SLA Engine creating and closing TASK SLA before update operation completes on Incident form

[manishnagar]

Hi All,

We are trying to achieve below steps, but ending up having SLA records created and completed in 0 seconds in between of the two Task SLA Records that gets created using reset condition with help of flag on incident form.

1. Attach SLA to check update on incident within a specific interval.

2. In case SLA got Breached, Business rule on Task SLA form will update flag on Incident form.

3. Reset condition in Task Definition will check this flag and create new SLA.

4. While closure of the current Task SLA, Before business rule on 0 order will reset the flag on incident form.

5. Step 4 will ensure that new Task SLA record should not get closed by SLA Definition Reset condition.

Issue : In step 5 new Task SLA record is getting created and closed as update operation to reset flag on incident form, while closure of old Task SLA record is some time taking too much time to update the flag on incident form. This behaviour is observed intermittently and not on consistent basis.

Looking forward to your support in resolving this issue.

Thank you!!

[williamsun]

Could you share with us a bit more behind the purpose of these two SLA's?

In ServiceNow there are usually several ways to do things, and maybe it can give a little help in understanding your plight.

[manishnagar]

Hi William,

We are trying to make the assignee to make update on incident at every 15 seconds.

In case, they fail we want the current Tasks SLA to be breached and new Task SLA should get created.

Thank you for your response !!

Regards,

Manish Nagar

[manishnagar]

Hi William,

Sorry I meant to say Every 30 minutes instead of 15 seconds.

Idea is for Priority 1, we want assignee to update the additional comments or work notes within 30 minutes and once SLA has been met a new SLA should start.

What we are observing is that when we use reset condition to start a new SLA, the new SLA is opened and immediately completed. Then a new SLA starts again and works correctly.

This doesn't happen every time but enough times to skew the reports. The SLA needs to be met for a percentage amount overall so having these 'fake' completed SLAs are not good.

It seems to be a timing issue where the time the SLA resets is before the task itself can update to no longer meet the completed condition.

Is there another way to create an 'update' SLA where the assignee needs to provide regular updates to the task?

[williamsun]

OK, I understand.

I would consider a different approach, as, for me, SLA is better used as a binary success, it was either breached or achieved, regardless of how many times it might have breached, in your scenario.

I have a similar set up, where I need to update my P1s every 30 minutes, and my P2s every hour, but I have set up so that when a specific template is used to send an email the "Follow up" field is reset with the current time plus 30 minutes, then I have a dashboard where P1 and P2 incidents are sorted by "follow up" time, so they can better monitor those INCs that require a notification.   Trying to concentrate on avoiding the failure.

Unfortunately I have not tried the details of cascading SLAs where one triggers another, but you need to pay attention at what is triggering each one, and then make sure that something is not overwriting the trigger condition before it is noticed, so maybe take a look at the "order" in which rules fire, and also take a look at maybe adding a delay somewhere to test if this is what is happening.

Hope this helps a bit.