol}
Here is the alarm defined by the customer:
XXX-Unix Servers:P2:Average total CPU usage is ${value}${unit} from source ${source} targeting ${target} has crossed the level ${level} threshold of ${threshold_sign}${threshold}${threshold_symbol}
But, if you place the cursor inside the window you will see the valid variables that can be used when you use the cursor and position it in front of the variable string.
There are several variables but there is no ‘threshold_sign’, ‘threshold_symbol’, nor ‘unit’ in the drop-down window.
But $unit works to display the threshold symbol as it is just percent.