I am going to assume you are using the processes probe for your monitoring but please do let us know for sure.
The processes probe will send an individual alarm for each failed service.
This then should intern based on your configuration generate a ticket in your support help desk software for each one.
(If you have a custom solution this may not be the case)
When the service is started the alarm on the UIM side should be cleared. Was this cleared?
If the alarm on the UIM side was not cleared then there is an issue with the probe that sent the alarm as it should have done that.
If the alarm on the UIM side WAS cleared then there is a problem with the probe or process that is sending the alarms to your ticketing system.
In either case you would need to increase the loglevel of the probe that is monitoring the service and the probe going to your ticket system to loglevel 5 with a logsize to 35000 or so and recreate the problem to see what might be going wrong.
You may want to open a support case for this.