Outage management and System Status page on the portal

[valkyrichek]

I'm implementing Outage management for a customer with a quite advanced process in this area, and I'm a bit puzzled about how to implement it in the best way using the existing functionality of the Services -> Service Offerings and the portal System Status page. Would be glad if you can share any experience or insights on this topic. 

Current design suggestion:

According to the CSDM 2.0 and the configuration of the newly released "Service Owner Workspace", the outages should be logged against service offerings, rather then Services, to enable the calculation of service availability, comparison to commitments etc. However, on the end-user side the "System status" page is configured to show only the outages linked to services (cmdb_ci_service). Luckily, this can be changed in the widget options. We can tell the service status widgets to use the table service_offering instead of cmdb_ci_service. And because Service Offering is a child table of Service, everything seems to be working correctly. And we might even automatically subscribe the users to outage notifications, based on their subscription to different service offerings. It seems that this approach is very logical and corresponds to the customer's structure of business services.

Problem:

However there is one trick that is puzzling me: usually the same outage affects several service commitments. One option to represent this fact is to add all the affected service commitments into the list of Affected CIs of the Outage. This is fine for the availability calculations, but it doesn't work for the System Status portal. All the widgets are only checking the CI that is directly assigned to the outage, so the other affected service commitments will not be visible, and the customers will not be informed. We can, of course, clone and change the widgets, and also change the notifications mechanism, but this is a significant customization, and we will not benefit from the further upgrades of this functionality. The other option would be to create as many outages as there are affected service commitments, but this is either a huge overhead to manage them all, or we need to implement a custom synchronisation mechanism with a kind of "parent-child" relationships between outages. And besides, multiple outage records may pollute the System Status page... 

Has anyone encountered such requirements? And what do you think would be the best approach?

Many thanks in advance for your opinions.

[Sameep Kodia]

@Valentina Lebedeva  - Your explanation and problem is spot on and this is a missed feature. Please log a CASE or PRB for this if you have an access to it. But you are right currently, the System Status Portal does not look at the Affected CI's

--

Sameep

P.S. Mark the answer helpful if it helped

[johnathanr]

This is a huge pet peeve of mine. The outage and other functionality only looks at the primary record, and not the related items.

[Sameep Kodia]

Yes, I'm really hoping, you all can log an idea, case or prb for this, so it get's prioritized.

[Chris D]

Good write-up. I've noticed the same flaw myself in trying to figure out how we can best implement the OOTB System Status page + outages.

Looks like somebody posted an idea to improve this: https://community.servicenow.com/community?id=view_idea&sysparm_idea_id=489a9df3db14541013b5fb243996...

No idea if we'll go this route, but my initial thought to resolve this issue was of business rules to create individual outage records for each affected CI on the respective INC/CHG/RLS record. Those aren't quite my areas of expertise (plus I have minimal experience with business rules), so maybe the affected CI list on INCs, etc. isn't OOTB, but if an INC has multiple affected CIs, can't you have a business rule to create a distinct outage for each CI? Then when the INC is resolved, a business rule could update all those outages with an end time?

This would only work if you were creating outages from those other applications though... not sure what you would if you were just manually creating outages...