Topic
1 reply Latest Post - ‏2013-06-17T18:51:03Z by IsaacGraf
techFanatic
techFanatic
32 Posts
ACCEPTED ANSWER

Pinned topic Incident Count SLA issue in TBSM 6.1

‏2013-06-17T13:57:39Z |

Hi,
I am facing an issue in TBSM 6.1.When my service is bad due to an alert ,any new alert that comes on the child of that parent service causes the Incident SLA count to increase by the number of new incoming alerts irrespective of whether the incoming alert makes my parent service bad/marginal or has no effect on it or the same alert comes with a higher severity but its status being visible in the Events part of the Service Tree under Service Availablity.
Is TBSM 6.1 supposed to function this way or am I doing any mistake?Is there any solution to this problem such that the Incident count may increase only for those alerts that cause the parent to go bad!

Whats the ideal functionality of working a TBSM Service,When should the incident SLA count increase?
Three scenarios arise,
Alert1 and Alert2 both affect the parent service as bad and Alert3 causes the parent to remain good(no effect).

Case 1.
Service is bad by Alert1 and Incident SLA count =1
Alert1 is cleared and Incident SLA count =1
Service is bad by Alert2 and Incident SLA count =2
Alert2 is cleared and Incident SLA count =2

Case 2.
Service is bad by Alert1 and Incident SLA count =1
Alert1 is cleared and Incident SLA count =1
Service is bad by Alert2 and Incident SLA count =2
Alert3 comes but does not affect service.
Service remains bad because of Alert2 and Incident SLA count =3

Case 3.
Service is bad by Alert1 and Incident SLA count =1
Alert2 comes and keeps service as bad
Service is bad by Alert1 and Alert2 and Incident SLA count =2
Alert3 comes but does not affect service.
Service remains bad because of Alert1 and Alert2 and Incident SLA count =3

Case 4.
Service is bad by Alert1 and Incident SLA count =1
Alert2 comes and keeps service as bad
Service is bad by Alert1 and Alert2 and Incident SLA count =1
Alert3 comes but does not affect service.
Service remains bad because of Alert1 and Alert2 and Incident SLA count =1

I want to replicate Case4 in TBSM 6.1 ,How can I achieve this?
Thanks and regards,
techFanatic

  • IsaacGraf
    IsaacGraf
    34 Posts
    ACCEPTED ANSWER

    Re: Incident Count SLA issue in TBSM 6.1

    ‏2013-06-17T18:51:03Z  in response to techFanatic

    Hi,

    I dont think what you want to do is possible with the incident SLA feature.

    As a possible other option, you could consider configuring a duration sla aswell as the incident sla and select the radio button that says "Duration violations only". If you set the violation threshold of the duration sla to be very small (2 seconds for example for violation) then when the service turns red it will almost immediately report an incident to the incident sla. However if more events come in while the service is still red they will not get reported as incidents. It sounds like you don't want to consider more than one concurrent incident, so that configuration should solve that for you.

    A downside to this is that you will have a duration sla that will be showing violation in the service viewer while the service is red.

    Isaac